2021-09-25 Akim Demaille version 3.8.2 * NEWS: Record release date. 2021-09-25 Akim Demaille tests: Apple clang issues warning on C++11 features $ cat /tmp/foo.cc using foo = int; foo f; $ clang++ -Wc++11-extensions -c /tmp/foo.cc /tmp/foo.cc:1:13: warning: alias declarations are a C++11 extension [-Wc++11-extensions] using foo = int; ^ 1 warning generated. $ clang++ --version Apple clang version 11.0.0 (clang-1100.0.33.17) Target: x86_64-apple-darwin18.7.0 Thread model: posix InstalledDir: /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/bin * tests/local.at (AT_COMPILE_CXX): Make sure -std=c++11 is passed when running glr2.cc. It may be overridden by another flag in CXXFLAGS afterwards. 2021-09-25 Akim Demaille avoid using atoi * cfg.mk: Disable sc_indent as auto indent is too invasive for now. Enable sc_prohibit_atoi_atof, except where we don't care. * src/location.c, src/muscle-tab.c: Use strtol instead of atoi. 2021-09-24 Akim Demaille news: update 2021-09-18 Akim Demaille gnulib: update To get commit 7818455627c5e54813ac89924b8b67d0bc869146 Author: Bruno Haible Date: Fri Sep 17 22:22:50 2021 +0200 threadlib: Avoid crashes in thread-related functions on Cygwin 3.2.0. Reported by Brian Inglis via Akim Demaille in . * m4/threadlib.m4 (gl_WEAK_SYMBOLS): Force a "guessing no" result on Cygwin. 2021-09-18 Akim Demaille c++: use YY_NOEXCEPT where it helps Suggested by Don Macpherson. * data/skeletons/c++.m4, data/skeletons/glr2.cc, * data/skeletons/lalr1.cc, data/skeletons/stack.hh: Use YY_NOEXCEPT where it helps constructors. 2021-09-13 Akim Demaille gnulib: update 2021-09-13 Akim Demaille glr2.cc: semantic_option: use a symbol * data/skeletons/c++.m4 (basic_symbol): Add assignment operators. * data/skeletons/glr2.cc (semantic_option): Replace yytoken, yyvalue and yylocation by yyla. 2021-09-13 Akim Demaille glr2.cc: don't publish move ctor to lalr1.cc These operators were introduced in "c++: add move assignments to the symbol type" (fdaedc780af0dd678a4f4fa3175a201a553be20a) for glr2.cc. * data/skeletons/c++.m4: Define them for glr2.cc only. 2021-09-12 Akim Demaille glr2.cc: simplify semantic_option * data/skeletons/glr2.cc (semantic_option): Simplify the rule-based constructor. This shows that it should be easy to use a symbol_kind, instead of the kind/value/location triple. 2021-09-12 Akim Demaille glr2.cc: we require C++11 * data/skeletons/glr2.cc: So no need for C++98 compatibility. 2021-09-12 Akim Demaille glr2.cc: introduse b4_glr2_cc_if * data/skeletons/bison.m4 (b4_glr2_cc_if): Intro. * data/skeletons/c++.m4: use. 2021-09-12 Akim Demaille glr2.cc: don't pass %parse-param to destroy I regret that %destructor and %printer give access to the %parse-param in lalr1.cc, since that prevents them from being implemented as a simple destructor and operator<<. Let's not repeat the mistake in glr2.cc. In addition, fixes a name conflict we have currently in tests 566 568 570 657: calc.cc:1395:85: error: declaration shadows a field of 'calc::parser' [-Werror,-Wshadow] void glr_state::destroy (char const* yymsg, calc::parser& yyparser, semantic_value *result, int *count, int *nerrs) ^ calc.hh:441:21: note: previous declaration is here semantic_value *result; ^ With this commit, the whole test suite passes for glr2.cc. * data/skeletons/glr2.cc (glr_state::destroy): Don't take the user arguments. 2021-09-12 Akim Demaille glr2.cc: kill trailing white spaces Fixes several calc tests. Tests 566 568 570 657 still fail because of a name clash when using %parse-param: calc.cc:1395:85: error: declaration shadows a field of 'calc::parser' [-Werror,-Wshadow] void glr_state::destroy (char const* yymsg, calc::parser& yyparser, semantic_value *result, int *count, int *nerrs) ^ calc.hh:441:21: note: previous declaration is here semantic_value *result; ^ * data/skeletons/glr2.cc: Fix indentation/trailing spaces. 2021-09-12 Akim Demaille glr2.cc: check linking conflicts * tests/headers.at (Several Headers): Link two glr2.cc parsers. 2021-09-12 Akim Demaille glr2.cc: move strong_index_alias into the unnamed namespace * data/skeletons/glr2.cc: here. 2021-09-12 Akim Demaille glr2.cc: use only symbol_kind_type, not yysymbol_kind_t * data/skeletons/glr2.cc: here. And prefer the unnamed namespace to static. 2021-09-12 Akim Demaille glr2.cc: prefer using, and remove useless type aliases * data/skeletons/glr2.cc: here. 2021-09-12 Akim Demaille glr2.cc: put glr_state_set and glr_stack_item in unnamed namespace * data/skeletons/glr2.cc: here. 2021-09-12 Akim Demaille glr2.cc: move state_stack into the unnamed namespace * data/skeletons/glr2.cc: here. Prefer `using` to `typedef`, this is C++11. Use the type alias we introduced. 2021-09-12 Akim Demaille glr2.cc: style: clarify control flow * data/skeletons/glr2.cc (yypreference): Be more functional-style. 2021-09-12 Akim Demaille glr2.cc: move glr_stack and glr_state into the parser class In order to be able to link several glr2.cc parser together, we cannot have glr_stack and glr_state be in no namespace. Putting them in the unnamed namespace is not doable, since we need to fwd declare them in the parser. Putting them in the specified namespace is not enough either, since some users would like to be able to put several parsers in the same name, only differing by the class name. * data/skeletons/glr2.cc (glr_state, glr_stack): Move into yy::parser. 2021-09-12 Akim Demaille glr2.cc: put create_state_set_index in unnamed namespace * data/skeletons/glr2.cc: here. 2021-09-12 Akim Demaille glr2.cc: prefer unnamed namespace to 'static' * data/skeletons/glr2.cc: here. 2021-09-12 Akim Demaille glr2.cc: put semantic_option into an unnamed namespace If we link several glr2.cc parsers together, we get linking failures because of duplicate symbols. * data/skeletons/glr2.cc (semantic_option::indexIn) (semantic_option::next): Remove the useless overloads. 2021-09-12 Akim Demaille glr2.cc: don't publish YY_EXCEPTIONS We don't need them in the header file. * data/skeletons/glr2.cc (YY_EXCEPTIONS): Define only in the implementation file. * tests/headers.at (Several Parsers): Also check glr2.cc. 2021-09-12 Akim Demaille c++: demonstrate custom error messages in the examples Let's use c++/glr to demonstrate custom error messages in C++ (not just in glr2.cc). * examples/c++/glr/c++-types.yy (report_syntax_error): New. * examples/c++/glr/c++-types.test: Adjust. * examples/c/bistromathic/parse.y: Comment changes. * tests/local.at (AT_YYERROR_DEFINE(c++)): Use a nicer way to print the lookakead's name. 2021-09-12 Akim Demaille glr2.cc: custom error messages Reported by Tom Shields . * data/skeletons/glr2.cc (context): New. Use it to generate the error messages. Add support for custom error messages. * tests/calc.at: Check support for custom error messages. 2021-09-12 Akim Demaille glr2.cc: start the transition to using symbol_type Currently glr2.cc uses three variables/struct members to denote the symbols' kind (or state), value and location. lalr1.cc has two types for "complete" symbols: symbol_type and stack_symbol_type. Let's use that model in glr2.cc too. For a start use yyla (a symbol_type) to denote the lookahead, instead of the triple yytoken, yylval and yylloc. This will make easier the introduction of the "context" subclass, used in parse.error=custom. It simplifies the code in several places. For instance from: symbol_kind_type yytoken_current = this->yytoken;]b4_variant_if([[ value_type yylval_current; ]b4_symbol_variant([this->yytoken], [yylval_current], [move], [this->yylval])], [[ value_type yylval_current = this->yylval;]])[]b4_locations_if([ location_type yylloc_current = this->yylloc;])[ to: symbol_type yyla_current = std::move (this->yyla); * data/skeletons/glr2.cc (yytoken, yylval, yylloc): Replace by... (yyla): this. Adjust all dependencies. (yyloc_default): Remove, unused. * tests/c++.at, tests/glr-regression.at, tests/types.at: C++11 is required for glr2.cc. Adjust to changes in glr2.cc. 2021-09-12 Akim Demaille c++: add move assignments to the symbol type This will be used in glr2.cc, which requires C++11. * data/skeletons/c++.m4 (basic_symbol, by_kind): Add move assignment. 2021-09-12 Akim Demaille c++: avoid using the obsolete names * data/skeletons/c++.m4: Don't define obsolete identifiers in the case of glr2.cc. Let's not start with technical debt. * data/skeletons/glr2.cc, data/skeletons/lalr1.cc, * data/skeletons/variant.hh: Use token_kind_type, not token_type. * tests/c++.at, tests/local.at: Use value_type, not semantic_type. 2021-09-12 Akim Demaille regen 2021-09-11 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2021-09-11 Akim Demaille version 3.8.1 * NEWS: Record release date. 2021-09-11 Akim Demaille portability: don't use strtof It is not available on HP-UX 11i. Reported by Larkin Nickle . Gnulib provides no replacement, but anyway we should be using doubles, since difftime uses doubles. * bootstrap.conf: We want portability on stdtod. * src/counterexample.c: Use double, not float, for time measurements. 2021-09-11 Akim Demaille yacc: declare yyerror/yylex only when POSIXLY_CORRECT The recent changes to comply with POSIX are breaking Automake's test suite. Reported by Kiyoshi Kanazawa. To limit the impact of POSIX changes, bind them to $POSIXLY_CORRECT. Suggested by Karl Berry. The existing `maintainer-check-posix` Make target checks these changes. * src/getargs.h, src/getargs.c (set_yacc): New. Use it. * data/skeletons/bison.m4 (b4_posix_if): New. * data/skeletons/yacc.c (b4_declare_yyerror_and_yylex): Use it. * doc/bison.texi, tests/local.at: Adjust. 2021-09-07 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2021-09-07 Akim Demaille version 3.8 * NEWS: Record release date. 2021-09-07 Akim Demaille news: update 2021-09-02 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2021-09-02 Akim Demaille version 3.7.91 * NEWS: Record release date. 2021-09-02 Akim Demaille tests: disable "Multistart reports" For some reason this test fails on Solaris/x86. But multistart is not part of 3.8, so we can postpone the debugging of this issue. Reported by Dagobert Michelsen. * tests/report.at (Multistart reports): Comment out. 2021-08-29 Akim Demaille style: c++: formatting changes * data/skeletons/variant.hh, data/skeletons/glr2.cc: Fix space before paren issue. 2021-08-29 Akim Demaille style: rename stmtMerge as stmt_merge Follow the GNU Coding Style. * doc/bison.texi, examples/c++/glr/c++-types.yy, * examples/c/glr/c++-types.y, tests/cxx-type.at: s/stmtMerge/stmt_merge/g. 2021-08-29 Akim Demaille c++: beware of Visual C++ limitations Contrary to what commit d8cc6b073e2802d871a16dc62418a7eb62ed2216 "c++: shorten the assertions that check whether tokens are correct" believed, MS Visual C++'s preprocessor limitation is not on the input line length, but on the size of the line holding the full C++ statement. Reported by Vince Huffaker * data/skeletons/variant.hh (_b4_symbol_constructor_define): Hide the assertion from Visual C++. 2021-08-29 Akim Demaille tests: don't postprocess stderr with sed On Solaris, sed throws away the NUL bytes from the stream, even in C locale. So instead of postprocessing bison's stderr to neutralize changes in value of `argv[0]`, use an envvar to actually neutralize variations of `argv[0]` during tests. Reported by Dagobert Michelsen. * src/main.c (main): Change `argv[0]` if BISON_PROGRAM_NAME is defined. * tests/bison.in: No longer mess with stderr, just pass the expected BISON_PROGRAM_NAME value. 2021-08-19 Akim Demaille CI: don't require Autoconf 2.71 2021-08-19 Akim Demaille examples: don't demonstrate multistart, which is not part of 3.8 Besides, for mysterious reasons, this fails on some environment. Reported by Dagobert Michelsen. * examples/c/lexcalc/lexcalc.test, examples/c/lexcalc/parse.y, * examples/c/lexcalc/scan.l: Revert to a single-start example. 2021-08-19 Akim Demaille tests: grep -E is not portable Reported by Dagobert Michelsen. * tests/calc.at: Use $EGREP. 2021-08-19 Akim Demaille glr2.cc: require C++11 Reported by Dagobert Michelsen. https://lists.gnu.org/r/bug-bison/2021-08/msg00006.html * m4/bison-cxx-std.m4 (_BISON_CXXSTD_98_snippet): We don't need vector::data, it was only for glr2.cc, which is C++11 anyway. (_BISON_CXXSTD_11_snippet): We need vector::data and std::swap on arrays. * m4/cxx.m4 (BISON_TEST_FOR_WORKING_CXX_COMPILER): We don't need vector::data. * tests/local.at (AT_COMPILE_CXX): Skip when glr2.cc and no support for C++11. 2021-08-13 Akim Demaille readme: add D 2021-08-13 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2021-08-13 Akim Demaille version 3.7.90 * NEWS: Record release date. 2021-08-13 Akim Demaille build: ignore syntax-check's file_system check on Doxyfile * cfg.mk: Here. 2021-08-12 Akim Demaille tests: fix incorrect test case In c2ba260487927a4cdf061fae10fdd1b387ca3d95 ("glr.c: fix signature when using custom error messages"), I meant to add a test case for C, not C++. It does not work in C++. * tests/calc.at: Run for glr.c, not glr.cc. 2021-08-12 Akim Demaille gnulib: update 2021-08-12 Akim Demaille news: be clear that glr2.cc is C++11 2021-08-12 Akim Demaille build: modernize to newer Autoconf releases * configure.ac: here. 2021-08-12 Akim Demaille build: reject C++ compilers that don't support std::vector::data GCC 4.2 on macOS claims to support C++98, but does not feature it. input.cc: In member function 'void state_stack::yycompressStack()': input.cc:1774: error: 'class std::vector >' has no member named 'data' Reported by Christopher Nielsen . . * m4/bison-cxx-std.m4 (_BISON_CXXSTD_98_snippet): Check for it. * m4/cxx.m4 (BISON_TEST_FOR_WORKING_CXX_COMPILER): Ditto. 2021-08-12 Akim Demaille build: check for C++ 20 and 2b C++20 is released. * configure.ac, m4/bison-cxx-std.m4, tests/atlocal.in, * tests/local.at: Replace 2a with 20, and add support for 2b. 2021-08-12 Akim Demaille tests: remove obsolete checks * tests/atlocal.in: It was an error for tests to depend on gnulib: they must not, as gnulib would hide portability issues that we want to catch. So this piece of code is no longer useful, and must not be useful. 2021-08-12 Akim Demaille tests: be robust to old versions of xlstproc Reported by Christopher Nielsen . * tests/report.at (AT_CHECK_HTML): New. Use it. 2021-08-11 Akim Demaille c: fix _Noreturn support detection Reported by Christopher Nielsen . and . * data/skeletons/c.m4: Fix typo. 2021-08-11 Akim Demaille lalr1.cc: style changes * data/skeletons/lalr1.cc: Formatting changes. Use more `const`, as in glr2.cc. 2021-08-11 Akim Demaille glr.c: fix signature when using custom error messages Reported by Tom Shields . * data/skeletons/glr.c (yypcontext_location): Fix return type. * tests/calc.at: Check the case pure, location, custom error messages. 2021-08-11 Akim Demaille tests: check symbol/token renumbering In some extreme situations, with lots of useless tokens, Bison was numbering them incorrectly, which resulted in a broken grammar. commit a774839ca873d1082f79ba3c4eecc1e242a28ce1. * tests/regression.at (Useless Tokens): New. 2021-08-09 Akim Demaille style: rename b4_lex as b4_yylex For consistency with b4_yyerror_formals, etc. * data/skeletons/bison.m4, data/skeletons/c.m4, data/skeletons/d.m4, * data/skeletons/glr.c, data/skeletons/glr2.cc, * data/skeletons/java.m4, data/skeletons/lalr1.cc, * data/skeletons/lalr1.d, data/skeletons/lalr1.java, * data/skeletons/yacc.c (b4_lex, b4_lex_formals): Rename as... (b4_yylex, b4_yylex_formals): these. 2021-08-09 Akim Demaille yacc: comply with recent POSIX updates: declare yyerror and yylex In POSIX Yacc mode, declare yyerror and yylex unless already #defined, or if YYERROR_IS_DECLARED/YYLEX_IS_DECLARED are defined (for consistency with Bison's YYSTYPE_IS_DECLARED/YYLTYPE_IS_DECLARED). See . * data/skeletons/c.m4 (b4_function_declare): Resurect. (b4_lex_formals): Since we will possibly expose this prototype in the header, take the prefix into account. * data/skeletons/yacc.c (b4_declare_yyerror_and_yylex): New. (b4_shared_declarations): Use it. * tests/local.at (AT_YACC_IF): New. When in Yacc mode, set the `yacc` Autotest keyword. (AT_YYERROR_DECLARE(c)): Don't declare in Yacc mode, to avoid clashes (since this signature is static). (AT_YYERROR_DEFINE(c)): Don't define as static in Yacc mode. * tests/regression.at (Early token definitions with --yacc): Specify that we are in Yacc mode. 2021-08-09 Akim Demaille build: enable -Wmismatched-dealloc * configure.ac (warn_common): Here. 2021-08-09 Akim Demaille gnulib: update 2021-08-08 Akim Demaille bistromathic: beware of portability issues with readline In some cases readline emits a trailing spaces after the last suggestion, which results in errors such as: ``` -( - atan cos exp ln number sin sqrt$ +( - atan cos exp ln number sin sqrt $ ``` Reported by Christopher Nielsen . * examples/test (run): Add support for -t. * examples/c/bistromathic/bistromathic.test: Use it. 2021-08-07 Akim Demaille d: prepare to be able to run LAC tests Unfortunately it seems to be quite difficult to have "LAC: Exploratory stack" run for D. * data/skeletons/lalr1.d: We need File when traces are enabled. * tests/local.at (AT_YYLEX_DEFINE(d)): New. * tests/regression.at: Prepare for D, but don't run it, it does not work. 2021-08-07 Akim Demaille tests: factor iterating over skeletons * tests/local.at (AT_FOR_EACH_SKEL): New. Use where appropriate. * data/skeletons/lalr1.d: Reject -d. * tests/input.at, tests/scanner.at: Also check D. 2021-08-07 Akim Demaille m4: catch suspicions of unevaluated macros Check in m4's output if there are sequences such as m4_foo or b4_foo, which are probably resulting from incorrect m4 processing. It actually already is useful: - it caught a leaking b4_lac_if leaking from glr.c, where LAC is not supported, hence b4_lac_if is not defined. - it also caught references to location.hh in position.hh when location.hh does not exist. - while making "Code injection" robust to these new warnings (it is its very purpose to let b4_canary pass unevaluated), I saw that it did not check lalr1.d, and when adding lalr1.d, it revealed it did underquote ocurrences of token value types. * src/scan-skel.l (macro): New abbreviation. Use it. * data/skeletons/glr.c: Don't use b4_lac_if, we don't have it. * data/skeletons/location.cc: Don't generate position.hh when we don't generate location.hh. * data/skeletons/d.m4 (b4_basic_symbol_constructor_define): Fix underquotation. * data/skeletons/bison.m4 (b4_canary): New. * tests/input.at (Code injection): Use it, and check lalr1.d too. 2021-08-07 Akim Demaille style: formatting changes in scan-code.l * src/scan-code.l: Fix indentation. 2021-08-07 Akim Demaille doc: avoid #define YYDEBUG in C++ * doc/bison.texi (Enabling Traces): here. 2021-08-06 Akim Demaille tests: extract AT_LOCATION_PRINT_DECLARE and AT_LOCATION_PRINT_DEFINE * tests/local.at (AT_LOCATION_PRINT_DECLARE) (AT_LOCATION_PRINT_DEFINE): New. 2021-08-06 Akim Demaille tests: rename AT_YACC_IF as AT_YACC_C_IF In data/, b4_yacc_if refers to %yacc, not yacc.c. * tests/local.at (AT_YACC_IF): Rename as... (AT_YACC_C_IF): this. 2021-08-06 Akim Demaille doc: more pointers to the examples * doc/bison.texi (Infix Calc): Here. 2021-08-06 Akim Demaille git: add diff patterns for Autotest and Texinfo 2021-08-06 Akim Demaille CI: fix the GCC 9 build 2021-08-04 Akim Demaille todo: d: push and token ctors are done 2021-08-04 Akim Demaille doc: refer to the examples * doc/bison.texi: Point to rpcalc, mfcalc, simple.y, calc++, c/glr and c++/glr. 2021-08-04 Akim Demaille examples: modernize the example Makefiles * examples/c++/Makefile, examples/c++/calc++/Makefile, * examples/c++/glr/Makefile, examples/c/bistromathic/Makefile, * examples/c/calc/Makefile, examples/c/glr/Makefile, * examples/c/lexcalc/Makefile, examples/c/mfcalc/Makefile, * examples/c/pushcalc/Makefile, examples/c/reccalc/Makefile, * examples/c/rpcalc/Makefile, examples/d/calc/Makefile, * examples/d/simple/Makefile, examples/java/calc/Makefile, * examples/java/simple/Makefile: Use --html to generate *.html directly. No longer demonstrate --xml. No longer show rules for xml to html. Use --header, not --defines. Use --graph without specifying the output file now that we generate *.gv by default. 2021-08-04 Akim Demaille doc: glr: document typed mergers See . * doc/bison.texi (Merging GLR Parses): document typed mergers. And avoid #define YYSTYPE. 2021-08-04 Akim Demaille style: tests: rebox comments * tests/glr-regression.at: here. 2021-08-04 Akim Demaille news: update In particular, announce lalr1.d. 2021-08-03 Akim Demaille doc: fix spello * doc/bison.texi (Multiple start-symbols): here. 2021-08-03 Akim Demaille scan: fix typo in UTF-8 escape We had: ``` -mbchar ...|\xF0[\x\90-\xBF]([\x80-\xBF]{2})|... +mbchar ...|\xF0[\x90-\xBF]([\x80-\xBF]{2})|... ``` so a precise sequence that matches the incorrect regex can let NUL bytes pass through, which triggers an assertion violation downstream. It is a pity that Flex does not report an error for such input. Reported by Ahcheong Lee . * src/scan-gram.l (mbchar): Fix the bad regex. * tests/input.at (Invalid inputs): Check that case. 2021-08-03 Akim Demaille todo: POSIX yacc and prototypes 2021-08-03 Akim Demaille doc: update Doxygen template file * doc/Doxyfile.in: here. 2021-08-03 Akim Demaille doc: a bit of editing * doc/bison.texi: Use @samp{...}, not "..." for pieces of code. Use @samp{...}, not @command{...} for command lines. Promote %header/--header over %defines/--defines. Spellcheck. 2021-08-03 Akim Demaille gnulib: update 2021-08-01 Alyssa Ross getargs: don't translate first line of --version * src/getargs.c (version): here. 2021-08-01 Akim Demaille all: fix confusion between token ctor and symbol ctor The symbol constructors are genuine constructors. Token constructors are plain functions that construct tokens. 2021-07-25 Akim Demaille d: minor clean up * doc/bison.texi: Use @samp, not "...", around pieces of code. * examples/d/calc/calc.y: Don't promote %union. 2021-06-23 Akim Demaille gnulib: update 2021-06-06 Adela Vais d: demonstrate the token constructors * examples/d/calc/calc.y: Use the token constructors in the 'calc' example. 2021-04-11 Adela Vais d: update documentation * doc/bison.texi: Various fixes. (D Push Parser Interface, D Complete Symbols): New sections. 2021-04-11 Adela Vais d: demonstrate the push parser * examples/d/calc/calc.y: Use a parser of type 'push' in the calc example. 2021-04-11 Adela Vais d: add push parser support Support the push-pull directive with the options pull, push and both. Pull remains the default option. * data/skeletons/d.m4: Add user aliases for the push parser's return values: PUSH_MORE, ABORT, ACCEPT. * data/skeletons/lalr1.d: Add push parser support. * tests/calc.at: Test it. 2021-03-26 Adela Vais d: add token constructors support The user can return from yylex() by calling the Symbol method of the same name as the TokenKind reported, and adding the parameters for value and location if necessary. These methods generate compile-time errors if the parameters are not correlated. Token constructors work with both %union and api.value.type union. * data/skeletons/d.m4: Here. * tests/calc.at: Test it. 2021-03-26 Akim Demaille d: add api.value.type union support The union of the values is handled by the backend. In D, unions can hold classes, structs, etc., so this is more similar to the C++ api.value.type variant. * data/skeletons/d.m4, data/skeletons/lalr1.d: Here. * tests/calc.at, tests/local.at: Test it. 2021-03-26 Adela Vais d: rewrite Symbol's constructors in M4 The D code was becoming too complex. M4 is easier to maintain in the long run. * data/skeletons/d.m4: Here. 2021-03-26 Akim Demaille gnulib: update 2021-03-10 Akim Demaille files: please syntax-check * src/files.c (string_free): syntax-check does not want us to cast arguments to free. 2021-03-10 Akim Demaille Merge 3.7.6 into master * maint: maint: post-release administrivia version 3.7.6 yacc: fix push parser tables: fix again the handling of useless tokens 2021-03-08 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2021-03-08 Akim Demaille version 3.7.6 * NEWS: Record release date. 2021-03-07 Akim Demaille doc: don't mention YY_LOCATION_PRINT * doc/bison.texi (Syntax Error Reporting Function): Don't refer to YY_LOCATION_PRINT, it is a private internal detail. 2021-03-07 Akim Demaille yacc: fix push parser When a pstate is used for multiple successive parses, some state may leak from one run into the following one. That was introduced in 330552ea499ca474f65967160e9d4e50265f9631 "yacc.c: push: don't clear the parser state when accepting/rejecting". Reported by Ryan https://lists.gnu.org/r/bug-bison/2021-03/msg00000.html * data/skeletons/yacc.c (yypush_parse): We reusing a pstate from a previous run, do behave as if it were the first run. * tests/push.at (Pstate reuse): Check this. 2021-03-07 Akim Demaille tables: fix again the handling of useless tokens The right-shift added in c22902e360e0fbbe9fd5657dcf107e03166da309 ("tables: fix handling for useless tokens") is incorrect. In particular, we need to reset the "new" bits. Reported by Balázs Scheidler. https://github.com/akimd/bison/issues/74 * src/tables.c (pos_set_set): Fix the right-shift. 2021-02-26 Akim Demaille html: fix memory leak * src/print-xml.c (print_html): Free allocated memory. 2021-02-26 Akim Demaille output: cache the mapped file names Don't repeatedly call malloc/free for each call to map_file_name. * bootstrap.conf: We need hash-map. * src/files.h, src/files.c (map_file_name): The caller must not free the result. Adjust callers. (mapped_dir_prefix, spec_mapped_header_file): Remove. * src/files.c (map_file_name): Rename as... (map_file_name_alloc): this. (mapped_files, map_file_name, string_equals, string_hash, string_free): New. 2021-02-26 Joshua Watt output: use mapped file name for symbols Applies the file name mapping before exporting it as a symbol. This allows the symbols to correctly respect the --file-prefix-map command line option. 2021-02-26 Akim Demaille style: formatting changes * src/files.c, src/files.h: Save horizontal space. Prefer `res` for returned values. Put the doc into the header. 2021-02-26 Akim Demaille gnulib: update 2021-02-25 Joshua Watt doc: fix documented function name * data/README.md: Fix the name of a referenced function to match the name in code. 2021-02-25 Akim Demaille regen 2021-02-09 Akim Demaille doc: printing locations Document YYLOCATION_PRINT. * doc/bison.texi (Printing Locations): New node. 2021-02-09 Akim Demaille examples: improve some function prototypes * examples/c/bistromathic/parse.y, examples/c/glr/c++-types.y, * examples/c/lexcalc/parse.y: Use const where appropriate. Avoid `yy` prefixes where it does not make sense. Avoid the `p` prefix for pointers. 2021-02-09 Akim Demaille examples: use YYLOCATION_PRINT * examples/c/bistromathic/parse.y, examples/c/glr/c++-types.y, * examples/c/lexcalc/parse.y: Don't use the private internal detail `YY_LOCATION_PRINT`, use `YYLOCATION_PRINT`. 2021-02-09 Akim Demaille c: rename YY_LOCATION_PRINT as YYLOCATION_PRINT and make it public It is very helpful to be able to look at the locations in the debug traces, let's provide the users with (i) a means to do that for their location types, and (ii) a public macro to print locations when debug traces are enabled. * data/skeletons/c.m4 (b4_yy_location_print_define): Rename as... (b4_yylocation_print_define): this. Define YYLOCATION_PRINT instead of YY_LOCATION_PRINT. Ensure backward compatibility for those who might have defined/used YY_LOCATION_PRINT in spite the warnings. Adjust dependencies. * data/skeletons/glr2.cc: We don't use YYLOCATION_PRINT here. 2021-02-09 Akim Demaille examples: do not rely on YY_LOCATION_PRINT * examples/c/bistromathic/parse.y (location_print): New. Use it. 2021-02-02 Akim Demaille c: stop defining YY_LOCATION_PRINT when locations are not enabled * data/skeletons/c.m4 (b4_yy_location_print_define): here. 2021-02-02 Akim Demaille graph: output foo.gv by default, instead of *.dot That change was started in Bison 3.4. The announcement for 3.7 stated that in Bison 3.8 we would use *.gv by default. * src/files.c (compute_output_file_names): spec_graph_file defaults too *.gv. * doc/bison.texi, examples/c++/calc++/local.mk, tests/output.at: Adjust. 2021-02-01 Akim Demaille tests: adjust to the removal of YYPRINT * tests/regression.at: here. 2021-02-01 Akim Demaille yacc: remove support for YYPRINT Its removal was annonced several times in NEWS (for Bison 3.5, 3.6, 3.7). * data/skeletons/c.m4, data/skeletons/yacc.c: Remove support for YYPRINT. * NEWS: Fix the mess introduced by the merge. Document the removal of YYPRINT. * doc/bison.texi (The YYPRINT Macro): Remove. 2021-02-01 Akim Demaille c++: issue a compile-time warning when #define YYSTYPE is used Using #define YYSTYPE has always been strongly discouraged in C++. Macros are dangerous and can result in subtle bugs. https://lists.gnu.org/r/bug-bison/2020-12/msg00007.html Maybe some people are currently using #define YYSTYPE. Instead of dropping support right now, first issue a warning. Bison can "see" if YYDEBUG is defined (it could even be on the command line), only the compiler knows. Unfortunately `#warning` is non-portable, and actually GCC even dies on it when `-pedantic` is enabled. So we need to use `#pragma message`. We must make it conditional as some compilers might not support it, but it doesn't matter if only _some_ compilers emit the warning: it should be enough to catch the attention of the developers. * data/skeletons/c++.m4: Issue a warning when the user defined YYSTYPE. * tests/actions.at: Don't #define YYSTYPE. * tests/headers.at (Several parsers): Ignore the YYSTYPE in the warning. 2021-02-01 Akim Demaille c++: make it clear that #define YYSTYPE is not supported We have been accepting this for years, but it is deprecated: people are expecting to define api.value.type instead. * doc/bison.texi: Make it clear that YYSTYPE and YYLTYPE are for C only. 2021-02-01 Akim Demaille c++: stop referring to YYSTYPE in yylex * data/skeletons/glr2.cc, data/skeletons/lalr1.cc (b4_lex): Don't refer to YYSTYPE, but to value_type. Note that this is "symbolic" in the sense that these values are never used by b4_function_call. 2021-02-01 Akim Demaille CI: try GCC 11 and Clang 11 2021-02-01 Akim Demaille gnulib: update 2021-01-31 Akim Demaille glr2.cc: no longer be part of b4_glr_cc_if We are fully independant of glr.c/glr.cc, we no longer need to pretend we are glr.cc. * data/skeletons/bison.m4: here. 2021-01-31 Akim Demaille m4: provide macros for api.value.type=union * data/skeletons/bison.m4 (b4_union_if): New. Remove support for "%define variant", which was deprecated long ago. * tests/local.at (AT_UNION_IF): New. 2021-01-30 Akim Demaille traces: display the Bison version * src/main.c (main): When traces are enabled, display the Bison version. * tests/conflicts.at, tests/report.at, tests/sets.at: Use AT_PACKAGE_VERSION (for package.m4) instead of post-processing the output. 2021-01-30 Akim Demaille tests: run the main test suite on "make check" Automake appears to not recognize "check-local" when it is not a isolated target. Fixes 77a8b84fc8bbe39dc231a1f657fd56a50fac5855. * tests/local.mk (check-local): Make it visible to Automake. 2021-01-30 Adela Vais d: tests: various style fixes * tests/calc.at, tests/scanner.at: Here. (yylex): Report values directly, without storing them to the union first. 2021-01-30 Paul Eggert Update URLs to prefer https: to http: Also, fix a few http: URLs that were no longer working. 2021-01-29 Paul Eggert Update URLs to prefer https: to http: Also, fix a few http: URLs that were no longer working. 2021-01-27 Akim Demaille tests: provide check-examples and check-tests * examples/local.mk (check-examples): New. * tests/local.mk (check-tests): New. * README-hacking.md: Document them. 2021-01-27 Akim Demaille bistromathic: use gettext for all the messages * examples/c/bistromathic/parse.y: Add missing calls to _. 2021-01-24 Adela Vais d: tests: avoid mixing output from reportSyntaxError and getExpectedTokens Function reportSyntaxError buffers and prints the message at the end. * tests/local.at: Here. 2021-01-24 Akim Demaille glr2.cc: fix destructor support Currently glr2.cc does not use 'symbol's everywhere, in various places it also uses yykind, yyval and yyloc "by hand". So we need two different calls for user-defined constructors: once for ~symbol, another for yy_destroy_. Both need to call the user destructors with different calling conventions. * data/skeletons/glr2.cc (b4_symbol_action): Rename as... (b4_symbol_action_for_yyval): this. (b4_symbol_action): New, taken from lalr1.cc. (yy_destroy_, yy_symbol_value_print_): Use b4_symbol_action_for_yyval. 2021-01-24 Akim Demaille glr2.cc: beware of types with angle brackets For C++98, avoid "yyval.as>". * data/skeletons/glr2.cc (b4_call_merger): Add spaces within "<...>". 2021-01-24 Akim Demaille package: fix details after merge with 3.7.5 * TODO, cfg.mk: Update. * src/parse-gram.c, src/parse-gram.h: Regen. 2021-01-24 Akim Demaille Merge tag 'v3.7.5' Three new commits: commit 8358090292e21c61a583da542bad9099ad65f355 Author: Paul Eggert Date: Wed Jan 20 18:30:16 2021 -0800 c: port to HP-UX 11.23 commit 2c294c132528ede23d8ae4959783a67e9ff05ac5 Author: Vincent Imbimbo Date: Sat Jan 23 13:25:18 2021 -0500 cex: fix state-item pruning commit c22902e360e0fbbe9fd5657dcf107e03166da309 Author: Akim Demaille Date: Sat Jan 23 18:40:15 2021 +0100 tables: fix handling for useless tokens 2021-01-24 Akim Demaille doc: fix typo Reported by Kaz Kylheku. https://lists.gnu.org/r/bison-patches/2020-11/msg00019.html * doc/bison.texi (Versioning): here. 2021-01-24 Kaz Kylheku doc: document best deployment practices. * doc/bison.texi (Versioning): New node about practices regarding dealing with multiple versions of Bison. 2021-01-24 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2021-01-24 Akim Demaille version 3.7.5 * NEWS: Record release date. 2021-01-24 Akim Demaille tables: fix handling for useless tokens In some rare conditions, the generated parser can be wrong when there are useless tokens. Reported by Balázs Scheidler. https://github.com/akimd/bison/issues/72 Balázs managed to prove that the bug was introduced in commit af1c6f973a60a51c609903713ff8f7fce0887025 Author: Theophile Ranquet Date: Tue Nov 13 10:38:49 2012 +0000 tables: use bitsets for a performance boost Suggested by Yuri at . The improvement is marginal for most grammars, but notable for large grammars (e.g., PosgreSQL's postgre.y), and very large for the sample.y grammar submitted by Yuri in http://lists.gnu.org/archive/html/bison-patches/2012-01/msg00012.html. Measured with --trace=time -fsyntax-only. parser action tables postgre.y sample.y Before 0,129 (44%) 37,095 (99%) After 0,117 (42%) 5,046 (93%) * src/tables.c (pos): Replace this set of integer coded as an unsorted array of integers with... (pos_set): this bitset. which was implemented long ago, but that I installed only recently (March 2019), first published in v3.3.90. That patch introduces a bitset to represent a set of integers. It managed negative integers by using a (fixed) base (the smallest integer to represent). It avoided negative accesses into the bitset by ignoring integers smaller than the base, under the asumption that these cases correspond to useless tokens that are ignored anyway. While it turns out to be true for all the test cases in the test suite (!), Balázs' use case demonstrates that it is not always the case. So we need to be able to accept negative integers that are smaller than the current base. "Amusingly" enough, the aforementioned patch was visibly unsure about itself: /* Store PLACE into POS_SET. PLACE might not belong to the set of possible values for instance with useless tokens. It would be more satisfying to eliminate the need for this 'if'. */ This commit needs several improvements in the future: - support from bitset for bit assignment and shifts - amortized resizing of pos_set - test cases * src/tables.c (pos_set_base, pos_set_dump, pos_set_set, pos_set_test): New. Use them instead of using bitset_set and bitset_test directly. 2021-01-24 Vincent Imbimbo cex: fix state-item pruning There were several bugs in pruning that would leave the state-item graph in an inconsistent state which could cause crashes later on: - Pruning now happens in one pass instead of two. - Disabled state-items no longer prune the state-items they transition to if that state-item has other states that transition to it. - State-items that transition to disabled state-items are always pruned even if they have productions. Reported by Michal Bartkowiak https://lists.gnu.org/r/bug-bison/2021-01/msg00000.html and Zartaj Majeed https://github.com/akimd/bison/issues/71 * src/state-item.c (prune_forward, prune_backward): Fuse into... (prune_state_item): this. Adjust callers. 2021-01-23 Akim Demaille package: pacify syntax-check * cfg.mk: Currently we cannot update gnulib because of portability issues with ancient versions of clang (https://lists.gnu.org/r/bug-gnulib/2021-01/msg00241.html). So skip the check about copyright date for gnulib. 2021-01-23 Akim Demaille news: update 2021-01-23 Akim Demaille package: bump copyrights to 2021 Run 'make update-copyright'. 2021-01-23 Akim Demaille %merge: associate it to its first definition, not the latest Currently each time we meet %merge we record this location as the defining location (and symbol). Instead, record the first definition. In the generated code we go from yy0->A = merge (*yy0, *yy1); to yy0->S = merge (*yy0, *yy1); where S was indeed the first symbol, and in the diagnostics we go from glr-regr18.y:30.18-24: error: result type clash on merge function 'merge': != 30 | sym2: sym3 %merge { $$ = $1; } ; | ^~~~~~~ glr-regr18.y:29.18-24: note: previous declaration 29 | sym1: sym2 %merge { $$ = $1; } ; | ^~~~~~~ glr-regr18.y:31.13-19: error: result type clash on merge function 'merge': != 31 | sym3: %merge { $$ = 0; } ; | ^~~~~~~ glr-regr18.y:30.18-24: note: previous declaration 30 | sym2: sym3 %merge { $$ = $1; } ; | ^~~~~~~ to glr-regr18.y:30.18-24: error: result type clash on merge function 'merge': != 30 | sym2: sym3 %merge { $$ = $1; } ; | ^~~~~~~ glr-regr18.y:29.18-24: note: previous declaration 29 | sym1: sym2 %merge { $$ = $1; } ; | ^~~~~~~ glr-regr18.y:31.13-19: error: result type clash on merge function 'merge': != 31 | sym3: %merge { $$ = 0; } ; | ^~~~~~~ glr-regr18.y:29.18-24: note: previous declaration 29 | sym1: sym2 %merge { $$ = $1; } ; | ^~~~~~~ where both duplicates are reported against definition 1, rather than using definition 1 as a reference when diagnosing about definition 2, and then 2 as a reference for 3. * src/reader.c (record_merge_function_type): Keep the first definition. * tests/glr-regression.at: Adjust. 2021-01-23 Akim Demaille %merge: fix compatibility with api.value.type=union Reported by Jot Dot. https://lists.gnu.org/r/help-bison/2020-12/msg00014.html * data/skeletons/glr.c, data/skeletons/glr2.cc (b4_call_merger): Use the symbol's slot, not its type. * examples/c/glr/c++-types.y: Use explicit per-symbol typing together with api.value.type=union. (yylex): Use yytoken_kind_t. 2021-01-23 Akim Demaille %merge: delegate the generation of calls to mergers to m4 Don't generate C code from bison, leave that to the skeletons. * src/output.c (merger_output): Emit invocations to b4_call_merger. * data/skeletons/glr.c (b4_call_merger): New. 2021-01-23 Akim Demaille %merge: let mergers record a typing-symbol, rather than a type Symbols are richer than types, and in M4 it is my simpler (and more common) to deal with symbols rather than types. So let's associate mergers to a symbol rather than a type name. * src/reader.h (merger_list): Replace the 'type' member by a symbol member. * src/reader.c (record_merge_function_type): Take a symbol as argument, rather than a type name. * src/output.c (merger_output): Adjust. 2021-01-23 Akim Demaille %merge: clearer tests on diagnostics * tests/glr-regression.at: Use caret errors. 2021-01-23 Akim Demaille skeletons: introduce "slot"s for symbols Extracted from d9cf99b6a5cb0345e91dfb90fe6d6473024ea97a, in the master branch. * data/skeletons/bison.m4 (b4_symbol_slot): New, with safer semantics than type and type_tag. 2021-01-23 Akim Demaille style: YYUSE is private, make it YY_USE This macro is not exposed to users, make start it with 'YY_'. * data/skeletons/bison.m4, data/skeletons/c.m4, data/skeletons/glr.c, * data/skeletons/glr.cc, data/skeletons/lalr1.cc, * src/parse-gram.c, tests/actions.at, tests/c++.at, tests/headers.at, * tests/local.at (YYUSE): Rename as... (YY_USE): this. 2021-01-23 Akim Demaille package: codespell * src/parse-gram.y: Fix spelling. 2021-01-23 Akim Demaille cex: fix traces: fix display of disabled items The display of disabled state items is incorrect. The item is stuttered, and lacks on end-of-line. From State 7: 1 exp: exp • "⊕" exp -> 1 exp: exp "⊕" • exp <- 1 exp: • exp "⊕" exp 2 exp: exp • "+" exp 2 exp: exp • "+" exp DISABLED 2 exp: exp "+" exp • <- 2 exp: exp "+" • exp 3 exp: exp • "+" exp 3 exp: exp • "+" exp DISABLED 3 exp: exp "+" exp • <- 3 exp: exp "+" • exp to State 7: 1 exp: exp • "⊕" exp -> 1 exp: exp "⊕" • exp <- 1 exp: • exp "⊕" exp 2 exp: exp • "+" exp DISABLED 2 exp: exp "+" exp • <- 2 exp: exp "+" • exp 3 exp: exp • "+" exp DISABLED 3 exp: exp "+" exp • <- 3 exp: exp "+" • exp * src/state-item.c (state_items_report): Don't issue disabled items twice, and issue two '\n' at their end. * tests/conflicts.at: Check it. 2021-01-23 Akim Demaille cex: fix traces: add missing end-of-lines In 430ca0fc632f5e8072fe468b8a99c640985f6926, I completely forgot that `puts` adds a `\n`. * src/lssi.c, src/state-item.c: Restore missing end-of-lines in the output. 2021-01-23 Akim Demaille cex: add support for $TIME_LIMIT * src/counterexample.c (TIME_LIMIT): Replace with... (time_limit): this. (counterexample_init): Check $TIME_LIMIT. * src/scan-gram.l: Reorder includes. 2021-01-23 Akim Demaille cex: send traces to stderr, not stdout When comparing traces from different machines, the mixture of stdout/stderr in the output are making things uselessly difficult. * src/lssi.c, src/state-item.c: Output debug traces on stderr. 2021-01-23 Akim Demaille c++: I'm tired of Flex's warnings * doc/bison.texi: Disable another warning I'm tired to see. New releases would be most welcome. 2021-01-23 Akim Demaille glr.cc: don't "leak" yyparse When using glr.cc, the C function yyparse is an internal detail that should not be exposed. Users might call it by accident (I did). * data/skeletons/glr.c (yyparse): When used for glr.cc, rename as yy_parse_impl. * data/skeletons/glr.cc: Adjust. 2021-01-23 Akim Demaille tables: avoid warnings and save bits The yydefgoto table uses -1 as an invalid for an impossible case (we never use yydefgoto[0], since it corresponds to the reduction to $accept, which never happens). Since yydefgoto is a table of state numbers, this -1 forces a signed type uselessly, which (1) might trigger compiler warnings when storing a value from yydefgoto into a state number (nonnegative), and (2) wastes bits which might result in using a int16 where a uint8 suffices. Reported by Jot Dot . https://lists.gnu.org/r/bug-bison/2020-11/msg00027.html * src/tables.c (default_goto): Use 0 rather than -1 as invalid value. * tests/regression.at: Adjust. 2021-01-23 Akim Demaille c++: use noexcept where appropriate Reported by Don Macpherson. https://github.com/akimd/bison/issues/63 https://github.com/akimd/bison/issues/64 * data/skeletons/c++.m4, data/skeletons/lalr1.cc: here. 2021-01-23 Martin Rehak examples: avoid "unbound variable" errors When the shell option `nounset` is set, we may get "unbound variable" errors. https://lists.gnu.org/r/bug-bison/2020-11/msg00013.html * examples/test (diff_opts): Be sure to initialize it. 2021-01-23 Akim Demaille autoconf: update 2021-01-23 Akim Demaille c: adjust _Noreturn to pedantic clang Reported by Joe Nelson . https://lists.gnu.org/r/help-bison/2021-01/msg00004.html Fixed by Paul Eggert in gnulib. https://lists.gnu.org/r/bug-gnulib/2021-01/msg00156.html * data/skeletons/c.m4 (b4_attribute_define): Adjust _Noreturn to pedantic clang. 2021-01-21 Paul Eggert c: port to HP-UX 11.23 Problem reported by Albert Chin in: https://lists.gnu.org/r/bug-bison/2021-01/msg00029.html * data/skeletons/c.m4 (b4_c99_int_type_define): Work around HP-UX bug. 2021-01-17 Akim Demaille d: examples: reduce scopes * data/skeletons/lalr1.d (YYLocation.step): New. * examples/d/calc/calc.y (Lexer): Reduce scopes to avoid uninitialized varibles. Factor the handling of locations. We don't need lenChars. 2021-01-17 Adela Vais d: examples: calc: remove Value from Lexer The Symbol constructor does not use it, so it is easier to not use Value at all. * examples/d/calc/calc.y: Here. 2021-01-17 Adela Vais d: examples: calc: use of std.conv.parse for location From Dlang v2.095.0 onwards, std.conv.parse reports the number of consumed characters. * examples/d/calc/calc.y: Here. 2021-01-17 Adela Vais d: examples: simple: fix style * examples/d/simple/calc.y: Formatting changes. 2021-01-17 Akim Demaille CI: run D tests * .travis.yml: here. 2021-01-16 Akim Demaille package: codespell * data/skeletons/glr2.cc, src/parse-gram.y: Fix spelling. 2021-01-16 Akim Demaille package: fixes * examples/c++/glr/local.mk, examples/c/glr/local.mk (CLEANFILES): Complete to fix distcheck. * src/output.c (merger_output): Obfuscate to pacify syntax-check's sc_space_before_open_paren. 2021-01-16 Akim Demaille regen 2021-01-16 Akim Demaille package: bump copyrights to 2021 Run 'make update-copyright'. 2021-01-16 Akim Demaille gnulib: update * src/output.c, src/print-xml.c: Adjust. 2021-01-16 Akim Demaille autoconf: update 2021-01-16 Akim Demaille cex: fix traces The display of disabled state items is incorrect. The item is stuttered, and lacks on end-of-line. From State 7: 1 exp: exp • "⊕" exp -> 1 exp: exp "⊕" • exp <- 1 exp: • exp "⊕" exp 2 exp: exp • "+" exp 2 exp: exp • "+" exp DISABLED 2 exp: exp "+" exp • <- 2 exp: exp "+" • exp 3 exp: exp • "+" exp 3 exp: exp • "+" exp DISABLED 3 exp: exp "+" exp • <- 3 exp: exp "+" • exp to State 7: 1 exp: exp • "⊕" exp -> 1 exp: exp "⊕" • exp <- 1 exp: • exp "⊕" exp 2 exp: exp • "+" exp DISABLED 2 exp: exp "+" exp • <- 2 exp: exp "+" • exp 3 exp: exp • "+" exp DISABLED 3 exp: exp "+" exp • <- 3 exp: exp "+" • exp * src/state-item.c (state_items_report): Don't issue disabled items twice, and issue two '\n' at their end. * tests/conflicts.at: Check it. 2021-01-16 Akim Demaille cex: fix traces In 430ca0fc632f5e8072fe468b8a99c640985f6926, I completely forgot that `puts` adds a `\n`. * src/lssi.c, src/state-item.c: Restore missing end-of-lines in the output. 2021-01-16 Akim Demaille c: add support for YYNOMEM Suggested by Joe Nelson . https://lists.gnu.org/r/help-bison/2020-12/msg00020.html * data/skeletons/glr.c, data/skeletons/yacc.c (YYNOMEM): New. Use it. (yyexhaustedlab): Rename as... (yynomemlab): this. * tests/calc.at: Check it. * doc/bison.texi: Document it. Fix incorrect statements about non-existing constants for YYERROR etc. 2021-01-16 Akim Demaille doc: spell check Gettext uses "catalog", not "catalogue". * doc/bison.texi, examples/c/README.md, * examples/c/bistromathic/Makefile, * examples/c/bistromathic/README.md, * examples/c/bistromathic/parse.y: Spell check. 2021-01-16 Akim Demaille yacc.c, glr.c: style: rename yyreturn and yyreturnlab * data/skeletons/glr.c, data/skeletons/yacc.c: here. 2021-01-16 Akim Demaille glr.c: style: formatting changes * data/skeletons/glr.c (YYCHK1): here. 2021-01-16 Akim Demaille tests: check YYACCEPT and YYABORT There are some tests that cover them, but nothing for all the skeletons. Let's do that in the calculator tests. * tests/calc.at: Check YYACCEPT and YYABORT. 2021-01-16 Akim Demaille glr2.cc: don't prepare for subclassing I am not aware of people subclassing the parser class, and I fail to see how this could be useful. Rather than leaving a badly baked feature (as in glr.cc currently), let's not support it at all, until someone comes and explains why and how it would be useful. * data/skeletons/glr2.cc (parser): We need no virtual function members. 2021-01-16 Akim Demaille glr2.cc: formatting changes * data/skeletons/glr2.cc: here. 2021-01-16 Akim Demaille glr2.cc: move strong_index_alias into implementation file * data/skeletons/glr2.cc: here. Fix coding style issues. 2021-01-16 Akim Demaille glr2.cc: remove some useless qualifications * data/skeletons/glr2.cc: Rely on the type-aliases provided by glr_stack. 2021-01-16 Akim Demaille glr2.cc: more free functions about the automaton into glr_stack * data/skeletons/glr2.cc (yypact_value_is_default) (yytable_value_is_error, yyisShiftAction, yyisErrorAction) (yyisDefaultedState, yydefaultAction): Move into... (glr_stack): here. Fix naming conventions. 2021-01-16 Akim Demaille glr2.cc: move free-functions into glr_stack * data/skeletons/glr2.cc (yypreference, yyLRgotoState): Move into... (glr_stack): here. 2021-01-16 Akim Demaille glr2.cc: make yygetToken a member of glr_stack It's on purpose that I keep the `this->` now. We'll see later if we want to remove them. * data/skeletons/glr2.cc (yygetToken): Move into... (glr_stack::yyget_token): this. (b4_lex): Adjust. 2021-01-16 Akim Demaille glr2.cc: move parser::parse into glr_stack Currently we have two classes that actually should be fused together: parser and glr_stack. Both carry part of the parsing: (i) parser contains `parse`, which is the top-level of the parsing process, it uses yygetToken, etc., and (ii) glr_stack takes care of all the details (dealing with the stack), and also calls yygetToken. However if we fuse them together, we would get a large parser class, in the header file. So it is probably better to split this large class using the pimpl idiom. But then it appears that glr_stack is very close from being the impl of parser. Let's improve this. For a start... * data/skeletons/glr2.cc (parser::parse): Move to... (glr_stack::parse): here. (parser::parse): Use it. 2021-01-14 Akim Demaille cex: add support for $TIME_LIMIT * src/counterexample.c (TIME_LIMIT): Replace with... (time_limit): this. (counterexample_init): Check $TIME_LIMIT. * src/scan-gram.l: Reorder includes. 2021-01-13 Akim Demaille c: adjust _Noreturn to pedantic clang Reported by Joe Nelson . https://lists.gnu.org/r/help-bison/2021-01/msg00004.html Fixed by Paul Eggert in gnulib. https://lists.gnu.org/r/bug-gnulib/2021-01/msg00156.html * data/skeletons/c.m4 (b4_attribute_define): Adjust _Noreturn to pedantic clang. 2021-01-13 Akim Demaille cex: send traces to stderr, not stdout When comparing traces from different machines, the mixture of stdout/stderr in the output are making things uselessly difficult. * src/lssi.c, src/state-item.c: Output debug traces on stderr. 2021-01-10 Akim Demaille glr2.cc: remove remains from glr.cc * data/skeletons/glr2.cc: We no longer play dirty tricks with parse-params, remove the now useless dance around them. We now have genuine objects for locations, leave the initial action alone. 2021-01-10 Akim Demaille glr2.cc: add support for api.token.constructor * data/skeletons/glr2.cc: Add support for api.token.constructor. * examples/c++/glr/c++-types.yy: Use it. * examples/c++/glr/c++-types.test: Adjust expectations for error messages. 2021-01-10 Akim Demaille glr2.cc: introduce the yytranslate_ member function * data/skeletons/c++.m4 (b4_yytranslate_define): Use static_cast rather than the YY_CAST macro. Avoids the need to define YY_CAST in the header. * data/skeletons/glr2.cc: Fix calls to b4_shared_declarations: pass the type of file we are in. Don't define YYTRANSLATE. (parser::yytranslate_): New, as in lalr1.cc. Adjust to use it. * tests/glr-regression.at: Adjust. 2021-01-10 Akim Demaille glr2.cc: tests: simplify * tests/glr-regression.at: Remove useless qualification for tokens. 2021-01-10 Akim Demaille glr2.cc: rely on symbol kinds rather than token kinds Instead of tracking the lookahead with yychar, use yytoken. This is consistent with lalr1.cc, saves us from calls to YYTRANSLATE (except when calling yylex), and makes it easier to migrate to using symbol_type. * data/skeletons/glr2.cc: Replace all uses of `int yychar` with `symbol_kind_type yytoken`. (yygetToken): Don't take/return the lookahead's token-kind and symbol-kind, just work directly on yystack's `yytoken` member. * tests/glr-regression.at (AT_PRINT_LOOKAHEAD_DECLARE) (AT_PRINT_LOOKAHEAD_DEFINE): New. Adjust to the fact that we have yytoken, not yychar, in glr2.cc. 2021-01-10 Akim Demaille glr2.cc: example: style changes * examples/c++/glr/c++-types.yy: We need C++ 11 (we use shared_ptr). 2021-01-10 Akim Demaille glr2.cc: use references to print symbols * data/skeletons/glr2.cc (b4_symbol_action): New, so that we use `yyval` and `yyloc` rather than the `yyvaluep` and `yylocationp` pointers. (yy_symbol_value_print_, yy_symbol_print_, yy_destroy_): Use references rather than pointers. Drop support for the undocumented, obsolete, `yyoutput` variable. Adjust callers. (glr_state::destroy): Don't use yy_symbol_print_ when we don't have a symbol. Rather, write dedicated code. 2021-01-10 Akim Demaille glr2.cc: fix memory leak * data/skeletons/glr2.cc (glr_stack_item::setState): Free the previous state before installing the new one. 2021-01-10 Akim Demaille glr: examples: fix locations The locations are actually false: they should include the location of the semicolon (we print statements), but they don't. * examples/c++/glr/c++-types.test, examples/c++/glr/c++-types.yy, * examples/c/glr/c++-types.test, examples/c/glr/c++-types.y, * tests/cxx-type.at: Fix locations and adjust expectations. 2021-01-10 Akim Demaille glr: tests: more macros This will be useful to support changes in glr2.cc. tests/glr-regression.at (Incorrect lookahead during deterministic GLR) (Incorrect lookahead during nondeterministc GLR): Introduce and use PRINT_LOOKAHEAD. 2021-01-10 Akim Demaille glr: tests: formatting changes * tests/glr-regression.at: here. Use %empty where applicable. 2021-01-10 Akim Demaille tests: fix definition for parser class name * tests/local.at (AT_PARSER_CLASS): Fix it. And use it. 2021-01-10 Akim Demaille CI: disable the installation of dmd Currently all the builds fail: $ mkdir -p ~/dlang && wget https://dlang.org/install.sh -O ~/dlang/install.sh --2021-01-10 06:05:39-- https://dlang.org/install.sh Resolving dlang.org (dlang.org)... 162.217.114.56, 2607:fc50:1:ff02::5:0 Connecting to dlang.org (dlang.org)|162.217.114.56|:443... connected. ERROR: cannot verify dlang.org's certificate, issued by ‘CN=R3,O=Let's Encrypt,C=US’: Unable to locally verify the issuer's authority. To connect to dlang.org insecurely, use `--no-check-certificate'. * .travis.yml: Disable dmd. 2021-01-07 Akim Demaille tests: remove some redundant tests * tests/calc.at: Cut the CI some slack. 2021-01-07 Adela Vais d: create getter for the number of errors from the parser * data/skeletons/lalr1.d: Here. 2021-01-07 Adela Vais d: remove support for parse.error verbose Without the history, D should not support this option. Before the removal, 'detailed' and 'verbose' options generated the same code. * data/skeletons/lalr1.d: Here. * doc/bison.texi: Adapt tests to use 'detailed' instead of 'verbose'. * tests/calc.at: Document it. 2021-01-07 Akim Demaille d: add support for %printer Currently we display the addresses of the semantic values. Instead, print the values. Add support for YY_USE across languages. * data/skeletons/c.m4, data/skeletons/d.m4 (b4_use): New. * data/skeletons/bison.m4 (b4_symbol_actions): Use b4_use to be portable to D. Add support for %printer, and use it. * data/skeletons/d.m4: Instead of duplicating what's already in c-like.m4, include it. (b4_symbol_action): New. Differs from the one in bison.m4 in that it uses yyval/yyloc instead of *yyvaluep and *yylocationp. * data/skeletons/lalr1.d (yy_symbol_print): Avoid calls to formatting, just call write directly. Use the %printer. * examples/d/calc/calc.y: Specify a printer. Enable traces when $YYDEBUG is set. * tests/calc.at: Fix the use of %printer with D. 2021-01-06 Adela Vais d: remove unnecessary imports * data/skeletons/lalr1.d: Here. 2021-01-06 Adela Vais d: remove yytnamerr usage It is a backwards-compatible feature for the other parsers. D should not support this option. * data/skeletons/d.m4: Here. 2021-01-06 Adela Vais d: add internationalisation support The D parser implements this feature similarly to the C parser, by using Gettext. Functions gettext() and dgettext() are imported using extern(C). The internationalisation uses yysymbol_name to report the name of the SymbolKinds. * data/skeletons/d.m4 (SymbolKind.toString.yytranslatable, SymbolKind.toString.yysymbol_name: New), data/skeletons/lalr1.d: Here. * doc/bison.texi: Document it. * tests/calc.at: Test it. 2021-01-05 Akim Demaille glr2.cc: add support for variants (Bison) Variants are extremely picky, which makes them both annoying (lots of micro-details must be taken care of) and precious (all the micro-details must be taken care of, in particular object lifetime). So (i) each time a semantic value is stored, it must be stored in a place that exists, and (ii) each time a semantic value is discarded, its place must have been emptied. Example of (i) - new (&yys.value ()) value_type (s->value ()); + {]b4_variant_if([[ + new (&yys.value ()) value_type (); + ]b4_symbol_variant([yy_accessing_symbol (s->yylrState)], + [yys.value ()], [copy], [s->value ()])], [[ + new (&yys.value ()) value_type (s->value ());]])[ + } Example of (ii) yyparser.yy_destroy_ ("Error: discarding", - yytoken, &yylval]b4_locations_if([, &yylloc])[); + yytoken, &yylval]b4_locations_if([, &yylloc])[);]b4_variant_if([[ + // Value type destructor. + ]b4_symbol_variant([[YYTRANSLATE (this->yychar)]], [[yylval]], [[template destroy]])])[ this->yychar = ]b4_namespace_ref[::]b4_parser_class[::token::]b4_symbol(empty, id)[; However, in some places we must not be "pure". In particular: glr_stack_item (const glr_stack_item& other) YY_NOEXCEPT YY_NOTHROW : is_state_ (other.is_state_) { std::memcpy (raw_, other.raw_, union_size); } still must use memcpy, because the constructor would change pred, and it must not. This constructor is used only when resizing the stack, in which case pred (which is relative) must not be "adjusted". The result works, but is messy. Its verbosity comes from at least two factors: - we don't have support for complete symbols (binding kind, value and location), and we should at least try to have it. That simplified lalr1.cc a lot. - I have not tried to be smart and use 'move' when possible. As a consequence many places have 'copy' and then 'destroy'. That kind of clean up can be done once everything appears to be solid. * data/skeletons/glr2.cc: Be more rigorous in object lifetime. In particular, don't forget to discard the lookahead when we're done with it. Call variant routines where needed. Deal with plenty of details. (b4_call_merger): Add support for variants. Use references in mergers, rather than pointers. * examples/c++/glr/c++-types.yy: Exercise variants. 2021-01-05 Akim Demaille glr: strengthen the tests On some experimentation I was running, the test suite was passing, yet the example crashed when run in verbose mode. Let's add this case to the test suite. * tests/cxx-type.at: Run all these tests in verbose mode too. 2021-01-03 Akim Demaille c++: I'm tired of Flex's warnings * doc/bison.texi: Disable another warning I'm tired to see. New releases would be most welcome. 2021-01-03 Akim Demaille glr: example: flush the output * examples/c/glr/c++-types.y: Flush stdout so that the logs (on stderr) and the effective output (on stdout) mix correctly. While at it, be a bit more const-correct. 2021-01-03 Akim Demaille style: YYUSE is private, make it YY_USE This macro is not exposed to users, make start it with 'YY_'. * data/skeletons/bison.m4, data/skeletons/c.m4, data/skeletons/glr.c, * data/skeletons/glr.cc, data/skeletons/glr2.cc, data/skeletons/lalr1.cc, * src/parse-gram.c, tests/actions.at, tests/c++.at, tests/headers.at, * tests/local.at (YYUSE): Rename as... (YY_USE): this. 2021-01-03 Akim Demaille glr.c: example: use a printer * examples/c++/glr/c++-types.yy: Here. 2021-01-03 Akim Demaille glr: consistently use the same wording in traces * data/skeletons/glr.c, data/skeletons/glr2.cc (yyglrReduce): Traces refer to "state 42", not to "state #42". 2021-01-02 Akim Demaille glr2.cc: also equip semantic_option with self check * data/skeletons/glr2.cc (semantic_option): Add MAGIC_, magic_ and check_ members. Use it. 2021-01-02 Akim Demaille glr2.cc: log the execution of deferred actions See "glr.c: log the execution of deferred actions". * data/skeletons/glr2.cc (yyuserAction): Take yyk as a new argument. Rename argument yyn as yyrule for clarity. Log before and after the user action. Adjust callers to not call YY_REDUCE_PRINT and YY_SYMBOL_PRINT. 2021-01-02 Akim Demaille glr2.cc: minor clean up * data/skeletons/glr2.cc (YYUNDEFTOK): Now useless. Formatting/coding style changes. 2021-01-02 Akim Demaille glr.c: log the execution of deferred actions Currently deferred reductions are not "verbose" at all: only immediate reductions are displayed in the YYDEBUG traces. I don't understand why. Besides it seems actually simpler the install the reduction traces right around the user action inside yyuserAction rather that around calls to yyuserAction. This only trouble is that yyuserAction does not know the stack number it works on, so we have to pass it. And pass -1 when we are actually running on a temporary stack. The glr example, on "T(x) + y;" as input, adds these logs, which allow to see when the `` is built: Stack 0 Entering state 26 Reduced stack 0 by rule 7 (line 108); action deferred. Now in state 7. Stack 0 Entering state 7 Reading a token Next token is token '+' (1.6: ) Stack 1 Entering state 27 Reduced stack 1 by rule 13 (line 123); action deferred. Now in state 12. Stack 1 Entering state 12 Next token is token '+' (1.6: ) Stack 1 dies. Removing dead stacks. On stack 0, shifting token '+' (1.6: ) Stack 0 now in state #14 +Reducing stack -1 by rule 6 (line 107): + $1 = token identifier (1.3: x) +-> $$ = nterm expr (1.3: x) +Reducing stack -1 by rule 7 (line 108): + $1 = token typename (1.0: T) + $2 = token '(' (1.2: ) + $3 = nterm expr (1.3: x) + $4 = token ')' (1.4: ) +-> $$ = nterm expr (1.0-3: (x,T)) Returning to deterministic operation. * data/skeletons/glr.c (yyuserAction): Take yyk as a new argument. Rename argument yyn as yyrule for clarity. Log before and after the user action. Adjust callers to not call YY_REDUCE_PRINT and YY_SYMBOL_PRINT. 2021-01-02 Akim Demaille glr.c: reorder routines The next commit wants to use YY_REDUCE_PRINT above its current definition. Move it higher. * data/skeletons/glr.c (yylhsNonterm, YY_REDUCE_PRINT): Make available earlier. 2021-01-02 Akim Demaille glr.c: example: use the exact same display as in the C++ example * examples/c/glr/c++-types.y: Add a space after the commas. * examples/c/glr/c++-types.test: Adjust expectations. 2021-01-02 Akim Demaille glr.c: example: several improvements * examples/c/glr/c++-types.y (node_print): New. Use YY_LOCATION_PRINT instead of duplicating it. And actually use it in the action instead of badly duplicating it. (main): Add proper option support. * examples/c/glr/c++-types.test: Adjust expectations on locations. * examples/c++/glr/c++-types.yy: Fix bad iteration. 2020-12-31 Akim Demaille glr2.cc: the example requires Bison 3.8 This will save us from generating the position.hh file. * src/parse-gram.y: Claim we are 3.8. * examples/c++/glr/c++-types.yy: Require 3.8. 2020-12-31 Akim Demaille %merge: associate it to its first definition, not the latest Currently each time we meet %merge we record this location as the defining location (and symbol). Instead, record the first definition. In the generated code we go from yy0->A = merge (*yy0, *yy1); to yy0->S = merge (*yy0, *yy1); where S was indeed the first symbol, and in the diagnostics we go from glr-regr18.y:30.18-24: error: result type clash on merge function 'merge': != 30 | sym2: sym3 %merge { $$ = $1; } ; | ^~~~~~~ glr-regr18.y:29.18-24: note: previous declaration 29 | sym1: sym2 %merge { $$ = $1; } ; | ^~~~~~~ glr-regr18.y:31.13-19: error: result type clash on merge function 'merge': != 31 | sym3: %merge { $$ = 0; } ; | ^~~~~~~ glr-regr18.y:30.18-24: note: previous declaration 30 | sym2: sym3 %merge { $$ = $1; } ; | ^~~~~~~ to glr-regr18.y:30.18-24: error: result type clash on merge function 'merge': != 30 | sym2: sym3 %merge { $$ = $1; } ; | ^~~~~~~ glr-regr18.y:29.18-24: note: previous declaration 29 | sym1: sym2 %merge { $$ = $1; } ; | ^~~~~~~ glr-regr18.y:31.13-19: error: result type clash on merge function 'merge': != 31 | sym3: %merge { $$ = 0; } ; | ^~~~~~~ glr-regr18.y:29.18-24: note: previous declaration 29 | sym1: sym2 %merge { $$ = $1; } ; | ^~~~~~~ where both duplicates are reported against definition 1, rather than using definition 1 as a reference when diagnosing about definition 2, and then 2 as a reference for 3. * src/reader.c (record_merge_function_type): Keep the first definition. * tests/glr-regression.at: Adjust. 2020-12-31 Akim Demaille %merge: test support for api.value.type=union * tests/glr-regression.at: here. 2020-12-31 Akim Demaille %merge: fix compatibility with api.value.type=union Reported by Jot Dot. https://lists.gnu.org/r/help-bison/2020-12/msg00014.html * data/skeletons/glr.c, data/skeletons/glr2.cc (b4_call_merger): Use the symbol's slot, not its type. * examples/c/glr/c++-types.y: Use explicit per-symbol typing together with api.value.type=union. (yylex): Use yytoken_kind_t. 2020-12-31 Akim Demaille %merge: delegate the generation of calls to mergers to m4 Don't generate C code from bison, leave that to the skeletons. * src/output.c (merger_output): Emit invocations to b4_call_merger. * data/skeletons/glr.c, data/skeletons/glr2.cc (b4_call_merger): New. 2020-12-31 Akim Demaille %merge: let mergers record a typing-symbol, rather than a type Symbols are richer than types, and in M4 it is my simpler (and more common) to deal with symbols rather than types. So let's associate mergers to a symbol rather than a type name. * src/reader.h (merger_list): Replace the 'type' member by a symbol member. * src/reader.c (record_merge_function_type): Take a symbol as argument, rather than a type name. * src/output.c (merger_output): Adjust. 2020-12-31 Akim Demaille %merge: clearer tests on diagnostics * tests/glr-regression.at: Use caret errors. 2020-12-28 Akim Demaille glr2.cc: style: quoting changes * data/skeletons/glr2.cc: Use stricter quoting rules. 2020-12-28 Akim Demaille lalr1.cc: style: quoting changes * data/skeletons/lalr1.cc: here. 2020-12-27 Akim Demaille glr2.cc: use references to the stack rather than pointers Now that the lookahead macros (that used yystackp) are out of the way, there is no reason to continue using a pointer. * data/skeletons/glr2.cc: Use yystack, a reference, rather that yystackp, a pointer. Fix tons of const-correctness issues. 2020-12-27 Akim Demaille glr2.cc: simplify names Now that we no longer play dangerous games with macros, we can give the lookahead's token kind its proper name. The content of yychar _is_ raw (as opposed to yytoken), there's no reason to pleonasmicate it (and thus to neologize). * data/skeletons/glr2.cc (glr_stack::yyrawchar): Rename as... (glr_stack::yychar): this. 2020-12-27 Akim Demaille glr2.cc: get rid of the macros wrapping the lookahead In glr.c, the macros yychar, yylval and yylloc allow to deal with api.pure: sometimes they point to global variables (impure), sometimes they point to the member variables (pure). There's no room for globals in glr2.cc. Besides, they map yychar to yyrawchar, yylval to yyval, etc. which obfuscates what is actually going on. * data/skeletons/glr2.cc (glr_stack::yyval, glr_stack::yyloc): Rename as... (glr_stack::yylval, glr_stack::yylloc): these, for clarity. (yynerrs, yychar, yylval, yylloc, yystackp): Remove these macros. (b4_yygetToken_call): Remove. 2020-12-26 Akim Demaille glr2.cc: reorganize the skeleton Restore a more natural order: first define the macros and then use them. Currently, some macros were defined between the moment the header is issued, and then the implementation file. As a result, it was possible for the header and the implementation to not use the same versions of the macros. * data/skeletons/glr2.cc: Define the macros first, then use them. * data/skeletons/lalr1.cc: Minor comment and quoting changes. 2020-12-26 Akim Demaille glr2.cc: example: simplify * examples/c++/glr/c++-types.yy: Formatting changes. Remove unused support for '@'. * examples/c/glr/c++-types.y: Ditto. 2020-12-26 Akim Demaille glr2.cc: make yyreportTree a member function of semantic_option * data/skeletons/glr2.cc (yy_accessing_symbol, yylhsNonterm): Define ealier. (state_stack::yyreportTree): Move to... (semantic_option::yyreportTree): here. Adjust dependencies. 2020-12-26 Akim Demaille glr2.cc: pass references to yyreportAmbiguity * data/skeletons/glr2.cc (yyreportAmbiguity): Use references. 2020-12-26 Akim Demaille style: use yyval only, not yysval * data/skeletons/glr.c, data/skeletons/glr2.cc: Use yyval, as in the other skeletons. 2020-12-26 Akim Demaille glr2.cc: pass location by const ref to yyglrShift * data/skeletons/glr2.cc (glr_state.yyglrShift): Take the location by const&. Remove useless `inline`. 2020-12-26 Akim Demaille style: rename semanticVal as value * data/skeletons/README-D.txt: Remove, now useless and obsolete. * data/skeletons/glr2.cc, examples/d/calc/calc.y, * tests/calc.at, tests/d.at, tests/scanner.at (semanticVal): Replace with... (value): this. 2020-12-26 Akim Demaille glr2.cc: remove dead comments * data/skeletons/glr2.cc: We no longer wrap glr.c here. 2020-12-26 Akim Demaille glr2.cc: use YYCDEBUG, not YY_DEBUG_STREAM * data/skeletons/glr2.cc (YY_DEBUG_STREAM): Rename as... (YYCDEBUG): this, as in lalr1.cc. 2020-12-26 Akim Demaille glr2.cc: formatting changes * data/skeletons/glr2.cc: here. Remove useless `inline`. 2020-12-26 Akim Demaille glr2.cc: don't use YYSTYPE/YYLTYPE at all * data/skeletons/glr2.cc: Define value_type and location_type where needed, and use them only. (yyuserMerge): Make it a member function of the glr_state class. 2020-12-26 Akim Demaille tests: don't require YYSTYPE/YYLTYPE to be defined in C++ * tests/glr-regression.at: Use AT_YYSTYPE/AT_YYLTYPE to generate yy::parser::value_type and yy::parser::location_type in C++. 2020-12-26 Akim Demaille c++: rename semantic_type as value_type We always refer to the triplet "kind, value, location". All of them are nouns, and we support api.value.type and api.location.type. On this regard, "semantic_type" was a poor choice. Make it "value_type". The test suite was not updated to use value_type, on purpose, to enforce backward compatibility. * data/skeletons/c++.m4, data/skeletons/glr.cc, data/skeletons/glr2.cc, * data/skeletons/variant.hh, doc/bison.texi: Define value_type rather than semantic_type. Add a backward compatibility typedef. * examples/c++/glr/c++-types.yy: Migrate. 2020-12-26 Akim Demaille doc: more about sanitizers * README-hacking.md: here. 2020-12-26 Akim Demaille glr2.cc: simplify * data/skeletons/glr2.cc (glr_state_set::yyremoveDeletes): Use vector::resize rather than vector::erase. (glr_state::copyFrom): Merge into... (glr_state::operator=): here. Valentin wanted each assignment to be explicit, hence copyFrom rather that operator=. But in 0a82316e54fead8c3f1cf41d2213c868ab88f508 (glr2.cc: example: use objects (not pointers) to represent the AST), in order to get real objects to be processed correctly, we had to introduce the assignment operator. Afterward, we also introduced a full implementation of the copy-ctor, independent of copyFrom. As a result, today the only invocation of copyFrom is from the assignment operator. Simplify this. 2020-12-26 Akim Demaille glr2.cc: fix warnings about uninitialized locations With GCC10, the CI shows tons of warnings such as (327. actions.at:374: testing Initial location: glr2.cc): input.cc: In member function 'YYRESULTTAG glr_stack::yyglrReduce(state_set_index, rule_num, bool)': input.cc:1357:11: error: '.glr_state::yyloc' may be used uninitialized in this function [-Werror=maybe-uninitialized] 1357 | yyloc = other.yyloc; | ~~~~~~^~~~~~~~~~~~~ This is because we don't have the constructors for locations. But we should have them! That's only because of glr.cc that ctors were not enabled by default. In glr2.cc, they should. That fixes all the warnings when Bison's locations are used. However, when user-defined locations without constructor are used, we still have: 550. calc.at:1409: testing Calculator glr2.cc %locations api.location.type={Span} ... calc.cc: In member function 'YYRESULTTAG glr_stack::yyglrReduce(state_set_index, rule_num, bool)': calc.cc:1261:11: error: '.glr_state::yyloc' may be used uninitialized in this function [-Werror=maybe-uninitialized] 1261 | yyloc = other.yyloc; | ~~~~~~^~~~~~~~~~~~~ To address this case, we need glr_state to explicily initialize its yyloc member. * data/skeletons/glr2.cc: Use genuine objects, with ctors, for position and location. (glr_state): Explicitly initialize yyloc in the constructors. 2020-12-25 Akim Demaille glr2.cc: provide glr_state with a genuine copy-constructor The copy constructor was (lazily) implemented by a call to copyFrom. Unfortunately copyFrom reads yyresolved from the destination (and source), and in the case of the copy-ctor this is random garbagge, which UBSAN catches: glr-regr2a.cc:1072:10: runtime error: load of value 7, which is not a valid value for type 'bool' Rather than defining yyresolved before calling copyFrom, let's just provide a genuine cpy-ctor for glr_state. * data/skeletons/glr2.cc (glr_state::glr_state): Implement properly. 2020-12-25 Akim Demaille glr2.cc: beware of self-assignment In yycompressStack: while (yyr != YY_NULLPTR) { nextFreeItem->check_ (); yyr->check_(); nextFreeItem->setState(*yyr); glr_state& nextFreeState = nextFreeItem->getState(); yyr = yyr->pred(); nextFreeState.setPred(&(nextFreeItem - 1)->getState()); setFirstTop(&nextFreeState); ++nextFreeItem; } it is possible that nextFreeItem and yyr are actually the same state. In which case `nextFreeItem->setState(*yyr)` does really bad things. * data/skeletons/glr2.cc (glr_stack_item::setState): Beware of self-assignment. 2020-12-25 Akim Demaille glr: comment changes * data/skeletons/glr.c: A bit more doc. (yypstates): Rename yyst (only occurrence) to yys (commonly used for yyGLRState). * data/skeletons/glr2.cc: Ditto. Prefer '\n' to "\n". 2020-12-21 Adela Vais d: remove unnecessary methods from the Lexer interface The complete symbol approach in yylex removes the need for the methods semanticVal, startPos and endPos, which were used when the values were reported separately. * data/skeletons/lalr1.d: Here. * doc/bison.texi: Remove sections about the three methods. * examples/d/calc/calc.y, examples/d/simple/calc.y: Remove the unused methods. * tests/calc.at, tests/d.at, tests/scanner.at: Test it. 2020-12-21 Adela Vais d: use Location and Position aliases in the backend * data/skeletons/lalr1.d: Here. 2020-12-21 Adela Vais d: remove unnecessary comparison from YYParser.parse() * data/skeletons/lalr1.d: Here. 2020-12-21 Akim Demaille style: address syntax-check diagnostics * examples/c/glr/c++-types.y: Formatting changes. * po/POTFILES.in: Add missing files. * src/reader.c: Remove useless include. * tests/calc.at: Avoid magic values for exit. Obfuscate calls to error. 2020-12-21 Adela Vais d: create alias Position for YYPosition * data/skeletons/d.m4 (b4_public_types_declare): Here. * data/skeletons/lalr1.d: Adjust. * doc/bison.texi: Document it. * examples/d/calc/calc.y: Use it. * tests/calc.at: Test it. 2020-12-21 Adela Vais d: create alias Value for YYSemanticType * data/skeletons/d.m4: Here. * data/skeletons/lalr1.d, examples/d/calc/calc.y, examples/d/simple/calc.y: Adjust. * tests/calc.at, tests/d.at, tests/scanner.at: Test it. 2020-12-21 Adela Vais d: create alias Location for YYLocation * data/skeletons/d.m4: Here. * doc/bison.texi: Document it. * examples/d/calc/calc.y: Adjust. * tests/calc.at: Test it. 2020-12-21 Adela Vais d: remove yychar from YYParse.parse() The yychar variable was keeping the external form of the token (the TokenKind). As the D parser translates the token to its internal form (the SymbolKind) inside the struct Symbol, there is no need for yychar anymore. * data/examples/lalr1.d (yychar): Remove. Use only yytoken. 2020-12-21 Adela Vais d: reduce verbosity for returning the location from yylex() * examples/d/calc/calc.y (start, end): Replace by this... (location): new member variable in the Lexer class. Use it. * tests/calc.at: Use the defined location variable. 2020-12-21 Adela Vais d: m4 style consistency fix * data/skeletons/lalr1.d: Here. 2020-12-21 Adela Vais d: create alias Symbol for YYParse.Symbol * data/skeletons/lalr1.d: Here. * doc/bison.texi: Document it. * examples/d/calc/calc.y, examples/d/simple/calc.y: Adjust. * tests/calc.at, tests/d.at, tests/scanner.at: Test it. 2020-12-21 Adela Vais d: change name of YYParser.Symbol's semanticValue() to value() Member value was renamed to value_ to avoid the name clash. * data/skeletons/d.m4: Change member names. * data/skeletons/lalr1.d: Adjust. 2020-12-20 Akim Demaille glr2.cc: formatting changes * data/skeletons/glr2.cc (glr_state): here. (glr_state::MAGIC): Make it easier to recognize in decimal. 2020-12-20 Akim Demaille glr2.cc: call the destructor in a way that complies with clang examples/c++/glr/c++-types.cc:721:24: error: expected the class name after '~' to name a destructor yysval.YYSTYPE::~semantic_type (); ^ Using a local typedef, for some reaon, result in clang complaining about a useless local typedef. Since anyway we don't want to keep on using YYSTYPE and YYLTYPE, it is time to introduce proper typedefs to reach these guys. And to be slightly in advance of the other skeletons: use value_type, not semantic_type. This is much more consistent with our use of the (kind, value, location) triplet. * data/skeletons/glr2.cc (glr_state::value_type) (glr_state::location_type): New. (glr_state::~glr_state): Use value_type to name the dtor. 2020-12-20 Akim Demaille glr2.cc: example: use objects (not pointers) to represent the AST Currently we are using pointers. The whole point of glr2.cc (vs. glr.cc) is precisely to allow genuine C++ objects to be semantic values. Let's make that work. * data/skeletons/glr2.cc (glr_state::glr_state): Be sure to initialize yysval. (glr_state): Add copy-ctor, assignment and dtor. (glr_state::copyFrom): Be sure to initialize the destination if it was not. (glr_state::~glr_state): Destroy the semantic value. * examples/c++/glr/ast.hh: Rewrite so that we use genuine objects, rather than a traditional OOP hierarchy that requires to deal with pointers. With help from Bruno Belanyi . * examples/c++/glr/c++-types.yy: Remove memory management. Use true objects. (main): Don't reach yydebug directly. * examples/c++/glr/local.mk: We need C++11. 2020-12-20 Akim Demaille glr2.cc: more checks * data/skeletons/glr2.cc: Clarify use of magic numbers. (glr_stack_item::getState): Check the returned state. 2020-12-20 Akim Demaille glr2.cc: fix GLR stack expansion When expanding the GLR stack, none of the pointers were updated to reflect the new location of the displaced objects. This fixes 748: Incorrect lookahead during nondeterministic GLR: glr2.cc * data/skeletons/glr2.cc (yyexpandGLRStack): Update the split point and the stack tops. (reduceToOneStack): Factor a bit. 2020-12-20 Akim Demaille glr2.cc: factor the computation of the accessing symbol * data/skeletons/glr2.cc (yy_accessing_symbol): New, as in glr.c. Use it. 2020-12-20 Akim Demaille glr2.cc: fix calling conventions for yyexpandGLRStackIfNeeded This test fails: 748: Incorrect lookahead during nondeterministic GLR: glr2.cc It consumes lots of stack space, so at some point we need to expand it. Because of Boolean logic mistakes, we then claim memory-exhausted (first error). Hence we jump to cleaning the stack (popall_), calling all the destructors, and at some point we crash with heap-use-after-free (second error). This commit fixes the first error. Unfortunately, even though we now do expand the stack, we crash again with (another) heap-use-after-free, not addressed here. Eventually, we should make sure popall_() properly works. * data/skeletons/glr2.cc (yyexpandGLRStackIfNeeded): Return true iff success (i.e., memory not exhausted). 2020-12-20 Akim Demaille glr2.cc: example: use streams and accept arguments From a debugger, it is easier to pass a file name than working on stdin. * examples/c++/glr/c++-types.yy: Reduce scopes. Avoid YYSTYPE/YYLTYPE: use the C++ types. (input, process): New. (main): Use them. 2020-12-20 Akim Demaille glr: formatting changes * data/skeletons/glr.c: Formatting changes. * data/skeletons/glr2.cc: Ditto. (glr_state_set::INITIAL_NUMBER_STATES): Remove, unused (and useless: let std::vector deal with that). 2020-12-20 Akim Demaille skeletons: better comments for some tables And also, remove the incorrect indentation of these comments: - /* YYR2[YYN] -- Number of symbols on the right hand side of rule YYN. */ +/* YYR2[RULE-NUM] -- Number of symbols on the right-hand side of rule RULE-NUM. */ static const yytype_int8 yyr2[] = { 0, 2, 4, 0, 2, 1, 1, 1, 3, 2, I don't remember why this indentation was added (in 0991e29b75f192570b6fac1ba23657ab4f3509ee), but it seems wrong, at least for yacc.c. I suspect this was done with lalr1.cc (where this is embeded in the class definition, so it should be indented), but today lalr1.cc uses other routines to output these comments. * data/skeletons/bison.m4 (b4_integral_parser_tables_map): Improve the wording of the comments of some tables. * data/skeletons/c.m4 (b4_integral_parser_table_define): Remove indentation. 2020-12-19 Akim Demaille glr2.cc: fix glr_stack_item::setState A glr_stack_item has "raw" memory to store either a glr_state or a semantic_option. glr_stack_item::setState stores a state using a copy assignment. However, this is more like a construction: we are starting from "raw" memory, so use the placement new operator instead. While it probably makes no difference when parse.assert is disabled, it does make one when it is: the constructor properly initialize the magic number, the assignment does not. So without these changes, the next commit (which stores genuine objects in semantic values) fails tests 712 and 730 because of incorrect magic numbers. * data/skeletons/glr2.cc (glr_stack_item::setState): Build the state, don't just copy it. 2020-12-19 Akim Demaille glr2.cc: more self checks * data/skeletons/glr2.cc: here. 2020-12-19 Akim Demaille glr2.cc: style: s/Type/Kind/g * data/skeletons/glr2.cc (YY_SYMBOL_PRINT): Use Kind, not Type. As in the other skeletons. 2020-12-19 Akim Demaille glr2.cc: example: style: add missing copyright headers * examples/c++/glr/ast.hh, examples/c++/glr/c++-types.yy: here. * examples/c++/glr/local.mk: Fix distribution of ast.hh. 2020-12-19 Akim Demaille glr2.cc: example: address Clang warnings ast.hh:24:7: error: 'Node' has no out-of-line virtual method definitions; its vtable will be emitted in every translation unit [-Werror,-Wweak-vtables] class Node ^ ast.hh:57:7: error: 'Nterm' has no out-of-line virtual method definitions; its vtable will be emitted in every translation unit [-Werror,-Wweak-vtables] class Nterm : public Node ^ ast.hh:102:7: error: 'Term' has no out-of-line virtual method definitions; its vtable will be emitted in every translation unit [-Werror,-Wweak-vtables] class Term : public Node ^ * examples/c++/glr/ast.hh: Define the destructors out of the class definition. This does not change anything, it is still in the header, but that does pacify clang. 2020-12-16 Akim Demaille glr: tests: add support for the YYDEBUG envvar When debugging these parsers, we really need debug traces. Enable them, and bind them to $YYDEBUG. * tests/glr-regression.at: Support the YYDEBUG envvar. As a consequence, now that syntactic ambiguities are reported, adjust the expected output. (No users destructors if stack 0 deleted): Don't return 0 on memory exhaustion, really return the parser's status, and adust expectations. 2020-12-14 Akim Demaille glr.c: comment changes * data/skeletons/glr.c (yycompressStack): Reduce scope, and import some nice comments from glr2.cc. 2020-12-14 Akim Demaille glr2.cc: make the example more C++ Currently the example really looks like C. Instead of a union of structs to implement the AST, use a hierarchy. It would be nice to feature a C++17 version with std variants. * examples/c++/glr/c++-types.yy (Node, free_node, new_nterm) (new_term): Move into... * examples/c++/glr/ast.hh: here, a proper C++ hierarchy. 2020-12-14 Akim Demaille glr2.cc: make yyparse a member function Amusingly enough, glr2.cc still had its core function, yyparse, being a free function instead of a member function. * data/skeletons/glr2.cc (yyparse): Remove this free function called from yyparser::parse. Inline its body into... (yyparser::parse): this member function. This requires moving a bit the yychar, etc. macros. Access to token can be simplified (the b4_namespace_ref::b4_parser_class prefix is no longer needed). 2020-12-14 Akim Demaille glr2.cc: being pure is not an option Remove the useless conditional b4_pure_if: the skeleton is, of course, pure (no global variables). Make glr2.cc intrinsically pure. * data/skeletons/glr2.cc (b4_pure_if): Remove definition and uses. (b4_lex): New. Stolen from lalr1.cc to avoid needing to use the one from c.m4. 2020-12-14 Akim Demaille glr2.cc: fix yycompressStack Currently, yycompressStack expects the free items to be states only. That's not the case. Fixes 712 and 730 pass. 748 still fails, but later and differently (heap-use-after-free). * data/skeletons/glr2.cc (glr_stack_item::setState): New. (glr_stack_item::yycompressStack): Use it. * tests/glr-regression.at: Adjust. 2020-12-14 Akim Demaille glr2.cc: fix pointer arithmethics A glr_state keeps tracks of its predecessor using an offset relative to itself (i.e., pointer subtraction). Unfortunately we sometimes have to compute offsets for pointers that live in different containers, in particular in yyfillin. In that case there is no reason for the distance between the two objects to be a multiple of the object size (0x40 on my machine), and the resulting ptrdiff_t may be "wrong", i.e., it does allow to recover one from the other. We cannot use "typed" pointer arithmetics here, the Euclidean division has it wrong. So use "plain" char* pointers. Fixes 718 (Duplicate representation of merged trees: glr2.cc) and examples/c++/glr/c++-types. Still XFAIL: 712: Improper handling of embedded actions and dollar(-N) in GLR parsers: glr2.cc 730: Incorrectly initialized location for empty right-hand side in GLR: glr2.cc 748: Incorrect lookahead during nondeterministic GLR: glr2.cc * data/skeletons/glr2.cc (glr_state::as_pointer_): New. (glr_state::pred): Use it. * examples/c++/glr/c++-types.test: The test passes. * tests/glr-regression.at (Duplicate representation of merged trees: glr2.cc): Passes. 2020-12-14 Akim Demaille glr2.cc: style fixes * data/skeletons/glr2.cc: Formatting changes. 2020-12-13 Akim Demaille glr2.cc: add sanity check in glr_state The use of YY_IGNORE_NULL_DEREFERENCE_BEGIN/END in `check_` is to please GCC 10: glr-regr8.cc: In member function 'YYRESULTTAG glr_stack::yyresolveValue(glr_state&)': glr-regr8.cc:1433:21: error: potential null pointer dereference [-Werror=null-dereference] 1433 | YYASSERT (this->magic_ == MAGIC); | ~~~~~~^~~~~~ glr-regr8.cc:905:40: note: in definition of macro 'YYASSERT' 905 | # define YYASSERT(Condition) ((void) ((Condition) || (abort (), 0))) | ^~~~~~~~~ * data/skeletons/glr2.cc (glr_state::check_): New. Use it in the member functions. 2020-12-13 Akim Demaille glr2.cc: add sanity checks in glr_stack_item We obviously have broken pointer arithmetics that hands us glr_stack_items that are not glr_stack_items. Have a simple check for this, to have earlier failures. * data/skeletons/glr2.cc (glr_stack_item::check_): New. Use it. (glr_stack_item::contents): Avoid the useless struct. Fix minor stylistic issues. 2020-12-06 Akim Demaille glr2.cc: run the glr-regression tests When installed on master as of 2020-12-05 (on top of "glr2.cc: fix when the stack is not expandable", almost all the GLR regression tests fail (with a SEGV): 709: Badly Collapsed GLR States: glr2.cc FAILED (glr-regression.at:130) 712: Improper handling of embedded actions and dollar(-N) in GLR parsers: glr2.cc FAILED (glr-regression.at:275) 715: Improper merging of GLR delayed action sets: glr2.cc FAILED (glr-regression.at:404) 718: Duplicate representation of merged trees: glr2.cc FAILED (glr-regression.at:502) 721: User destructor for unresolved GLR semantic value: glr2.cc FAILED (glr-regression.at:566) 724: User destructor after an error during a split parse: glr2.cc FAILED (glr-regression.at:624) 727: Duplicated user destructor for lookahead: glr2.cc FAILED (glr-regression.at:724) 730: Incorrectly initialized location for empty right-hand side in GLR: glr2.cc FAILED (glr-regression.at:823) 733: No users destructors if stack 0 deleted: glr2.cc FAILED (glr-regression.at:911) 736: Corrupted semantic options if user action cuts parse: glr2.cc FAILED (glr-regression.at:974) 739: Undesirable destructors if user action cuts parse: glr2.cc FAILED (glr-regression.at:1042) 742: Leaked semantic values if user action cuts parse: glr2.cc FAILED (glr-regression.at:1173) 748: Incorrect lookahead during nondeterministic GLR: glr2.cc FAILED (glr-regression.at:1546) 751: Leaked semantic values when reporting ambiguity: glr2.cc FAILED (glr-regression.at:1639) 754: Leaked lookahead after nondeterministic parse syntax error: glr2.cc FAILED (glr-regression.at:1710) 757: Uninitialized location when reporting ambiguity: glr2.cc FAILED (glr-regression.at:1794) 766: Predicates: glr2.cc FAILED (glr-regression.at:2045) These pass: 745: Incorrect lookahead during deterministic GLR: glr2.cc ok 760: Missed %merge type warnings when LHS type is declared later: glr2.cc ok 763: Ambiguity reports: glr2.cc ok With Valentin Tolmer's "glr2.cc: Fix memory corruption bug" commit, these test fail "gracefully": 712: Improper handling of embedded actions and dollar(-N) in GLR parsers: glr2.cc FAILED (glr-regression.at:268) 730: Incorrectly initialized location for empty right-hand side in GLR: glr2.cc FAILED (glr-regression.at:816) 748: Incorrect lookahead during nondeterministic GLR: glr2.cc FAILED (glr-regression.at:1539) And these do not end: 709: Badly Collapsed GLR States: glr2.cc FAILED (glr-regression.at:123) 715: Improper merging of GLR delayed action sets: glr2.cc FAILED (glr-regression.at:397) 718: Duplicate representation of merged trees: glr2.cc FAILED (glr-regression.at:495) 751: Leaked semantic values when reporting ambiguity: glr2.cc FAILED (glr-regression.at:1632) With "tests: glr2.cc: run the glr-regression tests", none loop, and 709, 715, and 751 pass. Only 718 still fails. * tests/glr-regression.at: Run all the tests with glr2.cc. * tests/local.at (AT_GLR2_CC_IF): New. 2020-12-06 Akim Demaille glr2.cc: use the same format for traces as glr.c * data/skeletons/glr2.cc: here. This allows to share the same expected output. 2020-12-06 Akim Demaille glr2.cc: add support for parse.assert * data/skeletons/glr2.cc: Fake support of parse.assert, so that the tests can use it without failing. 2020-12-06 Akim Demaille glr2.cc: fix yyresolveValue When "tests: glr2.cc: run the glr-regression tests" tests are run, before this commit the following tests used to loop endlessly: 709: Badly Collapsed GLR States: glr2.cc FAILED (glr-regression.at:123) 715: Improper merging of GLR delayed action sets: glr2.cc FAILED (glr-regression.at:397) 718: Duplicate representation of merged trees: glr2.cc FAILED (glr-regression.at:495) 751: Leaked semantic values when reporting ambiguity: glr2.cc FAILED (glr-regression.at:1632) After this commit, no test loops and 709, 715, and 751 pass. Only 718 still fails. * data/skeletons/glr2.cc (yyresolveValue): Add missing incrementation of the iteration variable. 2020-12-06 Valentin Tolmer glr2.cc: misc cleanups * data/skeletons/glr2.cc: Use 'const' on variables and applicable member functions. Improve comments. Use references where applicable. Enforce names_like_this, notLikeThis. Reduce scopes. 2020-12-06 Valentin Tolmer glr2.cc: fix memory corruption bug * data/skeletons/glr2.cc (yyremoveDeletes): Remove double-increment in the loop. (glr_state::copyFrom): Handle gracefully when other is resolved. 2020-12-06 Akim Demaille glr2.cc: turn some pointers into references * data/skeletons/glr2.cc: Prefer references to pointers. Add a few more const. 2020-12-06 Valentin Tolmer glr2.cc: use 'const' for some constant local variables * data/skeletons/glr2.cc: here. 2020-12-06 Akim Demaille glr2.cc: fix when the stack is not expandable * data/skeletons/glr2.cc (yyexpandGLRStackIfNeeded): Fix the implementation when !YYSTACKEXPANDABLE. 2020-12-06 Akim Demaille glr.c: fix line numbers in logs * data/skeletons/glr.c (yyglrReduce): Fix line numbers. * tests/glr-regression.at: Fix expectations. 2020-12-05 Akim Demaille tests: glr: run the glr regression tests with glr.cc * tests/glr-regression.at: Adjust the tests to be more independent of the language, and run them with glr.cc. Some tests relied on yychar, yylval and yylloc being global variables: pass arguments instead. 2020-12-05 Akim Demaille tests: glr: prepare for more languages * tests/glr-regression.at: Wrap each test in an AT_TEST. Call it with glr.c. 2020-12-05 Akim Demaille tests: glr: use AT_FULL_COMPILE * tests/glr-regression.at: Instead of using AT_BISON_CHECK and AT_COMPILE, use AT_FULL_COMPILE. This is shorter, and makes it easier to add support for other programming languages. 2020-12-05 Akim Demaille tests: glr: prefer directives to warnings * tests/glr-regression.at: Use %expect and %expect-rr in the grammar files, rather than accepting diagnostics. This will make it easier to support other programming languages. 2020-12-05 Akim Demaille tests: factor the access to token kinds * tests/local.at (AT_BISON_OPTION_PUSHDEFS): Define AT_TOKEN. (AT_BISON_OPTION_POPDEFS): Undefine it. * tests/actions.at, tests/c++.at, tests/calc.at: Use AT_TOKEN. 2020-12-05 Akim Demaille tests: formatting changes * tests/local.at: here. 2020-12-05 Akim Demaille glr.cc: don't "leak" yyparse When using glr.cc, the C function yyparse is an internal detail that should not be exposed. Users might call it by accident (I did). * data/skeletons/glr.c (yyparse): When used for glr.cc, rename as yy_parse_impl. * data/skeletons/glr.cc: Adjust. 2020-12-03 Akim Demaille portability: beware of GCC 4.6 src/reader.c: In function 'grammar_start_symbols_add': src/reader.c:67:24: error: declaration of 'dup' shadows a global declaration [-Werror=shadow] * src/reader.c (grammar_start_symbols_add): Rename dup as dupl. 2020-12-03 Akim Demaille tables: avoid warnings and save bits The yydefgoto table uses -1 as an invalid for an impossible case (we never use yydefgoto[0], since it corresponds to the reduction to $accept, which never happens). Since yydefgoto is a table of state numbers, this -1 forces a signed type uselessly, which (1) might trigger compiler warnings when storing a value from yydefgoto into a state number (nonnegative), and (2) wastes bits which might result in using a int16 where a uint8 suffices. Reported by Jot Dot . https://lists.gnu.org/r/bug-bison/2020-11/msg00027.html * src/tables.c (default_goto): Use 0 rather than -1 as invalid value. * tests/regression.at: Adjust. 2020-12-03 Akim Demaille c++: use noexcept where appropriate Reported by Don Macpherson. https://github.com/akimd/bison/issues/63 https://github.com/akimd/bison/issues/64 * data/skeletons/c++.m4, data/skeletons/lalr1.cc: here. 2020-11-30 Akim Demaille gnulib: update 2020-11-30 Akim Demaille multistart: check duplicates * src/symlist.h, src/symlist.c (symbol_list_find_symbol) (symbol_list_last): New. (symbol_list_append): Use symbol_list_last. * src/reader.c (grammar_start_symbols_add): Check and discard duplicates. * tests/input.at (Duplicate %start symbol): New. * tests/reduce.at (Bad start symbols): Add the multistart keyword. 2020-11-22 Akim Demaille style: change the format of a debugging function * src/symlist.c (symbol_list_syms_print): Use braces to make traces easier to read. 2020-11-22 Akim Demaille style: rename grammar_start_symbols_set as grammar_start_symbols_add * src/reader.h, src/reader.c (grammar_start_symbols_set): Rename as... (grammar_start_symbols_add): this. Adjust dependencies. 2020-11-22 Akim Demaille TODO: update 2020-11-21 Akim Demaille multistart: also use the api.prefix in the implementation We were declaring foo_parse_expr but implementing yyparse_expr. * data/skeletons/yacc.c (_b4_define_sub_yyparse): Use api.prefix for the parse function name. 2020-11-20 Akim Demaille doc: using stow to set up a bison herd * README-hacking.md (Stow): New. 2020-11-20 Adela Vais d: change YYLocation's type from class to struct This avoids heap allocation and gives minimal costs for the creation and destruction of the YYParser.Symbol struct if the location tracking is active. Suggested by H. S. Teoh. * data/skeletons/lalr1.d: Here. * doc/bison.texi: Document it. * examples/d/calc/calc.y: Adjust. * tests/calc.at: Test it. 2020-11-20 Adela Vais d: change the return value of yylex from TokenKind to YYParser.Symbol The complete symbol approach was deemed to be the right approach for Dlang. Now, the user can return from yylex() an instance of YYParser.Symbol structure, which binds together the TokenKind, the semantic value and the location. Before, the last two were reported separately to the parser. Only the user API is changed, Bisons's internal structure is kept the same. * data/skeletons/d.m4 (struct YYParser.Symbol): New. * data/skeletons/lalr1.d: Change the return value. * doc/bison.texi: Document it. * examples/d/calc/calc.y, examples/d/simple/calc.y: Demonstrate it. * tests/calc.at, tests/scanner.at: Test it. 2020-11-20 Martin Rehak examples: avoid "unbound variable" errors When the shell option `nounset` is set, we may get "unbound variable" errors. https://lists.gnu.org/r/bug-bison/2020-11/msg00013.html * examples/test (diff_opts): Be sure to initialize it. 2020-11-20 Akim Demaille gnulib: update * Makefile.am (gitsort): New. Use it. 2020-11-18 Adela Vais d: add support for lookahead correction When using lookahead correction, the method YYParser.Context.getExpectedTokens is not annotated with const, because the method calls yylacCheck, which is not const. Also, because of yylacStack and yylacEstablished, yylacCheck needs to be called from the context of the parser class, which is sent as parameter to the Context's constructor. * data/skeletons/lalr1.d (yylacCheck, yylacEstablish, yylacDiscard, yylacStack, yylacEstablished): New. (Context): Use it. * doc/bison.texi: Document it. * tests/calc.at: Check it. 2020-11-18 Adela Vais d: change the name of the custom error message function to reportSyntaxError Changed from syntax_error to reportSyntaxError to be similar to the Java parser. * data/skeletons/lalr1.d: Change the function name. * doc/bison.texi: Document it. * tests/local.at: Adjust. 2020-11-14 Akim Demaille Merge branch 'maint' * upstream/maint: maint: post-release administrivia version 3.7.4 2020-11-14 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2020-11-14 Akim Demaille version 3.7.4 * NEWS: Record release date. 2020-11-13 Akim Demaille style: prefer b4_symbol(empty, ...) to b4_symbol(-2, ...) * data/skeletons/c.m4: here. 2020-11-13 Akim Demaille Merge branch 'maint' * maint: c++: shorten the assertions that check whether tokens are correct c++: don't glue functions together lalr1.cc: YY_ASSERT should use api.prefix c++: don't use YY_ASSERT at all if parse.assert is disabled c++: style: follow the Bison m4 quoting pattern yacc.c: provide the Bison version as an integral macro regen style: make conversion of version string to int public %require: accept version numbers with three parts ("3.7.4") yacc.c: fix #definition of YYEMPTY gnulib: update doc: fix incorrect section title doc: minor grammar fixes in counterexamples section 2020-11-13 Akim Demaille c++: shorten the assertions that check whether tokens are correct Before: YY_ASSERT (tok == token::YYEOF || tok == token::YYerror || tok == token::YYUNDEF || tok == 120 || tok == 49 || tok == 50 || tok == 51 || tok == 52 || tok == 53 || tok == 54 || tok == 55 || tok == 56 || tok == 57 || tok == 97 || tok == 98); After: YY_ASSERT (tok == token::YYEOF || (token::YYerror <= tok && tok <= token::YYUNDEF) || tok == 120 || (49 <= tok && tok <= 57) || (97 <= tok && tok <= 98)); Clauses are now also wrapped on several lines. This is nicer to read and diff, but also avoids pushing Visual C++ to its arbitrary limits (640K and lines of 16380 bytes ought to be enough for anybody, otherwise make an C2026 error). The useless parens are there for the dummy warnings about precedence (in the future, will we also have to put parens in `1+2*3`?). * data/skeletons/variant.hh (_b4_filter_tokens, b4_tok_in, b4_tok_in): New. (_b4_token_constructor_define): Use them. 2020-11-13 Akim Demaille c++: don't glue functions together * data/skeletons/bison.m4 (b4_type_foreach): Accept a separator. * data/skeletons/c++.m4: Use it. And fix an incorrect comment. 2020-11-13 Akim Demaille lalr1.cc: YY_ASSERT should use api.prefix Working on the previous commit I realized that YY_ASSERT was used in the generated headers, so must follow api.prefix to avoid clashes when multiple C++ parser with variants are used. Actually many more macros should obey api.prefix (YY_CPLUSPLUS, YY_COPY, etc.). There was no complaint so far, so it's not urgent enough for 3.7.4, but it should be addressed in 3.8. * data/skeletons/variant.hh (b4_assert): New. Use it. * tests/local.at (AT_YYLEX_RETURN): Fix. * tests/headers.at: Make sure variant-based C++ parsers are checked too. This test did find that YY_ASSERT escaped renaming (before the fix in this commit). 2020-11-13 Akim Demaille c++: don't use YY_ASSERT at all if parse.assert is disabled In some extreme situations (about 800 tokens), we generate a single-line assertion long enough for Visual C++ to discard the end of the line, thus falling into parse ends for the missing `);`. On a shorter example: YY_ASSERT (tok == token::TOK_YYEOF || tok == token::TOK_YYerror || tok == token::TOK_YYUNDEF || tok == token::TOK_ASSIGN || tok == token::TOK_MINUS || tok == token::TOK_PLUS || tok == token::TOK_STAR || tok == token::TOK_SLASH || tok == token::TOK_LPAREN || tok == token::TOK_RPAREN); Whether NDEBUG is used or not is irrelevant, the parser dies anyway. Reported by Jot Dot . https://lists.gnu.org/r/bug-bison/2020-11/msg00002.html We should avoid emitting lines so long. We probably should also use a range-based assertion (with extraneous parens to pacify fascist compilers): YY_ASSERT ((token::TOK_YYEOF <= tok && tok <= token::TOK_YYUNDEF) || (token::TOK_ASSIGN <= tok && ...) But anyway, we should simply not emit this assertion at all when not asked for. * data/skeletons/variant.hh: Do not define, nor use, YY_ASSERT when it is not enabled. 2020-11-13 Akim Demaille c++: style: follow the Bison m4 quoting pattern * data/skeletons/variant.hh: here. 2020-11-11 Akim Demaille todo: more 2020-11-11 Akim Demaille yacc.c: provide the Bison version as an integral macro Suggested by Balazs Scheidler. https://github.com/akimd/bison/issues/55 * src/muscle-tab.c (muscle_init): Move/rename `b4_version` to/as... * src/output.c (prepare): `b4_version_string`. Also define `b4_version`. * data/skeletons/bison.m4, data/skeletons/c.m4, data/skeletons/d.m4, * data/skeletons/java.m4: Adjust. * doc/bison.texi: Document it. 2020-11-11 Akim Demaille regen 2020-11-11 Akim Demaille style: make conversion of version string to int public * src/parse-gram.y (str_to_version): Rename as/move to... * src/strversion.h, src/strversion.c (strversion_to_int): these new files. 2020-11-11 Akim Demaille %require: accept version numbers with three parts ("3.7.4") * src/parse-gram.y (str_to_version): Support three parts. * data/skeletons/location.cc, data/skeletons/stack.hh: Adjust. 2020-11-11 Todd C. Miller yacc.c: fix #definition of YYEMPTY When generating a C parser, YYEMPTY is present in enum yytokentype but there is no corresponding #define like there is for the other values. There is a special case for YYEMPTY in b4_token_enums but no corresponding case in b4_token_defines. * data/skeletons/c.m4 (b4_token_defines): Do define YYEMPTY. 2020-11-10 Akim Demaille gnulib: update 2020-11-10 Akim Demaille style: enforce java coding style * tests/scanner.at: here. 2020-11-10 Akim Demaille ielr: fix incorrect function call * src/ielr.c: s/rule_is_accepting/rule_is_initial/. 2020-11-10 Akim Demaille ielr: more comments and logs * src/ielr.c: More comments. (state_list_print): New. 2020-11-10 Akim Demaille multistart: fix IELR computations IELR needs to rule out the successors of the kernel items of the initial state (`$accept: input • $end`). In the case of multistart, this condition must be expressed differently: the mere item index does not suffice. * src/ielr.c (ielr_item_has_lookahead, ielr_compute_lookaheads): Don't rely on the item index to check whether is_successor_of_initial_item. It is certainly more costly than just checking the item index, but (i) we need to compute the rule anyway, so it's not very much more costly, and (ii) in ielr_item_has_lookahead, this situation is actually impossible, so an optimizing compiler reading the assertions should actually avoid this computation. 2020-11-10 Akim Demaille ielr: make some conditions about items easier to understand Checking that an item index is > 1 means ruling out `$accept: • input $end` and `$accept: input • $end`. But actually only the latter is possible there, i.e., we're checking whether this item is about a successor of a (kernel) item of the initial state ($accept: input • $end). * src/ielr.c (is_successor_of_initial_item): Use a variable to name this condition. 2020-11-10 Akim Demaille multistart: introduce and use rule_is_initial * src/gram.h (rule_is_initial): New. * src/graphviz.c, src/print-xml.c, src/print.c, src/lalr.c: Use it. Some of these occurrences were incorrect (checking whether this is rule 0), and not behaving properly in the case of multistart. 2020-11-08 Akim Demaille style: comment and formatting changes, and fixes * examples/c/lexcalc/parse.y: Fix option handling. * src/gram.h: Clarify comments. * src/ielr.c: Fix indentation. * src/print.c, src/state.h: More comments. 2020-11-08 Akim Demaille lalr: add assertions * src/lalr.c: Remove incorrect comment (subsumed anyway by the (correct) one in the header. (set_goto_map): More debug traces. (map_goto): Add an assertion. 2020-11-07 Akim Demaille glr2.cc: remove scaffolding for glr.c * data/skeletons/glr2.cc (b4_glr_cc_setup, b4_glr_cc_cleanup): Remove. 2020-11-07 Akim Demaille d: remove dead comment * data/skeletons/lalr1.d (reportSyntaxError): here. 2020-11-07 Adela Vais d: add the custom error message feature Parser.Context class returns a const YYLocation, so Lexer's method yyerror() needs to receive the location as a const parameter. Internal error reporting flow is changed to be similar to that of the other skeletons. Before, case YYERRLAB was calling yyerror() with the result of yysyntax_error() as the string parameter. As the custom error message lets the user decide if they want to use yyerror() or not, this flow needed to be changed. Now, case YYERRLAB calls yyreportSyntaxError(), that builds the error message using yysyntaxErrorArguments(). Then yyreportSyntaxError() passes the error message to the user defined syntax_error() in case of a custom message, or to yyerror() otherwise. In the tests in tests/calc.at, the order of the tokens needs to be changed in order of precedence, so that the D program outputs the expected tokens in the same order as the other parsers. * data/skeletons/lalr1.d: Add the custom error message feature. * doc/bison.texi: Document it. * examples/d/calc/calc.y: Adjust. * tests/calc.at, tests/local.at: Test it. 2020-11-07 Adela Vais d: examples: fix coding style * examples/d/calc/calc.y, examples/d/simple/calc.y: Fix whitespace issues and remove the @property attribute. 2020-11-07 Akim Demaille style: avoid explicit symbol numbers This should have been part of commit "symbols: stop dealing with YYEMPTY as b4_symbol(-2, ...)" (cd40ec9526df27d0e3e1c2d41e82a3e6a47254d1). Give names to all the special symbols: "eof", "error" and "undef". * data/skeletons/bison.m4 (b4_symbol): Let `b4_symbol(eof, ...)` mean `b4_symbol(0, ...)`, `b4_symbol(error, ...)` mean `b4_symbol(1, ...)`, and , `b4_symbol(undef, ...)` mean `b4_symbol(2, ...)`.. * data/skeletons/c.m4, data/skeletons/glr.c, data/skeletons/glr.cc, * data/skeletons/glr2.cc, data/skeletons/lalr1.cc, * data/skeletons/lalr1.d, data/skeletons/lalr1.java, * data/skeletons/yacc.c: Prefer symbols to numbers. 2020-11-07 Adela Vais d: fix Context class All methods are now declared as const. Method getExpectedTokens() has now the functionality of reporting only the number of expected tokens. * data/skeletons/lalr1.d: Fix Context class. 2020-11-07 Akim Demaille doc: fix typo Reported by Kaz Kylheku. https://lists.gnu.org/r/bison-patches/2020-11/msg00019.html * doc/bison.texi (Versioning): here. 2020-11-07 Adela Vais d: doc: add D Action Features section * doc/bison.texi (D Action Features section): New. 2020-11-07 Adela Vais d: add yyerrok In D's case, yyerrok() is a private method of the Parser class. It can be called directly as `yyerrok()` from the grammar rules section. * data/skeletons/lalr1.d: Add yyerrok(). * examples/d/calc/calc.y, examples/d/simple/calc.y: Demonstrate yyerrok(). * tests/calc.at: Update D tests to use yyerrok(). 2020-11-06 Akim Demaille java: lac: a stronger test for the exploratory stack * tests/local.at (AT_YYLEX_DEFINE(java)): Fix overquoting issue. Style changes. * tests/regression.at (LAC: Exploratory stack): Run for lalr1.java too. 2020-11-05 Kaz Kylheku doc: document best deployment practices. * doc/bison.texi (Versioning): New node about practices regarding dealing with multiple versions of Bison. 2020-11-04 Akim Demaille java: lac: more tests, and some doc * doc/bison.texi: C++ and Java support LAC. * tests/input.at (LAC: Errors for %define): Generalize the test, and apply it to Java. 2020-11-03 Akim Demaille java: avoid Integer(String), use parseInt examples/java/calc/Calc.java:1531: warning: [deprecation] Integer(String) in Integer has been deprecated yylval = new Integer(st.sval); ^ * examples/java/calc/Calc.y, examples/java/simple/Calc.y, * tests/calc.at, tests/scanner.at: Use Integer.parseInt. 2020-11-03 Akim Demaille java: prefer ArrayList to Vector Vector is synchronized, which is completely useless in our case (and not even relevant when concurrency matters). No seasoned Java programmer would use it. Reported by Uxio Prego. * data/skeletons/lalr1.java: Replace Vector with ArrayList. Unfortunately its API is not as rich, and lacks lastElement and setSize. 2020-11-03 Akim Demaille java: lac: check it * tests/calc.at: Add tests for LAC in pull and push parsers. Skip LAC: line from the logs. * tests/local.at (reportSyntaxError): Output the error message in a single call, to avoid having the error message on stderr be interrupted by the debug traces of LAC in getExpectedTokens. 2020-11-03 Akim Demaille java: add support for lookahead correction Shamelessly stolen from Adrian Vogelsgesang's implementation in lalr1.cc. * data/skeletons/lalr1.java (yycdebugNnl, yyLacCheck, yyLacEstablish) (yyLacDiscard, yylacStack, yylacEstablished): New. (Context): Use it. * examples/java/calc/Calc.y: Enable LAC. 2020-11-03 Akim Demaille style: java: more style fixes * data/skeletons/lalr1.java, tests/java.at: Avoid space before paren. And use braces as is customary in the Java. 2020-11-03 Akim Demaille style: comment changes in the skeletons * data/skeletons/lalr1.cc: Prepare for Java's LAC. * data/skeletons/lalr1.java: Style fixes. 2020-11-01 Akim Demaille doc: fix incorrect section title Reported by Gaurav Singh . https://lists.gnu.org/r/bug-bison/2020-11/msg00000.html * doc/bison.texi (Rpcalc Expr): Rename as... (Rpcalc Exp): this, as the nterm is named 'exp'. 2020-11-01 Akim Demaille doc: fix incorrect section title Reported by Gaurav Singh . https://lists.gnu.org/r/bug-bison/2020-11/msg00000.html * doc/bison.texi (Rpcalc Expr): Rename as... (Rpcalc Exp): this, as the nterm is named 'exp'. 2020-10-28 Nick Gasson doc: minor grammar fixes in counterexamples section * doc/bison.texi: Minor fixes in counterexamples section. 2020-10-28 Nick Gasson doc: minor grammar fixes in counterexamples section * doc/bison.texi: Minor fixes in counterexamples section. 2020-10-27 Adela Vais d: create the Parser.Context class This will provide the user an interface for creating custom error messages. * data/skeletons/lalr1.d: Add the Context class. * doc/bison.texi: Document it. 2020-10-14 Akim Demaille Merge branch 'maint' * upstream/maint: doc: fix typo maint: post-release administrivia version 3.7.3 build: don't link bison against libreadline gnulib: update glr.cc: fix: use symbol_name build: fix a concurrent build issue in examples 2020-10-14 Akim Demaille doc: fix typo * README: here. 2020-10-13 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2020-10-13 Akim Demaille version 3.7.3 * NEWS: Record release date. 2020-10-13 Akim Demaille build: don't link bison against libreadline Reported by Paul Smith . https://lists.gnu.org/r/bug-bison/2020-10/msg00001.html * src/local.mk (src_bison_LDADD): here. 2020-10-13 Akim Demaille gnulib: update 2020-10-07 Akim Demaille doc: D comes before Java in alphabetical order * doc/bison.texi: Keep languages sorted. 2020-10-07 Akim Demaille report: put the dot after %empty in items When printing items, it is clearer to put the dot after %emtpy rather than before: 0 $accept: . unit "end of file" 1 unit: . assignments exp - 2 assignments: . %empty + 2 assignments: %empty . 3 | . assignments assignment Also, use the Unicode characters if they are supported. * src/gram.c (item_print): Put the dot after %emtpy. * tests/conflicts.at, tests/reduce.at, tests/report.at: Adjust. 2020-10-03 Adela Vais d: change api.token.raw default value to true Generate consecutive values for enum TokenKind, as D's yylex() returns TokenKind and collisions can't happen. * data/skeletons/d.m4: Change default value. * tests/scanner.at, tests/d.at: Check it. 2020-10-03 Akim Demaille m4: have b4_percent_define_if_define apply default values lazily Currently `b4_percent_define_ifdef([foo])` assigns a default value to `foo` when invoked. As a consequence, skeletons such as lalr1.d cannot specify their specific default values: `foo` was defined in bison.m4. Instead, provide `foo` with a default value when `b4_foo_if` is invoked. I could not measure a runtime difference between both cases. * data/skeletons/bison.m4 (_b4_percent_define_define): New. Helps getting rid of spurious indentation that resulted in spurious white space in the output. (b4_percent_define_if_define): Move the definition to... (_b4_percent_define_if_define): when the defined macros is called. 2020-10-02 Adela Vais d: don't trigger GC in void toString * data/skeletons/d.m4 (b4_declare_symbol_enum): Here. 2020-10-02 Adela Vais d: document support * doc/bison.texi: Various fixes. (D Parsers): New section. 2020-10-02 Adela Vais d: remove the default prefix for SymbolKind's enum elements This is not needed in D, since the kinds are "scoped". * data/skeletons/d.m4: Remove the default prefix. * doc/bison.text: Document it. 2020-10-02 Adela Vais d: remove the @property attribute from interface Lexer D best practices is to not use it. * data/skeletons/lalr1.d: Remove attribute. 2020-09-29 Akim Demaille symbols: stop dealing with YYEMPTY as b4_symbol(-2, ...) * data/skeletons/bison.m4 (b4_symbol): Redirect `b4_symbol(empty, ...)` to `b4_symbol(-2, ...)`. Change all uses of the latter to the former. 2020-09-29 Akim Demaille tests: remove useless prefix for EOF in D * tests/calc.at (CALC_EOF): Rename as... (EOF): this. Since there is no risk of a clash with #define EOF here... 2020-09-28 Adela Vais d: support api.symbol.prefix and api.token.prefix The D skeleton was not properly supporting them. * data/skeletons/d.m4, data/skeletons/lalr1.d: Fix it. * tests/calc.at: Check it. 2020-09-28 Akim Demaille multistart: start more thorough testing * tests/local.at (AT_MULTISTART_IF): New. * tests/calc.at: Adjust to check multiple start symbols. * data/skeletons/yacc.c (yy_parse_impl): Fix. 2020-09-27 Akim Demaille tests: style: reorder the calculator test macros * tests/local.at (AT_TOKEN_TRANSLATE_IF): New, moved from... * tests/calc.at: here. Instead of sorting per feature (main, yylex, calc.y) and then by language, do the converse, so that C bits are together, etc. 2020-09-27 Akim Demaille tests: shorten the generated file * tests/synclines.at (_AT_SYNCLINES_COMPILE): Pull the comments out. 2020-09-27 Akim Demaille gnulib: update 2020-09-27 Akim Demaille multistart: also give access to yynerrs This is something that has always bothered me: with pure parsers (and they all should be) the user does not have an (easy) access to yynerrs at the end of the parse. In the case of error recovery, that's the only direct means to know if there were errors. The usual approach being having the user maintain a counter incremented each time yyerror is called. So here, also capture yynerrs in the return value of the start-symbol parsing functions. * data/skeletons/yacc.c (yy_parse_impl_t): New. (yy_parse_impl): Use it. (b4_accept): Fill it. * examples/c/lexcalc/parse.y, examples/c/lexcalc/scan.l: No longer pass nerrs as lex- and parse-param, just use the resulting yynerrs. bistromathic and reccalc both demonstrate %param. 2020-09-27 Akim Demaille yacc.c: also count calls to YYERROR in yynerrs * data/skeletons/yacc.c: here. 2020-09-27 Akim Demaille multistart: allow tokens as start symbols After all, why not? * src/reader.c (switching_token): Use symbol_id_get. (check_start_symbols): Require that the start symbol is a token only if it's the only one. * examples/c/lexcalc/parse.y: Let NUM be a start symbol. 2020-09-27 Akim Demaille multistart: use b4_accept instead of action post-processing For each start symbol, generate a parsing function with a richer return value than the usual of yyparse. Reserve a place for the returned semantic value, in order to avoid having to pass a pointer as argument to "return" that value. This also makes the call to the parsing function independent of whether a given start-symbol is typed. For instance, if the grammar file contains: %type expression %start input expression (so "input" is valueless) we get typedef struct { int yystatus; } yyparse_input_t; yyparse_input_t yyparse_input (void); typedef struct { int yyvalue; int yystatus; } yyparse_expression_t; yyparse_expression_t yyparse_expression (void); This commit also changes the implementation of the parser termination: when there are multiple start symbols, it is the initial rules that explicitly YYACCEPT. They do that after having exported the start-symbol's value (if it is typed): switch (yyn) { case 1: /* $accept: YY_EXPRESSION expression $end */ { ((*yyvalue).TOK_expression) = (yyvsp[-1].TOK_expression); YYACCEPT; } break; case 2: /* $accept: YY_INPUT input $end */ { YYACCEPT; } break; I have tried several ways to deal with termination, and this is the one that appears the best one to me. It is also the most natural. * src/scan-code.h, src/scan-code.l (obstack_for_actions): New. * src/reader.c (grammar_rule_check_and_complete): Generate the actions of the rules for each start symbol. * data/skeletons/bison.m4 (b4_symbol_slot): New, with safer semantics than type and type_tag. * data/skeletons/yacc.c (b4_accept): New. Generates the body of the action of the start rules. (_b4_declare_sub_yyparse): For each start symbol define a dedicated return type for its parsing function. Adjust the declaration of its parsing function. (_b4_define_sub_yyparse): Adjust the definition of the function. * examples/c/lexcalc/parse.y: Check the case of valueless symbols. * examples/c/lexcalc/lexcalc.test: Check start symbols. 2020-09-27 Akim Demaille multistart: adjust reader checks for generated rules So far we were not checking the generated rule 0 at all. Now there can be several of them. Instead of not checking at all, let's be more selective on the check to run on them. * src/reader.c (grammar_rule_check_and_complete): Don't check for value usage for generated rules, it is ok to have a valued start symbol, in which case it is ok for the generated rule ("accept: start $end {}") to not use $1. (packgram): Call grammar_rule_check_and_complete for all the rules. 2020-09-27 Akim Demaille todo: more 2020-09-27 Akim Demaille multistart: toy with it in lexcalc * examples/c/lexcalc/parse.y: Define several start symbols. * examples/c/lexcalc/lexcalc.test: Check support. 2020-09-27 Akim Demaille multistart: equip yacc.c * data/skeletons/yacc.c: Add support for multiple start symbols. 2020-09-27 Akim Demaille multistart: pass the list of start symbols to the backend * src/output.c (start_symbols_output): New. (muscles_output): Use it. 2020-09-27 Akim Demaille multistart: also check the HTML report We don't actually have checks for HTML, so let's do it for multistart. * tests/report.at: here. 2020-09-27 Akim Demaille multistart: adjust computation of initial core and adjust reports Currently the core of the initial state is limited to the single rule on $accept. * src/lr0.c (generate_states): There may now be several rules on $accept. * src/graphviz.c (conclude_red): Recognize "final" transitions by the fact that we reduce to "$accept". * src/print.c (print_reduction): Likewise. * src/print-xml.c (print_reduction): Likewise. 2020-09-27 Akim Demaille regen 2020-09-27 Akim Demaille multistart: turn start symbols into rules on $accept Now that the parser can read several start symbols, let's process them, and create the corresponding rules. * src/parse-gram.y (grammar_declaration): Accept a list of start symbols. * src/reader.h, src/reader.c (grammar_start_symbol_set): Rename as... (grammar_start_symbols_set): this. * src/reader.h, src/reader.c (start_flag): Replace with... (start_symbols): this. * src/reader.c (grammar_start_symbols_set): Build a list of start symbols. (switching_token, create_start_rules): New. (check_and_convert_grammar): Use them to turn the list of start symbols into a set of rules. * src/reduce.c (nonterminals_reduce): Don't complain about $accept, it's an internal detail. (reduce_grammar): Complain about all the start symbols that don't derive sentences. * src/symtab.c (startsymbol, startsymbol_loc): Remove, replaced by start_symbols. symbols_pack): Move the check about the start symbols to... * src/symlist.c (check_start_symbols): here. Adjust to multiple start symbols. * tests/reduce.at (Empty Language): Generalize into... (Bad start symbols): this. 2020-09-27 Akim Demaille regen 2020-09-27 Akim Demaille parser: expose a list of symbols * src/parse-gram.y (%type): Also use current_class. (symbol_decl.1): Rename as... (symbols.1): this. 2020-09-27 Akim Demaille reader: get ready to create several initial rules * src/reader.c (create_start_rule): New. Use it. 2020-09-27 Akim Demaille gram: more debugging information * src/gram.c (ritem_print): Show indices in ritem. 2020-09-27 Akim Demaille glr.cc: fix: use symbol_name * data/skeletons/glr.cc: here. 2020-09-26 Akim Demaille glr2.cc: add an example Currently this example crashes on input such as "T (x) + y;". The same example with glr.c works properly. * examples/c++/glr/Makefile, examples/c++/glr/README.md, * examples/c++/glr/c++-types.test, examples/c++/glr/c++-types.yy, * examples/c++/glr/local.mk, examples/c++/local.mk: New. Based on examples/c/glr/c++-types.y. 2020-09-26 Akim Demaille glr.cc: fix: use symbol_name * data/skeletons/glr.cc: here. 2020-09-26 Adela Vais d: change the return value of yylex from int to TokenKind * data/skeletons/lalr1.d: Change the return value. * examples/d/calc/calc.y, examples/d/simple/calc.y: Adjust. * tests/scanner.at: Adjust. * tests/calc.at (_AT_DATA_CALC_Y(d)): New, extracted from... (_AT_DATA_CALC_Y(c)): here. The two grammars have been sufficiently different to be separated. Still trying to be them together results in a maintenance burden. For the same reason, instead of specifying the results for D and for the rest, compute the expected results with D from the regular case. 2020-09-24 Adela Vais d: support api.parser.extends and api.parser.implements The D skeleton was not properly supporting them. * data/skeletons/d.m4: Fix it. * tests/d.at: Check it. * tests/local.mk, tests/testsuite.at: Adjust. 2020-09-21 Akim Demaille glr2.cc: also run all the calculator tests This revealed issues with yy_symbol_print and yy_reduce_print. These changes, in turn, reactivated GCC10 warnings: 559. calc.at:1258: testing Calculator glr2.cc %locations %header parse.error=verbose %debug api.prefix={calc} %verbose %parse-param {semantic_value *result}{int *count}{int *nerrs} ... tests/calc.at:1258: COLUMNS=1000; export COLUMNS; NO_TERM_HYPERLINKS=1; export NO_TERM_HYPERLINKS; bison --color=no -fno-caret -Wno-deprecated -o calc.cc calc.y tests/calc.at:1258: $CXX $CXXFLAGS $CPPFLAGS $LDFLAGS -o calc calc.cc calc-lex.cc calc-main.cc $LIBS stderr: calc.cc: In function 'void glr_stack::yyresolveLocations(glr_state*, int)': calc.cc:2623:46: error: potential null pointer dereference [-Werror=null-dereference] 2623 | yyrhsloc[0].getState().yyloc = yyoption->state()->yyloc; | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~ calc.cc:2623:46: error: potential null pointer dereference [-Werror=null-dereference] 2623 | yyrhsloc[0].getState().yyloc = yyoption->state()->yyloc; | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~ calc.cc:2623:46: error: potential null pointer dereference [-Werror=null-dereference] 2623 | yyrhsloc[0].getState().yyloc = yyoption->state()->yyloc; | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~ calc.cc:1177:10: error: potential null pointer dereference [-Werror=null-dereference] 1177 | return yypred ? &(asItem (this) - yypred)->getState () : YY_NULLPTR; | ^~~~~~ calc.cc:2623:46: error: potential null pointer dereference [-Werror=null-dereference] 2623 | yyrhsloc[0].getState().yyloc = yyoption->state()->yyloc; | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~ calc.cc:2623:46: error: potential null pointer dereference [-Werror=null-dereference] 2623 | yyrhsloc[0].getState().yyloc = yyoption->state()->yyloc; | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~ calc.cc: In member function 'YYRESULTTAG glr_stack::yyresolveValue(glr_state*)': calc.cc:2623:46: error: potential null pointer dereference [-Werror=null-dereference] 2623 | yyrhsloc[0].getState().yyloc = yyoption->state()->yyloc; | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~ * tests/calc.at (AT_CHECK_CALC_GLR_CC): Also check glr2.cc. * data/skeletons/glr2.cc: Don't pass the user arguments to yy_symbol_print and yy_reduce_print, since they have it in the parser object. (b4_user_formals_no_comma, b4_pure_args, b4_lpure_args) (b4_locuser_formals, b4_locuser_args): Remove, useless. (YY_IGNORE_NULL_DEREFERENCE_BEGIN): Enable for GCC >= 10 too. 2020-09-20 Akim Demaille tests: check the location of the right-hand side symbols The D skeleton was not properly supporting @1 etc. Reported by Adela Vais. https://lists.gnu.org/r/bison-patches/2020-09/msg00049.html * data/skeletons/d.m4 (b4_rhs_location): Fix it. * tests/calc.at: Check the support of @n for all the skeletons. 2020-09-20 Akim Demaille style: formatting changes * src/scan-code.l: here. 2020-09-20 Akim Demaille style: introduce parse_positional_ref * src/scan-code.l: here. 2020-09-20 Akim Demaille style: clarify the way state kernels (aka cores) are built Use state_list_append in a more natural way. * src/lr0.c (generate_states): Here. 2020-09-20 Akim Demaille style: reorder and comment * src/reader.h: here. 2020-09-19 Akim Demaille examples: add a demonstration of GLR parsers in C Based on the test case 668 (cxx-type.at:437) "GLR: Merge conflicting parses, pure, locations". * examples/c/glr/Makefile, examples/c/glr/README.md, * examples/c/glr/c++-types.test, examples/c/glr/c++-types.y, * examples/c/glr/local.mk: New. 2020-09-19 Akim Demaille glr: support api.header.include * data/skeletons/glr.c: here. 2020-09-19 Akim Demaille add support for --html * bootstrap.conf: We need the "execute" module. * src/files.h, src/files.c (spec_html_file, html_flag): New. * src/getargs.h, src/getargs.c (--html): New. * src/print-xml.h, src/print-xml.c (print_html): New. * src/main.c: Use them. * tests/output.at, tests/report.at: Check --html. 2020-09-19 Akim Demaille regen 2020-09-19 Akim Demaille deprecate %defines in favor of %header This is consistent with --defines being deprecated in favor of --header. The directive %defines is also too similar to %define. And %header matches nicely with api.header.name. * src/scan-gram.l (%defines): Deprecate to %header. (%header): Scan it. * src/parse-gram.y (PERCENT_DEFINES): Replace with... (PERCENT_HEADER): this. * data/skeletons/lalr1.java * doc/bison.texi * tests/actions.at, tests/c++.at, tests/calc.at, tests/conflicts.at, * tests/input.at, tests/java.at, tests/local.at, tests/output.at, * tests/synclines.at, tests/types.at: Convert most tests to check %header instead of %defines. 2020-09-19 Akim Demaille options: rename --defines as --header The name "defines" is incorrect, the generated file contains far more than just #defines. * src/getargs.h, src/getargs.c (-H, --header): New option. With optional argument, just like --defines, --xml, etc. (defines_flag): Rename as... (header_flag): this. Adjust dependencies. * data/skeletons/bison.m4, data/skeletons/c.m4, data/skeletons/glr.c, * data/skeletons/glr.cc, data/skeletons/glr2.cc, data/skeletons/lalr1.cc, * data/skeletons/yacc.c: Adjust. * examples, doc/bison.texi: Adjust. * tests/headers.at, tests/local.at, tests/output.at: Convert most tests from using --defines to using --header. 2020-09-17 Akim Demaille CI: beware of time limits * .travis.yml (GCC 8): Run only the part 1 of the tests. 2020-09-17 Valentin Tolmer glr2.cc: replace refs to parser::symbol_kind_type with yysymbol_kind_t * data/skeletons/glr2.cc: here. 2020-09-17 Valentin Tolmer glr2.cc: fix container out-of-bounds access Clang 10 with ASAN enabled reported errors in glr2.cc. * data/skeletons/glr2.cc: here. 2020-09-15 Akim Demaille glr2.cc: disable GCC 4.6 warning 231. conflicts.at:1096: testing Syntax error in consistent error state: glr2.cc ... tests/conflicts.at:1096: $CXX $CXXFLAGS $CPPFLAGS $LDFLAGS -o input input.cc $LIBS input.cc: In member function 'YYRESULTTAG glr_stack::yyresolveValue(glr_state*)': input.cc:2674:36: error: 'yysval' may be used uninitialized in this function [-Werror=uninitialized] Do not initialize the variable: this way ASAN can really make sure we do set it to a proper value. If we initialize it, ASAN would report nothing. * data/skeletons/c.m4 (YY_IGNORE_MAYBE_UNINITIALIZED_BEGIN): Disable GCC 4.6's -Wuninitialized. * data/skeletons/glr2.cc: Disable the warning locally. 2020-09-14 Akim Demaille glr2.cc: fix warning with GCC 4.7 and 4.8 231. conflicts.at:1096: testing Syntax error in consistent error state: glr2.cc ... tests/conflicts.at:1096: $CXX $CXXFLAGS $CPPFLAGS $LDFLAGS -o input input.cc $LIBS input.cc: In function 'int yyparse(yy::parser&)': input.cc:3147:41: error: 'yyarg' may be used uninitialized in this function [-Werror=maybe-uninitialized] return yytnamerr_ (yytname_[yysymbol]); ^ input.cc:2058:34: note: 'yyarg' was declared here yy::parser::symbol_kind_type yyarg[YYERROR_VERBOSE_ARGS_MAXIMUM]; ^ * data/skeletons/glr2.cc (yyreportSyntaxError): Initialize yyarg. 2020-09-14 Akim Demaille glr2.cc: simplify symbol kinds We emit code like if (yytoken != yy::parser::symbol_kind::symbol_kind::S_YYEMPTY) * data/skeletons/glr2.cc: Use b4_symbol correctly. 2020-09-14 Akim Demaille glr2.cc: fix warning about local variable vs. member Fix a warning triggered in GCC (at least from 4.6 to 4.9): input.cc: In constructor 'glr_stack_item::glr_stack_item(bool)': input.cc:1371:5: error: declaration of 'is_state' shadows a member of 'this' [-Werror=shadow] : is_state_(is_state) ^ * data/skeletons/glr2.cc (glr_stack_item): Alpha-convert. 2020-09-13 Akim Demaille c++: variants: minor simplification Do as Valentin Tolmer did in glr2.cc. * data/skeletons/variant.hh: The union does not need to be named. 2020-09-13 Akim Demaille glr2.cc: avoid type-punning issues On the CI, tests fail with GCC 4.6 to GCC 6 as follows: tests/synclines.at:440: COLUMNS=1000; export COLUMNS; NO_TERM_HYPERLINKS=1; export NO_TERM_HYPERLINKS; bison --color=no -fno-caret -o \"\\\"\".cc \"\\\"\".y tests/synclines.at:440: $CXX $CXXFLAGS $CPPFLAGS $LDFLAGS -o \"\\\"\" \"\\\"\".cc $LIBS stderr: "\"".cc: In member function 'glr_state& glr_stack_item::getState()': "\"".cc:1404:47: error: dereferencing type-punned pointer will break strict-aliasing rules [-Werror=strict-aliasing] return *reinterpret_cast(&raw_); ^ "\"".cc: In member function 'const glr_state& glr_stack_item::getState() const': "\"".cc:1408:53: error: dereferencing type-punned pointer will break strict-aliasing rules [-Werror=strict-aliasing] return *reinterpret_cast(&raw_); ^ "\"".cc: In member function 'semantic_option& glr_stack_item::getOption()': "\"".cc:1413:53: error: dereferencing type-punned pointer will break strict-aliasing rules [-Werror=strict-aliasing] return *reinterpret_cast(&raw_); ^ "\"".cc: In member function 'const semantic_option& glr_stack_item::getOption() const': "\"".cc:1417:59: error: dereferencing type-punned pointer will break strict-aliasing rules [-Werror=strict-aliasing] return *reinterpret_cast(&raw_); ^ See also be6fa942acae21a4a025bac5e339451be6ad136d. * data/skeletons/glr2.cc (glr_stack_item): Use a temporary void* variable to avoid type-punning issues with reinterpret_cast. 2020-09-13 Akim Demaille glr.cc, glr2.cc: don't publish compiler pragmas Currently the compiler attributes are defined in b4_shared_declarations (that can in the header if it exists, otherwise in the implementation file). This is not needed, only the implementation file needs them. Besides, glr2.cc was also defining these macros in the implementation file, so we had two definitions. * data/skeletons/glr.cc, data/skeletons/glr2.cc: Define the compiler attribute macros only in the implementation files. * tests/regression.at (Lex and parse params): Generate a header, to make it easy to check that the header is self-sufficient. 2020-09-13 Akim Demaille glr2.cc: disable incorrect warnings from GCC6 to 9 For instance with GCC8: 616. regression.at:1560: testing Lex and parse params: glr2.cc ... tests/regression.at:1560: COLUMNS=1000; export COLUMNS; NO_TERM_HYPERLINKS=1; export NO_TERM_HYPERLINKS; bison --color=no -fno-caret -o input.cc input.y tests/regression.at:1560: $CXX $CXXFLAGS $CPPFLAGS $LDFLAGS -o input input.cc $LIBS stderr: input.cc: In member function 'YYRESULTTAG glr_stack::yyresolveValue(glr_state*)': input.cc:1796:10: error: potential null pointer dereference [-Werror=null-dereference] return yypred ? &(asItem(this) - yypred)->getState() : YY_NULLPTR; ^~~~~~ input.cc:1796:10: error: potential null pointer dereference [-Werror=null-dereference] return yypred ? &(asItem(this) - yypred)->getState() : YY_NULLPTR; ^~~~~~ cc1plus: all warnings being treated as errors It complains that the implicit this in yypred might be null. It fears it because of loops such as for (glr_state* yys = firstTopState(); yys != yystateStack.yysplitPoint; yys = yys->pred()) yyn += 1; that could possibly set yys to null, since yys->pred might return null. However, the warning is incorrect, since in C++ `this` cannot be null. GCC 10 no longer emits this warning. GCC 7 also complains many times about glr_stack::yyresolveLocations when NDEBUG is enabled (when it is not, YYASSERT (yyoption != YY_NULLPTR) is probably enough to pacify GCC): 616. regression.at:1560: testing Lex and parse params: glr2.cc ... tests/regression.at:1560: COLUMNS=1000; export COLUMNS; NO_TERM_HYPERLINKS=1; export NO_TERM_HYPERLINKS; bison --color=no -fno-caret -o input.cc input.y tests/regression.at:1560: $CXX $CXXFLAGS $CPPFLAGS $LDFLAGS -o input input.cc $LIBS stderr: input.cc: In member function 'void glr_stack::yyresolveLocations(glr_state*, int)': input.cc:3061:46: error: potential null pointer dereference [-Werror=null-dereference] yyrhsloc[0].getState().yyloc = yyoption->state()->yyloc; ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~ input.cc:3061:46: error: potential null pointer dereference [-Werror=null-dereference] yyrhsloc[0].getState().yyloc = yyoption->state()->yyloc; ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~ input.cc:3061:46: error: potential null pointer dereference [-Werror=null-dereference] yyrhsloc[0].getState().yyloc = yyoption->state()->yyloc; ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~ input.cc: In member function 'YYRESULTTAG glr_stack::yyresolveValue(glr_state*)': input.cc:3061:46: error: potential null pointer dereference [-Werror=null-dereference] yyrhsloc[0].getState().yyloc = yyoption->state()->yyloc; ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~ input.cc:3061:46: error: potential null pointer dereference [-Werror=null-dereference] yyrhsloc[0].getState().yyloc = yyoption->state()->yyloc; ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~ input.cc:3061:46: error: potential null pointer dereference [-Werror=null-dereference] yyrhsloc[0].getState().yyloc = yyoption->state()->yyloc; ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~ input.cc:3061:46: error: potential null pointer dereference [-Werror=null-dereference] yyrhsloc[0].getState().yyloc = yyoption->state()->yyloc; ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~ input.cc:3061:46: error: potential null pointer dereference [-Werror=null-dereference] yyrhsloc[0].getState().yyloc = yyoption->state()->yyloc; ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~ input.cc:3061:46: error: potential null pointer dereference [-Werror=null-dereference] yyrhsloc[0].getState().yyloc = yyoption->state()->yyloc; ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~ * data/skeletons/glr2.cc (YY_IGNORE_NULL_DEREFERENCE_BEGIN) (YY_IGNORE_NULL_DEREFERENCE_BEGIN): New. (glr_state::pred, glr_stack::yyresolveLocations): Use them. 2020-09-12 Adela Vais d: make enum SymbolKind idiomatic D Taking into account comments from H. S. Teoh. https://lists.gnu.org/r/bison-patches/2020-09/msg00021.html * data/skeletons/d.m4, data/skeletons/lalr1.d (SymbolKind): Wrap the enum in a structure that contains its string representation. 2020-09-12 Akim Demaille glr2.cc: address warnings with G++ 4.8 input.cc: In constructor 'glr_stack_item::glr_stack_item(bool)': input.cc:1423:5: error: declaration of 'isState' shadows a member of 'this' [-Werror=shadow] : isState_(isState) { ^ test.cc:1165:45: error: declaration of 'begin' shadows a member of 'this' [-Werror=shadow] test.cc:1167:45: error: declaration of 'end' shadows a member of 'this' [-Werror=shadow] * data/skeletons/glr2.cc (isState): Rename as... (is_state): this. Formatting changes. (reduceToOneStack): Rename variables to avoid name clashes. 2020-09-12 Akim Demaille glr2.cc: get rid of the C indirection for yy_symbol_print * data/skeletons/glr2.cc (yy_symbol_print): Remove. Just use yyparser.yy_symbol_print_ directly. 2020-09-12 Akim Demaille glr2.cc: formatting changes * data/skeletons/glr2.cc: here. 2020-09-12 Akim Demaille glr2.cc: fix GCC10 warning For instance on test 433: "glr2.cc api.value.type={double}" tests/types.at:138: $CXX $CXXFLAGS $CPPFLAGS $LDFLAGS -o test test.cc $LIBS stderr: In file included from /opt/local/include/gcc10/c++/bits/stl_tempbuf.h:60, from /opt/local/include/gcc10/c++/bits/stl_algo.h:62, from /opt/local/include/gcc10/c++/algorithm:62, from test.cc:82: /opt/local/include/gcc10/c++/bits/stl_construct.h: In instantiation of 'constexpr decltype (::new(void*(0)) _Tp) std::construct_at(_Tp*, _Args&& ...) [with _Tp = glr_stack_item; _Args = {glr_stack_item}; decltype (::new(void*(0)) _Tp) = glr_stack_item*]': /opt/local/include/gcc10/c++/bits/alloc_traits.h:514:21: required from 'static constexpr void std::allocator_traits >::construct(std::allocator_traits >::allocator_type&, _Up*, _Args&& ...) [with _Up = glr_stack_item; _Args = {glr_stack_item}; _Tp = glr_stack_item; std::allocator_traits >::allocator_type = std::allocator]' /opt/local/include/gcc10/c++/bits/vector.tcc:115:30: required from 'std::vector<_Tp, _Alloc>::reference std::vector<_Tp, _Alloc>::emplace_back(_Args&& ...) [with _Args = {glr_stack_item}; _Tp = glr_stack_item; _Alloc = std::allocator; std::vector<_Tp, _Alloc>::reference = glr_stack_item&]' /opt/local/include/gcc10/c++/bits/stl_vector.h:1204:21: required from 'void std::vector<_Tp, _Alloc>::push_back(std::vector<_Tp, _Alloc>::value_type&&) [with _Tp = glr_stack_item; _Alloc = std::allocator; std::vector<_Tp, _Alloc>::value_type = glr_stack_item]' test.cc:1949:48: required from here /opt/local/include/gcc10/c++/bits/stl_construct.h:95:14: error: noexcept-expression evaluates to 'false' because of a call to 'glr_stack_item::glr_stack_item(const glr_stack_item&)' [-Werror=noexcept] 95 | noexcept(noexcept(::new((void*)0) _Tp(std::declval<_Args>()...))) | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ test.cc:1436:3: note: but 'glr_stack_item::glr_stack_item(const glr_stack_item&)' does not throw; perhaps it should be declared 'noexcept' 1436 | glr_stack_item(const glr_stack_item& other) | ^~~~~~~~~~~~~~ cc1plus: all warnings being treated as errors stdout: tests/types.at:138: exit code was 1, expected 0 433. types.at:138: 433. glr2.cc api.value.type={double} (types.at:138): FAILED (types.at:138) * data/skeletons/glr2.cc (glr_stack_item): Use YY_NOEXCEPT/YY_NOTHROW. 2020-09-12 Valentin Tolmer glr2.cc: coding style changes * data/skeletons/glr2.cc: Change some CamlCase to snake_case, and remove some yy prefixes for classes inside the namespace. 2020-09-12 Valentin Tolmer glr2.cc: move StrongIndexAlias into the namespace * data/skeletons/glr2.cc: here. 2020-09-12 Valentin Tolmer glr2.cc: remove usage of PTRDIFF_MAX * data/skeletons/glr2.cc: Use std::ptrdiff_t and numeric_limits. 2020-09-12 Valentin Tolmer glr2.cc: remove C-style casts * data/skeletons/glr2.cc: here. 2020-09-12 Valentin Tolmer glr2.cc: add copy constructor to yyGLRStackItem This silences the clang warning -Wdeprecated-copy. * data/skeletons/glr2.cc: here. 2020-09-07 Akim Demaille examples: d: remove unused token * examples/d/calc/calc.y, examples/d/simple/calc.y: Remove "=". 2020-09-06 Akim Demaille Merge branch 'maint' (i.e., Bison 3.7.2) * upstream/maint: maint: post-release administrivia version 3.7.2 build: disable syntax-check warning gnulib: update build: fix incorrect dependencies doc: updates gnulib: update tests: beware of sed portability issues 2020-09-06 Akim Demaille build: fix a concurrent build issue in examples Reported by Thomas Deutschmann . https://lists.gnu.org/r/bug-bison/2020-09/msg00010.html * examples/c/lexcalc/local.mk: scan.o depends on parse.[ch]. 2020-09-05 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2020-09-05 Akim Demaille version 3.7.2 * NEWS: Record release date. 2020-09-05 Akim Demaille build: disable syntax-check warning error_message_uppercase etc/bench.pl.in-419-static int yylex (@{[is_pure (@directive) ? "YYSTYPE *yylvalp" : "void"]}); * cfg.mk: here. 2020-09-05 Akim Demaille gnulib: update 2020-09-05 Akim Demaille build: fix incorrect dependencies Commit af000bab111768a04021bf5ffa4bbe91d44e231c ("doc: work around Texinfo 6.7 bug"), published in 3.4.91, added a dependency on the "all" target. This is a super bad idea, since "make all" will run this target *before* "all", which builds bison. It turns out that this new dependency actually needed bison to be built. So all the regular process (i) build $(BUILT_SOURCES) and then (ii) build bison, was wrecked since some of the $(BUILT_SOURCES) depended on bison... It was "easy" to see in the logs of "make V=1" because we were building bison files (such as src/files.o) *before* displaying the banner for "all-recursive". With this fix, we finally get again the proper sequence: rm -f examples/c/reccalc/scan.stamp examples/c/reccalc/scan.stamp.tmp /opt/local/libexec/gnubin/mkdir -p examples/c/reccalc touch examples/c/reccalc/scan.stamp.tmp flex -oexamples/c/reccalc/scan.c --header=examples/c/reccalc/scan.h ./examples/c/reccalc/scan.l mv examples/c/reccalc/scan.stamp.tmp examples/c/reccalc/scan.stamp rm -f lib/fcntl.h-t lib/fcntl.h && \ { echo '/* DO NOT EDIT! GENERATED AUTOMATICALLY! */'; \ ... } > lib/fcntl.h-t && \ mv lib/fcntl.h-t lib/fcntl.h ... mv -f lib/alloca.h-t lib/alloca.h make all-recursive Reported by Mingli Yu . https://github.com/akimd/bison/issues/31 https://lists.gnu.org/r/bison-patches/2020-05/msg00055.html Reported by Claudio Calvelli . https://lists.gnu.org/r/bug-bison/2020-09/msg00001.html https://bugs.gentoo.org/716516 * doc/local.mk (all): Rename as... (all-local): this. So that we don't compete with BUILT_SOURCES. 2020-09-05 Akim Demaille doc: simplify the extraction of example snippets * doc/bison.texi: Use qualified paths. * examples/extexi: Comment changes. 2020-09-05 Akim Demaille glr2.cc: style changes * data/skeletons/glr2.cc: Remove stray comment. 2020-09-05 Akim Demaille glr2.cc: get rid of the yyerror scaffolding The yyerror stand-alone function was used to bounce from glr.c's call to yyerror to glr.cc's parser.error. Now that glr.c is out of the way, just directly use parser.error. * data/skeletons/glr2.cc (yyerror): Remove. Adjust callers. (b4_yyerror_args, b4_lyyerror_args, b4_pure_formals): Remove. Now unused. 2020-09-04 Valentin Tolmer glr2.cc: avoid warnings about printf and shadowing * data/skeletons/glr2.cc: Migrate from using printf to std::cerr & co. Since the yyGLRStack has the user params, no need to pass them around. 2020-09-03 Adela Vais examples: d: demonstrate location tracking * examples/d/calc/calc.y: Track locations. * examples/d/calc/calc.test: Check locations. 2020-09-03 Adela Vais examples: d: duplicate the example as "simple" and "calc" * examples/d/Makefile, examples/d/calc.d, examples/d/calc.test, examples/d/calc/local.mk: Move into... * examples/d/calc, examples/d/simple: these new directories. 2020-09-02 Akim Demaille doc: updates * NEWS, TODO: here. 2020-09-02 Adela Vais examples: d: fix the handling of unary + It was interpreting "+exp" as "-exp". * examples/d/calc.y: Fix. * examples/d/calc.test: Check it. 2020-09-02 Akim Demaille cex: always show ε/%empty in counterexamples On a case such as %% exp : empty "a" | "a" empty empty : %empty we used to display warning: shift/reduce conflict on token "a" [-Wcounterexamples] Example: • "a" Shift derivation exp ↳ 2: • "a" empty ↳ 2: ε Example: • "a" Reduce derivation exp ↳ 1: empty "a" ↳ 3: • where the shift derivation shows an item "2: empty → ε", with an explicit "ε", but the reduce derivation shows "3: empty → •", without "ε". For consistency, let's always show ε/%empty in rules with an empty rhs: Reduce derivation exp ↳ 1: empty "a" ↳ 3: ε • * src/derivation.c (derivation_width, derivation_print_tree_impl): Always show ε/%empty in counterexamples. * tests/diagnostics.at: Check that case. * tests/conflicts.at, tests/counterexample.at: Adjust. 2020-08-30 Akim Demaille glr2.cc: avoid warnings about long long * data/skeletons/glr2.cc: Disable the warning before triggering it. 2020-08-30 Akim Demaille cex: display the rule numbers From Example: "if" expr "then" "if" expr "then" stmt • "else" stmt Shift derivation if_stmt ↳ "if" expr "then" stmt ↳ if_stmt ↳ "if" expr "then" stmt • "else" stmt Reduce derivation if_stmt ↳ "if" expr "then" stmt "else" stmt ↳ if_stmt ↳ "if" expr "then" stmt • to Example: "if" expr "then" "if" expr "then" stmt • "else" stmt Shift derivation if_stmt ↳ 3: "if" expr "then" stmt ↳ 2: if_stmt ↳ 4: "if" expr "then" stmt • "else" stmt Example: "if" expr "then" "if" expr "then" stmt • "else" stmt Reduce derivation if_stmt ↳ 4: "if" expr "then" stmt "else" stmt ↳ 2: if_stmt ↳ 3: "if" expr "then" stmt • * src/state-item.h, src/state-item.c (state_item_rule): New. * src/derivation.h, src/derivation.c (struct derivation): Add a rule member. Adjust dependencies. * src/counterexample.c, src/parse-simulation.c: Pass the rule to derivation_new. * src/derivation.c (fprintf_if): New. (derivation_width, derivation_print_tree_impl): Take the rule number into account. * tests/conflicts.at, tests/counterexample.at, tests/diagnostics.at, * tests/report.at: Adjust. * doc/bison.texi: Adjust. 2020-08-30 Akim Demaille gnulib: update 2020-08-30 Akim Demaille tests: beware of sed portability issues Reported by David Laxer . https://lists.gnu.org/r/bug-bison/2020-08/msg00027.html * tests/output.at: Don't use + with sed. While at it, fix a quotation problem hidden by the use of '#'. 2020-08-30 Akim Demaille gnulib: update 2020-08-30 Akim Demaille tests: beware of sed portability issues Reported by David Laxer . https://lists.gnu.org/r/bug-bison/2020-08/msg00027.html * tests/output.at: Don't use + with sed. While at it, fix a quotation problem hidden by the use of '#'. 2020-08-30 Akim Demaille glr2.cc: fix a few warnings * data/skeletons/glr2.cc: Fix some documentation. Be consistent between class/struct. (yydoAction, yyresolveAction): Avoid passing yyparser where useless. 2020-08-30 Valentin Tolmer glr2.cc: fork glr.cc to a c++ version This is a fork of glr.cc to be c++-first instead of a wrapper around glr.c. * data/skeletons/glr2.cc: New. * data/skeletons/bison.m4, data/skeletons/c++.m4: Adjust. * data/skeletons/c.m4 (b4_user_args_no_comma): New. * src/reader.c (grammar_rule_check_and_complete): glr2.cc is C++. * tests/actions.at, tests/c++.at, tests/calc.at, tests/conflicts.at, * tests/input.at, tests/local.at, tests/regression.at, tests/scanner.at, * tests/synclines.at, tests/types.at: Also check glr2.cc. 2020-08-30 Akim Demaille c: always use YYMALLOC/YYFREE Reported by Kovalex . https://lists.gnu.org/r/bug-bison/2020-08/msg00015.html * data/skeletons/yacc.c: Don't make direct calls to malloc/free. * tests/calc.at: Check it. 2020-08-30 Akim Demaille build: beware of POSIX mode Reported by Dennis Clarke. https://lists.gnu.org/r/bug-bison/2020-08/msg00013.html * examples/d/local.mk, examples/java/calc/local.mk, * examples/java/simple/local.mk: Pass bison's options before its argument, in case we're in POSIX mode. 2020-08-30 Akim Demaille doc: history of api.prefix Reported by Matthew Fernandez . https://lists.gnu.org/r/help-bison/2020-08/msg00015.html * doc/bison.texi (api.prefix): We move to {} in 3.0. 2020-08-11 Akim Demaille CI: intel moved the script for ICC * .travis.yml: Adjust. 2020-08-08 Akim Demaille fix: unterminated \-escape An assertion failed when the last character is a '\' and we're in a character or a string. Reported by Agency for Defense Development. https://lists.gnu.org/r/bug-bison/2020-08/msg00009.html * src/scan-gram.l: Catch unterminated escapes. * tests/input.at (Unexpected end of file): New. 2020-08-07 Akim Demaille fix: crash when redefining the EOF token Reported by Agency for Defense Development. https://lists.gnu.org/r/bug-bison/2020-08/msg00008.html On an empty such as %token FOO BAR FOO 0 %% input: %empty we crash because when we find FOO 0, we decrement ntokens (since FOO was discovered to be EOF, which is already known to be a token, so we increment ntokens for it, and need to cancel this). This "works well" when EOF is properly defined in one go, but here it is first defined and later only assign token code 0. In the meanwhile BAR was given the token number that we just decremented. To fix this, assign symbol numbers after parsing, not during parsing, so that we also saw all the explicit token codes. To maintain the current numbers (I'd like to keep no difference in the output, not just equivalence), we need to make sure the symbols are numbered in the same order: that of appearance in the source file. So we need the locations to be correct, which was almost the case, except for nterms that appeared several times as LHS (i.e., several times as "foo: ..."). Fixing the use of location_of_lhs sufficed (it appears it was intended for this use, but its implementation was unfinished: it was always set to "false" only). * src/symtab.c (symbol_location_as_lhs_set): Update location_of_lhs. (symbol_code_set): Remove broken hack that decremented ntokens. (symbol_class_set, dummy_symbol_get): Don't set number, ntokens and nnterms. (symbol_check_defined): Do it. (symbols): Don't count nsyms here. Actually, don't count nsyms at all: let it be done in... * src/reader.c (check_and_convert_grammar): here. Define nsyms from ntokens and nnterms after parsing. * tests/input.at (EOF redeclared): New. * examples/c/bistromathic/bistromathic.test: Adjust the traces: in "%nterm exp %% input: ...", exp used to be numbered before input. 2020-08-07 Akim Demaille style: fix missing space before paren * cfg.mk (_space_before_paren_exempt): Be less laxist. * src/output.c, src/reader.c: Fix space before paren issues. Pacify the warnings where applicable. 2020-08-07 Akim Demaille style: fix comments and more debug trace * src/location.c, src/symtab.h, src/symtab.c: here. 2020-08-07 Akim Demaille style: more uses of const * src/symtab.c: here. 2020-08-07 Akim Demaille bench: fix support for pure parser * etc/bench.pl.in (is_pure): New. (generate_grammar_calc): Use code provides where needed. Use is_pure to call yylex properly. Coding style fixes. 2020-08-03 Akim Demaille portability: multiple typedefs Older versions of GCC (4.1.2 here) don't like repeated typedefs. CC src/bison-parse-simulation.o src/parse-simulation.c:61: error: redefinition of typedef 'parse_state' src/parse-simulation.h:74: error: previous declaration of 'parse_state' was here make: *** [Makefile:7876: src/bison-parse-simulation.o] Error 1 Reported by Nelson H. F. Beebe. * src/parse-simulation.c (parse_state): Don't typedef, parse-simulation.h did it already. 2020-08-02 Akim Demaille style: revert "avoid warnings with GCC 4.6" This reverts commit d0bec3175ff5cf6582ffbf584b73ea6aaea838d0 (which should have read "We have a clash...", not "With have a clash..."). Now that `max()` was renamed `max_int()`, we can use `max` again, as elsewhere in the code. * src/counterexample.c (visited_hasher): Alpha reconversion. 2020-08-02 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2020-08-02 Akim Demaille version 3.7.1 * NEWS: Record release date. 2020-08-02 Akim Demaille portability: we use termios.h and sys/ioctl.h Reported by Maarten De Braekeleer. https://lists.gnu.org/r/bison-patches/2020-07/msg00079.html * bootstrap.conf (gnulib_modules): Add termios and sys_ioctl. 2020-08-02 Maarten De Braekeleer portability: rename accept to acceptsymbol because of MSVC MSVC already defines this symbol. * src/symtab.h, src/symtab.c (accept): Rename as... (acceptsymbol): this. Adjust dependencies. 2020-08-02 Akim Demaille regen 2020-08-02 Maarten De Braekeleer portability: use CHAR_LITERAL instead of CHAR because MSVC defines CHAR * src/parse-gram.y, src/scan-gram.l: here. 2020-08-02 Maarten De Braekeleer portability: use INT_LITERAL instead of INT because MSVC defines INT It is defined as a typedef, not a macro. https://lists.gnu.org/r/bison-patches/2020-08/msg00001.html * src/parse-gram.y, src/scan-gram.l: here. 2020-08-02 Akim Demaille portability: beware of max () with MSVC Reported by Maarten De Braekeleer. https://lists.gnu.org/r/bison-patches/2020-07/msg00080.html We don't want to use gnulib's min and max macros, since we use function calls in min/max arguments. * src/location.c (max_int, min_int): Move to... * src/system.h: here. * src/counterexample.c, src/derivation.c: Use max_int instead of max. 2020-08-02 Akim Demaille libtextstyle: be sure to have ostream_printf and hyperlink support Older versions of libtextstyle do not support them, rule them out. Reported by Lars Wendler https://lists.gnu.org/r/bug-bison/2020-07/msg00030.html and by Arnold Robbins https://lists.gnu.org/r/bug-bison/2020-07/msg00041.html and https://lists.gnu.org/mailman/private/gawk-devel/2020-July/003988.html and by Nelson H. F. Beebe https://lists.gnu.org/mailman/private/gawk-devel/2020-July/003993.html With support from Bruno Haible in gnulib https://lists.gnu.org/r/bug-gnulib/2020-08/msg00000.html thread starting at https://lists.gnu.org/r/bug-gnulib/2020-07/msg00148.html * configure.ac: Require libtextstyle 0.20.5. * gnulib: Update. 2020-08-01 Akim Demaille CI: comment changes 2020-08-01 Akim Demaille regen 2020-08-01 Akim Demaille diagnostics: better location for type redeclarations From foo.y:1.7-11: error: %type redeclaration for bar 1 | %type bar bar | ^~~~~ foo.y:1.7-11: note: previous declaration 1 | %type bar bar | ^~~~~ to foo.y:1.17-19: error: %type redeclaration for bar 1 | %type bar bar | ^~~ foo.y:1.13-15: note: previous declaration 1 | %type bar bar | ^~~ * src/symlist.h, src/symlist.c (symbol_list_type_set): There's no need for the tag's location, use that of the symbol. * src/parse-gram.y: Adjust. * tests/input.at: Adjust. 2020-07-30 Akim Demaille todo: updates for D 2020-07-29 Akim Demaille cex: style: comment changes * src/parse-simulation.c: here. 2020-07-29 Akim Demaille cex: style: prefer "res" for the returned value * src/derivation.c (derivation_new): here. 2020-07-29 Akim Demaille cex: style: prefer FOO_print to print_FOO * src/state-item.h, src/state-item.c (print_state_item): Rename as... (state_item_print): this. * src/counterexample.c (print_counterexample): Rename as... (counterexample_print): this. 2020-07-28 Akim Demaille scanner: don't crash on strings containing a NUL byte We crash if the input contains a string containing a NUL byte. Reported by Suhwan Song. https://lists.gnu.org/r/bug-bison/2020-07/msg00051.html * src/flex-scanner.h (STRING_FREE): Avoid accidental use of last_string. * src/scan-gram.l: Don't call STRING_FREE without calling STRING_FINISH first. * tests/input.at (Invalid inputs): Check that case. 2020-07-28 Akim Demaille doc: refer to cex from sections dealing with conflicts The documentation about -Wcex should be put forward. * doc/bison.texi: Refer to -Wcex from the sections about conflicts. 2020-07-28 Akim Demaille doc: factor ifnottex/iftex examples * doc/bison.texi: Factor the common bits out of ifnottex/iftex. 2020-07-28 Akim Demaille doc: fix colors The original Texinfo macros introducing colors were made for diagnostics, which are printed in bold. So by copy-paste accident the styles we introduced for counterexamples were also in bold. They should not. * doc/bison.texi: Separate the styling of diagnostics from the styling for counterexamples. Don't use bold in the latter case. 2020-07-28 Akim Demaille doc: fixes * doc/bison.texi: Fix spello. Fix missing colors, and factor. 2020-07-23 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2020-07-23 Akim Demaille version 3.7 * NEWS: Record release date. 2020-07-23 Akim Demaille style: avoid warnings with GCC 4.6 With have a clash with the "max" function. src/counterexample.c: In function 'visited_hasher': src/counterexample.c:720:48: error: declaration of 'max' shadows a global declaration [-Werror=shadow] src/counterexample.c:116:12: error: shadowed declaration is here [-Werror=shadow] * src/counterexample.c (visited_hasher): Alpha conversion. 2020-07-23 Akim Demaille doc: fix definition of -Wall * doc/bison.texi (Diagnostics): here. 2020-07-23 Akim Demaille gnulib: update * bootstrap.conf: We need stpncpy. 2020-07-23 Akim Demaille tests: fixes Fix 6b78e50cef3c2cd8e6f4e7938be987e8769f8eef, "cex: make "rerun with '-Wcex'" a note instead of a warning" * tests/conflicts.at (-W versus %expect and %expect-rr): Fix expectations. 2020-07-22 Akim Demaille cex: update NEWS for 3.7 * NEWS: Update to the current style of cex display. 2020-07-22 Akim Demaille doc: catch up with the current display of cex Unfortunately I found no way to use the ↳ glyph in Texinfo, so I used @arrow{} instead, which has a different width, so we have to have all the examples doubled, once for TeX, another for the rest of the world. * doc/bison.texi: Use the current display in the examples. * doc/calc.y, doc/ids.y, doc/if-then-else.y, doc/sequence.y: New. 2020-07-21 Akim Demaille cex: make "rerun with '-Wcex'" a note instead of a warning Currently the suggestion to rerun is a -Wother warning: warning: 2 shift/reduce conflicts [-Wconflicts-sr] warning: rerun with option '-Wcounterexamples' to generate conflict counterexamples [-Wother] Instead, let's attach it as a subnote of the diagnostic (in the current case, -Wconflicts-sr): warning: 2 shift/reduce conflicts [-Wconflicts-sr] note: rerun with option '-Wcounterexamples' to generate conflict counterexamples * src/conflicts.c (conflicts_print): Do that. Adjust the test suite. 2020-07-20 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2020-07-20 Akim Demaille version 3.6.93 * NEWS: Record release date. 2020-07-20 Akim Demaille cex: label all the derivations by their initial action From input.y: warning: reduce/reduce conflict on token $end [-Wcounterexamples] Example: A b . First derivation a `-> A b . Second derivation a `-> A b `-> b . to input.y: warning: reduce/reduce conflict on token $end [-Wcounterexamples] Example: A b . First reduce derivation a `-> A b . Second reduce derivation a `-> A b `-> b . * src/counterexample.c (print_counterexample): here. Compute the width of the labels to properly align the values. * tests/conflicts.at, tests/counterexample.at, tests/diagnostics.at, * tests/report.at: Adjust. 2020-07-20 Akim Demaille cex: improve readability of the subsections Now that the derivation is no longer printed on one line, aligning the example and the derivation is no longer useful. It can actually be harmful, as it makes the overall structure less clear. * src/derivation.h, src/derivation.c (derivation_print_leaves): Remove the `prefix` argument. * src/counterexample.c (print_counterexample): Put the example next to its label. * tests/conflicts.at, tests/counterexample.at, tests/diagnostics.at, * tests/report.at: Adjust. 2020-07-20 Akim Demaille cex: don't issue an empty line between counterexamples Now that we use complain, the "sections" are clearer. * src/counterexample.c (print_counterexample): Use the empty line only in reports. * tests/counterexample.at, tests/diagnostics.at, tests/report.at: Adjust. 2020-07-20 Akim Demaille cex: use usual routines for diagnostics about S/R conflicts See previous commit. We go from input.y: warning: 3 reduce/reduce conflicts [-Wconflicts-rr] Shift/reduce conflict on token "⊕": Example exp "+" exp • "⊕" exp Shift derivation exp ↳ exp "+" exp ↳ exp • "⊕" exp to input.y: warning: 3 reduce/reduce conflicts [-Wconflicts-rr] input.y: warning: shift/reduce conflict on token "⊕" [-Wcounterexamples] Example exp "+" exp • "⊕" exp Shift derivation exp ↳ exp "+" exp ↳ exp • "⊕" exp with an hyperlink on -Wcounterexamples. * src/counterexample.c (counterexample_report_shift_reduce): Use complain. * tests/counterexample.at, tests/diagnostics.at, tests/report.at: Adjust. 2020-07-20 Akim Demaille cex: use usual routines for diagnostics about R/R conflicts This is more consistent, and brings benefits: users know that these diagnostics are attached to -Wcounterexamples, and they can also click on the hyperlink if permitted by their terminal. We go from warning: 1 reduce/reduce conflict [-Wconflicts-rr] Reduce/reduce conflict on token $end: Example A b . First derivation a -> [ A b . ] Second derivation a -> [ A b -> [ b . ] ] to warning: 1 reduce/reduce conflict [-Wconflicts-rr] input.y: warning: reduce/reduce conflict on token $end [-Wcounterexamples] Example A b . First derivation a -> [ A b . ] Second derivation a -> [ A b -> [ b . ] ] with an hyperlink on -Wcounterexamples. * src/counterexample.c (counterexample_report_reduce_reduce): Use complain. * tests/counterexample.at, tests/diagnostics.at, tests/report.at: Adjust. 2020-07-19 Akim Demaille diagnostics: use hyperlinks to point to the only documentation * src/complain.c (begin_hyperlink, end_hyperlink): New. (warnings_print_categories): Use them. * tests/local.at (AT_SET_ENV): Disable hyperlinks in the tests, they contain random id's, and brackets (which is not so nice for M4). 2020-07-19 Akim Demaille doc: add anchors for warnings Unfortunately Texinfo somewhat mangles anchors such as `-Werror` into `g_t_002dWerror`, so let's not include the dash. * doc/bison.texi (Diagnostics): here. 2020-07-19 Akim Demaille glyphs: fix types The code was written on top of buffers of `char[26]`, and then was changed to use `char *`, yet was still using `sizeof buf`, which became `sizeof (char *)` instead of `sizeof (char[26])`. Reported by Dagobert Michelsen. https://lists.gnu.org/r/bug-bison/2020-07/msg00023.html * src/glyphs.h, src/glyphs.c: Get rid of uses of `char *`, use only glyph_buffer_t. 2020-07-19 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2020-07-19 Akim Demaille version 3.6.92 * NEWS: Record release date. 2020-07-19 Akim Demaille style: avoid strncpy syntax-check seems to dislike strncpy. The GNU Coreutils replaced their uses of strncpy with stpncpy. strlcpy is not an option. http://sources.redhat.com/ml/libc-alpha/2002-01/msg00159.html http://sources.redhat.com/ml/libc-alpha/2002-01/msg00011.html http://lists.gnu.org/archive/html/bug-gnulib/2004-09/msg00181.html * src/glyphs.c: Use stpncpy. 2020-07-18 Akim Demaille cex: display derivations as trees Sometimes, understanding the derivations is difficult, because they are serialized to fit in one line. For instance, the example taken from the NEWS file: %token ID %% s: a ID a: expr expr: expr ID ',' | "expr" gave First example expr • ID ',' ID $end Shift derivation $accept → [ s → [ a → [ expr → [ expr • ID ',' ] ] ID ] $end ] Second example expr • ID $end Reduce derivation $accept → [ s → [ a → [ expr • ] ID ] $end ] Printing as trees, it gives: First example expr • ID ',' ID $end Shift derivation $accept ↳ s $end ↳ a ID ↳ expr ↳ expr • ID ',' Second example expr • ID $end Reduce derivation $accept ↳ s $end ↳ a ID ↳ expr • * src/glyphs.h, src/glyphs.c (down_arrow, empty, derivation_separator): New. * src/derivation.c (derivation_print, derivation_print_impl): Rename as... (derivation_print_flat, derivation_print_flat_impl): These. (fputs_if, derivation_depth, derivation_width, derivation_print_tree) (derivation_print_tree_impl, derivation_print): New. * src/counterexample.c (print_counterexample): Adjust. * tests/conflicts.at, tests/counterexample.at, tests/diagnostics.at, * tests/report.at: Adjust. 2020-07-16 Akim Demaille cex: use the glyphs * src/derivation.c: here. * src/gram.h, src/gram.c (print_arrow, print_dot, print_fallback): Remove. 2020-07-16 Akim Demaille cex: factor the handling of graphical symbols * src/glyphs.h, src/glyphs.c: New. 2020-07-15 Akim Demaille cex: style changes * src/counterexample.c: here. 2020-07-15 Akim Demaille cex: simplify tests * tests/counterexample.at (AT_BISON_CHECK_CEX): Handle the keyword. Simplify the signature. 2020-07-15 Akim Demaille cex: more colors Provided by Daniela Becker. * data/bison-default.css: More colors. 2020-07-14 Akim Demaille style: comments changes * src/print.c: here. 2020-07-14 Akim Demaille doc: update GLR sections Reported by Christian Schoenebeck. * doc/bison.texi (GLR Parsers): Minor fixes. (Compiler Requirements for GLR): Remove, quite useless today. 2020-07-14 Akim Demaille cex: display shifts before reductions When reporting counterexamples for s/r conflicts, put the shift first. This is more natural, and displays the default resolution first, which is also what happens for r/r conflicts where the smallest rule number is displayed first, and "wins". * src/counterexample.c (counterexample): Add a shift_reduce member. (new_counterexample): Adjust. Swap the derivations when this is a s/r conflict. (print_counterexample): For s/r conflicts, prefer "Shift derivation" and "Reduce derivation" rather than "First/Second derivation". * tests/conflicts.at, tests/counterexample.at, tests/report.at: Adjust. * NEWS, doc/bison.texi: Ditto. 2020-07-14 Akim Demaille style: s/lookahead_tokens/lookaheads/g Currently we use both names. Let's stick to the short one. * src/AnnotationList.c, src/conflicts.c, src/counterexample.c, * src/getargs.c, src/getargs.h, src/graphviz.c, src/ielr.c, * src/lalr.c, src/print-graph.c, src/print-xml.c, src/print.c, * src/state-item.c, src/state.c, src/state.h, src/tables.c: s/lookahead_token/lookahead/gi. 2020-07-14 Akim Demaille cex: factor memory allocation * src/counterexample.c (counterexample_report_state): Allocate once per conflicted state, instead of once per r/r conflict. 2020-07-14 Akim Demaille cex: use state_item_number consistently * src/counterexample.c, src/state-item.c: here. (counterexample_report_state): While at it, prefer c2 to j/k, to match c1. 2020-07-14 Akim Demaille cex: more consistent memory allocation/copy * src/counterexample.c, src/parse-simulation.c: It is more usual in Bison to use sizeof on expressions than on types, especially for allocation. Let the compiler do it's job instead of calling memcpy ourselves. 2020-07-14 Akim Demaille cex: minor renaming * src/counterexample.c (has_common_prefix): Rename as... (have_common_prefix): this. 2020-07-14 Akim Demaille cex: use better type names There are too many gl_list_t in there, it's hard to understand what is going on. Introduce and use more precise types. I sure can be wrong in some places, it's hard to tell without proper tool support. * src/counterexample.c, src/lssi.c, src/lssi.h, src/parse-simulation.c, * src/parse-simulation.h, src/state-item.c, src/state-item.h (si_bfs_node_list, search_state_list, ssb_list, lssi_list) (state_item_list): New. 2020-07-14 Akim Demaille cex: minor style changes * src/counterexample.h, src/derivation.h, src/derivation.c: More comments. Use `out` for FILE*, as elsewhere. 2020-07-14 Akim Demaille tests: beware of version numbers from git describe * tests/report.at: Be robust to version numbers such as 3.6.4.133-fbac-dirty. 2020-07-14 Akim Demaille tests: fix expectations Broken in ee86ea88399ed02243fbceb2704c9ea322a12bf9. * tests/diagnostics.at: here. 2020-07-12 Akim Demaille doc: makeinfo wants @arrow{}, not @arrow * doc/bison.texi: here. 2020-07-11 Akim Demaille gnulib: update 2020-07-11 Akim Demaille cex: prefer → to ::= It does not make a lot of sense to use ::= in our counterexamples, that's not something that belongs to the Bison "vocabulary". Using the colon makes sense, but it's too discreet. Let's use the arrow, which we already use in some reports (HTML and Dot). * src/gram.h (print_dot_fallback): Generalize into... (print_fallback): this. (print_arrow): New. * src/derivation.c: Use it. * NEWS, tests/conflicts.at, tests/counterexample.at, * tests/diagnostics.at, tests/report.at: Adjust. * doc/bison.texi: Ditto. Unfortunately the literal `→` is output as `↦`. So we need to use @arrow. 2020-07-11 Akim Demaille style: cex: prefer the array notation Prefer `&foos[i]` to `foos + i` when `foos` is an array. IMHO, it makes the semantics clearer. * src/counterexample.c, src/lssi.c, src/parse-simulation.c, * src/state-item.c: With arrays, prefer the array notation rather than the pointer one. 2020-07-11 Akim Demaille style: cex: remove variables that don't make it simpler to read * src/counterexample.c: With arrays, prefer the array notation rather than the pointer one. 2020-07-11 Akim Demaille bistromathic: demonstrate caret-diagnostics * examples/c/bistromathic/parse.y (user_context): We need the current line. (yyreport_syntax_error): Quote the guilty line, with squiggles. * examples/c/bistromathic/bistromathic.test: Adjust. 2020-07-11 Akim Demaille bistromathic: do not display parse errors on completion Currently autocompletion on a line with errors leaks the error messages. It can be useful to let the user know, but GNU Readline does not provide us with an nice way to display the error. So we actually break into the current line of the user. So instead, do not show these errors. * examples/c/bistromathic/parse.y (user_context): New. Use %param to pass it to the parser and scanner. Keep quiet when in computing autocompletion. 2020-07-11 Akim Demaille bistromathic: don't stupidly reset the location for each token That quite defeats the whole point of locations... But anyway, we should not see these messages at all. * examples/c/bistromathic/parse.y (expected_tokens): Fix (useless) location tracking. 2020-07-11 Akim Demaille bistromathic: promote yytoken_kind_t * examples/c/bistromathic/parse.y: Use yytoken_kind_t rather than int. 2020-07-11 Akim Demaille html: capitalize titles * data/xslt/xml2xhtml.xsl: Use "State 0", not "state 0". As we do in text reports. 2020-07-11 Akim Demaille html: don't define several times the same anchors Currently when we output useless rules, they appear before the grammar, but using the same invocation. As a result, the anchor is defined twice, and the wrong one, being first, is honored. * data/xslt/xml2xhtml.xsl (rule): Take a new 'anchor' parameter to decide whether being an anchor, or a target. Let it be true when output the grammar. * tests/report.at: Adjust. 2020-07-11 Akim Demaille html: simplify * data/xslt/xml2xhtml.xsl: Merge two identical when-clauses. 2020-07-11 Akim Demaille reports: let html reports catch up with --report and --graph * data/xslt/xml2xhtml.xsl: Show the symbol types. * tests/report.at: Adjust. 2020-07-11 Akim Demaille reports: let xml reports catch up with --report and --graph The text and Dot reports are expected to be identical when generated directly (--report, --graph) or indirectly (via XML). The xml testsuite had not be run for ages, let it catch up a bit. * src/print-xml.c: Pass the type of the symbols. * data/xslt/xml2text.xsl Catch up with the new layout. Display the symbol types. Use '•', not '.' * tests/local.at: Smash '•' to '.' when matching against the direct text report. * tests/report.at: Adjust XML expectations. 2020-07-11 Akim Demaille reports: update html ouput * data/xslt/xml2xhtml.xsl: Improve indentation. Use ul/li rather that pre. 2020-07-11 Akim Demaille tests: check html * tests/report.at: here. 2020-07-11 Akim Demaille style: factor complex expressions * src/print-xml.c, src/print.c: Introduce a variable pointing to the current symbol. 2020-07-11 Akim Demaille maint: make it easier to update expectations * tests/local.mk (update-tests): New. 2020-07-09 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2020-07-09 Akim Demaille version 3.6.91 * NEWS: Record release date. 2020-07-09 Akim Demaille news: update 2020-07-09 Akim Demaille gnulib: update 2020-07-08 Akim Demaille examples: add license headers Prompted by Rici Lake. https://stackoverflow.com/questions/62658368/#comment110853985_62661621 Discussed with Paul Eggert. * doc/bison.texi, examples/c/bistromathic/parse.y, * examples/c/lexcalc/parse.y, examples/c/lexcalc/scan.l, * examples/c/pushcalc/calc.y, examples/c/reccalc/parse.y, * examples/c/reccalc/scan.l, examples/d/calc.y, * examples/java/calc/Calc.y, examples/java/simple/Calc.y: Install the GPL3+ header. 2020-07-05 Akim Demaille style: update comments * src/reader.c: action_obstack was removed in 2002... * src/parse-gram.y: Better names. * src/scan-code.h: More comments. 2020-07-05 Akim Demaille style: update comments in the skeletons * data/skeletons/c++.m4, data/skeletons/glr.c, data/skeletons/lalr1.d, * data/skeletons/lalr1.java, data/skeletons/yacc.c: Be more accurate about yychar and yytoken. Don't name local variables as if they were members. 2020-07-05 Akim Demaille doc: more details about symbols in m4 * data/README.md: here. * README-hacking.md (Vocabulary): More. 2020-07-05 Akim Demaille regen 2020-07-05 Akim Demaille examples: include the generated header * examples/c/bistromathic/parse.y, examples/c/lexcalc/parse.y, * examples/c/reccalc/parse.y: here. Add some comments. * src/parse-gram.y (api_version): Pull out of handle_require. Bump to 3.7. 2020-07-04 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2020-07-04 Akim Demaille version 3.6.90 * NEWS: Record release date. 2020-07-04 Akim Demaille news: update 2020-07-04 Akim Demaille package: fix syntax-check errors * examples/c/bistromathic/bistromathic.test, po/POTFILES.in: here. 2020-07-04 Akim Demaille gnulib: update 2020-07-04 Akim Demaille cex: give more details about -Wcex and -rcex * data/bison-default.css: Cobalt does not seem to be supported. * doc/bison.texi (Counterexamples): A new section. (Understanding): Show the counterexamples as it shows in the report: with its items. (Bison Options): Document -Wcex and -rcex. 2020-07-03 Akim Demaille news: update 2020-07-03 Akim Demaille dot: also use a dot in the output * src/print-graph.c (print_core): Use a dot instead of a point. * doc/figs/example-reduce.gv, doc/figs/example-reduce.txt, * doc/figs/example-shift.gv, doc/figs/example-shift.txt, * doc/figs/example.gv: Update. * tests/output.at, tests/report.at: Adjust. 2020-07-02 Akim Demaille doc: improve a sentence * doc/bison.texi: here. 2020-07-01 Akim Demaille news: formatting changes 2020-07-01 Akim Demaille news, todo: update 2020-06-30 Akim Demaille doc: clarify that the pcontext interface is *.c only Reported by Rici Lake. https://lists.gnu.org/r/bug-bison/2020-06/msg00054.html * doc/bison.texi (Syntax Error Reporting Function): Make it clear that this is not exported. Remove C++ details that landed in the C doc. 2020-06-30 Akim Demaille doc: use color in the cex examples * doc/bison.texi: here. And use smallexample when it no longer fits in PDF. 2020-06-30 Akim Demaille doc: repair the references to the Bibliography In commit c80cdf2db2b302db4137fabd4ae11e578fa51fca ("doc: simplify uses of @ref", Jan 27 2020, released in Bison 3.6), I broke the references to the Bibliography. For instance: For a more detailed exposition of the mysterious behavior in LALR parsers -and the benefits of IELR, @pxref{Bibliography,,Denny 2008 March}, and -@ref{Bibliography,,Denny 2010 November}. +and the benefits of IELR, @pxref{Bibliography}, and +@ref{Bibliography}. which results in "see Bibliography" twice, instead of the more precise reference. * doc/bison.texi (@pcite, @tcite): New. Use them instead of @ref to Bibliography. Cite only the first author (that's what we did for the other entries). 2020-06-30 Vincent Imbimbo doc: cex documentation * NEWS, doc/bison.texi: Add documentation for conflict counterexample generation. 2020-06-30 Akim Demaille doc: update Doxygen template file * doc/Doxyfile.in: here. 2020-06-29 Akim Demaille yacc.c: push: don't clear the parser state when accepting/rejecting Currently when a push parser finishes its parsing (i.e., it did not return YYPUSH_MORE), it also clears its state. It is therefore impossible to see if it had parse errors. In the context of autocompletion, because error recovery might have fired, the parser is actually already in a different state. For instance on `(1 + + ` in the bistromathic, because there's a `exp: "(" error ")"` recovery rule, `1 + +` tokens have already been popped, replaced by `error`, and autocompletions think we are ready for the closing ")". So here, we would like to see if there was a syntax error, yet `yynerrs` was cleared. In the case of a successful parse, we still have a problem: if error recovery succeeded, we won't know it, since, again, `yynerrs` is clearer. It seems much more natural to leave the parser state available for analysis when there is a failure. To reuse the parser, we should either: 1. provide an explicit means to reinitialize a parser state for future parses. 2. automatically reset the parser state when it is used in a new parse. Option 2 requires to check whether we need to reinitialize the parser each time we call `yypush_parse`, i.e., each time we give a new token. This seems expensive compared to Option 1, but benchmarks revealed no difference. Option 1 is incompatible with the documentation ("After `yypush_parse` returns a status other than `YYPUSH_MORE`, the parser instance `yyps` may be reused for a new parse."). So Option 2 wins, reusing the private `yynew` member to record that a parse was finished, and therefore that the state must reset in the next call to `yypull_parse`. While at it, this implementation now reuses the previously enlarged stacks from one parse to another. * data/skeletons/yacc.c (yypstate_new): Set up the stacks in their initial configurations (setting their bottom to the stack array), and use yypstate_clear to reset them (moving their top to their bottom). (yypstate_delete): Adjust. (yypush_parse): At the beginning, clear yypstate if needed, and at the end, record when yypstate needs to be clearer. * examples/c/bistromathic/parse.y (expected_tokens): Do not propose autocompletion when there are parse errors. * examples/c/bistromathic/bistromathic.test: Check that case. 2020-06-29 Akim Demaille bistromathic: don't display undefined locations Currently, completion when there is a syntax error shows broken locations. * examples/c/bistromathic/parse.y (expected_tokens): Initialize the location. * examples/c/bistromathic/bistromathic.test: Check that. 2020-06-29 Akim Demaille yacc.c: simplify initialization of push parsers The previous commit ("yacc.c: declare and initialize and the same time") made b4_initialize_parser_state_variables useless. * data/skeletons/yacc.c (b4_initialize_parser_state_variables): Inline into... (yypstate_clear): here. 2020-06-29 Akim Demaille regen 2020-06-29 Akim Demaille yacc.c: declare and initialize and the same time In order to factor the code of push and pull parsers, the declaration of the parser's state variable was common (being local variable in pull parsers, and struct members in push parsers). This result in rather poor style in pull parser, with first variable declarations, and then their initializations. The initialization is about to differ between push and pull parsers, so it is no longer worth keeping both cases together. * data/skeletons/yacc.c (b4_declare_parser_state_variables): Accept an argument, and when it is set, initialize the variables. Adjust dependencies. 2020-06-29 Akim Demaille yacc.c: style changes in push mode * data/skeletons/yacc.c: here. 2020-06-29 Akim Demaille yacc.c: simplify yypull_parse Currently yypull_parse takes a yypstate* as argument, and accepts it to be NULL. This does not seem to make a lot of sense: rather it is its callers that should do that. I believe this is historical: yypull_parse was introduced first (c3d503425f8014b432601a33b3398446d63b5963), with yyparse being a macro. So yyparse could hardly deal with memory allocation properly. In 7172e23e8ffb95b8cafee24c4f36c46ca709507f that yyparse was turned into a genuine function. At that point, it should have allocated its own yypstate*, which would have left yypull_parse deal with only one single non-null ypstate* argument. Fortunately, it is nowhere documented that it is valid to pass NULL to yypull_parse. It is now forbidden. * data/skeletons/yacc.c (yypull_parse): Don't allocate a yypstate. Needs a location to issue the error message. (yyparse): Allocate the yypstate. 2020-06-29 Akim Demaille doc: tidy the text files * etc/README: Rename/reformat as... * etc/README.md: this. And ship it. 2020-06-29 Akim Demaille bench: simplify the `rand` target * etc/bench.pl.in: There is no need to recompile the bench cases themselves. 2020-06-29 Akim Demaille bench: make it easy to edit the generated files * etc/bench.pl.in (&compile): Generate rules that compile the generated files independently of the source files. 2020-06-29 Akim Demaille tests: don't use $VERBOSE It is used by the test suite itself, which results in this test failing. * tests/c++.at: Use $DEBUG, not $VERBOSE. 2020-06-28 Akim Demaille doc: overhaul of the readmes * README-hacking.md (Working from the Repository): Make it first to make it easier to find the instructions to build from the repo. (Implementation Notes): New. * README: Provide more links. 2020-06-28 Akim Demaille java: rename package as api.package * data/skeletons/lalr1.java: here. * doc/bison.texi: Update. * src/muscle-tab.c: Ensure backward compat. * tests/java.at: Check it. 2020-06-28 Akim Demaille style: shift/reduce, not shift-reduce * src/reader.c: here. 2020-06-27 Akim Demaille style: rename endtoken as eoftoken * src/symtab.h, src/symtab.c (endtoken): Rename as... (eoftoken): this. Adjust dependencies. 2020-06-27 Akim Demaille news: fixes Reported by Jacob L. Mandelson. * NEWS: here. 2020-06-27 Akim Demaille style: use 'nonterminal' consistently * doc/bison.texi: Formatting changes. * src/gram.h, src/gram.c (nvars): Rename as... (nnterms): this. Adjust dependencies. (section): New. Use it. Replace "non terminal" and "non-terminal" by "nonterminal". 2020-06-27 Akim Demaille doc: parse.assert in C++ requires RTTI * doc/bison.texi (%define Summary): Say it. 2020-06-27 Akim Demaille c++: by default, use const std::string for file names Reported by Martin Blais and Yuriy Solodkyy. https://lists.gnu.org/r/help-bison/2020-05/msg00011.html https://lists.gnu.org/r/bug-bison/2020-06/msg00038.html While at it, modernize filename_type as api.filename.type and document it properly. * data/skeletons/c++.m4 (filename_type): Rename as... (api.filename.type): this. Default to const std::string. * data/skeletons/location.cc (position, location): Expose the filename_type type. Use api.filename.type. * doc/bison.texi (%define Summary): Document api.filename.type. (C++ Location Values): Document position::filename_type. * src/muscle-tab.c (muscle_percent_variable_update): Ensure backward compatibility. * tests/c++.at: Check that using const file names is ok. tests/input.at: Check backward compat. 2020-06-27 Akim Demaille ielr: fix crash on memory management Reported by Dwight Guth. https://lists.gnu.org/r/bug-bison/2020-06/msg00037.html * src/AnnotationList.c (AnnotationList__computePredecessorAnnotations): Beware that SBITSET__FOR_EACH nests _two_ for-loops, so "break" does not actually break out of it. That was the only occurrence in the code. * src/Sbitset.h (SBITSET__FOR_EACH): Warn passersby. 2020-06-25 Akim Demaille style: factor the access to a rule from its items * src/counterexample.c (item_rule): Move to... * src/counterexample.h: here. * src/AnnotationList.c, src/counterexample.c, src/ielr.c: Use it. 2020-06-25 Akim Demaille style: clean up nullable * src/nullable.c: Reduce scopes. Prefer `r` to `rules_ruleno`, which is truly an ugly name. 2020-06-25 Akim Demaille style: clean up ielr * src/AnnotationList.c, src/ielr.c: Fix include order. Prefer `res` to `result`. Reduce scopes. Be free of the oldish 76 cols limitation when it clutters too much the code. Denest when possible (we're starving for horizontal width). 2020-06-23 Akim Demaille don't use strlen to compute visual width * src/output.c (prepare_symbol_names): Use mbswidth. 2020-06-23 Akim Demaille doc: use dot/'•' rather than point/'.' AFAICT, "dotted rule" is a more frequent synonym of "item" than "pointed rule". So let's migrate to using "dot" only. * doc/bison.texi: Use dot/'•' rather than point/'.'. * src/print-xml.c (print_core): Use dot rather than point. This is not backward compatible, but AFAICT, we don't have actual user of the XML output (but ourselves). So... * data/xslt/xml2dot.xsl, data/xslt/xml2text.xsl, * data/xslt/xml2xhtml.xsl, tests/report.at: ... adjust. 2020-06-23 Akim Demaille cex: display all the S/R conflicts, not just one per (state, rule) Before this commit, on %% exp : "if" exp "then" exp | "if" exp "then" exp "else" exp | exp "+" exp | "num" we used to not display the third counterexample below: Shift/reduce conflict on token "+": Example exp "+" exp . "+" exp First derivation exp ::=[ exp ::=[ exp "+" exp . ] "+" exp ] Second derivation exp ::=[ exp "+" exp ::=[ exp . "+" exp ] ] Shift/reduce conflict on token "else": Example "if" exp "then" "if" exp "then" exp . "else" exp First derivation exp ::=[ "if" exp "then" exp ::=[ "if" exp "then" exp . ] "else" exp ] Second derivation exp ::=[ "if" exp "then" exp ::=[ "if" exp "then" exp . "else" exp ] ] Shift/reduce conflict on token "+": Example "if" exp "then" exp . "+" exp First derivation exp ::=[ exp ::=[ "if" exp "then" exp . ] "+" exp ] Second derivation exp ::=[ "if" exp "then" exp ::=[ exp . "+" exp ] ] Shift/reduce conflict on token "+": Example "if" exp "then" exp "else" exp . "+" exp First derivation exp ::=[ exp ::=[ "if" exp "then" exp "else" exp . ] "+" exp ] Second derivation exp ::=[ "if" exp "then" exp "else" exp ::=[ exp . "+" exp ] ] * src/counterexample.c (counterexample_report_state): Don't stop of the first conflicts. * tests/conflicts.at, tests/counterexample.at, tests/diagnostics.at, * tests/report.at: Adjust. 2020-06-22 Akim Demaille cex: don't display twice unifying examples if there is no color It makes no sense, and is actually confusing, to display twice the same example with no visible difference. * src/complain.h, src/complain.c (is_styled): New. * src/counterexample.c (print_counterexample): Display the unified example a second time only if it makes a difference. * tests/conflicts.at, tests/counterexample.at, tests/report.at: Adjust. * tests/diagnostics.at: Make sure we do display the unifying examples twice when colors are enabled. And check those colors. 2020-06-22 Vincent Imbimbo cex: fix reporting of null nonterminals I implemented this to print A ::= [ ], but A ::= [ %empty ] might be clearer. * src/parse-simulation.c (nullable_closure): Don't generate null nonterminal derivations as leaves. * src/derivation.c (derivation_print_impl): Don't print seperator spaces for null nonterminal. * tests/counterexample.at: Update test results. 2020-06-22 Akim Demaille cex: use the bullet in HTML * data/xslt/xml2xhtml.xsl: here. 2020-06-19 Akim Demaille cex: style changes * src/counterexample.c: Simplify a bit. * src/parse-simulation.c, src/parse-simulation.h: Enforce coding style. 2020-06-16 Akim Demaille c++: get rid of global_tokens_and_yystype This was a hack to make it easier for people to migrate from yacc.c to lalr1.cc and from glr.c to glr.cc: when set, YYSTYPE and YYLTYPE were `#defined`. It was never documented (just mentioned in NEWS for Bison 2.2, 2006-05-19), but was used to simplify the test suite. Stop that: adjust the test suite to the skeletons, not the converse. In C++ use yy::parser::semantic_type, yy::parser::location_type, and yy::parser::token::MY_TOKEN, instead of YYSTYPE, YYLTYPE and MY_TOKEN. * data/skeletons/glr.cc, data/skeletons/lalr1.cc: Remove its support. * tests/actions.at, tests/c++.at, tests/calc.at: Adjust. 2020-06-16 Akim Demaille cex: don't assume the terminal supports "•" Use of print_unicode_char suggested by Bruno Haible. https://lists.gnu.org/r/bug-gettext/2020-06/msg00012.html * src/gram.h (print_dot_fallback, print_dot): New. * src/gram.c, src/derivation.c: Use it. * tests/counterexample.at, tests/report.at: Adjust the test suite. * .travis.yml, README-hacking.md: Adjust. 2020-06-16 Akim Demaille cex: also include in the report on --report=counterexamples And let --report=all include the counterexamples. * src/getargs.h, src/getargs.c (report_cex): New. * src/main.c: Compute counterexamples when -rcex is specified. * src/print.c: Include the counterexamples when -rcex is specified. * tests/conflicts.at, tests/existing.at, tests/local.at: Adjust. 2020-06-16 Akim Demaille cex: also include the counterexamples in the report The report is the best place to show the details about counterexamples, since we have the state right under the nose. For instance: State 7 1 exp: exp . "⊕" exp 2 | exp . "+" exp 2 | exp "+" exp . [$end, "+", "⊕"] 3 | exp . "+" exp 3 | exp "+" exp . [$end, "+", "⊕"] "⊕" shift, and go to state 6 $end reduce using rule 2 (exp) $end [reduce using rule 3 (exp)] "+" reduce using rule 2 (exp) "+" [reduce using rule 3 (exp)] "⊕" [reduce using rule 2 (exp)] "⊕" [reduce using rule 3 (exp)] $default reduce using rule 2 (exp) Conflict between rule 2 and token "+" resolved as reduce (%left "+"). Shift/reduce conflict on token "⊕": 2 exp: exp "+" exp . 1 exp: exp . "⊕" exp Example exp "+" exp • "⊕" exp First derivation exp ::=[ exp ::=[ exp "+" exp • ] "⊕" exp ] Example exp "+" exp • "⊕" exp Second derivation exp ::=[ exp "+" exp ::=[ exp • "⊕" exp ] ] Reduce/reduce conflict on tokens $end, "+", "⊕": 2 exp: exp "+" exp . 3 exp: exp "+" exp . Example exp "+" exp • First derivation exp ::=[ exp "+" exp • ] Example exp "+" exp • Second derivation exp ::=[ exp "+" exp • ] Shift/reduce conflict on token "⊕": 3 exp: exp "+" exp . 1 exp: exp . "⊕" exp Example exp "+" exp • "⊕" exp First derivation exp ::=[ exp ::=[ exp "+" exp • ] "⊕" exp ] Example exp "+" exp • "⊕" exp Second derivation exp ::=[ exp "+" exp ::=[ exp • "⊕" exp ] ] * src/conflicts.h, src/conflicts.c (has_conflicts): New. * src/counterexample.h, src/counterexample.c (print_counterexample): Add a `prefix` argument. (counterexample_report_shift_reduce) (counterexample_report_reduce_reduce): Show the items when there's a prefix. * src/state-item.h, src/state-item.c (print_state_item): Add a `prefix` argument. * src/derivation.h, src/derivation.c (derivation_print) (derivation_print_leaves): Add a prefix argument. * src/print.c (print_state): When -Wcex is enabled, show the conflicts. * tests/report.at: Adjust. 2020-06-16 Akim Demaille cex: indent the diagnostics to highlight the structure Instead of Shift/reduce conflict on token D: Example A a • D First derivation s ::=[ A a a ::=[ b ::=[ c ::=[ • ] ] ] d ::=[ D ] ] Example A a • D Second derivation s ::=[ A a d ::=[ • D ] ] display Shift/reduce conflict on token D: Example A a • D First derivation s ::=[ A a a ::=[ b ::=[ c ::=[ • ] ] ] d ::=[ D ] ] Example A a • D Second derivation s ::=[ A a d ::=[ • D ] ] * src/counterexample.c (print_counterexample): Indent. * tests/counterexample.at: Adjust. 2020-06-16 Akim Demaille cex: don't report the items Showing the items (with the state numbers) is really something we should restrict to the report. * src/counterexample.c (counterexample_report_shift_reduce) (counterexample_report_reduce_reduce): Don't show the pointed rules, we will do that in the report. * tests/counterexample.at: Adjust. 2020-06-16 Akim Demaille cex: make sure traces go to stderr * src/parse-simulation.h, src/parse-simulation.c (print_parse_state): here. 2020-06-16 Akim Demaille cex: add an argument to the reporting functions to specify the stream * src/conflicts.c (find_state_item_number, report_state_counterexamples): Move to... * src/counterexample.h, src/counterexample.c (find_state_item_number) (counterexample_report_state): this. Add support for `out` as an argument. (counterexample_report_reduce_reduce, counterexample_report_shift_reduce): Accept an `out` argument, and be static. 2020-06-16 Akim Demaille style: more uses of const * src/print.c, src/state.h, src/state.c: here. 2020-06-16 Akim Demaille Merge 'maint' * upstream/maint: maint: post-release administrivia version 3.6.4 glr.cc: don't leak glr.c/glr.cc scaffolding to the user Some fixes were needed to adjust to recent changes in glr.cc and glr.c. * data/skeletons/glr.cc: Stop messing with the user's epilogue to insert glr.cc code. We need that code to be inserted _before_ the user's epilogue, not after. So define b4_glr_cc_pre_epilogue. * data/skeletons/glr.c: Use it. 2020-06-15 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2020-06-15 Akim Demaille version 3.6.4 * NEWS: Record release date. 2020-06-15 Akim Demaille glr.cc: don't leak glr.c/glr.cc scaffolding to the user Until we have a decent reimplementation of glr.cc, we have to use tricks to shoehorn C++ symbols to the C engine of glr.c. Some of them are done via #define. Unfortunately in Bison 3.6 some of these we done in the header file, which broke valid user code. Reported by Egor Pugin. https://lists.gnu.org/r/bug-bison/2020-06/msg00003.html * data/skeletons/glr.cc: Stop playing tricks with b4_pre_epilogue. (b4_glr_cc_setup, b4_glr_cc_cleanup): New. Much cleaner way to instal glr.cc's scaffolding around glr.c. * data/skeletons/glr.c: Adjust to use them. 2020-06-13 Akim Demaille reports: the column width differs from the byte count From "number" shift, and go to state 1 "Ñùṃéℝô" shift, and go to state 2 to "number" shift, and go to state 1 "Ñùṃéℝô" shift, and go to state 2 * src/print.c: Use mbswidth, not strlen, to compute visual columns. * tests/report.at: Adjust. 2020-06-13 Akim Demaille reports: don't escape the labels Currently we use "quotearg" to escape the strings output in Dot. As a result, if the user's locale is C for instance, all the non-ASCII are escaped. Unfortunately graphviz does not interpret this style of escaping. For instance: 5 -> 2 [style=solid label="\"\303\221\303\271\341\271\203\303\251\342\204\235\303\264\""] was displayed as a sequence of numbers. We now output: 5 -> 2 [style=solid label="\"Ñùṃéℝô\""] independently of the user's locale. * src/system.h (obstack_backslash): New. * src/graphviz.h, src/graphviz.c (escape): Remove, use obstack_backslash instead. * src/print-graph.c: Likewise. * tests/report.at: Adjust. 2020-06-13 Akim Demaille regen 2020-06-13 Akim Demaille parser: keep string aliases as the user wrote it Currently our scanner decodes all the escapes in the strings, and we later reescape the strings when we emit them. This is troublesome, as we do not respect the user input. For instance, when the user writes in UTF-8, we destroy her string when we write it back. And this shows everywhere: in the reports we show the escaped string instead of the actual alias: 0 $accept: . exp $end 1 exp: . exp "\342\212\225" exp 2 | . exp "+" exp 3 | . exp "+" exp 4 | . "number" 5 | . "\303\221\303\271\341\271\203\303\251\342\204\235\303\264" "number" shift, and go to state 1 "\303\221\303\271\341\271\203\303\251\342\204\235\303\264" shift, and go to state 2 This commit preserves the user's exact spelling of the string aliases, instead of interpreting the escapes and then reescaping. The report now shows: 0 $accept: . exp $end 1 exp: . exp "⊕" exp 2 | . exp "+" exp 3 | . exp "+" exp 4 | . "number" 5 | . "Ñùṃéℝô" "number" shift, and go to state 1 "Ñùṃéℝô" shift, and go to state 2 Likewise, the XML (and therefore HTML) outputs are fixed. * src/scan-gram.l (STRING, TSTRING): Do not interpret the escapes in the resulting string. * src/parse-gram.y (unquote, parser_init, parser_free, unquote_free) (handle_defines, handle_language, obstack_for_unquote): New. Use them to unquote where needed. * tests/regression.at, tests/report.at: Update. 2020-06-13 Akim Demaille tests: check reports with conflicts and UTF-8 This is to record the current state of the report, which escapes the UTF-8 characters (as parse.error="verbose" does), but shouldn't (as parse.error="detailed" does). * tests/report.at: here. 2020-06-13 Akim Demaille style: factor common bits about string scanning * src/scan-gram.l: here. 2020-06-13 Akim Demaille style: introduce & use STRING_1GROW * src/flex-scanner.h (STRING_1GROW): New. * src/scan-gram.l, src/scan-skel.l: Use it. 2020-06-13 Akim Demaille style: reduce scopes * src/scan-gram.l (STRING_GROW_ESCAPE): Move the static_assert about type sizes here. 2020-06-13 Akim Demaille style: prefer 'FOO ()' to 'FOO' for function-like macros * src/flex-scanner.h (STRING_GROW, STRING_FINISH, STRING_FREE): Make them function-like macros. Adjust dependencies. 2020-06-11 Akim Demaille regen 2020-06-10 Akim Demaille cex: suggest -Wcounterexamples when there are unexpected conflicts Suggesting -Wcounterexamples when there are conflicts is probably not what the user wants. If she knows her conflicts and has set %expect/%expect-rr appropriately, we shouldn't warn. The commit also swaps the counterexamples and the report of conflicts, into, IMHO, a more natural order: from Shift/reduce conflict on token B: 1: 3 a: A . 1: 8 y: A . B Example A • B C First derivation s ::=[ a ::=[ A • ] x ::=[ B C ] ] Example A • B C Second derivation s ::=[ y ::=[ A • B ] c ::=[ C ] ] input.y: warning: 1 shift/reduce conflict [-Wconflicts-sr] input.y:4.4: warning: rule useless in parser due to conflicts [-Wother] to input.y: warning: 1 shift/reduce conflict [-Wconflicts-sr] Shift/reduce conflict on token B: 1: 3 a: A . 1: 8 y: A . B Example A • B C First derivation s ::=[ a ::=[ A • ] x ::=[ B C ] ] Example A • B C Second derivation s ::=[ y ::=[ A • B ] c ::=[ C ] ] input.y:4.4: warning: rule useless in parser due to conflicts [-Wother] * src/conflicts.c (rule_conflicts_print): Rename as... (report_rule_expectation_mismatches): this. Move the handling of report_counterexamples to... (conflicts_print): Here. Display this warning when applicable. 2020-06-10 Akim Demaille cex: rename -Wcounterexample as -Wcounterexamples, and support -Wcex Plural vs. singular is always a problem... But we already have conflicts-sr and conflicts-rr, so counterexamples makes more sense than counterexample. Besides, -Wcounterexample will still be accepted as an unambiguous prefix of -Wcounterexamples. Add -Wcex as a convenient alias. While at it, use only "counterexample", never "counter example". * src/complain.h, src/complain.c (Wcounterexample, warning_counterexample): Rename as... (Wcounterexamples, warning_counterexamples): these. (argmatch_warning_docs): Rename -Wcounterexample as -Wcounterexamples. (argmatch_warning_args): Likewise. Add support for -Wcex. Adjust dependencies. 2020-06-09 Akim Demaille api.header.include: document it, and fix its default value While defining api.header.include worked as expected, its default value was incorrectly defined. As a result, by default, the generated parsers still duplicated the content of the generated header instead of including it. * data/skeletons/yacc.c (api.header.include): Fix its default value. * tests/output.at: Check it. * doc/bison.texi (%define Summary): Document api.header.include. While at it, move the definition of api.namespace at the proper place. 2020-06-07 Akim Demaille cex: color the counterexamples Use colors to show the counterexamples and the derivations in color, to highlight their structure. Align the outputs, and add i18n support. Reduce width by using a one-space separator instead of two-space. From Example A • B C First derivation s ::=[ a ::=[ A • ] x ::=[ B C ] ] Second derivation s ::=[ y ::=[ A • B ] c ::=[ C ] ] to Example A • B C First derivation s ::=[ a ::=[ A • ] x ::=[ B C ] ] Example A • B C Second derivation s ::=[ y ::=[ A • B ] c ::=[ C ] ] with colors. * data/bison-default.css (cex-dot, cex-0, cex-1, cex-2, cex-3, cex-4) (cex-5, cex-6, cex-7, cex-step, cex-leaf): New. * src/derivation.c (derivation_print_styled_impl): New. (derivation_print, derivation_print_leaves): Use it. * src/counterexample.c: Reformat the output. * tests/counterexample.at: Adjust. 2020-06-07 Akim Demaille cex: enforce case for tokens/nonterminals It's unfortunate that the traditions between formal language theory and Yacc differs, but here, tokens should be upper case, and nonterminals should be lower case. * tests/counterexample.at: Comply with this. 2020-06-07 Akim Demaille cex: reformat the s/r and r/r reports In Bison we refer to "shift/reduce" conflicts, not "shift-reduce" (in Bison 3.6.3 186 occurrences vs 15). Enforce consistency on this. Instead of "spending" a second line for each conflict to report the lookaheads, put that on the same line as the type of conflict. Also, prefer "token" to "symbol". Maybe we should even prefer "lookahead". While at it, enable internationalization, with plurals where appropriate. As a consequence, instead of Shift-Reduce Conflict: 6: 3 b: . %empty 6: 6 d: c . A On Symbol: A display Shift/reduce conflict on token A: 6: 3 b: . %empty 6: 6 d: c . A * NEWS, doc/bison.texi, src/conflicts.c: Spell it "shift/reduce", not "shift-reduce". * src/counterexample.c (counterexample_report_shift_reduce) (counterexample_report_reduce_reduce): Reformat and internationalize output. * tests/counterexample.at: Adjust expectations. 2020-06-07 Akim Demaille style: fix syntax-check issues * src/counterexample.c, src/files.c, src/files.h, src/lssi.c, * src/state-item.c: here. 2020-06-07 Akim Demaille all: show the rules in comments before the user actions For instance, in the case of Bison's own parser: - case 40: + case 40: /* grammar_declaration: "%code" "identifier" "{...}" */ { muscle_percent_code_grow ((yyvsp[-1].ID), (yylsp[-1]), translate_code_braceless ((yyvsp[0].BRACED_CODE), (yylsp[0])), (yylsp[0])); code_scanner_last_string_free (); } break; * data/skeletons/c.m4: Modified. * data/skeletons/d.m4: Modified. * data/skeletons/java.m4: Modified. * src/output.c (output_escaped): New. (quoted_output): Use it, and rename as... (output_quoted): this. Adjust dependencies. (rule_output): New. (user_actions_output): Use it. * data/skeletons/c.m4, data/skeletons/d.m4, data/skeletons/java.m4 (b4_case): Add support for $3, an optional comment. 2020-06-06 Akim Demaille CI: use GCC10 on ppc too We were still using GCC9, because GCC10 was failing. * .travis.yml (PPC64le): Use GCC10. While at it, use -O2 instead of -O3: it's certainly nicer for the CPUs, and allows to test different sets of compiler flags (we use -O3 in several other configurations). 2020-06-06 Akim Demaille examples: fix missing includes * examples/c/bistromathic/parse.y: Use abort rather than assert so that the "unused result" warning is silenced even with -DNDEBUG. 2020-06-03 Akim Demaille warnings: fix -Wmissing-prototypes issues * src/counterexample.c, src/lssi.c, src/parse-simulation.c, * src/state-item.c: Here. 2020-06-03 Akim Demaille Merge maint into HEAD * upstream/maint: maint: post-release administrivia version 3.6.3 build: check -Wmissing-prototypes tests: show logs c++: fix printing of state number on streams 2020-06-03 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2020-06-03 Akim Demaille version 3.6.3 * NEWS: Record release date. 2020-06-01 Akim Demaille build: check -Wstrict-aliasing * configure.ac (warn_common): Add -Wstrict-aliasing. 2020-06-01 Akim Demaille doc: using asan * README-hacking.md: here. 2020-06-01 Akim Demaille style: fix includes * src/fixits.c: Follow our usual pattern. * src/scan-code.l, src/scan-gram.l, src/scan-skel.l: Prefer "" to include src/ headers. * README-hacking.md: Document the pattern. 2020-06-01 Akim Demaille lists: fix various issues with the use of gnulib's list First, we should avoid code such as gl_list_iterator_t it = gl_list_iterator (deriv->children); derivation *child = NULL; while (gl_list_iterator_next (&it, (const void **) &child, NULL)) { derivation_print (child, f); because of -Wstrict-aliasing (whose job is to catch type-punning issues). See https://lists.gnu.org/r/bug-bison/2020-05/msg00039.html. Rather we need gl_list_iterator_t it = gl_list_iterator (deriv->children); const void **p = NULL; while (gl_list_iterator_next (&it, &p, NULL)) { derivation *child = (derivation *) p; derivation_print (child, f); Second, list iterators actually have destructors. Even though they are noop in the case of linked-lists, we should use them. Let's address both issues with typed wrappers (such as derivation_list_next) that take care of both issues, and besides allow to scope the iterators within the loop: derivation *child; for (gl_list_iterator_t it = gl_list_iterator (deriv->children); derivation_list_next (&it, &child); ) { derivation_print (child, f); * src/derivation.h, src/derivation.c (derivation_list_next): New. Use it where appropriate. * src/counterexample.c (search_state_list_next): New. Use it where appropriate. * src/parse-simulation.h, src/parse-simulation.c * src/state-item.h (state_item_list_next): New. Use it where appropriate. 2020-06-01 Akim Demaille build: check -Wmissing-prototypes pstate_clear is lacking a prototype. Reported by Ryan https://lists.gnu.org/r/bug-bison/2020-05/msg00101.html Besides, none of the C examples were compiled with the warning flags. * configure.ac (warn_c): Add -Wmissing-prototypes. * data/skeletons/yacc.c (pstate_clear): Make it static. * examples/local.mk (TEST_CFLAGS): New. * examples/c/bistromathic/local.mk, examples/c/calc/local.mk, * examples/c/lexcalc/local.mk, examples/c/mfcalc/local.mk, * examples/c/pushcalc/local.mk, examples/c/reccalc/local.mk, * examples/c/rpcalc/local.mk: Use it. GCC's warn_unused_result is not silenced by a cast to void, so we have to "use" scanf's result. https://gcc.gnu.org/bugzilla/show_bug.cgi?id=25509 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66425 Flex generated code produces too many warnings, including things such as, with ICC: examples/c/lexcalc/scan.c(1088): error #1682: implicit conversion of a 64-bit integral type to a smaller integral type (potential portability problem) 2259 YY_INPUT( (&YY_CURRENT_BUFFER_LVALUE->yy_ch_buf[number_to_move]), 2260 ^ 2261 2262 I am tired of trying to fix Flex's output. The project does not seem maintained. We ought to avoid it. So, for the time being, don't try to enable warnings with Flex. * examples/c/bistromathic/parse.y, examples/c/reccalc/scan.l: Fix warnings. * doc/bison.texi: Discard scanf's return value to defeat -Werror=unused-result. 2020-05-31 Akim Demaille gnulib: update 2020-05-25 Akim Demaille style: make item_index a truly different type from item_number See previous commit. * src/gram.h (item_index): Make it unsigned. Fix remaiming issues. 2020-05-25 Vincent Imbimbo style: decouple different uses of item_number item_number is used for elements of ritem as well as indices into ritem which is fairly confusing. Introduce item_index to represent indices into ritem. * src/gram.h (item_index): Introduce it for ritem indices. * src/closure.h, src/closure.c, src/ielr.c, src/lr0.c, * src/print-graph.c, src/state.h, src/state.h: Replace uses of item_number with item_index where appropriate. 2020-05-24 Joshua Watt bison: add command line option to map file prefixes Teaches bison about a new command line option, --file-prefix-map OLD=NEW (based on the -ffile-prefix-map option from GCC) which causes it to replace and file path of OLD in the text of the output file with NEW, mainly for header guards and comments. The primary use of this is to make builds reproducible with different input paths, and in particular the debugging information produced when the source code is compiled. For example, a distro may know that the bison source code will be located at "/usr/src/bison" and thus can generate bison files that are reproducible with the following command: bison --output=/build/bison/parse.c -d --file-prefix-map=/build/bison/=/usr/src/bison/ parse.y Importantly, this will change the header guards and #line directives from: #ifndef YY_BUILD_BISON_PARSE_H #line 100 "/build/bison/parse.h" to #ifndef YY_USR_SRC_BISON_PARSE_H #line 100 "/usr/src/bison/parse.h" which is reproducible. See https://lists.gnu.org/r/bison-patches/2020-05/msg00016.html * src/files.h, src/files.c (spec_mapped_header_file) (mapped_dir_prefix, map_file_name, add_prefix_map): New. * src/getargs.c (-M, --file-prefix-map): New option. * src/output.c (prepare): Define b4_mapped_dir_prefix and b4_spec_header_file. * src/scan-skel.l (@ofile@): Output the mapped file name. * data/skeletons/glr.c, data/skeletons/glr.cc, * data/skeletons/lalr1.cc, data/skeletons/location.cc, * data/skeletons/yacc.c: Adjust. * doc/bison.texi: Document. * tests/input.at, tests/output.at: Check. 2020-05-24 Akim Demaille tests: fix expectations Should have been part of 1ec93ca2a2b4718b5d94871475520a2688b4c5c8. * tests/counterexample.at: here. 2020-05-23 Akim Demaille cex: clean the display of conflicted symbols Instead of `On Symbols: {b,c,}`, display `On Symbols: b, c`. * src/counterexample.c (counterexample_report_reduce_reduce): We don't need braces. Use commas as a separator, not a terminator. * tests/counterexample.at: Adjust. 2020-05-23 Akim Demaille tests: show logs * examples/c/bistromathic/bistromathic.test, examples/test: here. 2020-05-23 Akim Demaille c++: fix printing of state number on streams Avoid this kind of display: LAC: checking lookahead identifier: R4 R3 G^B S5 * data/skeletons/lalr1.cc: Convert state_t to int before printing it. 2020-05-23 Akim Demaille c++: fix printing of state number on streams Avoid this kind of display: LAC: checking lookahead identifier: R4 R3 G^B S5 * data/skeletons/lalr1.cc: Convert state_t to int before printing it. 2020-05-23 Vincent Imbimbo cex: fix pruning crash Fixes a crash on Cim's grammar. https://lists.gnu.org/r/bison-patches/2020-05/msg00107.html * src/state-item.c (prune_disabled_paths): Prune forward and backwards paths in seperate passes. (prune_forward, prune_backward): New. (disable_state_item): Change function argument from state_item_number to state_item. (state_items_report): Add disabling to graph print-out. * src/conflicts.c (find_state_item_number, report_state_counterexamples): Add SI_DISABLED checks. 2020-05-23 Akim Demaille regen 2020-05-23 Akim Demaille kinds: use the symbol kinds where applicable Instead of generating switch statements with numbers, let's use the symbol kinds. Not only is this more readable, it also makes reading diff easier, as a change in symbol numbers won't have such a large effect on the implementation of symbol actions. * data/skeletons/bison.m4 (_b4_symbol_case): Use the symbol kind rather than its number. 2020-05-23 Akim Demaille kinds: also define the possibly qualified symbol kinds * data/skeletons/bison.m4 (b4_symbol_kind): Rename as... (b4_symbol_kind_base): this. (b4_symbol_kind): New, for fully qualified kind name. * data/skeletons/lalr1.cc (b4_symbol_kind): New. Adjust to use b4_symbol_kind where appropriate. * src/parse-gram.h, src/parse-gram.c: regen. 2020-05-23 Akim Demaille m4: simplify useless quotation * data/skeletons/bison.m4: The result of b4_symbol is "quoted" already, no need for m4_expand. 2020-05-23 Akim Demaille m4: use m4_shift2 etc. * data/skeletons/bison.m4 (m4_shift4): New. Use them where applicable. 2020-05-23 Akim Demaille tests: show logs * examples/c/bistromathic/bistromathic.test, examples/test: here. 2020-05-23 Akim Demaille style: spell fixes * Makefile.am (codespell): New. * doc/bison.texi: Fixes. Use @option for options. * src/lssi.c, src/lssi.h, src/parse-simulation.h, src/state-item.c: Fix spellos. 2020-05-23 Akim Demaille style: rename user_token_number as code This should have been done in 3.6, but I wanted to avoid introducing conflicts into Vincent's work on counterexamples. It turns out it's completely orthogonal. * data/README.md, data/skeletons/bison.m4, data/skeletons/c++.m4, * data/skeletons/c.m4, data/skeletons/glr.c, data/skeletons/java.m4, * data/skeletons/lalr1.d, data/skeletons/lalr1.java, * data/skeletons/variant.hh, data/skeletons/yacc.c, src/conflicts.c, * src/derives.c, src/gram.c, src/gram.h, src/output.c, * src/parse-gram.c, src/parse-gram.y, src/print-xml.c, src/print.c, * src/reader.c, src/symtab.c, src/symtab.h, tests/input.at, * tests/types.at: s/user_token_number/code/g. Plus minor changes. 2020-05-22 Akim Demaille Merge maint into master * upstream/maint: fix generated comments traces: provide a means to get short m4 traces traces: show the full m4 invocation 2020-05-22 Vincent Imbimbo cex: replace state-item data structures * src/state-item.h: Add trans, prods, and revs edges to state-item struct. (si_trans, si_revs, si_prods_lookup): Remove. * src/state-item.c, src/lssi.c, src/parse-simulation.c, * src/counterexample.c: Update state-item API usage accordingly. 2020-05-22 Vincent Imbimbo cex: fix bad reference counting * src/counterexample.c (si_bfs_free): Fix reference_count decrementing. 2020-05-22 Vincent Imbimbo cex: fix miscellaneous leaks * src/counterexample.c (unifying_counterexample): Always free stage3result when it exists. * src/conflicts.c (report_state_counterexamples): free leaked bitset. * src/state-item.c (prune_disabled_paths): free leaked queue. 2020-05-22 Vincent Imbimbo cex: fix counterexample leak * src/counterexample.c (free_counterexample): New. Free counterexamples after printing. 2020-05-22 Vincent Imbimbo cex: fix lssi leaks * src/lssi.c (shortest_path_from_start): Free the root of shortest_path_from_start search. Free eligible bitset. 2020-05-22 Vincent Imbimbo cex: fix parse state leaks * src/parse_simulation.c: Fix bug in parse_state_free. Free new_root when simulate_reduction generates zero states. * src/parse-simulation.c, src/parse-simulation.h (parse_state_list, parse_state_list_append): New. * src/parse-simulation.c, src/parse-simulation.h, * src/counterexample.c: Replace all uses of lists of parse states and appends to parse_state_lists with the new API. 2020-05-22 Vincent Imbimbo cex: derivation reference counting * src/derivation.h, src/derivation.c: Make derivation struct opaque. Add derivation_list type for clarity. (derivation_list_new): New. (derivation_list_append): New. (derivation_list_prepend): New. (derivation_new_leaf): New constructor for derivations with no children. * src/counterexample.c, src/parse-simulation.c, * src/parse-simulation.h: Replace uses of gl_list_t containing derivations with derivation_list and its API. Replace calls of dervation_new using null children with derivation_new_leaf. * src/parse-simulation.c: replace ps_chunk and its API with typed versions si_chunk and deriv_chunk. * src/parse-simlation.h, src/parse-simulation.c: Remove parse_state_retain_deriv in favor of derivation reference counting. * src/counterexample.c: Remove search_state_retain_deriv. 2020-05-22 Akim Demaille cex: style changes in parse-simulation * src/parse-simulation.c: Formatting changes. (parse_state_list_new): New. Use it. 2020-05-22 Akim Demaille cex: style: prefer res for returned value * src/lssi.c, src/parse-simulation.c: here. 2020-05-22 Akim Demaille cex: fix memory leaks when there are conflicts * src/counterexample.c (production_step, reduction_step): Release memory of temporary objects. 2020-05-22 Akim Demaille cex: be sure to always reclaim memory put in hashes One call to hash_initialize did not provide a function to free memory. * src/state-item.c (hash_pair_table_create): New. Use it. 2020-05-22 Akim Demaille cex: properly reclaim hash's allocated memory * src/state-item.c: Use hash_free where appropriate. 2020-05-22 Akim Demaille cex: avoid gratuitous heap allocations There's no need to go for the heap when using gnulib's hash module. * src/state-item.c (hash_pair_lookup, hash_pair_remove, state_sym_lookup): Use the heap Luke. That removes a leak from hash_pair_lookup. (init_prods): Use hash_pair_insert instead of duplicating it. 2020-05-22 Vincent Imbimbo cex: fix leaks * src/state-item.c: Various functions were using heap allocated locals and not freeing them. 2020-05-22 Akim Demaille style: use hash_xinsert * gnulib: Update to get hash_xinsert. Use it where appropriate. 2020-05-22 Akim Demaille cex: style changes in state-item * src/state-item.h, src/state-item.c (state_item): Make the state const. (state_item_set): Make it clearer that it works in the state_items global array. 2020-05-22 Akim Demaille cex: stylistic changes * src/counterexample.c: Use 'res' as a variable name for returned value, as elsewhere. Avoid uninitialized variables, especially pointers. Avoid assignment where possible. 2020-05-22 Akim Demaille cex: avoid uninitialized variables * src/counterexample.c (item_rule_bounds): Split into... (item_rule_start, item_rule_end): these. Adjust dependencies. * src/conflicts.c (find_state_item_number): New. Use it to avoid uninitialized variables. 2020-05-22 Akim Demaille cex: isolate missing API from gl_list * src/counterexample.c (list_get_end): New. Use it. Reduce scopes. 2020-05-22 Akim Demaille cex: tests: be robust to variations in time limit reports The CI has "failures" such as (253, "Null nonterminals"): @@ -21,7 +21,7 @@ 3: 3 b: . %empty 3: 4 c: . %empty On Symbols: {A,} -time limit exceeded: 6.000000 +time limit exceeded: 11.000000 First Example c • c A A $end First derivation $accept ::=[ a ::=[ c d ::=[ a ::=[ b ::=[ • ] d ::=[ c A A ] ] ] ] $end ] Second Example c • A $end * tests/counterexample.at (AT_BISON_CHECK_CEX): New. Use it to neutralize differences in timeout values. 2020-05-22 Vincent Imbimbo cex: fix stack overflow * src/parse-simulation.c: Replace reference counting with parse_state_retain everywhere. (free_parse_state): Make this function iterative instead of recursive. Long parse_state chains were causing stack exhaustion. * tests/counterexample.at: Fix expectations. 2020-05-22 Vincent Imbimbo cex: fix crash from zombie result Fixes the SEGV in test 247 (counterexample.at:195): "S/R after first token". * src/counterexample.c: here. * tests/counterexample.at: Fix expectations. 2020-05-22 Akim Demaille cex: fixes, and enable tests * src/counterexample.c, src/derivation.c: Do not output diagnostics on stdout, that's the job of stderr, and the testsuite heavily depend on this. Do not leave trailing spaces in the output. * tests/counterexample.at: Use AT_KEYWORDS. Specify the expected outputs. * tests/local.mk: Add counterexample.at. 2020-05-22 Akim Demaille cex: fix a crash * src/state-item.c (init_state_items): If the rule has no reductions at all, don't read at all in its list of reduced rules. 2020-05-22 Vincent Imbimbo cex: add tests * tests/counterexample.at: New. 2020-05-22 Vincent Imbimbo cex: bind counterexample generation * src/complain.h, src/complain.c: Add support for -Wcounterexample. * src/conflicts.c (report_counterexamples): New. (rule_conflicts_print): Use it when -Wcounterexample is given. * src/getargs.h, src/getargs.c: Add support for --trace=cex. * src/main.c (main): Init and deinit counterexample generation. 2020-05-22 Vincent Imbimbo cex: introduce counterexample search * src/counterexample.h, src/counterexample.c: New. 2020-05-22 Vincent Imbimbo cex: introduce the parse simulator * src/derivation.h, src/derivation.c, * src/parse-simulation.h, src/parse-simulation.c: New. 2020-05-22 Vincent Imbimbo cex: add support for state-item pair graph generation * src/lssi.h, src/lssi.c, src/state-item.h, src/state-item.c: New. 2020-05-22 Akim Demaille cex: add gnulib dependencies * bootstrap.conf (gnulib_modules): Add linked-list. 2020-05-21 Akim Demaille fix generated comments In Bison 3.6.2, the comments with brackets lose their brackets, for improper m4 quotation. * data/skeletons/bison.m4 (b4_gsub): New. * data/skeletons/c-like.m4 (_b4_comment): Use it. * tests/m4.at: Check b4_gsub. 2020-05-21 Akim Demaille traces: provide a means to get short m4 traces Let --trace=m4-early dump all the logs from the start (as --trace=m4 used to do), and have --trace=m4 now start traces only when actually working of the user's grammar. Can make a big difference in the case of small inputs. E.g. $ bison -S tests/testsuite.dir/001/input.m4 tests/testsuite.dir/001/input.y --trace=m4 |& wc 3952 19446 251068 $ bison -S tests/testsuite.dir/001/input.m4 tests/testsuite.dir/001/input.y --trace=m4-early |& wc 19491 131904 1830495 * data/skeletons/traceon.m4: New. * src/getargs.h, src/getargs.c: Introduce --trace=m4-early. * src/output.c (output_skeleton): Adjust for --trace=m4 and --trace=m4-early. 2020-05-21 Akim Demaille traces: show the full m4 invocation Unfortunately the effect of -dV is still position independent. * src/output.c (output_skeleton): here. 2020-05-20 Thomas Petazzoni src: make path to m4 relocatable Commit a4ede8f85b0c9a254fcb01e5888cee1983095669 ("package: make bison a relocatable package") made Bison relocatable, but in fact it still contains one absolute reference: the M4 variable, which points to the M4 program. Let's fix that by using relocate(), see if an M4 binary is available at the relocated location, and otherwise fallback to the original M4 location. See https://lists.gnu.org/r/bison-patches/2020-05/msg00078.html, and https://lists.gnu.org/r/bison-patches/2020-05/msg00087.html. * src/files.h, src/files.c (m4path): New. * src/output.c: Use it. 2020-05-17 Akim Demaille CI: fix PPC recipe 2020-05-17 Akim Demaille Merge branch 'maint' * upstream/maint: maint: post-release administrivia version 3.6.2 tests: improve update-test CI: add GCC 10 and Clang 10 fix: do not emit nested comments todo: update examples: use markdown hyperlinks tests: don't use == to compare const char *... gnulib: update 2020-05-17 Akim Demaille c: more fixes for _Noreturn The previous fix was insufficient. tests/types.at:366: $CXX $CXXFLAGS $CPPFLAGS $LDFLAGS -o test test.cc $LIBS ++ ccache clang++-mp-9.0 -Qunused-arguments -ggdb -Wall -Wextra -Wcast-align -Wchar-subscripts -fparse-all-comments -Wdocumentation -Wformat -Wimplicit-fallthrough -Wnull-dereference -Wno-sign-compare -Wno-tautological-constant-out-of-range-compare -Wpointer-arith -Wshadow -Wwrite-strings -Wextra-semi -Wold-style-cast -Wundefined-func-template -Wweak-vtables -Wunreachable-code -Wundef -pedantic -Wconversion -Wdeprecated -Wsign-compare -Wsign-conversion -Wtautological-constant-out-of-range-compare -fno-color-diagnostics -Wno-keyword-macro -Werror -std=c++98 -I/Users/akim/src/gnu/bison/tests -isystem /opt/gostai/include -isystem /opt/local/include -L/opt/gostai/lib -L/opt/local/lib -o test test.cc /Users/akim/src/gnu/bison/_build/c9d/lib/libbison.a -lintl -Wl,-framework -Wl,CoreFoundation stderr: test.cc:955:1: error: _Noreturn functions are a C11-specific feature [-Werror,-Wc11-extensions] _Noreturn static void ^ test.cc:963:1: error: _Noreturn functions are a C11-specific feature [-Werror,-Wc11-extensions] _Noreturn static void ^ 2 errors generated. * data/skeletons/c.m4 (b4_attribute_define): Do not use _Noreturn at all in C++, clang or not. 2020-05-17 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2020-05-17 Akim Demaille version 3.6.2 * NEWS: Record release date. 2020-05-17 Akim Demaille tests: improve update-test * build-aux/update-test: When given a directory, use the testsuite.log which it contains. Do not accept empty "from"s, as substituting the empty string with something is rarely a good idea. 2020-05-17 Akim Demaille CI: add GCC 10 and Clang 10 * .travis.yml: Here. * tests/input.at, tests/regression.at: Beware of clang's -Wdocumentation. 2020-05-17 Akim Demaille fix: do not emit nested comments With input such as %token yVL_CLOCK "/*verilator sc_clock*/" we generate yVL_CLOCK = 610, /* "/*verilator sc_clock*/" */ which is invalid since the comment will actually be closed on the first "*/". Let's turn "*/" into "*\/" to avoid this. But GCC will also warn about "/*" inside a comment, so let's "escape" it too. Reported by Huang Rui. https://github.com/akimd/bison/issues/38 * data/skeletons/c-like.m4 (_b4_comment): Escape comment delimiters in comments. * tests/input.at (Torturing the Scanner): Check thes cases. * tests/m4.at: New. 2020-05-16 Akim Demaille c: restore definition of _Noreturn as [[noreturn]] in C++ c.m4 contains a definition of _Noreturn which is modeled after gnulib's lib/_Noreturn.h. The latter was recently changed (b61bf2f0e8bdc1e522ae8e97d57d5625163b42ea) to not using [[noreturn]] at all, because the uses of _Noreturn in gnulib are sometimes incompatible with the rules of [[noreturn]]. As a result glr.cc started to use _Noreturn in C++, which clang refuses (all the glr.cc tests currently fail with Clang++). * data/skeletons/c.m4 (b4_attribute_define): Restore the definition of _Noreturn as [[noreturn]] in modern C++. The generated code uses _Noreturn in places where [[noreturn]] is valid. 2020-05-16 Akim Demaille examples: don't promote unchecked function calls * etc/bench.pl.in, examples/c/bistromathic/parse.y, * examples/c/calc/calc.y, examples/c/pushcalc/calc.y: Check scanf's return value. * doc/bison.texi: Likewise, but only for the second example, to avoid cluttering the very simple case. 2020-05-16 Akim Demaille gnulib: update 2020-05-15 Akim Demaille todo: update 2020-05-14 Akim Demaille examples: use markdown hyperlinks * examples/c++/README.md, examples/c++/calc++/README.md, * examples/c/README.md: here. 2020-05-14 Akim Demaille tests: don't use == to compare const char *... Reported by Dagobert Michelsen. https://lists.gnu.org/r/bug-bison/2020-05/msg00091.html * tests/c++.at: here. 2020-05-14 Akim Demaille gnulib: update 2020-05-13 Akim Demaille tests: improve update-test * build-aux/update-test: When given a directory, use the testsuite.log which it contains. Do not accept empty "from"s, as substituting the empty string with something is rarely a good idea. 2020-05-10 Akim Demaille Merge branch maint * maint: news: update maint: post-release administrivia version 3.6.1 c++: style: reorder generated code c++: provide yy::parser::symbol_type::name c++: make parser::symbol_name public examples: beware of ~/.inputrc build: also provide lzip compressed tarballs style: minor fixes yacc.c: restore ansi-c compatibility 2020-05-10 Akim Demaille news: update 2020-05-10 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2020-05-10 Akim Demaille version 3.6.1 * NEWS: Record release date. 2020-05-10 Akim Demaille bench: add support to randomize the order of execution It's amazing how much the order matters. To a point that many of these benches are meaningless. For instance (some of the benches where run with `make -C benches/latest rand BENCHFLAGS=--benchmark_min_time=3`): compiler: g++ -std=c++11 -O2 0. %define nofinal 1. -------------------------------------------------- Benchmark Time CPU Iterations -------------------------------------------------- BM_y0 1543 ns 1541 ns 441660 BM_y1 1521 ns 1520 ns 456535 -------------------------------------------------- BM_y0 1531 ns 1530 ns 440584 BM_y1 1512 ns 1511 ns 457591 -------------------------------------------------- BM_y0 1539 ns 1538 ns 2749330 BM_y1 1516 ns 1515 ns 2771500 -------------------------------------------------- BM_y0 1571 ns 1570 ns 2600782 BM_y1 1542 ns 1541 ns 2708349 -------------------------------------------------- BM_y0 1530 ns 1529 ns 2670363 BM_y1 1519 ns 1518 ns 2764096 -------------------------------------------------- Benchmark Time CPU Iterations -------------------------------------------------- BM_y1 1529 ns 1528 ns 451937 BM_y0 1508 ns 1507 ns 453944 -------------------------------------------------- BM_y1 1525 ns 1524 ns 2750684 BM_y0 1516 ns 1515 ns 2794034 -------------------------------------------------- BM_y1 1526 ns 1525 ns 2749620 BM_y0 1515 ns 1514 ns 2808112 -------------------------------------------------- BM_y1 1524 ns 1523 ns 4475844 BM_y0 1502 ns 1501 ns 4611665 * etc/bench.pl.in: here. 2020-05-10 Akim Demaille bench: use a Makefile This makes it much easier to toy with the benchs. * etc/bench.pl.in: Generate a Makefile instead of directly compiling the files. 2020-05-10 Akim Demaille examples: use markdown hyperlinks * examples/c++/README.md, examples/c++/calc++/README.md, * examples/c/README.md: here. 2020-05-10 Akim Demaille don't use stdnoreturn Reported by Paul Eggert. * src/getargs.c: We don't need it anyway, since we use _Noreturn. * data/skeletons/c.m4: While at it, update the definition of _Noreturn stolen from gnulib. 2020-05-10 Akim Demaille c++: style: reorder generated code The implementation of yy::parser::symbol_name is emitted even before the implementation of yy::parser::parser. This makes little sense. * data/skeletons/lalr1.cc (symbol_name): Move its implementation in the same place as in the class definition: after "error" and before "context". 2020-05-10 Akim Demaille c++: provide yy::parser::symbol_type::name * data/skeletons/c++.m4 (yy::parser::basic_symbol::name): New. * data/skeletons/lalr1.cc (yy_print_): Use it. * doc/bison.texi: Document. * tests/c++.at: Check. 2020-05-10 Akim Demaille c++: make parser::symbol_name public Reported by Martin Blais . https://lists.gnu.org/r/help-bison/2020-05/msg00005.html * data/skeletons/lalr1.cc (symbol_name): Make it public. Add a private hidden hook to enable testing of private parts. * tests/local.at (AT_DATA_GRAMMAR_PROLOGUE): Help Emacs find the right language mode. * tests/c++.at (C++ Variant-based Symbols Unit Tests): Check that we can read symbol_name. 2020-05-10 Akim Demaille examples: beware of ~/.inputrc * examples/c/bistromathic/bistromathic.test: here. 2020-05-10 Akim Demaille build: also provide lzip compressed tarballs Suggested by Matias Fonzo . * cfg.mk: Post announcements to bison-announce. * configure.ac: Build lzip packages. * .travis.yml: Build only xz, we don't care about the other formats here. 2020-05-10 Akim Demaille style: minor fixes * examples/c/README.md: here. 2020-05-09 Akim Demaille yacc.c: restore ansi-c compatibility Reported by neok-m4700. https://github.com/akimd/bison/issues/37 * data/skeletons/yacc.c: Don't use // comments. 2020-05-09 Akim Demaille bench: use *.cc for C++ Using *.c is simpler, but triggers annoying warnings with Clang++. * etc/bench.pl.in: Please the dictator. 2020-05-09 Akim Demaille style: minor fixes * examples/c/README.md: here. 2020-05-09 Akim Demaille gnulib: update 2020-05-08 Akim Demaille gnulib: update, and use the attribute module * gnulib: Update. * bootstrap.conf: Use attribute. * src/system.h: Remove macros for attributes. Adjust dependencies. * src/scan-gram.l (DEPRECATED): Rename as... (DEPRECATED_DIRECTIVE): this, to avoid the clash with the DEPRECATED macro. 2020-05-08 Akim Demaille build: also provide lzip compressed tarballs Suggested by Matias Fonzo . * cfg.mk: Post announcements to bison-announce. * configure.ac: Build lzip packages. * .travis.yml: Build only xz, we don't care about the other formats here. 2020-05-08 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2020-05-08 Akim Demaille version 3.6 * NEWS: Record release date. 2020-05-08 Akim Demaille examples: beware of portability issue on Windows Reported by Jannick. https://lists.gnu.org/r/bug-bison/2020-05/msg00040.html https://lists.gnu.org/r/bug-bison/2020-05/msg00066.html * examples/test (diff_opts): Use --strip-trailing-cr if supported, to avoid \n vs. \r\n issues. * examples/c/bistromathic/bistromathic.test: When on MSYS, don't try to check autocompletion. 2020-05-08 Akim Demaille regen 2020-05-08 Akim Demaille doc: fix the generation of the man page When there is no bison.1 at all, the procedure fails. * doc/local.mk (bison.1): Be robust to cold starts. 2020-05-08 Akim Demaille news: prepare for 3.6 2020-05-08 Akim Demaille doc: complete the table of symbols * doc/bison.texi: Add YYEMPTY, YYEOF and YYUNDEF. 2020-05-07 Akim Demaille doc: clarify the glossary item about kinds * doc/bison.texi (Glossary): here. 2020-05-06 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2020-05-06 Akim Demaille version 3.5.94 * NEWS: Record release date. 2020-05-06 Akim Demaille doc: document yypstate_expected_tokens * doc/bison.texi (Push Parser Interface): Here. 2020-05-06 Akim Demaille doc: restructure the push parser documentation I don't think it's fair to have yypstate_new, yypstate_delete, yypush_parse and yypull_parse to have their own section, on par with yyparse and yylex. Let them be in a single section about push parsers. And show new/delete first. * doc/bison.texi (Push Parser Interface): New. Fuse the aforementioned sections into it. 2020-05-06 Akim Demaille all: fix the interface of yyexpected_tokens The user gives yyexpected_tokens a limit: the max number of tokens she wants to hear about. That's because an error message that reports a bazillion of possible tokens is useless. In that case yyexpected_tokens returned 0, so the user would not know if there are too many expected tokens or none (yes, that's possible). There are several ways to tell the user in which situation she's in: - return some E2MANY, a negative value. Then it makes the pattern int argsize = yypcontext_expected_tokens (ctx, arg, ARGS_MAX); if (argsize < 0) return argsize; no longer valid, as for E2MANY (i) the user must generate the error message anyway, and (ii) she should not return E2MANY - return ARGS_MAX + 1. Then it makes it dangerous for the user, as she has to iterate update `min (ARGS_MAX, argsize)`. Returning 0 is definitely simpler and safer for the user, as it tells her "this is not an error, just generate your message without a list of expecting tokens". So let's still return 0, but set arg[0] to the empty token when the list is really empty. * data/skeletons/glr.c, data/skeletons/lalr1.cc, data/skeletons/lalr1.java * data/skeletons/yacc.c (yyexpected_tokens): Put the empty symbol first if there are no possible tokens at all. * examples/c/bistromathic/parse.y: Demonstrate how to use that. 2020-05-05 Akim Demaille examples: fix handling of syntax errors The shell grammar does not allow empty statements in then/else part of an if, but examples/test failed to catch the syntax errors from the script it ran. So exited with success anyway. You would expect 'set -e' to suffice, but with bash 3.2 actually it does not. As a matter of fact, I could find a way to have this behave properly: $ cat test.sh set -e cleanup () { status=$? echo "cleanup: $status" exit $status } trap cleanup 0 1 2 13 15 . $1 s=$? echo "test.sh: $s" exit $s $ cat bistro.test if true; then fi $ /bin/sh ./test.sh ./bistro.test ./bistro.test: line 2: syntax error near unexpected token `fi' cleanup: 0 $ echo $? 0 Remove the set -e (or the trap), and tada, it works... So we have to deal with the error by hand. * examples/test ($exit): Replace with... ($status): this. Preserve the exit status of the test case. * examples/c/bistromathic/bistromathic.test: Fix syntax error. 2020-05-04 Akim Demaille doc: beware of timestamp issues on Haiku On Haiku, help2man is fired on a freshly extracted tarball. Reported by Bruno Haible. https://lists.gnu.org/r/bug-bison/2020-05/msg00055.html * doc/local.mk (bison.1): Be robust to a missing help2man. 2020-05-04 Akim Demaille tests: beware of wchar_t portability issues on AIX https://lists.gnu.org/r/bug-bison/2020-05/msg00050.html Reported by Bruno Haible. * tests/diagnostics.at: here. 2020-05-04 Akim Demaille glr.c: beware of portability issues with PTRDIFF_MAX For instance test 386, "glr.cc api.value.type={double}": types.at:366: $CXX $CXXFLAGS $CPPFLAGS $LDFLAGS -o test test.cc $LIBS stderr: test.cc: In function 'ptrdiff_t yysplitStack(yyGLRStack*, ptrdiff_t)': test.cc:490:4: error: 'PTRDIFF_MAX' was not declared in this scope (PTRDIFF_MAX < SIZE_MAX ? PTRDIFF_MAX : YY_CAST (ptrdiff_t, SIZE_MAX)) ^ test.cc:1805:37: note: in expansion of macro 'YYSIZEMAX' ptrdiff_t half_max_capacity = YYSIZEMAX / 2 / state_size; ^~~~~~~~~ test.cc:490:4: note: suggested alternative: '__PTRDIFF_MAX__' (PTRDIFF_MAX < SIZE_MAX ? PTRDIFF_MAX : YY_CAST (ptrdiff_t, SIZE_MAX)) ^ test.cc:1805:37: note: in expansion of macro 'YYSIZEMAX' ptrdiff_t half_max_capacity = YYSIZEMAX / 2 / state_size; ^~~~~~~~~ The failing tests are using glr.cc only, which I don't understand, the problem is rather in glr.c, so I would expect glr.c tests to also fail. Reported by Bruno Haible. https://lists.gnu.org/archive/html/bug-bison/2020-05/msg00053.html * data/skeletons/yacc.c: Move the block that defines YYPTRDIFF_T/YYPTRDIFF_MAXIMUM, YYSIZE_T/YYSIZE_MAXIMUM, and YYSIZEOF to... * data/skeletons/c.m4 (b4_sizes_types_define): Here. (b4_c99_int_type): Also take care of the #undefinition of short. * data/skeletons/yacc.c, data/skeletons/glr.c: Use b4_sizes_types_define. * data/skeletons/glr.c: Adjust to use YYPTRDIFF_T/YYPTRDIFF_MAXIMUM, YYSIZE_T/YYSIZE_MAXIMUM. 2020-05-04 Akim Demaille todo: update 2020-05-04 Akim Demaille examples: beware of strnlen portability issues One function missing on Solaris 10 Sparc: CCLD examples/c/bistromathic/bistromathic Undefined first referenced symbol in file strnlen examples/c/bistromathic/bistromathic-parse.o ld: fatal: symbol referencing errors. No output written to examples/c/bistromathic/bistromathic Reported by Dagobert Michelsen. https://lists.gnu.org/r/bug-bison/2020-05/msg00048.html * examples/c/bistromathic/parse.y (xstrndup): Don't use strnlen. xstrndup is assembled from gnulib's xstrndup, strndup and strnlen... 2020-05-04 Akim Demaille examples: beware of portability issues with sh's trap On AIX 7.2, when invoking "exit 77", we actually exit with 127. The "cleanup" function, called via trap, received an incorrect exit status, something described in Autoconf's doc. Reported by Bruno Haible. https://lists.gnu.org/archive/html/bug-bison/2020-05/msg00029.html https://lists.gnu.org/archive/html/bug-bison/2020-05/msg00047.html * examples/test (skip): New. * examples/c/bistromathic/bistromathic.test, * examples/c/reccalc/reccalc.test: Use it, to ensure $? is set to 77 when the trap is called. 2020-05-04 Akim Demaille tests: beware of portability issues with diff -u AIX 7.1 supports diff -u, but its output does not match the expected one. Reported by Bruno Haible. https://lists.gnu.org/r/bug-bison/2020-05/msg00049.html * tests/atlocal.in (DIFF_U_WORKS): New. * tests/local.at (AT_DIFF_U_CHECK): New. * tests/existing.at (_AT_TEST_EXISTING_GRAMMAR): Use AT_DIFF_U_CHECK. 2020-05-03 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2020-05-03 Akim Demaille version 3.5.93 * NEWS: Record release date. 2020-05-03 Akim Demaille news: update for 3.5.93 2020-05-03 Akim Demaille gnulib: update 2020-05-03 Akim Demaille tests: really skip tricky multichar test on Cygwin In Autotest, anything outside AT_SETUP/AT_CLEANUP is discarded. * tests/diagnostics.at (AT_TEST): Accept a skip-if test. Use it to skip on cygwin. 2020-05-03 Akim Demaille bistromathic: beware of portability issues of readline on AIX Readline may emit escape sequences before the prompt. Reported by Bruno Haible. https://lists.gnu.org/r/platform-testers/2020-05/msg00001.html. * examples/c/bistromathic/bistromathic.test: Trust readline _only_ if we get what we expect on some reference computation. 2020-05-03 Akim Demaille examples: beware of portability issues with cmp As someone wrote nearly 20 years ago in Autoconf's documentation, don't use cmp to compare text files, but diff. https://git.savannah.gnu.org/cgit/autoconf.git/commit/?id=abad4f0576a7dc361e5385e19c7681449103cdb1 Reported by Jannick. * examples/test: Use diff, not cmp. 2020-05-03 Akim Demaille build: fix warnings (shown on IRIX) Appearing on IRIX with gcc -mabi=n32. Reported by Bruno Haible. https://lists.gnu.org/r/bug-bison/2020-05/msg00039.html * examples/c++/variant-11.yy, examples/c/bistromathic/parse.y: Don't give chars to isdigit, cast them to unsigned char before. * src/complain.c: Use c_isdigit. * src/fixits.c (fixits_run): Avoid casts. * src/lalr.c (goto_print): Use %zu for a size_t. 2020-05-03 Akim Demaille c++: be compatible with the pre-3.6 way to get a symbol's kind Reported by Pramod Kumbhar. https://lists.gnu.org/r/bug-bison/2020-05/msg00025.html * data/skeletons/c++.m4: here. 2020-05-03 Akim Demaille bistromathic: beware of portability issues with strndup Reported by Dagobert Michelsen. https://lists.gnu.org/r/bug-bison/2020-05/msg00026.html * examples/c/bistromathic/parse.y (xstrndup): New. Use it. 2020-05-03 Akim Demaille flex: fix incorrect use of Automake conditional AM_CONDITIONAL does _not_ define a shell variable... Reported privately by Denis Excoffier. * configure.ac (LEX_CXX_WORKS): Fix its definition. 2020-05-03 Bruno Haible package: fix a link error on IRIX https://lists.gnu.org/r/bug-bison/2020-05/msg00035.html * src/local.mk (src_bison_LDADD): Mention libbison.a before, not after, the system libraries. 2020-05-03 Akim Demaille bistromathic: beware of portability of readline Don't try to build bistromathic if we don't have readline. Reported by Bruno Haible. https://lists.gnu.org/r/bug-bison/2020-05/msg00028.html * configure.ac (ENABLE_BISTROMATHIC): New. * examples/c/bistromathic/local.mk: Use it. * examples/c/bistromathic/bistromathic.test: Exit 77 for skip. 2020-05-03 Akim Demaille tests: beware of portability issues of sh "foo || bar" does not invoke bar on AIX 7.2 when foo does not exist. It just dies. Reported by Bruno Haible. https://lists.gnu.org/r/bug-bison/2020-05/msg00029.html * examples/c/reccalc/reccalc.test: Check for seq in a subshell. 2020-05-03 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2020-05-03 Akim Demaille version 3.5.92 * NEWS: Record release date. 2020-05-03 Akim Demaille news: update for 3.5.92 2020-05-03 Akim Demaille gnulib: update 2020-05-03 Akim Demaille java: demonstrate push parsers * data/skeletons/lalr1.java (Location): Make it a static class. (Lexer.yylex, Lexer.getLVal, Lexer.getStartPos, Lexer.getEndPos): These are not needed in push parsers. * examples/java/calc/Calc.y: Demonstrate push parsers in the Java. * doc/bison.texi: Push parsers have been supported for a long time, remove incorrect statements stating the opposite. 2020-05-03 Akim Demaille doc: clarify what a location is Reported by Arthur Schwarz https://lists.gnu.org/r/help-bison/2013-12/msg00009.html * doc/bison.texi (Location Type): here. 2020-05-03 Akim Demaille tests: beware of mbswidth portability issues Shy away from these issues on Cygwin. Reported Denis Excoffier. https://lists.gnu.org/r/bug-bison/2020-05/msg00003.html * tests/diagnostics.at (Tabulations and multibyte characters): Split in two. 2020-05-03 Akim Demaille examples: beware of intl portability issues Reported by Horst von Brand. https://lists.gnu.org/r/bug-bison/2020-04/msg00033.html * examples/c/bistromathic/Makefile: libintl might not be needed, but libm probably is. * examples/c/bistromathic/parse.y: Include locale.h. 2020-05-03 Akim Demaille examples: beware of portability issues with readline On OpenBSD 6.5, the prompt is repeated, but not the actual command line... Don't try to cope with that. Reported by Bruno Haible. https://lists.gnu.org/r/bug-bison/2020-05/msg00015.html * examples/c/bistromathic/bistromathic.test: Skip when readline behave this way. 2020-05-03 Akim Demaille examples: beware of the portability of flex --header-file The option --header was introduced in version 2.5.6. The option --header-file was introduced in version 2.6.4. Reported by Bruno Haible. https://lists.gnu.org/r/bug-bison/2020-05/msg00013.html So use --header, and do bother with versions that don't support it. * m4/flex.m4: Check whether flex supports --header. * configure.ac (FLEX_WORKS, FLEX_CXX_WORKS): Set to false if it doesn't. * * examples/c/reccalc/local.mk, examples/c/reccalc/Makefile: Use --header rather than --header-file. 2020-05-03 Akim Demaille c++: provide backward compatibility on by_type To write unit tests for their scanners, some users depended on symbol_type::token(): Lexer lex("12345"); symbol_type t = lex.nextToken(); assert(t.token() == token::INTLIT); assert(t.value.as() == 12345); But symbol_type::token() was removed in Bison 3.5 because it relied on a conversion table. So users had to find other patterns, such as assert(t.type_get() == by_type(token::INTLIT).type_get()); which relies on several private implementation details. As part of transitioning from "token type" to "token kind", and making this a public and documented interface, "by_type" was renamed "by_kind" and "type_get()" was renamed as "kind()". The latter had backward compatibility mechanisms, not the former. In Bison 3.6 none of this should be used, but rather assert(t.kind() == symbol_kind::S_INTLIT); Reported by Pramod Kumbhar. https://lists.gnu.org/r/bug-bison/2020-05/msg00012.html * data/skeletons/c++.m4 (by_type): Make it an alias to by_kind. 2020-05-02 Akim Demaille yacc.c: improve formatting of the generated code * data/skeletons/yacc.c (yy_reduce_print): here. 2020-05-02 Akim Demaille doc: java supports push parsers since 3.0 (2013-07-25) * doc/bison.texi: Clarify this. 2020-05-02 Akim Demaille java: fix coding style I don't plan to fix everything in one go. But this was in the way of the next commit. * data/skeletons/lalr1.java: Avoid space before parens. * tests/java.at: Adjust. 2020-05-02 Akim Demaille style: comment changes * tests/java.at: here. 2020-05-02 Akim Demaille todo: more 2020-05-02 Akim Demaille style: more documentation about errs Suggested by Angelo Borsotti. https://lists.gnu.org/r/bug-bison/2014-02/msg00003.html * src/state.h: here. 2020-05-01 Akim Demaille doc: document the exit status Suggested by Alexandre Duret-Lutz. https://lists.gnu.org/r/bug-bison/2013-09/msg00015.html * doc/bison.texi (Invocation): Here. 2020-05-01 Akim Demaille java: add missing i18n requests * data/skeletons/lalr1.java (reportSyntaxError): Here. 2020-05-01 Akim Demaille java: style: fix coding style of yyerror/reportSyntaxError * data/skeletons/lalr1.java: here. 2020-05-01 Akim Demaille java: avoid useless work * data/skeletons/lalr1.java (yySymbolPrint): Avoid the computation of the argument if useless. While at it, fix Java coding style. 2020-05-01 Akim Demaille java: comment changes * data/skeletons/lalr1.java, examples/java/calc/Calc.y: here. 2020-05-01 Akim Demaille news: make it more consistent * NEWS: Use the same pattern for titles. 2020-05-01 Akim Demaille c++: use modern idioms to make classes non-copyable Reported by Don Macpherson. https://lists.gnu.org/r/bug-bison/2019-05/msg00015.html https://github.com/akimd/bison/issues/36 * data/skeletons/lalr1.cc, data/skeletons/stack.hh, * data/skeletons/variant.hh: Delete the copy-ctor and the copy operator. 2020-04-30 Akim Demaille yacc.c: avoid the use of a temporary * data/skeletons/yacc.c: Use YYLLOC_DEFAULT directly with the final destination. 2020-04-29 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2020-04-29 Akim Demaille version 3.5.91 * NEWS: Record release date. 2020-04-29 Akim Demaille build: fix syntax-check issues * cfg.mk: We do want to gettextize the examples. * po/POTFILES.in: Adjust. 2020-04-29 Akim Demaille gnulib: update 2020-04-29 Akim Demaille doc: document YYEOF, YYUNDEF and YYerror * doc/bison.texi (Special Tokens): New. * examples/c/bistromathic/parse.y: Formatting changes. 2020-04-29 Akim Demaille package: fix distcheck Bison emits strings to translate in the generated code, for builtin tokens. So they appear only in generated parsers, which are not shipped, so they are not in the src tree, so we cannot use them in our POTFILE. Except src/parse-gram.c, which is in the source tree. And even in the git repo. But to avoid useless diffs in the repo, we do not keep the src/parse-gram.c _with_ the #lines. This is done in a dist-hook which regenerates src/parse-gram.c when we run "make dist". Unfortunately, then, update-po traverses the whole tree and sees that the location of the strings to translate in src/parse-gram.c have changed, so the bison.pot is to be updated. And that is not possible in the "make dist" which is run within "make distcheck" (not the one preparing the dist for distcheck, the one run by distcheck to check that a distributed tarball can build a tarball) because then the src tree is read-only. So let's not put src/parse-gram.c in the POTFILE, and expose these strings to gettextize by hand. * src/i18n-strings.c: New. * po/POTFILES.in: Add it, and remove src/parse-gram.c. 2020-04-29 Akim Demaille style: avoid gettextize warnings * src/scan-gram.l, src/scan-skel.l: Help it see the start and end of "character literals". 2020-04-29 Akim Demaille tests: beware of portability of readline * examples/test: here. 2020-04-28 Akim Demaille yacc.c: install backward compatibility for YYERRCODE Some people have been using that symbol. Some even have #defined it themselves. https://lists.gnu.org/r/bison-patches/2020-04/msg00138.html Let's provide backward compatibility, having it point to YYUNDEF, so that an error message is generated. * data/skeletons/yacc.c (YYERRCODE): New, at the exact same location it was defined before. 2020-04-28 Akim Demaille style: c++: s/type/kind/ where appropriate These are internal details. `type_get ()` is still there to ensure backward compatibility, `kind ()` being the modern way. * data/skeletons/c++.m4 (by_type, by_type::type): Rename as... (by_kind, by_kind::kind_): this. Adjust dependencies. 2020-04-28 Akim Demaille java: clean up the definition of token kinds From public interface Lexer { /* Token kinds. */ /** Token number, to be returned by the scanner. */ static final int YYEOF = 0; /** Token number, to be returned by the scanner. */ static final int YYERRCODE = 256; /** Token number, to be returned by the scanner. */ static final int YYUNDEF = 257; /** Token number, to be returned by the scanner. */ static final int BANG = 258; ... /** Deprecated, use b4_symbol(0, id) instead. */ public static final int EOF = YYEOF; to public interface Lexer { /* Token kinds. */ /** Token "end of file", to be returned by the scanner. */ static final int YYEOF = 0; /** Token error, to be returned by the scanner. */ static final int YYerror = 256; /** Token "invalid token", to be returned by the scanner. */ static final int YYUNDEF = 257; /** Token "!", to be returned by the scanner. */ static final int BANG = 258; ... /** Deprecated, use YYEOF instead. */ public static final int EOF = YYEOF; * data/skeletons/java.m4 (b4_token_enum): Display the symbol's tag in comment. * data/skeletons/lalr1.java: Address overquotation issue. * examples/java/calc/Calc.y, examples/java/simple/Calc.y: Use YYEOF, not EOF. 2020-04-28 Akim Demaille error: rename the error token from YYERRCODE to YYerror See https://lists.gnu.org/r/bison-patches/2020-04/msg00162.html. * data/skeletons/bison.m4, data/skeletons/c.m4, data/skeletons/glr.cc, * data/skeletons/lalr1.java, doc/bison.texi, * examples/c/bistromathic/parse.y, src/scan-gram.l, src/symtab.c (YYERRCODE): Rename as... (YYerror): this. Adjust dependencies. 2020-04-27 Akim Demaille dogfooding: use YYERRCODE in our scanner * src/scan-gram.l: Use it. * tests/input.at: Adjust. 2020-04-27 Akim Demaille scanner: avoid spurious errors about empty character literals On an invalid character literal such as "'\777'" we used to produce two errors: input.y:2.9-12: error: invalid number after \-escape: 777 input.y:2.8-13: error: empty character literal Get rid of the second one. * src/scan-gram.l (STRING_GROW_ESCAPE): New. * tests/input.at: Adjust. 2020-04-27 Akim Demaille scanner: bad character literals are errors * src/scan-gram.l: These are errors, not warnings. * tests/input.at: Adjust. 2020-04-27 Akim Demaille regen 2020-04-26 Akim Demaille todo: update 2020-04-26 Akim Demaille all: don't emit an error message when the scanner returns YYERRCODE I'm quite pleased to see that the tricky case of glr.c was already prepared by the changes to support syntax_error exceptions. Better yet, it is actually syntax_error that becomes a special case of the general pattern: make yytoken be YYERRCODE. * data/skeletons/glr.c (YYFAULTYTOK): Remove the now useless (Basil) Faulty token. Instead, use the error token. * data/skeletons/lalr1.d, data/skeletons/lalr1.java: When computing the action, first check the case of the error token. * tests/calc.at: Check cases for the error token symbols before and after it. 2020-04-26 Akim Demaille c: don't emit an error message when the scanner returns YYERRCODE * data/skeletons/yacc.c (yyparse): When the scanner returns YYERRCODE, go directly to error recovery (yyerrlab1). However, don't keep the error token as lookahead, that token is too special. * data/skeletons/lalr1.cc: Likewise. * examples/c/bistromathic/parse.y (yylex): Use that feature to report nicely invalid characters. * examples/c/bistromathic/bistromathic.test: Check that. * examples/test: Neutralize gratuitous differences such as rule position. * tests/calc.at: Check that case in C only. The other case seem to be working, but that's an illusion that the next commit will address (in fact, they can enter endless loops, and report the error several times anyway). 2020-04-26 Akim Demaille examples: bistromathic: demonstrate error recovery * examples/c/bistromathic/parse.y: here. * examples/c/bistromathic/bistromathic.test: Check it. Included a stupid case where the error is actually ignored. 2020-04-26 Akim Demaille examples: bistromathic: when quitting, close the current line When the user ctrl-d the line, we left the cursor not at col 0. Let's fix that. This revealed a few short-comings in the testing framework. * examples/test (run): Also display the diffs. And support -n. * examples/c/bistromathic/bistromathic.test * examples/c/bistromathic/parse.y 2020-04-26 Akim Demaille examples: bistromathic: comment changes * examples/c/bistromathic/parse.y: here. 2020-04-26 Akim Demaille doc: hacking tricks * README-hacking.md: Here. 2020-04-26 Akim Demaille c++: make valid to print the empty symbol * data/skeletons/lalr1.cc (yy_print_): here. 2020-04-26 Akim Demaille c++: always define symbol_name * data/skeletons/lalr1.cc (symbol_name): Always define it, even when it's actually yytname which is used. 2020-04-26 Akim Demaille c++: fix a few style issues * data/skeletons/lalr1.cc (yystack_print_, yy_reduce_print_): Add missing const. (yystack_print_): Rename as... (yy_stack_print_): this. * data/skeletons/glr.cc (yy_symbol_value_print_, yy_symbol_print_): Add missing const. 2020-04-26 Akim Demaille all: prefer YYERRCODE to YYERROR We will not keep YYERRCODE anyway, it causes backward compatibility issues. So as a first step, let all the skeletons use that name, until we have a better one. * data/skeletons/bison.m4, data/skeletons/glr.c, * data/skeletons/glr.cc, data/skeletons/lalr1.cc, * data/skeletons/lalr1.d, data/skeletons/lalr1.java, * data/skeletons/yacc.c, doc/bison.texi, tests/headers.at, * tests/input.at: here. 2020-04-26 Akim Demaille style: glr.c: clarify * data/skeletons/glr.c: Make the code a bit clearer. 2020-04-26 Akim Demaille style: prefer b4_has_translations_if * data/skeletons/glr.c, data/skeletons/yacc.c: here. 2020-04-26 Akim Demaille style: glr.c: fix indentation issue * data/skeletons/glr.c (yyparse): here. 2020-04-26 Akim Demaille style: fix a few remaining 'type' instead of 'kind' * data/skeletons/glr.c, data/skeletons/yacc.c (YY_SYMBOL_PRINT): Here. 2020-04-26 Akim Demaille skeletons: make the warning about implementation details clearer * data/skeletons/bison.m4 (b4_disclaimer): Here. * data/skeletons/lalr1.d, data/skeletons/lalr1.java: Use it. 2020-04-25 Akim Demaille style: c: fix a few minor issues about indentation of cpp directives * README-hacking.md: More about cpp. * data/skeletons/c.m4, data/skeletons/yacc.c: Style changes. 2020-04-25 Akim Demaille bench: store in benches/012 rather than in benches/12 * etc/bench.pl.in ($basedir): New. Format $count with a least three digits. 2020-04-25 Akim Demaille bench: minor improvements * etc/bench.pl.in: Don't force parse.error=detailed Use a simpler way to display the pseudo %bison directive. (&bench_with_gbenchmark): Give details about the compiler. 2020-04-25 Akim Demaille style: clarify #endif We could try to avoid the weird "#if 1", but then the indentation of the inner #if would be wrong. Let' keep it this way. * data/skeletons/yacc.c: here. Also, avoid sticking the comment to the directive. 2020-04-25 Akim Demaille style: minor fixes * data/skeletons/bison.m4, doc/bison.texi: Spell check. * examples/c/bistromathic/parse.y (N_): Remove, now useless. 2020-04-24 Akim Demaille examples: bistromathic: shorten token description * examples/c/bistromathic/parse.y: "number" is enough. * doc/bison.texi: Likewise. 2020-04-24 Akim Demaille examples: bistromathic: demonstrate internationalization Currently it was only using stubs. Let's actually translate the strings using gettext. * examples/c/bistromathic/local.mk: Define LOCALEDIR, BISON_LOCALEDIR and link with libintl. * examples/c/bistromathic/parse.y: Use them. Remove useless includes. Take ENABLE_NLS into account. (error_format_string): New. (yyreport_syntax_error): Rewrite to rely on a format string, which is more appropriate for internationalization. * examples/c/bistromathic/Makefile: We no longer use Flex. We need readline and intl. * doc/bison.texi: Point to bistromathic for a better option for internationalization. * po/POTFILES.in: Add bistromathic. 2020-04-24 Akim Demaille todo: update for YYERRCODE 2020-04-24 Akim Demaille regen 2020-04-24 Akim Demaille diagnostics: fix a typo * src/complain.c: here. 2020-04-20 Akim Demaille c, c++: provide a default definition for N_ In C/C++, N_ is a no-op. Define it if the user didn't. Suggested by Frank Heckenbach. https://lists.gnu.org/r/bug-bison/2020-04/msg00010.html * src/output.c (prepare_symbol_names): Rename has_translations as has_translations_flag. * data/skeletons/bison.m4 (b4_has_translations_if): New. * data/skeletons/java.m4 (b4_trans): Use it. * data/skeletons/glr.c, data/skeletons/lalr1.cc, data/skeletons/yacc.c (N_): Provide a default definition. 2020-04-19 Akim Demaille i18n: also look in src/parse-gram.c Some strings appears in the generated file only, e.g., translation of "end of file". So don't forget to go and see there. 2020-04-19 Akim Demaille style: fix comments * data/skeletons/glr.c, data/skeletons/lalr1.cc, * data/skeletons/yacc.c: here. 2020-04-19 Akim Demaille tokens: clean up the translation of special symbols * src/output.c (prepare_symbol_names): Don't play tricks with the symbols, it's quite too late. (has_translations): Move to... * src/symtab.c: here. (symbols_pack): Use it to enable translation for special symbols. 2020-04-19 Akim Demaille news: fix typo Reported by Frank Heckenbach. 2020-04-19 Akim Demaille tests: beware of portability issues with wc On macOS, wc -l always prepends the result with a tab, even when fed by stdin. But anyway, we should have used `grep -c -v`, which appears to be portable according to Autoconf's "Limitations of Usual Tools" section. Reported by Denis Excoffier. https://lists.gnu.org/r/bug-bison/2020-04/msg00009.html * tests/calc.at (_AT_CHECK_CALC): Use grep's -c instead. 2020-04-18 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2020-04-18 Akim Demaille version 3.5.90 * NEWS: Record release date. 2020-04-18 Akim Demaille examples: beware of readline on macOS macOS' version of readline does not repeat stdin on stdout in non-interactive mode, contrary to the current version of GNU readline. * examples/test: Add support for strip_prompt. * examples/c/bistromathic/bistromathic.test (strip_prompt): Set it when needed. Early exit when needed. 2020-04-18 Akim Demaille c++: give public access to the symbol kind symbol_type::token () was removed: it returned the token kind of a symbol. To do that, one needs to convert from the symbol kind to the token kind, which requires a table. This broke some users' unit tests for scanners, see https://lists.gnu.org/r/bug-bison/2020-01/msg00001.html https://lists.gnu.org/r/bug-bison/2020-03/msg00020.html https://lists.gnu.org/r/help-bison/2020-04/msg00005.html Instead of making this possible again, let's check the symbol's kind instead. So give proper access to a symbol's kind. That feature existed, undocumented, as 'type_get()'. Let's rename this as 'kind()'. * data/skeletons/c++.m4, data/skeletons/glr.cc, * data/skeletons/lalr1.cc (type_get): Rename as... (kind): This. (type_get): Install a backward compatibility alias. * doc/bison.texi (Complete Symbols): Document symbol_type and symbol_type::kind. 2020-04-17 Akim Demaille doc: token_kind_type in C++ * data/skeletons/c++.m4: Define the old names in terms on the new ones, instead of the converse. * doc/bison.texi (C++ Parser Interface): Be more extensive about token_kind_type. 2020-04-16 Akim Demaille doc: updates for 3.6 * doc/bison.texi: More s/token type/token kind/. * NEWS: Update. 2020-04-16 Akim Demaille skeletons: use symbol(-2, kind) Not all the symbols have a fixed symbol code. UNDEF's one is fixed: -2. * data/skeletons/glr.c, data/skeletons/lalr1.cc, data/skeletons/lalr1.d, * data/skeletons/yacc.c: here. 2020-04-16 Akim Demaille style: comments changes about error handling * data/skeletons/glr.c, data/skeletons/lalr1.cc, data/skeletons/lalr1.d, * data/skeletons/lalr1.java, data/skeletons/yacc.c: here. * data/skeletons/lalr1.cc: Reduce scope. 2020-04-14 Akim Demaille examples: bistro: don't be lazy with switch * examples/c/bistromathic/parse.y (yylex): Use the switch to discriminate all the cases. 2020-04-13 Akim Demaille doc: spell check * doc/bison.texi, NEWS, README-hacking.md: here. And elsewhere. 2020-04-13 Akim Demaille gnulib: update 2020-04-13 Akim Demaille doc: more about the coding style * README-hacking.md: here. (Troubleshooting): New. 2020-04-13 Akim Demaille java: promote YYEOF rather that Lexer.EOF * doc/bison.texi: here. * data/skeletons/lalr1.java: Use YYEOF. 2020-04-13 Akim Demaille java: fix names * data/skeletons/lalr1.java (yySymbolPrint): There are no pointers here, remove the `p` suffix. Use the appropriate type for locations. 2020-04-13 Akim Demaille doc: java: SymbolKind, etc. Why didn't I think about this before??? symbolName should be a method of SymbolKind. * data/skeletons/lalr1.java (YYParser::yysymbolName): Move as... * data/skeletons/java.m4 (SymbolKind::getName): this. Make the table a static final table, not a local variable. Adjust dependencies. * doc/bison.texi (Java Parser Interface): Document i18n. (Java Parser Context Interface): Document SymbolKind. * examples/java/calc/Calc.y, tests/local.at: Adjust. 2020-04-13 Akim Demaille style: java: get closer to the Java style * examples/java/calc/Calc.y, examples/java/simple/Calc.y: here. 2020-04-13 Akim Demaille doc: c++: document parser::context * doc/bison.texi (C++ Parser Context): New. * data/skeletons/lalr1.cc (parser::yysymbol_name): Rename as... (parser::symbol_name): this. (A Complete C++ Example): Promote LAC, now that we have it. Promote parse.error detailed over verbose. * examples/c++/calc++/calc++.test, tests/local.at: Adjust. 2020-04-13 Akim Demaille doc: promote YYEOF * NEWS (Deep overhaul of the symbol and token kinds): New. * doc/bison.texi: Promote YYEOF over "0" in scanners. (Token Decl): No longer show YYEOF here, it now works by default. (Token I18n): More details about YYEOF here. (Calc++): Just use YYEOF. 2020-04-13 Akim Demaille d: put YYEMPTY in the TokenKind * data/skeletons/d.m4, data/skeletons/lalr1.d (b4_token_enums): Rename YYTokenType as TokenKind. Define YYEMPTY. * examples/d/calc.y, tests/calc.at, tests/scanner.at: Adjust. 2020-04-13 Akim Demaille regen 2020-04-13 Akim Demaille c, c++: also define YYEMPTY in yytoken_kind_t I have been hesitating a lot before doing it ---after all the user must not use this kind, so what's the point of showing it in yytoken_kind_t. And eventually I chose to play it safe with the typing system and make it possible to use yytoken_kind_t for all the tokens, even the "empty token". * data/skeletons/c.m4: Give an id and a tag to YYEMPTY. (b4_token_enums): Define YYEMPTY. * data/skeletons/c++.m4 (b4_token_enums): Define YYEMPTY. * data/skeletons/glr.c, data/skeletons/glr.cc, data/skeletons/yacc.c: (YYEMPTY): Remove. Use b4_symbol(-2, id) instead. 2020-04-12 Akim Demaille doc: use "code", not "number", for token (and symbol) kinds "Number" is too much about arithmethics. "Code" conveys better the "enum" nature of token kinds. And of symbol kinds. * doc/bison.texi: Here. 2020-04-12 Akim Demaille doc: promote yytoken_kind_t, not yytokentype * data/skeletons/c.m4 (yytoken_kind_t): New. * data/skeletons/c++.m4, data/skeletons/lalr1.cc (yysymbol_kind_type): New. * examples/c/lexcalc/parse.y, examples/c/reccalc/parse.y, * tests/regression.at: Use them. * doc/bison.texi: Replace "enum yytokentype" by "yytoken_kind_t". (api.token.raw): Explain that it forces "yytoken_kind_t" to coincide with "yysymbol_kind_t". (Calling Convention): Mention YYEOF. (Table of Symbols): Add entries for "yytoken_kind_t" and "yysymbol_kind_t". (Glossary): Add entries for "Kind", "Token kind" and "Symbol kind". 2020-04-12 Akim Demaille doc: document yypcontext_t, and api.symbol.prefix * doc/bison.texi (%define Summary): Document api.symbol.prefix. (Syntax Error Reporting Function): Document yypcontext_t, yypcontext_location, yypcontext_token, yypcontext_expected_tokens, and yysymbol_kind_t. 2020-04-12 Akim Demaille c: rename yyexpected_tokens as yypcontext_expected_tokens The user should think of yypcontext fields as accessible only via yypcontext_* functions. So let's rename yyexpected_tokens to reflect that. Let's _not_ rename yyreport_syntax_error, as the user may define this function, and is not allowed to access directly the fields of yypcontext_t: she *must* use the "accessors". This is comparable to the case of C++/Java where the user defines parser::report_syntax_error, not parser::context::report_syntax_error. * data/skeletons/glr.c, data/skeletons/yacc.c (yyexpected_tokens): Rename as... (yypcontext_expected_tokens): this. Adjust dependencies. 2020-04-12 Akim Demaille skeletons: clarify the tag of special tokens From GRAM_EOF = 0, /* $end */ GRAM_ERRCODE = 1, /* error */ GRAM_UNDEF = 2, /* $undefined */ to GRAM_EOF = 0, /* "end of file" */ GRAM_ERRCODE = 1, /* error */ GRAM_UNDEF = 2, /* "invalid token" */ * src/output.c (symbol_tag): New. Use it to pass the token names and the symbol tags to the skeletons. * tests/input.at: Adjust. 2020-04-12 Akim Demaille skeletons: use "invalid token" instead of "$undefined" * src/output.c (prepare_symbol_names): Also handle undeftoken. * tests/actions.at, tests/calc.at, tests/regression.at: Adjust. 2020-04-12 Akim Demaille skeletons: make the eof token translatable if i18n is enabled * src/output.c (has_translations): New. (prepare_symbol_names): Translate endtoken if the user already translated tokens. * examples/c/bistromathic/parse.y, src/parse-gram.y: Simplify. 2020-04-12 Akim Demaille skeletons: use "end of file" instead of "$end" The name "$end" is nice in the report, in particular it avoids that pointed-rules (aka items) be too long. It also helps keeping them "standard". But it is bad in error messages, we should report "end of file" (or maybe "end of input", this is debatable). So, unless the user already defined the alias for the error token herself, make it "end of file". It should even be translated if the user already translated some tokens, so that there is now no strong reason to redefine the $end token. * src/output.c (prepare_symbol_names): Issue "end of file" instead of "$end". * data/skeletons/lalr1.java (yytnamerr_): Remove the renaming hack. * build-aux/update-test: Accept files with names containing a "+", such as c++.at. * tests/actions.at, tests/c++.at, tests/conflicts.at, * tests/glr-regression.at, tests/regression.at, tests/skeletons.at: Adjust. 2020-04-12 Akim Demaille diagnostics: replace "user token number" by "token code" Yet, don't change the structure identifier to avoid introducing conflicts in Vincent Imbimbo's PR (which, amusingly enough, is about conflicts). * src/symtab.c: here. * tests/diagnostics.at, tests/input.at: Adjust. 2020-04-12 Akim Demaille c++: remove the yy prefix from some functions yy::parser features a parse() function, not a yyparse() one. * data/skeletons/lalr1.cc (yyreport_syntax_error) (context::yyexpected_tokens): Rename as... (report_syntax_error, context::expected_tokens): these. 2020-04-12 Akim Demaille tokens: properly define the YYEOF token kind Currently EOF is handled in an adhoc way, with a #define YYEOF 0 in the implementation file. As a result, the user has to define her own EOF token if she wants to use it, which is a pity. Give the $end token a visible kind name, YYEOF. Except that in C, where enums are not scoped, we would have collisions between all the definitions of YYEOFs in the header files, so in C, make it EOF. * data/skeletons/c.m4 (YYEOF): Override its name to avoid collisions. Unless the user already gave it a different name. * data/skeletons/glr.c (YYEOF): Remove. Use ]b4_symbol(0, [id])[ instead. Add support for "pre_epilogue", for glr.cc. * data/skeletons/glr.cc: Remove dead code (never emitted #undefs). * data/skeletons/yacc.c * src/parse-gram.c * src/reader.c * src/symtab.c * tests/actions.at * tests/input.at 2020-04-12 Akim Demaille tokens: define the "$undefined" token kind * data/skeletons/bison.m4 (b4_symbol_token_kind): Give a definition to $undefined. (b4_token_visible_if): $undefined has an id. * src/output.c (prepare_symbol_definitions): Stop lying: $undefined _is_ a token. * tests/input.at: Adjust. 2020-04-12 Akim Demaille tokens: properly define the "error" token kind There are people out there that do use YYERRCODE (the token kind of the error token). See for instance https://github.com/borbolla-automation/SPC_Machines/blob/3812012bb782bfdfe7b325950a35cd337925fcad/unixODBC-2.3.2/Drivers/nn/yylex.c. Currently, YYERRCODE is defined by yacc.c in an adhoc way as a #define in the *.c file only. It belongs with the other token kinds. YYERRCODE is not a nice name, it does not fit in our naming scheme. YYERROR would be more logical, but it collides with the YYERROR macro. Shall we keep the same name in all the skeletons? Besides, to avoid collisions in C, we need to apply the api prefix: YYERRCODE is actually ERRCODE. This is not needed in the other languages. * data/skeletons/bison.m4 (b4_symbol_token_kind): New. Map the error token to "YYERRCODE". * data/skeletons/yacc.c (YYERRCODE): Don't define it, it's handled by... * src/output.c (prepare_symbol_definitions): this. * tests/input.at (Redefining the error token): Check it. 2020-04-12 Akim Demaille tokens: style: minor fixes * data/skeletons/bison.m4 (b4_symbol_kind): Dispatch on the UNDEF token number rather than its name. * data/skeletons/c++.m4, data/skeletons/c.m4, data/skeletons/java.m4: Comment changes. 2020-04-12 Akim Demaille glr.cc: remove dead code * data/skeletons/glr.cc: here. 2020-04-12 Akim Demaille c++: fix generated headers A forthcoming commit (tokens: properly define the "error" token kind) revealed a problem in the C++ generated headers: they are not self-contained. With this file: %language "c++" %define api.value.type variant %code { static int yylex (yy::parser::semantic_type *lvalp); } %token X %% exp: X { printf ("x\n"); } ; %% void yy::parser::error (const std::string& m) { std::cerr << m << '\n'; } static int yylex (yy::parser::semantic_type *lvalp) { static int const input[] = {yy::parser::token::X, 0}; static int toknum = 0; return input[toknum++]; } int main (int argc, char const* argv[]) { yy::parser p; return p.parse (); } the generated header fails to compile cleanly (foo.cc just #includes the generated header): $ clang++-mp-9.0 -c -Wundefined-func-template foo.cc In file included from foo.cc:1: bar.tab.hh:550:12: warning: instantiation of function 'yy::parser::basic_symbol::basic_symbol' required here, but no definition is available [-Wundefined-func-template] struct symbol_type : basic_symbol ^ bar.tab.hh:436:7: note: forward declaration of template entity is here basic_symbol (basic_symbol&& that); ^ bar.tab.hh:550:12: note: add an explicit instantiation declaration to suppress this warning if 'yy::parser::basic_symbol::basic_symbol' is explicitly instantiated in another translation unit struct symbol_type : basic_symbol ^ 1 warning generated. * data/skeletons/c++.m4 (b4_public_types_define): Move the implementation of the basic_symbol move-ctor to... (b4_public_types_define): here, its declaration. * tests/headers.at (Sane headers): Use a declared token so that the corresponding token constructor is declared. Which triggers the aforementioned issue. 2020-04-10 Akim Demaille style: rename YYNOMEM as YYENOMEM This is clearer. * data/skeletons/glr.c, data/skeletons/yacc.c (YYNOMEM): Rename as... (YYENOMEM): here. 2020-04-10 Akim Demaille todo: update 2020-04-10 Akim Demaille c++: improvements on symbol kinds Instead of /// (Internal) symbol kind. enum symbol_kind_type { YYNTOKENS = 5, ///< Number of tokens. YYSYMBOL_YYEMPTY = -2, YYSYMBOL_YYEOF = 0, // END_OF_FILE YYSYMBOL_YYERROR = 1, // error YYSYMBOL_YYUNDEF = 2, // $undefined YYSYMBOL_TEXT = 3, // TEXT YYSYMBOL_NUMBER = 4, // NUMBER YYSYMBOL_YYACCEPT = 5, // $accept YYSYMBOL_result = 6, // result YYSYMBOL_list = 7, // list YYSYMBOL_item = 8 // item }; generate /// Symbol kinds. struct symbol_kind { enum symbol_kind_type { YYNTOKENS = 5, ///< Number of tokens. S_YYEMPTY = -2, S_YYEOF = 0, // END_OF_FILE S_YYERROR = 1, // error S_YYUNDEF = 2, // $undefined S_TEXT = 3, // TEXT S_NUMBER = 4, // NUMBER S_YYACCEPT = 5, // $accept S_result = 6, // result S_list = 7, // list S_item = 8 // item }; }; * data/skeletons/c++.m4 (api.symbol.prefix): Define to S_. Adjust all the uses. (b4_public_types_declare): Nest the enum inside 'struct symbol_kind'. * data/skeletons/glr.cc, data/skeletons/lalr1.cc, * tests/headers.at, tests/local.at: Adjust. 2020-04-10 Akim Demaille d: improvements on symbol kinds public enum SymbolKind { S_YYEMPTY = -2, /* No symbol. */ S_YYEOF = 0, /* $end */ S_YYERROR = 1, /* error */ S_YYUNDEF = 2, /* $undefined */ S_EQ = 3, /* "=" */ * data/skeletons/d.m4 (api.symbol.prefix): Default to S_. Output the symbol kind definitions with a comment. 2020-04-10 Akim Demaille symbols: minor fixes * data/skeletons/bison.m4 (b4_symbol_kind): Series of _ are useless, one is enough. * data/skeletons/c.m4 (b4_token_enum): Fix overquoting. 2020-04-07 Akim Demaille skeletons: introduce api.symbol.prefix * data/skeletons/bison.m4 (b4_symbol_prefix): New. (b4_symbol_kind): Use it. * data/skeletons/c++.m4, data/skeletons/c.m4, data/skeletons/d.m4 * data/skeletons/java.m4 (api.symbol.prefix): Provide a default value. * data/skeletons/glr.c, data/skeletons/glr.cc, data/skeletons/lalr1.cc, * data/skeletons/lalr1.d, data/skeletons/lalr1.java, data/skeletons/yacc.c: Adjust: use b4_symbol_prefix instead of YYSYMBOL_. 2020-04-07 Akim Demaille java: also emit documenting comments for symbol kinds * data/skeletons/java.m4 (b4_symbol_enum): here. And stop defined YYSYMBOL_YYEMPTY, we no longer use it. 2020-04-06 Akim Demaille todo: update * TODO (YYERRCODE): Remove, handled by YYSYMBOL_ERROR. 2020-04-06 Akim Demaille skeletons: beware not to use yyarg when it's null Reported by Adrian Vogelsgesang. * data/skeletons/glr.c, data/skeletons/lalr1.cc, * data/skeletons/lalr1.java, data/skeletons/yacc.c: Here. 2020-04-06 Akim Demaille java: document new features * data/skeletons/lalr1.java: More comments. (Context.EMPTY): Remove. * doc/bison.texi (Java Parser Context Interface): New. 2020-04-06 Akim Demaille java: prefer null to YYSYMBOL_YYEMPTY That's one nice benefit from using enums. * data/skeletons/lalr1.java (YYSYMBOL_YYEMPTY): No longer define it. Use 'null' instead. * examples/java/calc/Calc.y, tests/local.at: Adjust. 2020-04-06 Akim Demaille java: rename Lexer.yyreportSyntaxError as reportSyntaxError * data/skeletons/lalr1.java: here. * examples/java/calc/Calc.y, tests/local.at: Adjust. 2020-04-06 Akim Demaille java: use getExpectedTokens, not yyexpectedTokens * data/skeletons/lalr1.java, examples/java/calc/Calc.y, tests/local.at: here. 2020-04-06 Akim Demaille java: style: fix coding style * data/skeletons/java.m4: Indent by two. * data/skeletons/lalr1.java (yynnts_): Remove. (yyfinal_, yyntokens_, yylast_, yyempty_): Rename as... (YYFINAL_, YYNTOKENS_, YYLAST_, YYEMPTY_): these, they are constants. 2020-04-06 Akim Demaille c: make the symbol kind definition nicer to read From enum yysymbol_kind_t { YYSYMBOL_YYEMPTY = -2, YYSYMBOL_YYEOF = 0, YYSYMBOL_YYERROR = 1, YYSYMBOL_YYUNDEF = 2, to enum yysymbol_kind_t { YYSYMBOL_YYEMPTY = -2, YYSYMBOL_YYEOF = 0, /* "end of file" */ YYSYMBOL_YYERROR = 1, /* error */ YYSYMBOL_YYUNDEF = 2, /* $undefined */ * data/skeletons/bison.m4 (b4_last_symbol): New. (b4_symbol_enum, b4_symbol_enums): Reformat the output. * data/skeletons/c.m4 2020-04-06 Akim Demaille c: make the token kind definition nicer to read From enum gram_tokentype { GRAM_EOF = 0, STRING = 3, TSTRING = 4, PERCENT_TOKEN = 5, To enum gram_tokentype { GRAM_EOF = 0, /* "end of file" */ STRING = 3, /* "string" */ TSTRING = 4, /* "translatable string" */ PERCENT_TOKEN = 5, /* "%token" */ * data/skeletons/bison.m4 (b4_last_enum_token): New. * data/skeletons/c.m4 (b4_token_enum, b4_token_enums): Show the corresponding symbol. 2020-04-06 Akim Demaille c: make the generated YYSTYPE nicer to read From union GRAM_STYPE { /* precedence_declarator */ assoc precedence_declarator; /* "string" */ char* STRING; /* "translatable string" */ char* TSTRING; /* "{...}" */ char* BRACED_CODE; /* "%?{...}" */ to union GRAM_STYPE { assoc precedence_declarator; /* precedence_declarator */ char* STRING; /* "string" */ char* TSTRING; /* "translatable string" */ char* BRACED_CODE; /* "{...}" */ * data/skeletons/c.m4 (b4_symbol_type_register): Use m4_format to align the comments. * src/parse-gram.h: Regen. 2020-04-05 Akim Demaille regen 2020-04-05 Akim Demaille bison: use consistently "token kind", not "token type" * src/output.c, src/reader.c, src/scan-gram.l, src/tables.c: here. 2020-04-05 Akim Demaille skeletons: use consistently "kind" instead of "type" in the code * data/skeletons/bison.m4, data/skeletons/c++.m4, data/skeletons/c.m4, * data/skeletons/glr.cc, data/skeletons/lalr1.cc, * data/skeletons/lalr1.d, data/skeletons/lalr1.java: Refer to the "kind" of a symbol, not its "type", where appropriate. 2020-04-05 Akim Demaille doc: refer to the token kind rather than the token type * doc/bison.texi: Replace occurrences of "token type" with "token kind". Stop referring to the "macro definitions" of the token kinds, just name them "definitions". 2020-04-05 Akim Demaille m4: we don't need undef_token_number It's replaced by YYSYMBOL_YYUNDEF. 2020-04-05 Akim Demaille m4: rename b4_symbol_sid as b4_symbol_kind * data/skeletons/bison.m4, data/skeletons/c++.m4, data/skeletons/c.m4, * data/skeletons/d.m4, data/skeletons/java.m4 (b4_symbol_sid): Rename as... (b4_symbol_kind): this. Adjust dependencies. * data/README.md: Document the kind. 2020-04-05 Akim Demaille d, java: rename SymbolType as SymbolKind See https://lists.gnu.org/r/bison-patches/2020-04/msg00031.html. * data/skeletons/d.m4, data/skeletons/lalr1.d, * data/skeletons/java.m4, data/skeletons/lalr1.java (SymbolType): Rename as... (SymbolKind): this. Adjust dependencies. 2020-04-05 Akim Demaille c, c++: rename yysymbol_type_t as yysymbol_kind_t See https://lists.gnu.org/r/bison-patches/2020-04/msg00031.html * data/skeletons/c.m4, data/skeletons/glr.c, data/skeletons/yacc.c (yysymbol_type_t): Rename as... (yysymbol_kind_t): this. Adjust dependencies. * data/skeletons/c++.m4, data/skeletons/glr.cc, data/skeletons/lalr1.cc (symbol_type_type): Rename as... (symbol_kind_type): this. Adjust dependencies. 2020-04-05 Akim Demaille doc: remove obsolete release instructions * README-hacking.md: here. 2020-04-05 Akim Demaille Merge branch 'maint' * maint: maint: post-release administrivia version 3.5.4 examples: reccalc: really compile cleanly in C99 news: announce that Bison 3.6 drops YYERROR_VERBOSE news: update for 3.5.4 style: fix spellos typo: succesful -> successful package: improve the readme java: check and fix support for api.token.raw java: style: prefer 'int[] foo' to 'int foo[]' build: fix syntax-check issues tests: recheck: work properly when the test suite was interrupted doc: c++: promote api.token.raw build: fix compatibility with old compilers examples: reccalc: compile cleanly in C99 2020-04-05 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2020-04-05 Akim Demaille version 3.5.4 * NEWS: Record release date. 2020-04-05 Akim Demaille news: update the yyreport_syntax_error example * examples/c/bistromathic/parse.y, tests/local.at (yyreport_syntax_error): Fix use of YYSYMBOL_YYEMPTY. * NEWS: Update. 2020-04-05 Akim Demaille todo: update 2020-04-05 Akim Demaille style: rename yysyntax_error_arguments as yy_syntax_error_arguments It's a private implementation detail. * NEWS, data/skeletons/glr.c, data/skeletons/lalr1.cc, * data/skeletons/yacc.c, doc/bison.texi: here. 2020-04-05 Akim Demaille examples: reccalc: really compile cleanly in C99 The previous fix does not suffice, and actually managed to make things worse by defining yyscan_t twice in parse.y... Reported by kencu. https://trac.macports.org/ticket/59927#comment:29 * examples/c/reccalc/parse.y (yyscan_t): Define it with the same guards as used by Flex. 2020-04-04 Akim Demaille c: rename yyparse_context_t as yypcontext_t The first name is too long. We already have `yypstate`, so `yypcontext` is ok. We are also migrating to using `*_t` for our types. * NEWS, data/skeletons/glr.c, data/skeletons/yacc.c, doc/bison.texi, * examples/c/bistromathic/parse.y, src/parse-gram.y, tests/local.at: (yyparse_context_t, yyparse_context_location, yyparse_context_token): Rename as... (yypcontext_t, yypcontext_location, yypcontext_token): these. 2020-04-04 Akim Demaille readme: more about the coding style 2020-04-04 Akim Demaille java: fixes in SymbolType Reported by Paolo Bonzini. https://github.com/akimd/bison/pull/34#issuecomment-609029634 * data/skeletons/java.m4 (SymbolType): Use 'final' where possible. (get): Rewrite on top of an array instead of a switch. 2020-04-04 Akim Demaille java: use SymbolType The Java enums are very different from the C model. As a consequence, one cannot "build" an enum directly from an integer, we must retrieve it. That's the purpose of the SymbolType.get class method. * data/skeletons/java.m4 (b4_symbol_enum, b4_case_code_symbol) (b4_declare_symbol_enum): New. * data/skeletons/lalr1.java: Use SymbolType, SymbolType.YYSYMBOL_YYEMPTY, etc. * examples/java/calc/Calc.y, tests/local.at: Adjust. 2020-04-04 Akim Demaille examples: java: use explicit token identifiers * examples/java/calc/Calc.y: Declare all the tokens, so that we are compatibile with api.token.raw. * examples/java/calc/Calc.test: Adjust. 2020-04-04 Akim Demaille news: announce that Bison 3.6 drops YYERROR_VERBOSE * NEWS: here. 2020-04-04 Akim Demaille news: update for 3.5.4 2020-04-04 Akim Demaille style: fix spellos * src/complain.c, src/print.c, src/print-xml.c, src/symtab.h: here. 2020-04-04 Adrian Vogelsgesang typo: succesful -> successful * tests/calc.at: Here. 2020-04-04 Akim Demaille package: improve the readme * README: Describe what Bison is. 2020-04-04 Akim Demaille java: check and fix support for api.token.raw * tests/local.at (AT_LANG_MATCH, AT_YYERROR_DECLARE(java)) (AT_YYERROR_DECLARE_EXTERN(java), AT_PARSER_CLASS): New. (AT_MAIN_DEFINE(java)): Use AT_PARSER_CLASS. * tests/scanner.at: Add a test for Java. * data/skeletons/lalr1.java (yytranslate_): Cast the result. 2020-04-04 Akim Demaille d: use the SymbolType enum for symbol kinds * data/skeletons/d.m4 (b4_symbol_enum, b4_declare_symbol_enum): New. * data/skeletons/lalr1.d: Use them. Use SymbolType, SymbolType.YYSYMBOL_YYEMPTY etc. where appropriate. (undef_token_, token_number_type, yy_error_token_): Remove. 2020-04-04 Akim Demaille java: style: prefer 'int[] foo' to 'int foo[]' * data/skeletons/java.m4 (b4_typed_parser_table_define): Here. 2020-04-04 Akim Demaille build: fix syntax-check issues * src/system.h, tests/local.mk: Fix indentation. 2020-04-02 Akim Demaille tests: recheck: work properly when the test suite was interrupted * tests/local.mk (recheck): Look at the per-test logs, not the overall log, which, when interrupted, contains only information about... the tests that passed. 2020-04-02 Akim Demaille doc: c++: promote api.token.raw * doc/bison.texi (Calc++ Parser): Here. 2020-04-02 Akim Demaille build: fix compatibility with old compilers GCC 4.2 dies with src/InadequacyList.c: In function 'InadequacyList__new_conflict': src/InadequacyList.c:37: error: #pragma GCC diagnostic not allowed inside functions src/InadequacyList.c:37: error: #pragma GCC diagnostic not allowed inside functions src/InadequacyList.c:40: error: #pragma GCC diagnostic not allowed inside functions Reported by Evan Lavelle. See https://lists.gnu.org/r/bug-bison/2020-03/msg00021.html and https://trac.macports.org/ticket/59927. * src/system.h (GCC_VERSION): New. Use it to control IGNORE_TYPE_LIMITS_BEGIN and IGNORE_TYPE_LIMITS_END. 2020-04-02 Akim Demaille examples: reccalc: compile cleanly in C99 See https://trac.macports.org/ticket/59927. * examples/c/reccalc/parse.y: C99 does not allow multiple typedefs. 2020-04-01 Akim Demaille c++: replace symbol_number_type with symbol_type_type * data/skeletons/c++.m4, data/skeletons/glr.cc, * data/skeletons/lalr1.cc: here. 2020-04-01 Akim Demaille c++: also use symbol_type_type Because of the insane current implementation of glr.cc, things are a bit nasty. We will rename symbol_number_type as symbol_type_type later, to keep this commit small. * data/skeletons/c++.m4 (b4_declare_symbol_enum): New. Also define YYNTOKENS to avoid type clashes when yyntokens_ was actually defined in another enum. Use it. (symbol_number_type): Be an alias of symbol_type_type. Use YYSYMBOL_YYEMPTY and the like. Use symbol_number_type where appropriate. (empty_symbol): Remove. (yytranslate_): Use symbol_number_type, not token_number_type. * data/skeletons/lalr1.cc: Use symbol_number_type where appropriate. Adjust to the replacement of empty_symbol by YYSYMBOL_YYEMPTY. (yy_error_token_, yy_undef_token_, yyeof_, yyntokens_): Remove. Adjust dependencies. * data/skeletons/glr.cc: Use symbol_number_type where appropriate. Forward definitions of YYSYMBOL_YYEMPTY, etc. to glr.c. * tests/headers.at: Accept YYNTOKENS and other YYSYMBOL_*. * tests/local.at (AT_YYERROR_DEFINE(c++)): Use symbol_number_type. 2020-04-01 Akim Demaille glr.c: remove the yySymbol alias * data/skeletons/glr.c: Use yysymbol_type_t only. 2020-04-01 Akim Demaille regen 2020-04-01 Akim Demaille glr.c, yacc.c: propagate yysymbol_type_t Now that yacc.c and glr.c both know yysymbol_type_t, convert the common routines. * data/skeletons/c.m4 (yydestruct, yy_symbol_value_print) (yy_symbol_print): Use yysymbol_type_t instead of int. * data/skeletons/glr.c: Use yySymbol where appropriate. * data/skeletons/yacc.c (YY_ACCESSING_SYMBOL): New wrapper around yystos. Use it. * tests/local.at (yyreport_syntax_error): Use yysymbol_type_t where appropriate. 2020-04-01 Akim Demaille glr.c: use yysymbol_type_t, YYSYMBOL_YYEOF etc. Apply the same changes as in yacc.c. Now yySymbol and yysymbol_type_t are aliases. We will remove the former later, to avoid cluttering this commit. * data/skeletons/glr.c: Use b4_declare_symbol_enum. Use YYSYMBOL_YYEOF etc. where appropriate. (YYUNDEFTOK, YYTERROR): Remove. (YYTRANSLATE, yySymbol, yyexpected_tokens, yysyntax_error_arguments): Adjust. (yy_accessing_symbol): New. Use it where appropriate. 2020-04-01 Akim Demaille regen 2020-04-01 Akim Demaille yacc.c: fix more errors from make maintainer-check-g++ * data/skeletons/yacc.c (yyexpected_tokens): Use casts where needed. 2020-04-01 Akim Demaille regen 2020-04-01 Akim Demaille yacc.c: revert to not using yysymbol_type_t in the yytranslate table This triggers warnings with several compilers. For instance ICC fills the logs with pages and pages of input.c(477): error: a value of type "int" cannot be used to initialize an entity of type "const yysymbol_type_t={yysymbol_type_t}" 0, 2, 2, 2, 2, 2, 2, 2, 2, 2, ^ input.c(477): error: a value of type "int" cannot be used to initialize an entity of type "const yysymbol_type_t={yysymbol_type_t}" 0, 2, 2, 2, 2, 2, 2, 2, 2, 2, ^ And so does G++9 when compiling yacc.c's (C) output input.c:545:8: error: invalid conversion from 'int' to 'yysymbol_type_t' [-fpermissive] 545 | 0, 5, 9, 2, 2, 2, 2, 2, 2, 2, | ^ | | | int input.c:545:15: error: invalid conversion from 'int' to 'yysymbol_type_t' [-fpermissive] 545 | 0, 5, 9, 2, 2, 2, 2, 2, 2, 2, | ^ | | | int Clang++ is no exception input.c:545:8: error: cannot initialize an array element of type 'const yysymbol_type_t' with an rvalue of type 'int' 0, 5, 9, 2, 2, 2, 2, 2, 2, 2, ^ input.c:545:15: error: cannot initialize an array element of type 'const yysymbol_type_t' with an rvalue of type 'int' 0, 5, 9, 2, 2, 2, 2, 2, 2, 2, ^ At some point we could use yysymbol_type_t's enumerators to define yytranslate. Meanwhile... * data/skeletons/yacc.c (yytranslate): Use the original integral type to define it. (YYTRANSLATE): Cast the result into yysymbol_type_t. 2020-04-01 Akim Demaille regen 2020-04-01 Akim Demaille yysymbol_type_t: always assign an enumerator Currently we define enumerators only for symbols that have an identifier. That rules out tokens such as '+', and nonterminals such as foo-bar and foo.bar. As a consequence we are taking chances: the compiler might compile yysymbol_type_t as too small an integral type for some symbol codes. * data/skeletons/bison.m4 (b4_symbol_sid): Forge a unique symbol identifier for symbols that don't have an ID. 2020-04-01 Akim Demaille bistromathic: use symbol numbers instead of YYTRANSLATE * examples/c/bistromathic/parse.y: here. 2020-04-01 Akim Demaille regen 2020-04-01 Akim Demaille yacc.c: prefer YYSYMBOL_YYERROR to YYSYMBOL_error * data/skeletons/bison.m4 (b4_symbol_sid): Map "error" to YYSYMBOL_YYERROR. * data/skeletons/yacc.c: Adjust. 2020-04-01 Akim Demaille regen 2020-04-01 Akim Demaille yacc.c: also define a symbol number for the empty token This is not only cleaner, it also protects us from mixing signed values (YYEMPTY is #defined as -2) with unsigned types (the yysymbol_type_t enum is typically compiled as a small unsigned). For instance GCC 9: input.c: In function 'yyparse': input.c:1107:7: error: conversion to 'unsigned int' from 'int' may change the sign of the result [-Werror=sign-conversion] 1107 | yyn += yytoken; | ^~ input.c:1107:10: error: conversion to 'int' from 'unsigned int' may change the sign of the result [-Werror=sign-conversion] 1107 | yyn += yytoken; | ^~~~~~~ input.c:1108:47: error: comparison of integer expressions of different signedness: 'yytype_int8' {aka 'const signed char'} and 'yysymbol_type_t' {aka 'enum yysymbol_type_t'} [-Werror=sign-compare] 1108 | if (yyn < 0 || YYLAST < yyn || yycheck[yyn] != yytoken) | ^~ input.c:702:25: error: operand of ?: changes signedness from 'int' to 'unsigned int' due to unsignedness of other operand [-Werror=sign-compare] 702 | #define YYEMPTY (-2) | ^~~~ input.c:1220:33: note: in expansion of macro 'YYEMPTY' 1220 | yytoken = yychar == YYEMPTY ? YYEMPTY : YYTRANSLATE (yychar); | ^~~~~~~ input.c:1220:41: error: unsigned conversion from 'int' to 'unsigned int' changes value from '-2' to '4294967294' [-Werror=sign-conversion] 1220 | yytoken = yychar == YYEMPTY ? YYEMPTY : YYTRANSLATE (yychar); | ^ Eventually, it might be interesting to move away from -2 (which is the only possible negative symbol number) and use the next available number, to save bits. We could actually even simply use "0" and shift the rest, which would allow to write "!yytoken" to mean really "yytoken != YYEMPTY". * data/skeletons/c.m4 (b4_declare_symbol_enum): Define YYSYMBOL_YYEMPTY. * data/skeletons/yacc.c: Use it. * src/parse-gram.y (yyreport_syntax_error): Use YYSYMBOL_YYEMPTY, not YYEMPTY, when dealing with a symbol. * tests/regression.at: Adjust. 2020-04-01 Akim Demaille yacc.c: use yysymbol_type_t instead of int for yytoken Now that we have a proper type for internal symbol numbers, let's use it. More code needs conversion, e.g., printers and destructors, but they are shared with glr.c, which is not ready yet for this change. It will also help us deal with warnings such as (GCC9 on GNU/Linux): input.c: In function 'int yyparse()': input.c:475:37: error: enumeral and non-enumeral type in conditional expression [-Werror=extra] 475 | (0 <= (YYX) && (YYX) <= YYMAXUTOK ? yytranslate[YYX] : YYSYMBOL_YYUNDEF) | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ input.c:1024:17: note: in expansion of macro 'YYTRANSLATE' 1024 | yytoken = YYTRANSLATE (yychar); | ^~~~~~~~~~~ * data/skeletons/yacc.c (yytranslate, yysymbol_name) (yyparse_context_t, yyexpected_tokens, yypstate_expected_tokens) (yysyntax_error_arguments): Use yysymbol_type_t instead of int. 2020-04-01 Akim Demaille regen 2020-04-01 Akim Demaille yacc.c: introduce an enum that defines the symbol's number There's a number of advantage in exposing the symbol (internal) numbers: - custom error messages can use them to decide how to represent a given symbol, or a set of symbols. - we need something similar in uses of yyexpected_tokens. For instance, currently, bistromathic's completion() reads: int ntokens = expected_tokens (line, tokens, YYNTOKENS); [...] for (int i = 0; i < ntokens; ++i) if (tokens[i] == YYTRANSLATE (TOK_VAR)) [...] else if (tokens[i] == YYTRANSLATE (TOK_FUN)) [...] else [...] - now that it's a compile-time expression, we can easily build static tables, switch, etc. - some users depended on the ability to get the token number from a symbol to write test cases for their scanners. But Bison 3.5 removed the table this feature depended upon (a reverse yytranslate). Now they can check against the actual symbol number, without having pay (space and time) a conversion. See https://lists.gnu.org/r/bug-bison/2020-01/msg00001.html, and https://lists.gnu.org/archive/html/bug-bison/2020-03/msg00015.html. - it helps us clearly separate the internal symbol numbers from the external token numbers, whose difference is sometimes blurred in the code when values coincide (e.g. "yychar = yytoken = YYEOF"). - it allows us to get rid of ugly macros with inconsistent names such as YYUNDEFTOK and YYTERROR, and to group related definitions together. - similarly it provides a clean access to the $accept symbol (which proves convenient in a current experimentation of mine with several %start symbols). Let's declare this type as a private type (in the *.c file, not the *.h one). So it does not need to be influenced by the api prefix. * data/skeletons/bison.m4 (b4_symbol_sid): New. (b4_symbol): Use it. * data/skeletons/c.m4 (b4_symbol_enum, b4_declare_symbol_enum): New. * data/skeletons/yacc.c: Use b4_declare_symbol_enum. (YYUNDEFTOK, YYTERROR): Remove. Use the corresponding symbol enum instead. 2020-03-30 Akim Demaille style: comment changes about token numbers * data/skeletons/bison.m4, data/skeletons/c.m4: here. 2020-03-30 Akim Demaille tests: recheck: work properly when the test suite was interrupted * tests/local.mk (recheck): Look at the per-test logs, not the overall log, which, when interrupted, contains only information about... the tests that passed. 2020-03-30 Akim Demaille java: move away from _ for internationalization The "_" is becoming a keyword in Java, which causes tons of warnings currently in our test suite. GNU Gettext is now using "i18n" instead of "_" (https://git.savannah.gnu.org/gitweb/?p=gettext.git;a=commitdiff;h=e89fea36545f27487d9652a13e6a0adbea1117d0). * data/skeletons/java.m4: Use "i18n", not "_". * examples/java/calc/Calc.y, tests/calc.at: Adjust. 2020-03-28 Akim Demaille regen 2020-03-28 Akim Demaille c: use YYNOMEM instead of -2 See 84b1972c96060866b4bd94a33b97711f8f7d0b6c. * data/skeletons/glr.c, data/skeletons/yacc.c (YYNOMEM): New. Use it. 2020-03-28 Akim Demaille todo: update * TODO (Token Number): We have to clean this. (Naming conventions, Symbol numbers): New. (Bad styling): Addressed in e21ff47f5d0b64da693a47b7dd200a1a44a5bbeb. 2020-03-28 Akim Demaille regen 2020-03-28 Akim Demaille java: make yysyntaxErrorArguments a private detail * data/skeletons/lalr1.java (yysyntaxErrorArguments): Move it from the context, to the parser object. Generate only for detailed and verbose error messages. * tests/local.at (AT_YYERROR_DEFINE(java)): Use yyexpectedTokens instead. 2020-03-28 Akim Demaille skeletons: make yysyntax_error_arguments a private detail We could just "inline yysyntax_error_arguments back" in the routines it was originally extracted from, but I think the code is nicer to read this way. * data/skeletons/glr.c (yysyntax_error_arguments): Generate only for detailed and verbose error messages. * data/skeletons/yacc.c: Likewise. * data/skeletons/lalr1.cc (parser::context::yysyntax_error_arguments): Move as... (parser::yysyntax_error_arguments_): this. And only for detailed and verbose error messages. 2020-03-28 Akim Demaille lalr1.cc: avoid using yysyntax_error_arguments * data/skeletons/lalr1.cc (context::token): New. * tests/local.at (yyreport_syntax_error): Don't use yysyntax_error_arguments. 2020-03-28 Akim Demaille bison: avoid using yysyntax_error_arguments * src/parse-gram.y (yyreport_syntax_error): Use yyparse_context_token and yyexpected_tokens. 2020-03-28 Akim Demaille tests: yacc.c: avoid yysyntax_error_arguments Because glr.c shares the same testing routines, we also need to convert it. * data/skeletons/glr.c (yyparse_context_token): New. * tests/local.at (yyreport_syntax_error): here. 2020-03-28 Akim Demaille examples: don't use yysyntax_error_arguments Suggested by Adrian Vogelsgesang. https://lists.gnu.org/archive/html/bison-patches/2020-02/msg00069.html * data/skeletons/lalr1.java (Context.EMPTY, Context.getToken): New. (Context.yyntokens): Rename as... (Context.NTOKENS): this. Because (i) all the Java coding styles recommend upper case for constants, and (ii) the Java Skeleton exposes Lexer.EOF, not Lexer.YYEOF. * data/skeletons/yacc.c (yyparse_context_token): New. * examples/c/bistromathic/parse.y (yyreport_syntax_error): Don't use yysyntax_error_arguments. * examples/java/calc/Calc.y (yyreportSyntaxError): Likewise. 2020-03-28 Akim Demaille skeletons: fix incorrect type for translatable tokens * data/skeletons/glr.c, data/skeletons/lalr1.c, data/skeletons/yacc.c: Fix confusion between the "translatable" and the "translate" tables. 2020-03-23 Akim Demaille yacc.c: use negative numbers for errors in auxiliary functions yyparse returns 0, 1, 2 since ages (accept, reject, memory exhausted). Some of our auxiliary functions such as yy_lac and yyreport_syntax_error also need to return error codes and also use 0, 1, 2. Because it uses yy_lac, yyexpected_tokens also needs to return "problem", "memory exhausted", but in case of success, it needs to return the number of tokens, so it cannot use 1 and 2 as error code. Currently it uses -1 and -2, which is later converted into 1 and 2 as yacc.c expects it. Let's simplify this and use consistently -1 and -2 for auxiliary functions that are not exposed (or not yet exposed) to the user. In particular this will save the user from having to convert yyexpected_tokens's -2 into yyreport_syntax_error's 2: both return -1 or -2. * data/skeletons/yacc.c (yy_lac, yyreport_syntax_error) (yy_lac_stack_realloc): Return -1, -2 for errors instead of 1, 2. Adjust callers. * examples/c/bistromathic/parse.y (yyreport_syntax_error): Do take error codes into account. Issue a syntax error message even if we ran out of memory. * src/parse-gram.y, tests/local.at (yyreport_syntax_error): Adjust. 2020-03-23 Akim Demaille style: reduce length of private constant * data/skeletons/glr.c, data/skeletons/lalr1.cc, data/skeletons/yacc.c (YYERROR_VERBOSE_ARGS_MAXIMUM): Rename as... (YYARGS_MAX): this. * src/parse-gram.y (YYERROR_VERBOSE_ARGS_MAXIMUM): Rename as... (ARGS_MAX): this. 2020-03-23 Akim Demaille doc: c++: promote api.token.raw * doc/bison.texi (Calc++ Parser): Here. 2020-03-22 Akim Demaille bench: calc: no need for super long inputs * etc/bench.pl.in ($iterations): Restore initial value, -1, meaning "at least one second". ($calc_input): There is no need to generate 400 lines. 2020-03-22 Akim Demaille bench: calc: work on a string instead of a file The cost of the file layer is large and makes benchmarks too coarse, as seen for in following example, first with a file, then with a literal string: 0. %skeleton "yacc.c" %define parse.lac full 1. %skeleton "yacc-v1.c" %define nofinal %define parse.lac full 2. %skeleton "yacc-v2.c" %define nofinal %define parse.lac full 3. %skeleton "yacc-v3.c" %define nofinal %define parse.lac full 4. %skeleton "yacc.c" 5. %skeleton "yacc-v1.c" %define nofinal 6. %skeleton "yacc-v2.c" %define nofinal 7. %skeleton "yacc-v3.c" %define nofinal -------------------------------------------------- Benchmark Time CPU Iterations -------------------------------------------------- BM_y0 32558 ns 32537 ns 21228 BM_y1 32400 ns 32369 ns 21233 BM_y2 33485 ns 33464 ns 20625 BM_y3 32139 ns 32125 ns 21446 BM_y4 31343 ns 31329 ns 21747 BM_y5 31344 ns 31317 ns 22035 BM_y6 31287 ns 31255 ns 22039 BM_y7 31387 ns 31373 ns 22178 -------------------------------------------------- Benchmark Time CPU Iterations -------------------------------------------------- BM_y0 10642 ns 10634 ns 63601 BM_y1 10657 ns 10654 ns 63625 BM_y2 10441 ns 10432 ns 65957 BM_y3 10558 ns 10554 ns 64546 BM_y4 9521 ns 9516 ns 72011 BM_y5 9179 ns 9157 ns 75028 BM_y6 9360 ns 9356 ns 73770 BM_y7 9365 ns 9359 ns 72609 Of course, at the same time it is less realistic: most users read files rather that strings, so it might lead to us to pay attention to costs most people don't see. * etc/bench.pl.in (&calc_input): Output into a file given as argument. Output in C syntax. (&generate_grammar_calc): Use it. Simplify the grammar: remove operators we don't care about. Rewrite the scanner to work on a char* instead of a FILE*. 2020-03-22 Akim Demaille bench: add a "latest" symlink * etc/bench.pl.in: here. 2020-03-22 Akim Demaille bench: use the same prefix in both bench methods * etc/bench.pl.in (&bench_with_timethese): Also use y$i, as in &bench_with_gbenchmark. (&generate_grammar_calc): Don't add a prefix, let the callers do it. 2020-03-22 Akim Demaille bench: use a C++-11 compiler See https://github.com/google/benchmark#a-faster-keeprunning-loop. * etc/bench.pl.in ($cxx): Be C++11. (&bench_with_gbenchmark): Adjust. 2020-03-22 Akim Demaille bench: create a README file with benches * etc/bench.pl.in (&bench_with_gbenchmark): Here. 2020-03-21 Akim Demaille bench: calc: add support for google benchmark * etc/bench.pl.in (&compiler): New, extracted from... (&compile): here. Don't link when using gbm. (&calc_input): Don't make massive input for micro benchmarks. (&generate_grammar_calc): When using gbm, use api.prefix to avoid name collisions. Be ready to issue BENCHMARKS instead of a main. (&bench): Rename as... (&bench_with_timethese): this. (&bench_with_gbenchmark): New. (&bench): New. Dispatch on these two. 2020-03-21 Akim Demaille bench: better error messages on invalid input * etc/bench.pl.in: here. 2020-03-21 Akim Demaille bench: simplify the calc grammar * etc/bench.pl.in (generate_grammar_calc): We don't need global_result etc. 2020-03-21 Akim Demaille bench: die clearly on incorrect --grammar arguments * etc/bench.pl.in (getopt): here. 2020-03-17 Akim Demaille regen 2020-03-17 Akim Demaille yacc.c: style: prefer switch to if * data/skeletons/yacc.c: Prefer switch to decode yy_lac's return value. 2020-03-17 Akim Demaille yacc.c: yypstate_expected_tokens In push parsers, when asking for the list of expected tokens at some point, it makes no sense to build a yyparse_context_t: the yypstate alone suffices (the only difference being the lookahead). Instead of forcing the user to build a useless shell around yypstate, let's offer yypstate_expected_tokens. See https://lists.gnu.org/r/bison-patches/2020-03/msg00025.html. * data/skeletons/yacc.c (yypstate): Declare earlier, so that we can use it for... (yypstate_expected_tokens): this new function, when in push parsers. Adjust dependencies. * examples/c/bistromathic/parse.y: Simplify: use yypstate_expected_tokens. Style fixes. Reduce scopes (reported by Joel E. Denny). 2020-03-09 Akim Demaille examples: bistromathic: simplify * examples/c/bistromathic/parse.y (expected_tokens): Remove useless "break". 2020-03-08 Akim Demaille merge branch 'maint' * upstream/maint: maint: post-release administrivia version 3.5.3 news: update for 3.5.3 yacc.c: make sure we properly propagated the user's number for error diagnostics: don't crash because of repeated definitions of error style: initialize some struct members diagnostics: beware of zero-width characters diagnostics: be sure to close the styling when lines are too short muscles: fix incorrect decoding of $ code: be robust to reference with invalid tags build: fix typo doc: update recommandation for libtextstyle style: comment changes examples: use consistently the GFDL header for readmes style: remove useless declarations typo: succesful -> successful README: point to tests/bison, and document --trace gnulib: update maint: post-release administrivia 2020-03-08 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2020-03-08 Akim Demaille version 3.5.3 * NEWS: Record release date. 2020-03-08 Akim Demaille news: update for 3.5.3 2020-03-08 Akim Demaille yacc.c: make sure we properly propagated the user's number for error * data/skeletons/yacc.c (YYERRCODE): Be truthful. * tests/input.at (Redefining the error token): Check that. 2020-03-08 Akim Demaille diagnostics: don't crash because of repeated definitions of error According to https://www.unix.com/man-page/POSIX/1posix/yacc/, the user is allowed to specify her user number for the error token: The token error shall be reserved for error handling. The name error can be used in grammar rules. It indicates places where the parser can recover from a syntax error. The default value of error shall be 256. Its value can be changed using a %token declaration. The lexical analyzer should not return the value of error. I think this feature is useless, the user should not have to deal with that. The intend is probably to give the user a means to use 256 if she wants to, but provided "error" cleared the path first by being assigned another number. In the case of Bison, 256 is assigned to "error" at the end if the user did not use it for a token of hers. So this feature is useless. Yet it is valid, and if the user assigns twice a token number to "error", then the second time we want to complain about it and want to show the original definition. At this point, we try to display the built-in definition of "error", whose location is NULL, and we crash. Rather, the location of the first user definition of "error" should become its defining location. Reported byg Ahcheong Lee. https://lists.gnu.org/r/bug-bison/2020-03/msg00007.html * src/symtab.c (symbol_class_set): If this is a declaration and the symbol was not declared yet, keep this as defining location. * tests/input.at (Redefining the error token): New. 2020-03-08 Akim Demaille style: initialize some struct members * src/symtab.c (sym_content_new): Initialize all the location members. Not needed by the code, but disturbing values when using a debugger. 2020-03-08 Akim Demaille diagnostics: beware of zero-width characters Currenly we rely on (visual) width of the characters to decide where to open and close the styling of the quoted lines. This breaks when we deal with zero-width characters: we cannot just rely on (visual) columns, we need to know whether we are before, inside, or after the highlighted portion. * src/location.c (location_caret): col_end: no longer add 1, "regular" characters have a width of 1, only 0-width characters have 0-width. opened: replace with 'state', a three-valued enum. Don't reopen the style if we already did. * tests/diagnostics.at (Zero-width characters): New. 2020-03-07 Akim Demaille diagnostics: be sure to close the styling when lines are too short bar.y:4.12-17: error: redefining user token number of foo - 4 | %token foo 123 + 4 | %token foo 123 | ^~~~~~ * src/location.c (location_caret): Be sure to close. * tests/diagnostics.at (Line is too short, and then you die): New. 2020-03-07 Akim Demaille muscles: fix incorrect decoding of $ Bug introduced in 458171e6df5a0110a35ee45ad8b2e9f6fb426f1d. https://lists.gnu.org/archive/html/bison-patches/2013-11/msg00009.html Reported by Ahcheong Lee. https://lists.gnu.org/r/bug-bison/2020-03/msg00010.html * src/muscle-tab.c (COMMON_DECODE): "$" is coded as "$][", not "$[][". * tests/input.at ("%define" enum variables): Check that case. 2020-03-06 Akim Demaille code: be robust to reference with invalid tags Because we want to support $b>$, we must accept -> in type tags, and reject $<->$, as it is unfinished. Reported by Ahcheong Lee. * src/scan-code.l (yylex): Make sure "tag" does not end with -, since -> does not close the tag. * tests/input.at (Stray $ or @): Check this. 2020-03-06 Akimn Demaille build: fix typo * build-aux/cross-options.pl: here. 2020-03-06 Akim Demaille doc: update recommandation for libtextstyle * README: here. 2020-03-06 Akim Demaille style: comment changes * src/symtab.h, src/lr0.c: here. 2020-03-06 Akim Demaille examples: use consistently the GFDL header for readmes * examples/c++/README.md, examples/c++/calc++/README.md, * examples/c/calc/README.md, examples/c/lexcalc/README.md, * examples/c/reccalc/README.md: Prefer the GFDL banner to the GPL one. 2020-03-06 Akim Demaille style: remove useless declarations * src/reader.h: Don't duplicate what parse-gram.h already exposes. * src/lr0.h: Remove useless include. 2020-03-06 Adrian Vogelsgesang typo: succesful -> successful * data/skeletons/lalr1.cc: here * etc/bench.pl.in: here * src/location.c: and here. 2020-03-06 Akim Demaille README: point to tests/bison, and document --trace Reported by Victor Morales Cayuela. * README, README-hacking.md: here. 2020-03-06 Akim Demaille gnulib: update 2020-03-05 Akim Demaille README: point to tests/bison, and document --trace Reported by Victor Morales Cayuela. * README, README-hacking.md: here. 2020-03-05 Akim Demaille yacc.c: simplify yyparse_context_t member names * data/skeletons/yacc.c (yyparse_context_t): Rename yyes_p and yyes_capacity_p as... (yyes, yyes_capacity): These. 2020-03-05 Akim Demaille yacc.c: yyerror_range does not need to be preserved accross calls * data/skeletons/yacc.c (b4_parse_state_variable_macros): Don't define yyerror_range. (yyparse): Add yyerror_range as local variable. 2020-03-05 Akim Demaille yacc.c: push: undefine the pstate macros for the epilogue * data/skeletons/yacc.c (b4_macro_define, b4_macro_undef) (b4_pstate_macro_define, b4_parse_state_variable_macros): New. Use them. * examples/c/bistromathic/parse.y: Remove now useless undefs. 2020-03-05 Akim Demaille yacc.c: push: initialize the pstate variables in pstate_new Currently pstate_new does not set up its variables, this task is left to yypush_parse. This was probably to share more code with usual pull parsers, where these (local) variables are indeed initialized by yyparse. But as a consequence yyexpected_tokens crashes at the very beginning of the parse, since, for instance, the stacks are not even set up. See https://lists.gnu.org/r/bison-patches/2020-03/msg00001.html. The fix could have very simple, but the documentation actually makes it very clear that we can reuse a pstate for several parses: After yypush_parse returns a status other than YYPUSH_MORE, the parser instance yyps may be reused for a new parse. so we need to restore the parser to its pristine state so that (i) it is ready to run the next parse, (ii) it properly supports yyexpected_tokens for the next run. * data/skeletons/yacc.c (b4_initialize_parser_state_variables): New, extracted from the top of yyparse/yypush_parse. (yypstate_clear): New. (yypstate_new): Use it when push parsers are enabled. Define after the yyps macros so that we can use the same code as the regular pull parsers. (yyparse): Use it when push parsers are _not_ enabled. * examples/c/bistromathic/bistromathic.test: Check the completion on the beginning of the line. 2020-03-04 Akim Demaille style: formatting changes * data/skeletons/yacc.c, tests/torture.at: here. 2020-03-04 Akim Demaille bistromathic: properly compute the lcp, as expected by readline Currently completion on "at" proposes only "atan", but does not actually complete "at" into "atan". * examples/c/bistromathic/parse.y (completion): Install the lcp in matches[0]. * examples/c/bistromathic/bistromathic.test: Check that case. 2020-03-04 Akim Demaille bistromathic: don't require spaces after operators for completion Currently "(1+" does not work as expected, because "+" is not a word breaking character. * examples/c/bistromathic/parse.y (init_readline): Specify our word breaking characters. * examples/c/bistromathic/bistromathic.test: Avoid trailing spaces. 2020-03-02 Akim Demaille bistromathic: check completion * examples/c/bistromathic/bistromathic.test: here. * examples/c/bistromathic/parse.y (expected_tokens): Fix a memory leak. 2020-03-02 Akim Demaille m4: remove b4_function_define and b4_function_declare * data/skeletons/c.m4: here. 2020-03-02 Akim Demaille m4: decommission b4_function_declare * data/skeletons/glr.c, data/skeletons/glr.cc, data/skeletons/yacc.c: Stop using b4_function_declare. 2020-03-02 Akim Demaille m4: decommission function generating macro These macros have been extremely useful when we had to support K&R C, which we dropped long ago. Now, they merely make the code uselessly hard to read. * data/skeletons/c.m4, data/skeletons/glr.c, data/skeletons/glr.cc, * data/skeletons/yacc.c: Stop using b4_function_define. 2020-03-01 Akim Demaille examples: bistromathic: demonstrate the use of yyexpected_tokens Let's use GNU readline and its TAB autocompletion to demonstrate the use of yyexpected_tokens. This shows a number of weaknesses in our current approach: - some macros (yyssp, etc.) from push parsers "leak" in user code, we need to undefine them - the context needed by yyexpected_tokens does not need the token, yypstate actually suffices - yypstate is not properly setup when first allocated, which results in a crash of yyexpected_tokens if fired before a first token was read. We should move initialization from yypush_parse into yypstate_new. * examples/c/bistromathic/parse.y (yylex): Take input as a string, not a file. (EXIT): New token. (input): Adjust to work only on a line. (line): Remove. (symbol_count, process_line, expected_tokens, completion) (init_readline): New. * examples/c/bistromathic/bistromathic.test: Adjust expectations. 2020-03-01 Akim Demaille examples: use consistently the GFDL header for readmes * examples/c++/README.md, examples/c++/calc++/README.md, * examples/c/calc/README.md, examples/c/lexcalc/README.md, * examples/c/pushcalc/README.md, examples/c/reccalc/README.md: Prefer the GFDL banner to the GPL one. 2020-03-01 Akim Demaille gnulib: use readline 2020-02-29 Akim Demaille examples: bistromathic: don't use Flex This example will soon use GNU readline, so its scanner should be easy to use (concurrently) on strings, not streams. This is not a place where Flex shines, and anyway, these are examples of Bison, not Flex. There's already lexcalc and reccalc that demonstrate the use of Flex. * examples/c/bistromathic/scan.l: Remove. * examples/c/bistromathic/parse.y (yylex): New. Adjust dependencies. 2020-02-29 Akim Demaille examples: bistromathic: strengthen tests * examples/c/bistromathic/bistromathic.test: here. * examples/test: Be clearer on failing tests. 2020-02-29 Akim Demaille examples: lexcalc: demonstrate location tracking The bistromathic example should not use Flex, it makes it too complex. But it was the only example to show location tracking with Flex. * examples/c/lexcalc/lexcalc.test, examples/c/lexcalc/parse.y, * examples/c/lexcalc/scan.l: Demonstrate location tracking as is done in bistromathic. 2020-02-27 Akim Demaille c++: don't copy the lookahead The current implementation of parser::context keeps a copy of the lookahead. This is troublesome since we support move-only types. Besides, while GCC is happy with the current implementation, Clang complains that the ctor it needs to build the copy of the lookahead is not yet available. 461. calc.at:1120: testing Calculator C++ %defines %locations parse.error=verbose %name-prefix "calc" %verbose ... calc.at:1120: COLUMNS=1000; export COLUMNS; bison --color=no -fno-caret -Wno-deprecated -o calc.cc calc.y calc.at:1120: $CXX $CXXFLAGS $CPPFLAGS $LDFLAGS -o calc calc.cc calc-lex.cc calc-main.cc $LIBS stderr: In file included from calc-lex.cc:7: calc.hh:351:12: error: instantiation of function 'calc::parser::basic_symbol::basic_symbol' required here, but no definition is available [-Werror,-Wundefined-func-template] struct symbol_type : basic_symbol ^ calc.hh:273:7: note: forward declaration of template entity is here basic_symbol (const basic_symbol& that); ^ calc.hh:351:12: note: add an explicit instantiation declaration to suppress this warning if 'calc::parser::basic_symbol::basic_symbol' is explicitly instantiated in another translation unit struct symbol_type : basic_symbol ^ 1 error generated. In file included from calc-main.cc:7: calc.hh:351:12: error: instantiation of function 'calc::parser::basic_symbol::basic_symbol' required here, but no definition is available [-Werror,-Wundefined-func-template] struct symbol_type : basic_symbol ^ calc.hh:273:7: note: forward declaration of template entity is here basic_symbol (const basic_symbol& that); ^ calc.hh:351:12: note: add an explicit instantiation declaration to suppress this warning if 'calc::parser::basic_symbol::basic_symbol' is explicitly instantiated in another translation unit struct symbol_type : basic_symbol ^ 1 error generated. stdout: calc.at:1120: exit code was 1, expected 0 461. calc.at:1120: 461. Calculator C++ %defines %locations parse.error=verbose %name-prefix "calc" %verbose (calc.at:1120): FAILED (calc.at:1120) * data/skeletons/lalr1.cc (context::yyla_): Make it a const-ref. Move the implementation out of the declaration. 2020-02-27 Akim Demaille c++: minor fixes Address compiler warnings such as warning: declaration of 'yyla' shadows a member of 'yy::parser::context' [-Wshadow] * data/skeletons/lalr1.cc (context): Don't use the same names for variables and members. Use foo_ for private members, as in parser. Also, use the + trick in array accesses to please ICC and provide it with an int. 2020-02-27 Adrian Vogelsgesang c++: add support for parse.error=custom * data/skeletons/lalr1.cc: added support here * tests/calc.at: added test cases * tests/local.at: added yyreport_syntax_error implementation for C++ test cases 2020-02-27 Adrian Vogelsgesang c++: add parser::context for syntax error handling * data/skeletons/lalr1.cc: here 2020-02-27 Adrian Vogelsgesang c++: add support for parse.error=detailed * data/skeletons/lalr1.cc: added support here * tests/calc.at: added a test case 2020-02-27 Adrian Vogelsgesang skeletons: prefer b4_parse_error_{case,bmatch} over manual solution Prefer b4_parse_error_case over the adhoc solution `m4_case + b4_percent_define_get`. Same for b4_parse_error_bmatch. * data/skeletons/glr.c: here * data/skeletons/yacc.c: here 2020-02-27 Adrian Vogelsgesang typo: succesful -> successful * data/skeletons/lalr1.cc: here * etc/bench.pl.in: here * src/location.c: here * tests/calc.at: and here 2020-02-24 Akim Demaille bench.pl: clean up the dust * etc/bench.pl.in: Adjust to the current use of %define's values. Don't use %error-verbose. Prefer Bison to CPP (e.g., api.value.type). Avoid returning characters directly, so that %define api.token.raw works. 2020-02-23 Akim Demaille style: comment changes * src/symtab.h, src/lr0.c: here. 2020-02-23 Akim Demaille style: avoid using 'this' as an identifier LLDB insists on parsing 'this' as a C++ keyword, even when debugging a C program. * src/symtab.c: Please the dictator. 2020-02-23 Akim Demaille style: remove useless declarations * src/reader.h: Don't duplicate what parse-gram.h already exposes. * src/lr0.h: Remove useless include. 2020-02-19 Akim Demaille examples: fix c/calc * examples/c/calc/calc.y: Remove experiment traces. 2020-02-19 Akim Demaille todo: update 2020-02-15 Akim Demaille doc: update recommandation for libtextstyle * README: here. 2020-02-15 Akimn Demaille build: fix typo * build-aux/cross-options.pl: here. 2020-02-15 Akim Demaille doc: simplify the cross references * doc/bison.texi: here. 2020-02-15 Akim Demaille doc: document token internationalization * doc/bison.texi (Parser Internationalization): Move most of its content into... (Enabling I18n): this new node. (Token I18n): New. (Token Decl): Refer to token internationalization. (Error Reporting Function): Promote parse.error detailed. 2020-02-15 Akim Demaille regen 2020-02-15 Victor Morales Cayuela diagnostics: modernize the display of submessages Since Bison 2.7, output was indented four spaces for explanatory statements. For example: input.y:2.7-13: error: %type redeclaration for exp input.y:1.7-11: previous declaration Since the introduction of caret-diagnostics, it became less clear. Remove the indentation and display submessages as in GCC: input.y:2.7-13: error: %type redeclaration for exp 2 | %type exp | ^~~~~~~ input.y:1.7-11: note: previous declaration 1 | %type exp | ^~~~~ * src/complain.h (SUB_INDENT): Remove. (warnings): Add "note" to the enum. * src/complain.h, src/complain.c (complain_indent): Replace by... (subcomplain): this. Adjust all dependencies. * tests/actions.at, tests/diagnostics.at, tests/glr-regression.at, * tests/input.at, tests/named-refs.at, tests/regression.at: Adjust expectations. 2020-02-13 Akim Demaille doc: simplify uses of references This reverts "doc: work around problems with PDF generation", commit d810aa3d8f76b1a4d7d402072f45a0662152ffd4. Upstream issue is fixed. https://lists.gnu.org/r/bug-texinfo/2020-02/msg00006.html * gnulib: Update. * doc/bison.texi: Simplify. 2020-02-13 Akim Demaille java: provide a Context ctor This is really a private auxiliary inner class, so it should not matter. But it's better style. * data/skeletons/lalr1.java: here. 2020-02-13 Akim Demaille Merge tag 'v3.5.2' bison 3.5.2 * tag 'v3.5.2': version 3.5.2 news: 3.5.2 gnulib: update doc: update Doxygen template java: avoid trailing white spaces m4: fix b4_token_format doc: clearly state that %yacc only makes sense with yacc.c doc: spell check examples: be more robust to spaces in paths larlr1.cc: Reject unsupported values for parse.lac 2020-02-13 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2020-02-13 Akim Demaille version 3.5.2 * NEWS: Record release date. 2020-02-13 Akim Demaille news: 3.5.2 * NEWS: Update. 2020-02-13 Akim Demaille gnulib: update 2020-02-13 Akim Demaille doc: update Doxygen template * Doxyfile.in: Run doxygen -u on it. 2020-02-13 Akim Demaille java: avoid trailing white spaces * data/skeletons/java.m4 (b4_maybe_throws): Issue a space before when needed. * data/skeletons/lalr1.java: Avoid trailing spaces. 2020-02-13 Akim Demaille m4: fix b4_token_format We used to emit: /** Token number,to be returned by the scanner. */ static final int NUM = 258; /** Token number,to be returned by the scanner. */ static final int NEG = 259; with no space after the comma. Fix that. * data/skeletons/bison.m4 (b4_token_format): Quote where appropriate. 2020-02-13 Akim Demaille doc: clearly state that %yacc only makes sense with yacc.c * doc/bison.texi: here. 2020-02-13 Akim Demaille doc: spell check * doc/bison.texi: here. 2020-02-12 Akim Demaille examples: bistromathic: demonstrate named references * examples/c/bistromathic/parse.y: here. 2020-02-12 Akim Demaille c++: simplify * data/skeletons/stack.hh (ssize): Remove, same as size. 2020-02-12 Akim Demaille tests: check calls to yyerror from the user actions This revealed a number of things I had not realized: - the Java location tracking was aliasing the same pair of positions for all the symbols (see previous commit). - in impure parsers, it's quite easy to use incorrect locations for diagnostics, since yyerror uses yylloc, which is the location of the lookahead, not that of the current lhs. So we need something like { YYLTYPE old_yylloc = yylloc; yylloc = @$; yyerror (]AT_PARAM_IF([result, count, nerrs, ])[buf); yylloc = old_yylloc; } Maybe we should do that little yylloc dance in the skeleton instead of leaving it to the user? It might be costly... But that's only for users of the impure parsers, which are asking for trouble anyway. - in glr.cc invoking yyerror is somewhat cumbersome: the C++ interface is not available as we are in yyparse (which in C), and yyerror is used by glr.cc itself to bind it to the user's parser::error. If we call yyerror, we need: yyerror (]AT_LOCATION_IF([[&@$, ]])[yyparser, ]AT_PARAM_IF([result, count, nerrs, ])[msg); However calling yy::parser::error is easier, once we know that the current parser object is available as 'yyparser'. Which also saves us from having to pass the parse-params ourselves: yyparser.error (]AT_LOCATION_IF([[@$, ]])[msg); * tests/calc.at: Invoke yyerror by hand, instead of using fprintf etc. Adjust expectations. 2020-02-11 Akim Demaille java: beware not to alias the locations of the various symbols * examples/java/calc/Calc.y, tests/calc.at, tests/local.at (getStartPos, getEndPos): Always return a new object. * doc/bison.texi: Clarify this. 2020-02-11 Akim Demaille java: check that parse.error custom|detailed work with push parsers * tests/calc.at: here. 2020-02-11 Akim Demaille java: don't expose the Context's members * data/skeletons/lalr1.java (Context): Make data members private. (Context.getLocation): New. * examples/java/calc/Calc.y, tests/java.at, tests/local.at: Adjust. 2020-02-10 Akim Demaille build: pacify syntax-check * src/complain.c: Fix indentation. * cfg.mk: Using strcmp is ok in the tests. Test cases and examples don't need Bison's PO support. 2020-02-10 Akim Demaille regen 2020-02-10 Akim Demaille gnulib: update 2020-02-10 Akim Demaille build: prefer %D% and %C% to hard coded values * doc/local.mk: here. 2020-02-10 Akim Demaille parse.error: document and diagnose the incompatibility with %token-table * doc/bison.texi (Tokens from Literals): Move to code using %token-table to... (Decl Summary: %token-table): here. * data/skeletons/bison.m4: Implement mutual exclusion. * tests/input.at: Check it. * doc/local.mk: Be robust to the removal of doc/. 2020-02-10 Akim Demaille doc: spell check * doc/bison.texi: here. 2020-02-10 Akim Demaille doc: formatting changes * doc/bison.texi: here. 2020-02-10 Akim Demaille doc: work around problems with PDF generation With texinfo.tex 2019-09-24.13, node names with + are not properly handled. https://lists.gnu.org/r/bug-texinfo/2020-02/msg00004.html * doc/bison.texi: Always use the three-argument form for references to node with a + in the name. 2020-02-10 Akim Demaille java: revert "style: avoid useless initializers" This reverts commit ebab1ffca8a728158051481795ae798231cfd93d. This commit removed "useless" initializers, going from /* YYPACT[STATE-NUM] -- Index in YYTABLE of the portion describing STATE-NUM. */ private static final byte yypact_[] = yypact_init (); private static final byte[] yypact_init () { return new byte[] { 25, -7, -8, 37, -8, 40, -8, 20, -8, 61, -8, -8, 3, 9, 51, -8, -8, -2, -2, -2, -2, -2, -2, -8, -8, -8, 1, 66, 66, 3, 3, 3 }; } to /* YYPACT[STATE-NUM] -- Index in YYTABLE of the portion describing STATE-NUM. */ private static final byte[] yypact_ = { 25, -7, -8, 37, -8, 40, -8, 20, -8, 61, -8, -8, 3, 9, 51, -8, -8, -2, -2, -2, -2, -2, -2, -8, -8, -8, 1, 66, 66, 3, 3, 3 }; But it turns out that this was on purpose, to work around the 64KB limitation in JVM methods. It was introduced on the 2008-11-10 by Di-an Jan in 09ccae9b18a7c09ebf7bb8df2a18c8c4a6def248: "Work around Java's ``code too large'' problem for parser tables". See https://lists.gnu.org/r/help-bison/2008-11/msg00004.html. A real test, where we would hit the JVM limitation, would be nice. To avoid further regressions, add comments. 2020-02-10 Akim Demaille skeletons: avoid b4_error_verbose_if, which is confusing parse.error has more than two possible values. * data/skeletons/bison.m4 (b4_error_verbose_if, b4_error_verbose_flag): Remove. (b4_parse_error_case, b4_parse_error_bmatch): New. Adjust dependencies. 2020-02-10 Akim Demaille skeletons: decorelate %token-table from verbose error messages Reported by Adrian Vogelsgesang. * data/skeletons/bison.m4: Here. * data/skeletons/lalr1.cc: Adjust. 2020-02-10 Akim Demaille doc: formatting changes * doc/bison.texi: here. 2020-02-09 Akim Demaille doc: clearly state that %yacc only makes sense with yacc.c * doc/bison.texi: here. * tests/calc.at: Stop testing %yacc with non yacc.c skeletons. 2020-02-09 Adrian Vogelsgesang style: stylistic cleanups in the C skeletons * data/skeletons/glr.c, data/skeletons/yacc.c: Avoid duplicated declaration of yysymbol_name. 2020-02-08 Akim Demaille java: provide Context with a more OO interface * data/skeletons/lalr1.java (yyexpectedTokens) (yysyntaxErrorArguments): Make them methods of Context. (Context.yysymbolName): New. * tests/local.at: Adjust. 2020-02-08 Akim Demaille java: add support for parse.error custom * data/skeletons/lalr1.java: Add support for custom parse errors. (yyntokens_): Make it public. Under... (yyntokens): this name. (Context): Capture the location too. * examples/c/bistromathic/parse.y, * examples/c/bistromathic/bistromathic.test: Improve error message. * examples/java/calc/Calc.test, examples/java/calc/Calc.y: Use custom error messages. * tests/calc.at, tests/local.at: Check custom error messages. 2020-02-08 Akim Demaille java: let the Context give access to yyntokens * data/skeletons/lalr1.java (Context.yytokens): New. 2020-02-08 Akim Demaille java: make the syntax error format string translatable The error format should be translated, but contrary to the case of C/C++, we cannot just depend on macros to adapt on the presence/absence of '_'. Let's consider that the message format is to be translated iff there are some internationalized tokens. * src/output.c (prepare_symbol_names): Define b4_has_translations. * data/skeletons/java.m4 (b4_trans): New. * data/skeletons/lalr1.java: Use it to emit translatable or not the format string. 2020-02-08 Akim Demaille java: introduce yyexpectedTokens And allow syntax error messages for 'detailed' and 'verbose' to be translated. * data/skeletons/lalr1.java (Context, yyexpectedTokens) (yysyntaxErrorArguments): New. (yysyntax_error): Use it. 2020-02-08 Akim Demaille java: add support for parse.error=detailed In Java there is no need for N_ and yytranslate_. So instead of hard-coding the use of N_ in the table of the symbol names, rely on b4_symbol_translate. * src/output.c (prepare_symbol_names): Use b4_symbol_translate instead of N_. * data/skeletons/c.m4 (b4_symbol_translate): New. * data/skeletons/lalr1.java (yysymbolName): New. Use it. * examples/java/calc/Calc.y: Use parse.error=detailed. * tests/calc.at: Check parse.error=detailed. 2020-02-08 Akim Demaille m4: fix b4_token_format We used to emit: /** Token number,to be returned by the scanner. */ static final int NUM = 258; /** Token number,to be returned by the scanner. */ static final int NEG = 259; with no space after the comma. Fix that. * data/skeletons/bison.m4 (b4_token_format): Quote where appropriate. 2020-02-05 Akim Demaille java: tests: remove now redundant tests * tests/javapush.at: here. 2020-02-05 Akim Demaille java: tests: check push parsers like the others Currently in javapush.at. * tests/calc.at: Here. 2020-02-05 Akim Demaille java: tests: remove now redundant tests * tests/java.at: Calculator tests are now in calc.at. 2020-02-05 Akim Demaille java: tests: check location tracking in the calculator Unfortunately in the Java skeleton the user cannot override the way locations are displayed, and locations don't know the structure of the positions. So they cannot implement the tricks used in the C/C++ skeletons to display "1.1" instead of "1.1-1.2". * tests/local.at (Java): Add support for column tracking in the locations, as we did in examples/java/calc. * tests/calc.at: Use AT_CALC_YYLEX. 2020-02-05 Akim Demaille java: tests: prepare the replacement of calculator tests Soon calculator tests for Java will move from java.at to calc.at. Which implies improving the Java testing infrastructure in local.at (for instance really tracking columns in positions, not just token number). Detach java.at from local.at. * tests/java.at (AT_JAVA_POSITION_DEFINE_OLD): New. Use it. 2020-02-05 Akim Demaille java: style: avoid useless initializers Instead of /* YYPACT[STATE-NUM] -- Index in YYTABLE of the portion describing STATE-NUM. */ private static final byte yypact_[] = yypact_init (); private static final byte[] yypact_init () { return new byte[] { 25, -7, -8, 37, -8, 40, -8, 20, -8, 61, -8, -8, 3, 9, 51, -8, -8, -2, -2, -2, -2, -2, -2, -8, -8, -8, 1, 66, 66, 3, 3, 3 }; } generate /* YYPACT[STATE-NUM] -- Index in YYTABLE of the portion describing STATE-NUM. */ private static final byte[] yypact_ = { 25, -7, -8, 37, -8, 40, -8, 20, -8, 61, -8, -8, 3, 9, 51, -8, -8, -2, -2, -2, -2, -2, -2, -8, -8, -8, 1, 66, 66, 3, 3, 3 }; I have no idea what motivated the previous approach. * data/skeletons/java.m4 (b4_typed_parser_table_define): Here. 2020-02-05 Akim Demaille java: style: prefer putting the square brackets on the type * examples/java/calc/Calc.y, examples/java/simple/Calc.y, * tests/calc.at, tests/local.at: here. 2020-02-05 Akim Demaille java: examples: fix the tracking of locations * examples/java/calc/Calc.y: The StreamTokenizer cannot "peek" for the next character, it reads it, and keeps it for the next call. So the current location is one passed the end of the current token. To avoid this, keep the previous position, and use it to end the current token. * examples/java/calc/Calc.test: Adjust. 2020-02-05 Akim Demaille java: examples: prefer switch to chains of else-if * examples/java/calc/Calc.y, examples/java/simple/Calc.y: here. * examples/java/simple/Calc.y: Use the tokenizer's handling of blanks. 2020-02-05 Akim Demaille java: examples: split in two * examples/java: Split in... * examples/java/simple, examples/java/calc: these. 2020-02-05 Akim Demaille traces: don't print the stack before the gotos The C, C++ and D skeletons used to show the stack right after popping the stack during the reduction. Now that the stack is printed after reaching a new state, that has become useless: Entering state 1 Stack now 0 1 Reducing stack by rule 5 (line 83): $1 = token "number" (1) -> $$ = nterm exp (1) Stack now 0 Entering state 8 Stack now 0 8 Remove the "Stack now 0" line. * data/skeletons/lalr1.cc, data/skeletons/lalr1.d, * data/skeletons/lalr1.java, data/skeletons/yacc.c: Here. 2020-02-05 Akim Demaille traces: show the stack after reading a token Currently, if we have long rules and series of shift, we stack states without showing stack. Let's be more incremental, and do how the Java skeleton does. * data/skeletons/lalr1.cc, data/skeletons/lalr1.d, * data/skeletons/yacc.c: Here. Adjust test cases. * tests/torture.at (AT_DATA_STACK_TORTURE): Disable stack traces: this test produces a very large stack, and showing the stack each time we shift a token goes quadatric. 2020-02-04 Akim Demaille traces: write the "Reading a token" alone on its line The Java skeleton displays Reading a token: Next token is token "number" (1) while the other display Reading a token: Next token is token "number" (1) When generating logs in the scanner, the first part is separated from the second, and the end of the scanner logs have the second part pasted in. So let's propagate the Java way, but with the colon. * data/skeletons/glr.c, data/skeletons/lalr1.cc, data/skeletons/lalr1.d, * data/skeletons/lalr1.java, data/skeletons/yacc.c: Do it. Adjust test cases and doc. 2020-02-02 Akim Demaille java: use the same calc tests as the other skeletons * tests/local.at (AT_LANG_MATCH): New. (AT_YYERROR_DECLARE(java), AT_YYERROR_DECLARE_EXTERN(java)): New. * tests/calc.at: The grammar file for Java is quite different for the others, and continuing to assemble it from pieces makes the grammar file hard to understand. Let's also dispatch on the language to assemble it, and isolate Java from the others. Most of this comes from java.at. 2020-02-02 Akim Demaille java: add access to the number of errors * data/skeletons/lalr1.java (yynewrrs, getNumberOfErrors): New. Formatting changes. 2020-02-02 Akim Demaille java: formatting changes * data/skeletons/java.m4, data/skeletons/lalr1.java: here. 2020-02-02 Akim Demaille java: avoid trailing white spaces * data/skeletons/java.m4 (b4_maybe_throws): Issue a space before when needed. * data/skeletons/lalr1.java: Avoid trailing spaces. 2020-02-02 Akim Demaille java: example: properly track the locations This example, so far, was tracking the current token number, not the current column number. This is not nice for an example... * examples/java/Calc.y (PositionReader): New. Use it. * examples/java/Calc.test: Check the output. 2020-02-02 Akim Demaille java: example: improve * examples/java/Calc.y: Propagate the exit status. Support -p. 2020-02-02 Akim Demaille java: example: rely on autoboxing AFAICT, autoboxing/unboxing was added in Java 5 (September 30, 2004). I think we can afford to use it. It should help us merge some Java tests with the main ones. However, beware that != does not unbox: it compares the object addresses. * examples/java/Calc.y, tests/java.at: Simplify. * examples/java/Calc.test, tests/java.at: Improve tests. 2020-02-02 Akim Demaille tests: comment changes * tests/calc.at: Shorten titles and reduce redundancy. 2020-02-02 Akim Demaille skeletons: add support for %code epilogue When building the test cases, emitting code in the epilogue is very constraining. Let's make it simpler thanks to %code epilogue. However, I don't want to document this: it is bad style to use it (we should avoid having too many ways to write the same thing, TI!MTOWTDI), just put your code in the true epilogue section. * data/skeletons/glr.c, data/skeletons/lalr1.d, data/skeletons/lalr1.java, * data/skeletons/yacc.c: Implement support for %code epilogue. Remove useless comments. * tests/calc.at, tests/java.at: Simplify. 2020-02-02 Akim Demaille examples: bistromathic: fix location tracking * examples/c/bistromathic/scan.l (LOCATION_STEP): New. Use to properly ignore blanks. * examples/c/bistromathic/bistromathic.test: Check that case. 2020-02-02 Akim Demaille doc: document new features of parse.error * doc/bison.texi (Error Reporting): Rename as... (Error Reporting Function): this. Adjust dependencies. Make it a subsection of this... (Error Reporting): new section. (Syntax Error Reporting Function): New. (parse.error): Update description. 2020-01-29 Akim Demaille glr.c: add support for parse.error=custom * data/skeletons/glr.c (yyreportSyntaxError): Call the user's yyreport_syntax_error in custom mode. * tests/calc.at: Check it. 2020-01-29 Akim Demaille glr.c: add support for parse.error=detailed * data/skeletons/glr.c (yystrlen, yysymbol_name): New. Implement parse.error detailed. * tests/calc.at: Check it. 2020-01-29 Akim Demaille glr.c: introduce yyexpected_tokens and yysyntax_error_arguments Modeled after what was done in yacc.c, yet somewhat different since yyGLRStack play the role of the full yyparse_context_t. It will probably be defined as a alias, to make interfaces compatible. LAC is not supported here. And is somewhat tricky. * data/skeletons/glr.c (yyexpected_tokens, yysyntax_error_arguments): New. 2020-01-29 Akim Demaille glr.c: move code around * data/skeletons/glr.c: Move the handling of error messages after the definition of types. Especially after yyGLRStack, which we will need in forthcoming patches. Simplify: yyGLRStack is defined at this point. 2020-01-29 Akim Demaille glr.c: rename yyStateNum as yy_state_t * data/skeletons/glr.c: here. For consistency with yacc.c. 2020-01-29 Akim Demaille yacc.c: fix misleading indentation * data/skeletons/yacc.c: here. 2020-01-27 Akim Demaille doc: simplify uses of @ref The PDF output is more consistent: some nodes were not pointed to with their title. The HTML output becomes "see section Foo" instead of "see Foo", but this should be addressed in the next Texinfo release. Info output is simplified, as it uses only the node name and not its title. But it's considered easier to read this way. See https://lists.gnu.org/r/help-texinfo/2020-01/msg00031.html. * doc/bison.texi: Set @xrefautomaticsectiontitle on. Simplify all uses of ref. 2020-01-27 Akim Demaille examples: be more robust to spaces in paths Reported by Nikki Valen. https://lists.gnu.org/r/bug-bison/2020-01/msg00032.html * examples/test ($prog): Remove, replaced by... (prog): This new function, which pays attention to quoting shell variables. 2020-01-27 Akim Demaille doc: don't pretend trigonometry is part of arithmetics * doc/bison.texi (arith_funs): Rename as... (funs): this. 2020-01-27 Akim Demaille doc: update Doxygen template * Doxyfile.in: Run doxygen -u on it. 2020-01-27 Akim Demaille examples: add a complete example with all the bells and whistles * examples/c/bistromathic/Makefile, * examples/c/bistromathic/README.md, * examples/c/bistromathic/bistromathic.test, * examples/c/bistromathic/local.mk, * examples/c/bistromathic/parse.y, * examples/c/bistromathic/scan.l: New. * Makefile.am (AM_YFLAGS_WITH_LINES): Add -Wdangling-alias. * examples/test: Make failure errors easier to read. 2020-01-26 Akim Demaille examples: add an example of a push parser Add an example to demonstrate the use of push parser. I'm pleasantly surprised: parse.error=detailed works like a charm with push parsers. * examples/c/local.mk, examples/c/pushcalc/Makefile * examples/c/pushcalc/README.md, examples/c/pushcalc/calc.test, * examples/c/pushcalc/calc.y, examples/c/pushcalc/local.mk: New. 2020-01-26 Akim Demaille examples: more tests * examples/c/mfcalc/mfcalc.test: here. 2020-01-26 Akim Demaille examples: clean up * examples/c/calc/calc.y: Restore to its original state, with parse.error=detailed instead of parse.error=custom (this example should be simple). * examples/c/calc/calc.test: Check syntax errors. * examples/c/lexcalc/parse.y: Add comments. 2020-01-26 Akim Demaille tests: check custom error messages and push parsers * tests/local.at (AT_LAC_IF): New. * tests/calc.at: And also check the suppot for LAC. 2020-01-26 Akim Demaille style: formatting changes * data/skeletons/lalr1.java: here. 2020-01-26 Akim Demaille todo: update 2020-01-26 Akim Demaille bison: pretend to 3.6 already * src/parse-gram.y: here. 2020-01-26 Akim Demaille git: update ignores * lib/.gitignore: here. 2020-01-23 Akim Demaille regen 2020-01-23 Akim Demaille diagnostics: modernize bison's syntax errors We used to display the unexpected token first: $ bison foo.y foo.y:1.8-13: error: syntax error, unexpected %token, expecting character literal or identifier or 1 | %token %token | ^~~~~~ GCC uses a different format: $ gcc-mp-9 foo.c foo.c:1:5: error: expected identifier or '(' before ')' token 1 | int()()() | ^ and so does Clang: $ clang-mp-9.0 foo.c foo.c:1:5: error: expected identifier or '(' int()()() ^ 1 error generated. They display the unexpected token last (or not at all). Also, they don't waste width with "syntax error". Let's try that. It gives, for the same example as above: $ bison foo.y foo.y:1.8-13: error: expected character literal or identifier or before %token 1 | %token %token | ^~~~~~ * src/complain.h, src/complain.c (syntax_error): New. * src/parse-gram.y (yyreport_syntax_error): Use it. 2020-01-23 Akim Demaille regen 2020-01-23 Akim Demaille diagnostics: report syntax errors in color * src/parse-gram.y (parse.error): Set to 'custom'. (yyreport_syntax_error): New. * data/bison-default.css (.expected, .unexpected): New. * tests/diagnostics.at: Adjust. 2020-01-23 Akim Demaille regen 2020-01-23 Akim Demaille diagnostics: translate bison's own tokens As a test case, support translations in Bison itself. * src/parse-gram.y: Mark the translatable tokens. While at it, use clearer names. * tests/input.at: Adjust expectations. 2020-01-22 Akim Demaille diagnostics: handle -fno-caret in the called functions Don't force callers of location_caret to have to deal with flags that disable it. * src/location.h, src/location.c (location_caret) (location_caret_suggestion): Early return if disabled. * src/complain.c: Simplify. 2020-01-22 Akim Demaille yacc.c: fixes * data/skeletons/yacc.c: Avoid warnings about unused functions. Fix typo. 2020-01-21 Akim Demaille examples: be more robust to spaces in paths Reported by Nikki Valen. https://lists.gnu.org/r/bug-bison/2020-01/msg00032.html * examples/test ($prog): Remove, replaced by... (prog): This new function, which pays attention to quoting shell variables. 2020-01-21 Adrian Vogelsgesang larlr1.cc: Reject unsupported values for parse.lac Just as the yacc.c skeleton, the lalr1.cc skeleton should reject invalid values for parse.lac. * data/skeletons/lalr1.cc: check validity of parse.lac * tests/input.at: new test cases 2020-01-21 Adrian Vogelsgesang larlr1.cc: Reject unsupported values for parse.lac Just as the yacc.c skeleton, the lalr1.cc skeleton should reject invalid values for parse.lac. * data/skeletons/lalr1.cc: check validity of parse.lac * tests/input.at: new test cases 2020-01-19 Akim Demaille parsers: issue tname with i18n markup Some users would like to avoid having to "parse" the *.y file to find the strings to translate. Let's issue the translatable tokens with N_ to allow "parsing" the generated parsers instead. See https://lists.gnu.org/archive/html/bison-patches/2019-01/msg00015.html * src/output.c (prepare_symbol_names): Issue symbol_names with N_() markup. 2020-01-19 Akim Demaille tests: check token internationalization * tests/calc.at: Check it. 2020-01-19 Akim Demaille regen 2020-01-19 Akim Demaille parsers: support translatable token aliases In addition to %token NUM "number" accept %token NUM _("number") in which case the token will be translated in error messages. Do not use _() in the output if there are no translatable tokens. * src/symtab.h, src/symtab.c (symbol): Add a 'translatable' member. * src/parse-gram.y (TSTRING): New token. (string_as_id.opt): Replace with... (alias): this. Use it. * src/scan-gram.l (SC_ESCAPED_TSTRING): New start conditions, to match TSTRINGs. * src/output.c (prepare_symbols): Define b4_translatable if there are translatable strings. * data/skeletons/glr.c, data/skeletons/lalr1.cc, * data/skeletons/yacc.c (yytnamerr): Receive b4_translatable, and use it. 2020-01-19 Akim Demaille tests: check that detailed error messages preserve UTF-8 characters * tests/regression.at: here. 2020-01-19 Akim Demaille yacc.c: escape trigraphs in detailed parse.error * src/output.c (escape_trigraphs, xescape_trigraphs): New. (prepare_symbol_names): Use it. * tests/regression.at: Check the handling of trigraphs with parse.error = detailed. 2020-01-19 Akim Demaille regen 2020-01-19 Akim Demaille bison: use detailed error messages * #: . 2020-01-19 Akim Demaille regen 2020-01-19 Akim Demaille yacc.c: tests: check detailed error messages * tests/local.at (AT_ERROR_DETAILED_IF): New. (AT_ERROR_SIMPLE_IF): Adjust. * tests/calc.at: Check parse.error=detailed. 2020-01-19 Akim Demaille yacc.c: add support for parse.error detailed "detailed" error messages are almost like "verbose", except that we don't double escape them, they don't get inner quotes, we don't use yytnamerr, and we hide the table. "custom" is exposed with the "detailed" tokens, not the "verbose" ones: they are not double-quoted. Because there's a risk that some people use yytname even without "verbose", let's keep yytname (instead of yys_name) in "simple" parse.error. * src/output.c (prepare_symbol_names): Be ready to output symbol names unquoted. (prepare_symbol_names): Output both the old tname table, and the new symbol_names one. * data/skeletons/bison.m4: Accept 'detailed'. * data/skeletons/yacc.c: When parse.error is 'detailed', don't emit yytname and yytnamerr, just yysymbol_name with the table inside. * tests/calc.at: Adjust. 2020-01-19 Akim Demaille c: use yysymbol_name in traces Only parse.error verbose and simple will get the original yytname: the other options will rely on a different table. So let's move on top of the yysymbol_name function. * data/skeletons/c.m4 (yy_symbol_print): Use yysymbol_name. * data/skeletons/glr.c (yytokenName): Rename as... (yysymbol_name): this. The change of naming scheme is unfortunate, but it's definitely glr.c which is "wrong". 2020-01-19 Akim Demaille glr.c: move some functions after the definition of types Currently yy_symbol_print is defined before yytokenName, although it should use it instead of read yytname directly. Move blocks around to avoid this. * data/skeletons/glr.c (yy_symbol_print): Move its definition after that of yytokenName. 2020-01-19 Akim Demaille Merge branch 'maint' * maint: maint: post-release administrivia version 3.5.1 news: update CI: use ICC again warnings: pacify ICC in lalr1.cc test: report.at: avoid tiny new failure git: update ignores 2020-01-19 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2020-01-19 Akim Demaille version 3.5.1 * NEWS: Record release date. 2020-01-19 Akim Demaille news: update 2020-01-19 Akim Demaille CI: use ICC again See https://github.com/nemequ/icc-travis/issues/15. Thanks to Jeff Hammond and Evan Nemerson for their help. * configure.ac (warn_common): Disable dubious warnings. * .travis.yml: Use ICC again. 2020-01-19 Akim Demaille warnings: pacify ICC in lalr1.cc See 139d0655947c87f90af08718618feaaca0e558d7. * data/skeletons/yacc.c: If I might be a char, write a[+I] instead of a[I], so that ICC does not complain. 2020-01-19 Jim Meyering test: report.at: avoid tiny new failure Be robust to newer versions of Autoconf where the package URL defaults to https instead of http. * configure.ac (AC_INIT): Use https. * tests/report.at: Adjust expected output s/http/https/ to match updated URL. 2020-01-19 Akim Demaille git: update ignores 2020-01-17 Akim Demaille regen 2020-01-17 Akim Demaille yacc.c: portability to G++ 4.8 Currently we get warnings with GCC 4.8 when running the maintainer-check-g++ tests: 143. skeletons.at:85: testing Installed skeleton file names ... ../../tests/skeletons.at:120: COLUMNS=1000; export COLUMNS; bison --color=no -fno-caret --skeleton=yacc.c -o input-cmd-line.c input-cmd-line.y ../../tests/skeletons.at:121: $CC $CFLAGS $CPPFLAGS $LDFLAGS -o input-cmd-line input-cmd-line.c $LIBS stderr: input-cmd-line.c: In function 'int yysyntax_error(long int*, char**, const yyparse_context_t*)': input-cmd-line.c:977:52: error: conversion to 'int' from 'long int' may alter its value [-Werror=conversion] YYSIZEOF (yyarg) / YYSIZEOF (*yyarg)); ^ cc1plus: all warnings being treated as errors stdout: ../../tests/skeletons.at:121: exit code was 1, expected 0 and 429. calc.at:823: testing Calculator parse.error=custom %locations api.prefix={calc} ... ../../tests/calc.at:823: COLUMNS=1000; export COLUMNS; bison --color=no -fno-caret -Wno-deprecated -o calc.c calc.y ../../tests/calc.at:823: $CC $CFLAGS $CPPFLAGS $LDFLAGS -o calc calc.c $LIBS stderr: calc.y: In function 'int yyreport_syntax_error(const yyparse_context_t*)': calc.y:157:58: error: conversion to 'int' from 'long unsigned int' may alter its value [-Werror=conversion] int n = yysyntax_error_arguments (ctx, arg, sizeof arg / sizeof *arg); ^ cc1plus: all warnings being treated as errors stdout: ../../tests/calc.at:823: exit code was 1, expected 0 We could use a cast to avoid the warning, but it becomes too cluttered. We can also use YYPTRDIFF_T, but that forces the user to use YYPTRDIFF_T too, although this is an array of tokens, which is limited by YYNTOKENS, an int. So let's completely avoid this warning. * data/skeletons/yacc.c, tests/local.at (yyreport_syntax_error): Avoid relying on sizeof to compute the array capacity. 2020-01-17 Akim Demaille yacc.c: pass the parse-params to yyreport_syntax_error Enhance the calculator tests: show that passing arguments to yyerror works. * tests/calc.at: Add a new parse-param, nerrs, which counts the number of syntax errors in a run. * tests/local.at: Adjust to handle the new 'nerrs' argument, when present. The custom error reporting function show sees the user's additional arguments. Let's experiment with passing them as arguments to yyreport_syntax_error, but maybe storing them in the context would be a bettter alternative. * data/skeletons/yacc.c (yyreport_syntax_error): Handle the parse-params. * tests/calc.at, tests/local.at: Adjust. 2020-01-17 Akim Demaille tests: a clearer test for parse-params Currently the parse-params are tested in calc.at by checking that the global variable and the parse-params have the same value. But it does not check that value, that could remain being 0 just as well. * tests/calc.at: Don't define the params when they are not used. Check the final value of result and count. Also, do count the number of line of logs. 2020-01-17 Akim Demaille yacc.c: check custom error messages with parse-params * tests/calc.at: Check with prefix and parse-params. 2020-01-17 Akim Demaille yacc.c: let custom error messages see the location * data/skeletons/yacc.c (yyparse_context_t): Add yylloc when applicable. (yyparse_context_location): New. * tests/local.at (AT_YYERROR_DEFINE(c)): Handle the location. * tests/calc.at: Check it. 2020-01-17 Akim Demaille yacc.c: isolate yyexpected_tokens Provide users with a means to query for the currently allowed tokens. Could be used for autocompletion for instance. * data/skeletons/yacc.c (yyexpected_tokens): New, extracted from yysyntax_error_arguments. * examples/c/calc/calc.y (PRINT_EXPECTED_TOKENS): New. Use it. 2020-01-17 Akim Demaille tests: compute verbose error messages from the custom ones We use a different format to check parse.error custom. Compute the "verbose" one from it instead of forcing the test author to provide the various formats of expected error messages. * tests/calc.at (_AT_CHECK_CALC_ERROR): Handle this transformation when needed. Simplify callers. 2020-01-17 Akim Demaille yacc.c: check custom error messages * tests/local.at (AT_ERROR_CUSTOM_IF, AT_ERROR_VERBOSE_IF) (AT_ERROR_SIMPLE_IF): New. (AT_YYERROR_DEFINE(c)): Generate yyreport_syntax_error. * tests/calc.at (_AT_CHECK_CALC_ERROR): Accept custom error messages as additional test case. Use it. Add a new test case for %define parse.error custom. 2020-01-17 Akim Demaille yacc.c: add custom error message generation When parse.error is custom, let users define a yyreport_syntax_error function, and use it. * data/skeletons/bison.m4 (b4_error_verbose_if): Accept 'custom'. * data/skeletons/yacc.c: Implement it. * examples/c/calc/calc.y: Experiment with it. 2020-01-17 Akim Demaille yacc.c: style: avoid macros * data/skeletons/yacc.c (YYSYNTAX_ERROR): Remove, the call is now sufficiently small so that we can afford to duplicate it. 2020-01-17 Akim Demaille yacc.c: store token numbers, not token strings That allows users to cover more cases, such as easily filtering some arguments they don't want to expose. But they now have to call yysymbol_name explicitly. * data/skeletons/yacc.c (yysyntax_error_arguments, yysyntax_error): Deal with symbol numbers instead of symbol names. 2020-01-17 Akim Demaille yacc.c: extract yyerror_message_arguments Isolate a function that returns the list of expected and unexpected tokens. It will be exposed to users willing to customize their error messages. * data/skeletons/yacc.c (yyparse_context_t): New. (yyerror_message_arguments): New, extracted from yysyntax_error. 2020-01-17 Akim Demaille regen 2020-01-15 Akim Demaille tests: make AT_PARSE_PARAMS usable at the end of arguments When not empty, AT_PARSE_PARAMS was guaranteed to end with a comma. Remove the trailing comma, so that we can use AT_PARSE_PARAMS at the end of the arguments, not only at the beginning. * tests/local.at: here. Unfortunately, m4_append relies on the macro not being defined whereas we would have preferred it to check for emptiness. So use m4_define/m4_undefine instead of m4_pushdef/m4_popdef. 2020-01-15 Akim Demaille tests: fix AT_BISON_OPTION_PUSHDEFS, AT_BISON_OPTION_POPDEFS pairs * tests/glr-regression.at, tests/java.at, tests/javapush.at: Close properly what is opened. Do not nest. 2020-01-15 Akim Demaille d, java: use traces more alike that of C Same order, same places, same content. * data/skeletons/lalr1.d, data/skeletons/lalr1.java: here. 2020-01-15 Akim Demaille c++: report the stack at the same places as in C Let's have C be the reference, and match it elsewhere. Maybe C is too verbose and some adjustments are needed, but then that would be done in another batch of patches. * data/skeletons/lalr1.cc: Print the stack once we popped after YYERROR, and before emptying the stack at the end of parsing. 2020-01-15 Akim Demaille c++: display the stack in the same order as in C Currently the C and C++ parse traces differ in the order in which the stack is displayed: bottom up in C, top down in C++. Let's stick to the C order. * data/skeletons/stack.hh (stack::iterator, stack::const_iterator) (begin, end): Be forward, not backward. 2020-01-15 Akim Demaille style: avoid redundancy in the tests * tests/local.at (m4_rpatsubst): New. Use it to handle %parse-params. * tests/calc.at: Use %parse-params with several arguments. 2020-01-11 Akim Demaille yacc.c: comment changes In particular, import Adrian Vogelsgesang's comments about LAC from lalr1.cc. * data/skeletons/yacc.c: here. 2020-01-11 Akim Demaille yacc.c: style: double-quote the argument of b4_percent_define_get * data/skeletons/yacc.c: Here, for consistency. 2020-01-11 Akim Demaille yacc.c: introduce yysymbol_name Provide the users with a public API to get the name of the tokens. A thin wrapper around yytname. * data/skeletons/yacc.c (yysymbol_name): New. Use it. 2020-01-11 Akim Demaille CI: check on PPC64le, ARM64 and s390x I was hoping it would help us catch warnings when char is unsigned (see 78bb152a63f711af65364881c434af4c198e1ee0), but it does not seem to help. It's a pity that the compiler is the same all over the place, I would have preferred testing others. * .travis.yml: here. 2020-01-11 Akim Demaille todo: update 2020-01-11 Akim Demaille Merge branch 'maint' into HEAD * maint: gnulib: update lalr1.cc: avoid static_cast glr.c: add missing cast regen package: bump copyrights to 2020 gitignore: update 2020-01-11 Akim Demaille gnulib: update 2020-01-10 Akim Demaille lalr1.cc: avoid static_cast Reported by donmac703. Fixes https://github.com/akimd/bison/issues/20. * data/skeletons/lalr1.cc: here. 2020-01-10 Akim Demaille glr.c: add missing cast Reported by psjo. Fixes https://github.com/akimd/bison/issues/19. * data/skeletons/glr.c (yyprocessOneStack): Here. 2020-01-10 Akim Demaille regen 2020-01-10 Akim Demaille package: bump copyrights to 2020 Run 'make update-copyright'. 2020-01-10 Akim Demaille gitignore: update 2020-01-10 Akim Demaille regen 2020-01-09 Akim Demaille yacc.c: simplify use of YYDPRINTF * data/skeletons/yacc.c (YYDPRINTF): Expand to no-op (instead of nothing) when disabled. Simplify callers. 2020-01-05 Akim Demaille package: bump copyrights to 2020 Run 'make update-copyright'. 2020-01-05 Akim Demaille gitignore: update 2020-01-04 Akim Demaille doc: YYERROR_VERBOSE is no longer supported * doc/bison.texi (Table of Symbols): Remove last reference to it. * NEWS: Be clear about that. 2020-01-04 Akim Demaille glr.c: no longer support YYERROR_VERBOSE * data/skeletons/glr.c: Rather, dispatch directly on parse.error's value. 2020-01-04 Akim Demaille yacc.c: no longer support YYERROR_VERBOSE Supporting YYERROR_VERBOSE via cpp is a nuisance: m4 is in charge of handling alternatives. When adding more options for %define parse.error, supporting both CPP and M4 is too complex. Anyway, YYERROR_VERBOSE was deprecated long ago. * data/skeletons/yacc.c: Use m4 only to handle verbose/simple error messages. 2020-01-03 Akim Demaille yacc.c: avoid negations * data/skeletons/yacc.c (yyerrlab): here. 2019-12-31 Akim Demaille glr.c: clarify yyreportSyntaxError See the previous commit. * data/skeletons/glr.c (yyreportSyntaxError): First compute the arguments of the error message, _then_ th error message size. 2019-12-31 Akim Demaille yacc: restructure and fix yysyntax_error I would like to offer new ways to build the error message. As a first step, let's simplify yysyntax_error whose first loop does two things at the same time: (i) collect the tokens to be reported in the error message, and (ii) accumulate their sizes and possibly return "overflow". Let's pull (ii) in a second step. Then test 525 (regression.at:1193: parse.error=verbose overflow) failed. This test checks that we correctly report "memory overflow" when the error message is too large. However the test is mistaken: it is triggered in a place where there are five (large) expected tokens, so anyway we would not display them, so there is no (memory) overflow here! Transform this test to (i) check that indeed there is no overflow, and (ii) create syntax_error3 which does check the intended behavior, but with four expected tokens. * data/skeletons/yacc.c (yysyntax_error): First compute the list of arguments, then compute yysize. * tests/regression.at (parse.error=verbose overflow): Enhance and fix. 2019-12-31 Akim Demaille tests: also check -Wchar-subscripts GCC's -Wchar-subscripts may report issues on platforms where char is unsigned. Unfortunately the current CI does not reproduce the problem. But that would allow contributors to report issues if the warning appears somewhere. See 139d0655947c87f90af08718618feaaca0e558d7. Problem reported by Andy Fiddaman in: https://lists.gnu.org/r/bug-bison/2019-12/msg00021.html * configure.ac (warn_common): Add -Wchar-subscripts. 2019-12-30 Akim Demaille CI: do not specify the language When we give travis the langugage, it overrides our envvars. Instead of the MATRIX_EVAL trick, just stop specifying the language. 2019-12-30 Akim Demaille CI: remove ICC support, we can no longer use it https://github.com/nemequ/icc-travis/issues/15 2019-12-29 Akim Demaille doc: clean up the description of YYDEBUG * doc/bison.texi: Make it clearer that %define parse.trace is the preferred options. Fix a typo about api.prefix. 2019-12-29 Akim Demaille glr.cc: avoid compiler warnings 381. types.at:366: testing glr.cc api.value.type={double} ... test.cc:207:57: error: "__clang_major__" is not defined, evaluates to 0 [-Werror=undef] 207 | #if defined __APPLE__ && YY_CPLUSPLUS < 201103L && 4 <= __clang_major__ | ^~~~~~~~~~~~~~~ * data/skeletons/glr.cc: Check __clang_major__ before using it. 2019-12-18 Paul Eggert warnings: pacify ‘gcc -Wchar-subscripts’ in yacc.c Problem reported by Andy Fiddaman in: https://lists.gnu.org/r/bug-bison/2019-12/msg00021.html * data/skeletons/yacc.c (yy_reduce_print, yy_lac, yysyntax_error) (yyreturn): If I might be a char, write a[+I] instead of a[I], so that ‘gcc -Wchar-subscripts’ does not complain. 2019-12-14 Akim Demaille doc: formatting changes * doc/bison.texi: No output changes. 2019-12-14 Akim Demaille tests: don't fail if seq is no available As is the case on Solaris. Reported by Dennis Clarke. https://lists.gnu.org/archive/html/bug-bison/2019-12/msg00011.html * examples/c/reccalc/reccalc.test: Skip if there is no seq. 2019-12-11 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2019-12-11 Akim Demaille version 3.5 * NEWS: Record release date. 2019-12-10 Akim Demaille news: prepare for 3.5 2019-12-08 Akim Demaille glr.cc: disable warnings from Clang on macOS $ cat test.cc #include #include ptrdiff_t half_max_capacity = PTRDIFF_MAX; $ clang++-mp-9.0 -pedantic -std=c++98 /tmp/test.cc -c /tmp/test.cc:4:31: warning: 'long long' is a C++11 extension [-Wc++11-long-long] ptrdiff_t half_max_capacity = PTRDIFF_MAX; ^ /Library/Developer/CommandLineTools/SDKs/MacOSX.sdk/usr/include/stdint.h:149:23: note: expanded from macro 'PTRDIFF_MAX' #define PTRDIFF_MAX INT64_MAX ^ /Library/Developer/CommandLineTools/SDKs/MacOSX.sdk/usr/include/stdint.h:75:26: note: expanded from macro 'INT64_MAX' #define INT64_MAX 9223372036854775807LL ^ 1 warning generated. * data/skeletons/glr.cc: here. 2019-12-08 Akim Demaille api.token.raw: fix it in C++ Another breakage revealed by vcsn. * data/skeletons/c++.m4 (yytranslate_): Do not hard code "yy" and "parser", both can be changed by the user. Actually, since we are in the parser itself, there's really no need to qualify the type. 2019-12-08 Akim Demaille c++: fix comments for %code blocks In a project of mine, vcsn, this commit fixes the following comments. --- /tmp/parse.hh 2019-12-08 15:51:24.792934703 +0100 +++ lib/vcsn/rat/parse.hh 2019-12-08 16:00:59.137107503 +0100 @@ -43,7 +43,7 @@ #ifndef YY_YY_USERS_AKIM_SRC_LRDE_2_LIB_VCSN_RAT_PARSE_HH_INCLUDED # define YY_YY_USERS_AKIM_SRC_LRDE_2_LIB_VCSN_RAT_PARSE_HH_INCLUDED -// // "%code requires" blocks. +// "%code requires" blocks. #line 20 "/Users/akim/src/lrde/2/lib/vcsn/rat/parse.yy" #include @@ -1851,7 +1851,7 @@ -// // "%code provides" blocks. +// "%code provides" blocks. #line 60 "/Users/akim/src/lrde/2/lib/vcsn/rat/parse.yy" #define YY_DECL_(Class) \ * data/skeletons/bison.m4 (b4_percent_code_get): Pass an expanded string to b4_comment. 2019-12-08 Akim Demaille parser: pretend we are Bison 3.5 * src/parse-gram.y: Accept we're Bison 3.5. 2019-12-08 Akim Demaille c++: fix spello * data/skeletons/lalr1.cc: here. 2019-12-08 Akim Demaille todo: update * TODO: Schedule some features for 3.6. Remove obsolete stuff. 2019-12-08 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2019-12-08 Akim Demaille version 3.4.92 * NEWS: Record release date. 2019-12-08 Akim Demaille news: fixes Reported by Paul Eggert. https://lists.gnu.org/archive/html/bison-patches/2019-12/msg00014.html * NEWS: here. 2019-12-07 Akim Demaille doc: minor changes * README-hacking.md: here. 2019-12-07 Akim Demaille gnulib: update 2019-12-07 Akim Demaille doc: clearly deprecate YYPRINT * doc/bison.texi (Prologue): Stop using YYPRINT as an example. (The YYPRINT Macro): Clearly show this macro is deprecated. 2019-12-07 Akim Demaille doc: formatting changes * doc/bison.texi: here. No change in content. 2019-12-07 Akim Demaille news: update 2019-12-07 Akim Demaille d: obey parse.error * data/skeletons/lalr1.d (yysyntax_error): Let the dispatch be bison-time, not runtime. 2019-12-07 Akim Demaille c++: also prefer YY_ASSERT to YYASSERT Like the other skeletons. * data/skeletons/variant.hh: here. 2019-12-07 Akim Demaille glr.c: obey the parse.assert %define variable * data/skeletons/glr.c (YYASSERT): Rename as... (YY_ASSERT): this, for consistency with yacc.c, and also to emphasize the fact that this is not for the end user (YY_ prefix). * tests/glr-regression.at: Define parse.assert. 2019-12-07 Akim Demaille c++: beware of short ranges for state numbers Now that we use small integral types, possibly unsigned (e.g., unsigned char), to store state numbers, using -1 to denote an empty state (i.e., a state that stores no semantical value) is very dangerous: it will be confused with state 255, which might be non-empty. Rather than allocating a larger range of state numbers to keep the empty-state apart, let's use the number of a state known to store no value. The initial state, numbered 0, seems to fit perfectly the job. Reported by Frank Heckenbach. https://lists.gnu.org/archive/html/bug-bison/2019-11/msg00016.html * data/skeletons/lalr1.cc (empty_state): Be 0. 2019-12-07 Akim Demaille api.token.raw: check it against api.token.constructor * tests/scanner.at: here. 2019-12-06 Akim Demaille regen 2019-12-06 Akim Demaille warnings: enable -Wuseless-cast, and eliminate warnings Prompted by Frank Heckenbach. https://lists.gnu.org/archive/html/bug-bison/2019-11/msg00016.html. * configure.ac (warn_cxx): Add -Wuseless-cast. * data/skeletons/c.m4 (b4_attribute_define): Define YY_IGNORE_USELESS_CAST_BEGIN and YY_IGNORE_USELESS_CAST_END. * data/skeletons/glr.c (YY_FPRINTF): New, replaces YYFPRINTF, wrapped with YY_IGNORE_USELESS_CAST_BEGIN and YY_IGNORE_USELESS_CAST_END. (YY_DPRINTF): Likewise. * tests/actions.at: Remove useless cast. * tests/headers.at: Adjust. 2019-12-02 Akim Demaille diagnostics: style changes * src/complain.h, src/complain.c: Comment changes. * src/scan-skel.l: Reduce scopes. * data/skeletons/bison.m4: Factor diagnostic functions. 2019-12-02 Akim Demaille glr.c: style changes * data/skeletons/glr.c (yysplitStack): Reduce scopes. * tests/atlocal.in: Formatting changes. 2019-12-01 Akim Demaille c++: get rid of symbol_type::token () It is not used. And its implementation was wrong when api.token.raw was defined, as it was still mapping to the external token numbers, instead of the internal ones. Besides it was provided only when api.token.constructor is defined, yet always declared. * data/skeletons/c++.m4 (by_type::token): Remove, useless. 2019-12-01 Akim Demaille c++: remove useless cast about user_token_number_max_ Reported by Frank Heckenbach. https://lists.gnu.org/archive/html/bug-bison/2019-11/msg00016.html The cast is needed when yytranslate_'s argument type is token_type, i.e., when api.token.constructor is defined. 373. types.at:138: testing lalr1.cc api.value.type=variant api.token.constructor ... ======== Testing with C++ standard flags: '' ../../tests/types.at:138: bison --color=no -fno-caret -o test.cc test.y ../../tests/types.at:138: $CXX $CXXFLAGS $CPPFLAGS $LDFLAGS -o test test.cc $LIBS stderr: test.cc:966:16: error: result of comparison of constant 257 with expression of type 'yy::parser::token_type' (aka 'yy::parser::token::yytokentype') is always true [-Werror,-Wtautological-constant-out-of-range-compare] else if (t <= user_token_number_max_) ~ ^ ~~~~~~~~~~~~~~~~~~~~~~ 1 error generated. It is because it is expected that when api.token.constructor is defined, only symbol constructors will be used, that yytranslate_ then takes a token_type. But it is wrong: we still allow literal characters in this case, as demonstrated by test 373 for instance. %define api.value.type variant %define api.token.constructor %token > '1' '2'; [...] static yy::parser::symbol_type yylex () { static char const input[] = "12"; int res = input[toknum++]; typedef yy::parser::symbol_type symbol; if (res) return symbol (res, std::make_pair (res - '0', res - '0' + 1)); else return symbol (res); } So let yytranslate_ always take an int, which makes the cast truly useless. * data/skeletons/c++.m4, data/skeletons/lalr1.cc (yytranslate_): here. 2019-12-01 Akim Demaille c++: clean a few issues wrt special tokens The C++ implementation of LAC did not skip the $undefined token, probably because it was not exposed. Expose it, and use clearer names. * data/skeletons/c++.m4: Don't define undef_token_ in yytranslate_, but... * data/skeletons/lalr1.cc (yy_undef_token_): here. Use a more precise type to define yy_undef_token_ and yy_error_token_. Unfortunately we move from a compile-time value defined via an enum to a static const member. Eventually we should make it constexpr. Make LAC implementation more alike yacc.c's one. 2019-12-01 Akim Demaille d, java: improve yytranslate and neighbors * data/skeletons/lalr1.d, data/skeletons/lalr1.java: Don't expose yyuser_token_number_max_ and yyundef_token_. Do as in C++: scope them into yytranslate_, and only when api.token.raw is not defined. (yyterror_): Rename as... (yy_error_token_): this. * data/skeletons/lalr1.d (token_number_type): New. Use it. Can't be done in the Java backend, as Java does not have type aliases. 2019-12-01 Akim Demaille d, java: get rid of a useless table * data/skeletons/lalr1.d, data/skeletons/lalr1.java (yytoken_number_): Remove, useless. Was used in ancient C skeletons to support YYPRINT, long obsoleted by %printer. 2019-11-30 Akim Demaille c++, d, java: remove yyerrcode It is not used at all. We will remove it also from yacc.c, but later (see TODO). * data/skeletons/lalr1.cc, data/skeletons/lalr1.d, * data/skeletons/lalr1.java (yyerrcode_): Remove. 2019-11-30 Akim Demaille c++: improve typing * data/skeletons/lalr1.cc (yysyntax_error_): symbol_type::type_get returns a symbol_number_type (which is indeed an int). 2019-11-30 Akim Demaille c++: remove useless cast about yyeof_ Reported by Frank Heckenbach. https://lists.gnu.org/archive/html/bug-bison/2019-11/msg00016.html * data/skeletons/c++.m4 (b4_yytranslate_define): Don't use yyeof_ as if it had two different types. It is used once against the input argument, which is the value returned by yylex, which is an "external token number", typically an int. It is also used as output type, an "internal symbol number". It turns out that in both cases we mean "0", but let's keep yyeof_ only for the case "internal symbol number", i.e., _after_ conversion by yytranslate. This frees us from one cast. 2019-11-30 Akim Demaille glr: style change * data/skeletons/glr.c (YYDPRINTF): Expand into an empty statement, instead of nothing. Simplify callers. 2019-11-30 Akim Demaille glr: remove useless casts Reported by GCC's -Wuseless-cast. * data/skeletons/glr.c: Don't cast to yybool, it's useless. 2019-11-29 Akim Demaille yacc.c, glr.c: fix crash when reporting errors in consistent states The current code for yysyntax_error for %define parse.error verbose is fishy (given that YYEMPTY is -2, invalid argument for yytname[]): static int yysyntax_error ([...]) { YYPTRDIFF_T yysize0 = yytnamerr (YY_NULLPTR, yytname[yytoken]); [...] if (yytoken != YYEMPTY) A nearby comment reports The only way there can be no lookahead present (in yychar) is if this state is a consistent state with a default action. Thus, detecting the absence of a lookahead is sufficient to determine that there is no unexpected or expected token to report. In that case, just report a simple "syntax error". So it _is_ possible to call yysyntax_error with yytoken == YYEMPTY, albeit quite difficult when meaning to, so virtually impossible by accident (after all, there was never a bug report about this). I failed to produce a test case, but Joel E. Denny provided me with one (added to the test suite below). The yacc.c skeleton fails on this, and once fixed dies on a second problem. The glr.c skeleton was also dying, but immediately of this second problem. Indeed we were not allocating space for the error message's final \0. This was hidden by the fact that we only had error messages with at least an unexpected token displayed, so with at least one "%s" in the format string, whose size (2) was included (incorrectly) in the final size of the message (where the %s have been replaced by the actual content). * data/skeletons/glr.c, data/skeletons/yacc.c (yysyntax_error): Do not invoke yytnamerr on YYEMPTY. Clarify the computation of the length of the _final_ error message, with the NUL terminator but without the '%s's. * tests/conflicts.at (Syntax error in consistent error state): New, contributed by Joel E. Denny. 2019-11-26 Akim Demaille tests: avoid creating files whose name collide with standard headers Having a file named "exception" is risky: the compiler might use that file in #include. Reported by 马俊 . * tests/local.at (AT_SKIP_IF_EXCEPTION_SUPPORT_IS_POOR): Generate 'exceptions', not 'exception'. 2019-11-22 Akim Demaille doc: more details about the test suite * README-hacking.md: here. 2019-11-20 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2019-11-20 Akim Demaille version 3.4.91 * NEWS: Record release date. 2019-11-20 Akim Demaille style: pacify syntax-check * cfg.mk: No need to translate *.md files. * data/skeletons/glr.c, data/skeletons/yacc.c: Fix space issues. 2019-11-19 Akim Demaille gnulib: update 2019-11-18 Akim Demaille doc: don't promote dangling aliases String literals as tokens serve two distinct purposes: freeing from having to implement the keyword matching in the scanner, and improving error messages. Most of the time both can be achieved at the same time, but on occasions, it does not work so well. We promote their use for error messages. We will also still support the former case, but it is _not_ the recommended approach. * doc/bison.texi (Tokens from Literals): Clearly state that we don't recommend looking up the token types in the list of token names. 2019-11-17 Akim Demaille diagnostics: complain about undeclared string tokens String literals, which allow for better error messages, are (too) liberally accepted by Bison, which might result in silent errors. For instance %type cond "condition" does not define “condition” as a string alias to 'cond' (nonterminal symbols do not have string aliases). It is rather equivalent to %nterm cond %token "condition" i.e., it gives the type 'exVal' to the "condition" token, which was clearly not the intention. Introduce -Wdangling-alias to catch this. * src/complain.h, src/complain.c: Add support for -Wdangling-alias. (argmatch_warning_args): Sort. * src/symtab.c (symbol_check_defined): Complain about dangling aliases. * doc/bison.texi: Document it. * tests/input.at (Dangling aliases): New test. 2019-11-17 Akim Demaille diagnostics: yacc reserves %type to nonterminals On %token TOKEN1 %type TOKEN1 TOKEN2 't' %token TOKEN2 %% expr: bison -Wyacc gives input.y:2.15-20: warning: POSIX yacc reserves %type to nonterminals [-Wyacc] 2 | %type TOKEN1 TOKEN2 't' | ^~~~~~ input.y:2.29-31: warning: POSIX yacc reserves %type to nonterminals [-Wyacc] 2 | %type TOKEN1 TOKEN2 't' | ^~~ input.y:2.22-27: warning: POSIX yacc reserves %type to nonterminals [-Wyacc] 2 | %type TOKEN1 TOKEN2 't' | ^~~~~~ The messages appear to be out of order, but they are emitted when the error is found. * src/symtab.h (symbol_class): Add pct_type_sym, used to denote symbols appearing in %type. * src/symtab.c (complain_pct_type_on_token): New. (symbol_class_set): Check that %type is not applied to tokens. (symbol_check_defined): pct_type_sym also means undefined. * src/parse-gram.y (symbol_decl.1): Set the class to pct_type_sym. * src/reader.c (grammar_current_rule_begin): pct_type_sym also means undefined. * tests/input.at (Yacc's %type): New. 2019-11-16 Akim Demaille doc: promote %nterm over %type As an extension to POSIX Yacc, Bison's %type accepts tokens. Unfortunately with string literals as implicit tokens, this is misleading, and led some users to write %type cond "condition" believing that "condition" would be associated to the 'cond' nonterminal (see https://github.com/apache/httpd/pull/72). * doc/bison.texi: Promote %nterm rather than %type to declare the type of nonterminals. 2019-11-16 Akim Demaille doc: formatting changes * doc/bison.texi: No visible changes. 2019-11-16 Akim Demaille doc: work around warnings when Flex C output is compiled in C++ * doc/bison.texi (calc++/scanner.ll): here. While at it, clarify clang vs. warnings. 2019-11-16 Akim Demaille tests: be robust to old Perl versions on Cygwin Reported by Denis Excoffier. https://lists.gnu.org/archive/html/bug-bison/2019-11/msg00008.html. * tests/output.at: Be sure to remove back up files. 2019-11-16 Akim Demaille regen 2019-11-12 kaneko y gram.c: Fix condition of aver * src/gram.c (grammar_dump): Fix condition of aver. What we want to check is that rhs is followed by its rule. 2019-11-11 Akim Demaille doc: clarify build instructions * README: A few fixes. Explain how to install color support. * README-hacking: Rename as... * README-hacking.md: this, and convert to Markdown. Improve typography. Improve explanations about update-test. 2019-11-11 Akim Demaille gnulib: update 2019-11-11 Yuichiro Kaneko gram.c: also print terminals in grammar_dump * src/gram.c (grammar_dump): Print terminals likewise non terminals. * tests/sets.at (Reduced Grammar): Update test case to catch up the change and add a test case where prec and assoc are used. 2019-11-10 Akim Demaille doc: work around Texinfo 6.7 bug When @code is used in a @deftype... definition, it issues quotes. Remove them. See https://lists.gnu.org/archive/html/help-texinfo/2019-11/msg00004.html. * doc/local.mk: here. 2019-11-09 Akim Demaille doc: formatting changes * doc/bison.texi: Wrap lines. No semantical difference. 2019-11-09 Akim Demaille doc: use upper case for tokens * doc/bison.texi: here. 2019-11-07 Akim Demaille doc: type-face fixes * doc/bison.texi: Use @code for types in function definitions. 2019-11-06 Akim Demaille c++: expose the type used to store line and column numbers * data/skeletons/location.cc (position::counter_type) (location::counter_type): New. Use them. * doc/bison.texi (C++ position, C++ location): Adjust. 2019-11-03 Akim Demaille tests: fix comment and adjust to locale names on GNU/Linux Reported by Denis Excoffier. * tests/diagnostics.at: here. 2019-11-03 Akim Demaille tests: really check complaints from m4 * tests/diagnostics.at (Locations from M4, Tabulations and multibyte characters from M4): These tests are actually checking a message coming from C, not from M4. Replace with... (Complaints from M4): This. 2019-11-03 Akim Demaille tests: simplify prologue * tests/testsuite.h: We no longer load gnulib in the tests. 2019-11-03 Akim Demaille diagnostics: add missing translation * src/muscle-tab.c (muscle_percent_define_check_kind): Here. 2019-11-02 Akim Demaille c++: fix old cast warnings We still have a few old C casts in lalr1.cc, let's get rid of them. Reported by Frank Heckenbach. Actually, let's monitor all our casts using easy to grep macros. Let's use these macros to use the C++ standard casts when we are in C++. * data/skeletons/c.m4 (b4_cast_define): New. * data/skeletons/glr.c, data/skeletons/glr.cc, * data/skeletons/lalr1.cc, data/skeletons/stack.hh, * data/skeletons/yacc.c: Use it and/or its casts. * tests/actions.at, tests/cxx-type.at, * tests/glr-regression.at, tests/headers.at, tests/torture.at, * tests/types.at: Use YY_CAST instead of C casts. * configure.ac (warn_cxx): Add -Wold-style-cast. * doc/bison.texi: Disable it. 2019-11-01 Akim Demaille tests: be robust to tput errors Reported by Denis Excoffier. * tests/bison.in: here. 2019-11-01 Akim Demaille git: update ignores I don't understand what happened in 10acc148bb90fac8a52a5d35f2bd18bd824c1639. 2019-10-29 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2019-10-29 Akim Demaille version 3.4.90 * NEWS: Record release date. 2019-10-29 Akim Demaille C++: finish propagating the unsigned->signed conversion in locations * data/skeletons/location.cc: Remove the u (for unsigned) suffix from the initial line and column. * NEWS: AFAICT, only C++ backends have their location types changed. 2019-10-29 Akim Demaille style: fix cpp indentation Reported by syntax-check. * src/system.h: here. 2019-10-29 Akim Demaille style: glr.c: comment changes * data/skeletons/glr.c: here. 2019-10-26 Akim Demaille CI: pass -O1 to GCC8 with sanitizers This build never finishes in the 50min credit given by Travis. See if with optimizations it works better. * .travis.yml: here. 2019-10-26 Akim Demaille reader: reduce the "scope" of global variables We have too many global variables, adding structure would help. For a start, let's hide some of the variables closer to their usage. * src/getargs.c, src/files.h (current_file): Move to... * src/scan-gram.c: here. * src/scan-gram.h (gram_in, gram__flex_debug): Remove, make them private to the scanner. * src/reader.h, src/reader.c (reader): Take a grammar file as argument. Move the handling of scanner variables to... * src/scan-gram.l (gram_scanner_open, gram_scanner_close): here. (gram_scanner_initialize): Remove, replaced by gram_scanner_open. * src/main.c: Adjust. 2019-10-26 Akim Demaille regen 2019-10-26 Akim Demaille parser: use grammar_file instead of current_file * src/parse-gram (%initial-action): here. (handle_skeleton): Don't depend on the current file name to look for "local" skeletons (subject to changes coming from "#lines"): depend only on the initial file name, the one given on the command line. 2019-10-26 Akim Demaille diagnostics: use grammar_file instead of current_file Currently there are two globals denoting the input file: grammar_file is the one from the command line, and current_file which might change because of #line. Use only the former. * src/complain.c (error_message): here. * tests/diagnostics.at: Adjust. 2019-10-25 Akim Demaille reader: let symtab deal with the symbols * src/reader.c (reader): Move the setting up of the builtin symbols to... * src/symtab.c (symbols_new): here. 2019-10-25 Akim Demaille style: remove incorrect comment Reported by Paul Eggert. * src/system.h: here. 2019-10-24 Akim Demaille lalr1.cc: fix previous commit: printing of state numbers * data/skeletons/lalr1.cc: Printing a char prints... a char. Print ints instead. 2019-10-24 Akim Demaille lalr1.cc: use computed state types This skeleton uses a single stack of state structures, so it is less likely to benefit from a stack size reduction than yacc.c (which uses several stacks: state number, value and location). But it will reduce the size of the LAC stack. This skeleton was already using int for state numbers, so, contrary to yacc.c, this brings nothing for large automata. Overall, it is still nicer to make the skeletons alike. * data/skeletons/lalr1.cc (state_type): Here. 2019-10-24 kaneko y README: Fix a typo * README: Fix a typo. Git command name is submodule. 2019-10-24 Akim Demaille examples: fix missing dependencies Reported by Thomas Petazzoni. https://lists.gnu.org/archive/html/bug-bison/2019-08/msg00000.html * examples/c/reccalc/local.mk: Complete dependencies, including for earlier versions of Automake (for sake of our CI, on top of Ubuntu Xenial/Bionic, which feature only Automake 1.15). (%D%/scan.c %D%/scan.h): Upgrade to the full version provided in Automake's documentation. 2019-10-24 Akim Demaille diagnostics: simplify location handling Locations start at line 1. Don't accept line 0. * src/location.c (location_print): Don't print locations with line 0. (location_caret): Simplify. 2019-10-24 Akim Demaille build: reenable -Wtype-limits See https://lists.gnu.org/archive/html/bug-bison/2019-10/msg00061.html to https://lists.gnu.org/archive/html/bug-bison/2019-10/msg00073.html. Paul Eggert's changes in gnulib do fix the issue for modern GCCs (7, 8, 9) on macOS. Unfortunately these warnings are back on the CI (GNU/Linux) with GCC 4.6, 4.7, (not 4.8) and 4.9. Disable the warning locally. * configure.ac (warn_common, warn_tests): Remove -Wtype-limits. * src/system.h (IGNORE_TYPE_LIMITS_BEGIN, IGNORE_TYPE_LIMITS_END): New. * src/InadequacyList.c, src/parse-gram.c, src/parse-gram.y, * src/symtab.c: Use it. 2019-10-24 Akim Demaille build: remove dmalloc support Today sanitizers are a better alternative. * m4/dmalloc.m4: Remove. * configure.ac, src/system.h: Adjust. 2019-10-23 Akim Demaille gitignore: update 2019-10-23 Paul Eggert build: update gnulib submodule to latest 2019-10-23 Yuichiro Kaneko style: update comment in reader.c rrhs and rlhs were removed by b2ed6e5826e772162719db595446b2c58e4ac5d6. * src/reader.c (packgram): Update comment. 2019-10-22 kaneko y yacc.c: fix a typo * data/skeletons/yacc.c (yysetstate): fix comment. 2019-10-22 Akim Demaille style: pacify syntax-check * doc/.gitignore, src/complain.c, src/getargs.c, * src/output.c: here. 2019-10-21 Akim Demaille main: also free memory on errors * src/derives.c (derives_free): Beware of NULL. * src/main.c (main): Let the 'finish' label include memory release. 2019-10-21 Akim Demaille gnulib: update To get bitset_free accept NULL. See https://lists.gnu.org/archive/html/bug-gnulib/2019-10/msg00054.html 2019-10-21 Akim Demaille style: reduce scope in derives * src/derives.c: here. And prefer prefix to postfix increment. 2019-10-21 Akim Demaille build: disable -Wtautological-constant-out-of-range-compare Also see e31f92495ce14a5d924b148c8ea1470003cc47c1 and https://lists.gnu.org/archive/html/bug-bison/2019-10/msg00061.html * configure.ac (warn_common): Disable -Wtautological-constant-out-of-range-compare. (warn_tests): Restore it. 2019-10-21 Akim Demaille CI: formatting changes * .travis.yml: Use the single line form of lists, when reduced to a singletons. 2019-10-21 Akim Demaille CI: rename jobs * .travis.yml (compile, test): Rename as... (dist, check): these, which are more traditional for GNU projects. 2019-10-21 Akim Demaille doc: update README * README: Be clearer that README-hacking _must_ be read. Convert to Markdown. 2019-10-21 Akim Demaille bootstrap: relieve developpers from Gettext version mismatch issues * .travis.yml (compile): Move the workaround from here... * bootstrap.conf (bootstrap_epilogue): to there. 2019-10-20 Akim Demaille tests: beware of GCC9 warnings in push mode This is really weird: GCC points to the LHS of the assignment... 260. headers.at:184: testing Sane headers: api.pure api.push-pull=both ... tests/headers.at:184: COLUMNS=1000; export COLUMNS; bison --color=no -fno-caret -d -o input.c input.y tests/headers.at:184: $CC $CFLAGS $CPPFLAGS -c -o input.o input.c stderr: input.c: In function 'yyparse': input.c:1276:16: error: 'yylval' may be used uninitialized in this function [-Werror=maybe-uninitialized] 1276 | yylval = *yypushed_val; | ~~~~~~~^~~~~~~~~~~~~~~ input.c: In function 'yypull_parse': input.c:1276:16: error: 'yylval' may be used uninitialized in this function [-Werror=maybe-uninitialized] 1276 | yylval = *yypushed_val; | ~~~~~~~^~~~~~~~~~~~~~~ cc1: all warnings being treated as errors stdout: tests/headers.at:184: exit code was 1, expected 0 See also d87c8ac79ab844d6a7a4f5103dcf7a842d18b611 and 9645a2b20ee7cbfa8bb4ac2237f87d598afe349c. * tests/headers.at (Several parsers, Several parsers): Disable these warnings when in push parser. 2019-10-20 Akim Demaille CI: try GCC9 and Clang9 The logs show: Disallowing sources: llvm-toolchain-bionic-8, ubuntu-toolchain-r-test To add unlisted APT sources, follow instructions in https://docs.travis-ci.com/user/installing-dependencies#Installing-Packages-with-the-APT-Addon * .travis.yml: Remove a few apt sources which are ignored in Bionic (e.g., see https://github.com/travis-ci/apt-source-safelist/issues/410). Where needed, use sources/sourceline instead. Also, don't use -DNDEBUG with older builds. 2019-10-20 Akim Demaille parser: clarify version checking * src/parse-gram.y: Use the same conventions for gnulib as elsewhere: . (str_to_version): New. (handle_require): Use it. Prefer < to >. 2019-10-20 Akim Demaille build: disable -Wtype-limits, except in the test suite The current implementation of lib/intprops.h results in "unsigned < 0" comparisons, which triggers warnings. See https://lists.gnu.org/archive/html/bug-bison/2019-10/msg00061.html * configure.ac (warn_common): Disable -Wtype-limits. (warn_tests): Restore it. 2019-10-17 Paul Eggert c++: port to Sun C++ 5.12 The documentation for Oracle Solaris Studio 12.3 (Sun C++ 5.12 2011/11/16) says it supports C++03. This compiler rejects the location.cc use of std::max for some reason; I don’t know why since I don’t use C++ as a rule. The simplest workaround is to open-code ‘max’. * data/skeletons/location.cc (add_): Do max by hand rather than relying on std::max. Don’t include ; no longer needed. 2019-10-17 Paul Eggert regen 2019-10-17 Paul Eggert tests: port to Solaris 10 grep * tests/scanner.at (Token numbers: $1): Use $EGREP, not grep -E. 2019-10-17 Paul Eggert tests: port to Solaris 10 sed As documented in the Autoconf manual, Solaris 10 sed rejects script labels contianing more than 7 characters. POSIX requires support for at least 8 characters, but we might as well be portable to Solaris 10 which is still supported. * tests/local.at (AT_SETS_CHECK): Use only the first 7 characters in sed labels. 2019-10-17 Paul Eggert bison: check for int overflow in token numbers * src/symtab.c: Include intprops.h (symbol_user_token_number_set): Don’t allow user_token_number == INT_MAX because too much other code adds 1 to the user token number. (symbols_token_translations_init): Complain on integer overflow instead of indulging in undefined behavior. 2019-10-17 Paul Eggert bison: check for int overflow when scanning * src/scan-gram.l: Include errno.h, for errno. (scan_integer, handle_syncline): Check for integer overflow. * tests/input.at (too-large.y): Adjust to match new diagnostics. 2019-10-17 Paul Eggert bison: check version numbers more carefully * src/parse-gram.y: Include intprops.h. (handle_require): Don’t indulge in undefined behavior if the major or minor number is out of range. Instead, check that the resulting value is nonnegative, fits in int, and that the minor number is less than 100. Also, check that a number was parsed. 2019-10-17 Paul Eggert c: port YY_ATTRIBUTE_UNUSED to Sun C 5.12 Sun C 5.12 defines __SUNPRO_C to 0x5120 but diagnoses ‘__attribute__ ((__unused__))’. Change the ifdefs to use the same method as Gnulib in this area. * data/skeletons/c.m4 (YY_ATTRIBUTE): Remove, since not all attributes were added in the same compiler version. (YY_ATTRIBUTE_PURE, YY_ATTRIBUTE_UNUSED): Use specific GCC version for each attribute. Pay no attention to __SUNPRO_C. * tests/headers.at (Several parsers): Tighten tests accordingly. 2019-10-17 Paul Eggert c: improve port of stdint.h usage to pre-C99 Oracle Solaris Studio 12.3 (Sun C 5.12 2011/11/16) by default does not conform to C99; it defines __STDC_VERSION__ to be 199409L, so the Bison code does not include (not required by C89 amendment 1) even though this compiler does have . On this platform defines INT_LEAST8_MAX (POSIX allows this) so the skeleton got confused and thought that had been included even though it wasn’t. * data/skeletons/c.m4 (b4_c99_int_type_define) [!__PTRDIFF_MAX__]: Always include . (YY_STDINT_H): Define when was included. All uses of expressions like ‘defined INT_LEAST8_MAX’ changed to ‘defined YY_STDINT_H’, since Sun C 5.12 defines macros like INT_LEAST8_MAX but does not declare types like int_least8_t. 2019-10-17 Paul Eggert gnulib:update 2019-10-17 Paul Eggert autoconf:update 2019-10-15 Akim Demaille TODO: more updates 2019-10-15 Akim Demaille TODO: update 2019-10-15 Akim Demaille yacc: rename types for states * data/skeletons/yacc.c (yy_state_num): Rename as... (yy_state_t): this. (yy_state_fast_t): New. Use it. 2019-10-15 Akim Demaille glr: style changes * data/skeletons/glr.c (yytnamerr): here. (yyprocessOneStack): Initialize variables. 2019-10-15 Akim Demaille yacc: style changes * data/skeletons/yacc.c: Move call to lac discard to clarify the shifting of the token. Like in lalr1.cc. 2019-10-15 Akim Demaille tests: avoid $(...) Reported by Paul Eggert. * tests/local.at (AT_DATA_NO_FINAL_EOL): here. 2019-10-14 Akim Demaille tests: use a portable 'truncate' implementation Suggested by Paul Eggert. https://lists.gnu.org/archive/html/bison-patches/2019-10/msg00044.html * tests/local.at (AT_DATA_NO_FINAL_EOL): Use dd instead of perl. 2019-10-13 Akim Demaille tests: factor the generation of files without the final eol AFAICT Autotest 2.69 still does not support AT_DATA without the final eol. * tests/local.at (AT_DATA_NO_FINAL_EOL): New. * tests/input.at: Use it. 2019-10-13 Akim Demaille tests: refactor the handling of Perl Let's make a difference between places where Perl is required for the test (AT_PERL_REQUIRE), and the places where it's used to run the test, but it's not not to run the test (AT_PERL_CHECK). * tests/local.at (AT_REQUIRE): New. (AT_PERL_CHECK, AT_PERL_REQUIRE): New. Use them where appropriate. * tests/local.mk ($(TESTSUITE)): Beware not to start the line with '-pi' if Perl is empty, as Make understands this as "it's ok to fail". Which it is not. 2019-10-12 Akim Demaille d: comment changes * data/skeletons/lalr1.d: Here. 2019-10-12 Akim Demaille i18n: don't push too hard for '…' Suggested by Paul Eggert. * src/location.c (ellipsis): Clarify comment for translators. 2019-10-11 Akim Demaille regen 2019-10-11 Akim Demaille glr: display line numbers in traces Suggested by Lars Maier. * data/skeletons/glr.c: Also display rule locations when rules are deferred, and rejected. 2019-10-11 Akim Demaille tests: be really robust to Perl missing My previous tests (with ./configure PERL=false) have been fooled by configure, that managed to find perl anyway. This time, I ran this on a Fedora in Docker, without Perl. * tests/calc.at, tests/diagnostics.at, tests/headers.at, * tests/input.at, tests/local.at, tests/named-refs.at, * tests/output.at, tests/regression.at, tests/skeletons.at, * tests/synclines.at, tests/torture.at: Don't require Perl. 2019-10-10 Akim Demaille configure: perl is not required But it's used in various places, including in some tests. * configure.ac: here. 2019-10-10 Akim Demaille news: update 2019-10-10 Akim Demaille diagnostics: prefer "…" to "..." if the locale supports it * src/location.c (ellipsis, ellipsize): New. Use them. 2019-10-10 Paul Eggert c: improve patch for UCHAR_MAX etc. problem * data/skeletons/c.m4 (b4_c99_int_type_define): Reorder to put the signed types first, since they’re simpler and this keeps similar code closer. For signed types, don’t bother checking whether the type promotes to int since the type must be signed anyway. For unsigned types, protect a test like ‘UCHAR_MAX <= INT_MAX’ with ‘!defined __UINT_LEAST8_MAX__’, as otherwise the logic is wrong for oddball platforms; and once we do that, there should no need for ‘defined INT_MAX’ so remove that. 2019-10-10 Akim Demaille tests: do not depend on config.h Currently we face test suite failures in different environments, because of a conflict between the definitions of isnan by gnulib, and by the C++ library: 262. headers.at:186: testing Sane headers: %locations %debug c++ ... ./headers.at:186: COLUMNS=1000; export COLUMNS; bison --color=no -fno-caret -d -o input.cc input.y ./headers.at:186: $CXX $CXXFLAGS $CPPFLAGS -c -o input.o input.cc stderr: In file included from /usr/include/c++/4.8.2/cmath:44:0, from /usr/include/c++/4.8.2/random:38, from /usr/include/c++/4.8.2/bits/stl_algo.h:65, from /usr/include/c++/4.8.2/algorithm:62, from location.hh:41, from input.hh:90, from input.cc:50: /u/cs/fac/eggert/src/gnu/bison/lib/math.h: In function 'bool isnan(double)': /u/cs/fac/eggert/src/gnu/bison/lib/math.h:2849:1: error: new declaration 'bool isnan(double)' _GL_MATH_CXX_REAL_FLOATING_DECL_2 (isnan, isnan, bool) ^ In file included from /usr/include/features.h:375:0, from /usr/include/c++/4.8.2/x86_64-redhat-linux/bits/os_defines.h:39, from /usr/include/c++/4.8.2/x86_64-redhat-linux/bits/c++config.h:2097, from /usr/include/c++/4.8.2/cstdlib:41, from input.hh:48, from input.cc:50: /usr/include/bits/mathcalls.h:235:1: error: ambiguates old declaration 'int isnan(double)' __MATHDECL_1 (int,isnan,, (_Mdouble_ __value)) __attribute__ ((__const__)); ^ There might be something to do in gnulib about this, but I believe that gnulib should not be used in the test suite in the first place. The test suite should work with other compilers than the one used to compile the package. For a start, Bison sources are more demanding (C99) than the generated parsers. Last time I tried, tcc for example, was not able to compile Bison, yet our generated parsers should compile cleanly with it. Besides the problem at hand is with the C++ compiler, with is not the one used to set up gnulib at configuration-time (config.h is mainly built from probing the C compiler). We should really not depend on gnulib in tests. This was introduced in 2001 to check whether including stdlib.h/string.h is safe thanks to STDC_HEADERS (2ce1014469742b5c6618daf8506b69e38787c7d5). Today, we assume at least a C90 compiler, it should be safe enough. * tests/local.at, tests/testsuite.h: Do not include config.h. * tests/atlocal.in (conftest.cc): Likewise. (CPPFLAGS): Do not expose lib/, as because of this we might picked up gnulib replacement headers for system headers. * tests/input.at: Use int instead of ptrdiff_t, for easier portability (some machine on the CI did not find ptrdiff_t). * tests/c++.at: Add missing include for getchar. 2019-10-10 Akim Demaille doc: spell check * doc/bison.texi: Remove the index about yyoutput, it is no longer documented. Spell check. 2019-10-10 Akim Demaille tests: style changes * tests/actions.at: Prefer printf to fprintf. Prefer yyo to yyoutput in %printer. 2019-10-10 Akim Demaille tests: formatting changes * tests/actions.at, tests/local.at: here. 2019-10-10 Akim Demaille tests: add missing includes * tests/actions.at, tests/c++.at, tests/headers.at, * tests/regression.at: here. 2019-10-10 Akim Demaille c: don't assume that UCHAR_MAX, etc. are defined A number of portability issues with GCC 4.6 .. 4.9 (inclusive): input.c:184:7: error: "UCHAR_MAX" is not defined [-Werror=undef] #elif UCHAR_MAX <= INT_MAX ^ input.c:184:20: error: "INT_MAX" is not defined [-Werror=undef] #elif UCHAR_MAX <= INT_MAX ^ input.c:202:7: error: "USHRT_MAX" is not defined [-Werror=undef] #elif USHRT_MAX <= INT_MAX ^ input.c:202:20: error: "INT_MAX" is not defined [-Werror=undef] #elif USHRT_MAX <= INT_MAX ^ * data/skeletons/c.m4 (b4_c99_int_type_define): Don't assume they are defined. 2019-10-09 Akim Demaille configure: don't require Flex Flex should not be required to build Bison or run the test suite (of course it is needed for maintaining Bison). Yet the Automake conditional FLEX_WORKS does not work. * m4/flex.m4 (_AC_PROG_LEX_YYTEXT_DECL): Since this is called conditionally, don't define LEX_IS_FLEX here, but rather... (AC_PROG_LEX): here. * configure.ac: Be more cautious about possibly undefined variables. 2019-10-07 Paul Eggert Move the integer-type selection into c.m4 That way, glr.c can use it too. * data/skeletons/c.m4 (b4_int_type): Do not special-case ‘char’; it’s not worth the trouble, as clang complains about char subscripts. (b4_c99_int_type, b4_c99_int_type_define): New macros, taken from yacc.c. * data/skeletons/glr.c: Use b4_int_type_define. * data/skeletons/yacc.c (b4_int_type): Remove, since there’s no longer any need to redefine it. Use b4_c99_int_type_define rather than its body. 2019-10-07 Paul Eggert Use “least” types for integers in Yacc tables This changes the Yacc skeleton to use “least” integer types to keep tables smaller on some platforms, which should lessen cache pressure. Since Bison uses the Yacc skeleton, it follows suit. * data/skeletons/yacc.c: Include limits.h and stdint.h if this seems to be needed. (yytype_uint8, yytype_int8, yytype_uint16, yytype_int16): If available, use GCC predefined macros __INT_MAX__ etc. to select a “least” type, as this avoids namespace hassles. Otherwise, if available fall back on selecting a “least” type via the C99 macros INT_MAX, INT_LEAST8_MAX, etc. Otherwise, fall further back on one of the builtin C99 types signed char, short, and int. Make sure that any selected type promotes to int. Ignore any macros YYTYPE_INT16, YYTYPE_INT8, YYTYPE_UINT16, YYTYPE_UINT8 defined by the user. (ptrdiff_t, PTRDIFF_MAX): Simplify in the light of the above. (yytype_uint8, yytype_uint16): Do not assume that unsigned char and unsigned short promote to int, as this isn’t true on some platforms (e.g., TI TMS320C55x). * src/parse-gram.y (YYTYPE_INT16, YYTYPE_INT8, YYTYPE_UINT16) (YYTYPE_UINT8): Remove, as these are no longer effective. 2019-10-06 Paul Eggert Port better to C++ platforms * data/skeletons/yacc.c (YYPTRDIFF_T, YYPTRDIFF_MAXIMUM): Default to long, not int. (yy_lac_stack_realloc, yy_lac, yytnamerr, yyparse): Avoid casts to YYPTRDIFF_T that were masking the problem. 2019-10-06 Paul Eggert Work around GCC 4.8 false alarms without casts * data/skeletons/yacc.c (yyparse): Initialize yyes_capacity with a signed expression. * tests/local.at (AT_YYLEX_DEFINE(c)): Use enum to avoid cast. 2019-10-06 Akim Demaille regen 2019-10-06 Akim Demaille tests: make recheck * tests/local.mk (recheck): New. 2019-10-06 Akim Demaille diagnostics: also show suggested %empty * src/reader.c (grammar_rule_check_and_complete): Suggest to add %empty. * tests/actions.at, tests/diagnostics.at: Adjust expectations. 2019-10-06 Akim Demaille diagnostics: sort symbols per location Because the checking of the grammar is made by phases after the whole grammar was read, we sometimes have diagnostics that look weird. In some case, within one type of checking, the entities are not checked in the order in which they appear in the file. For instance, checking symbols is done on the list of symbols sorted by tag: foo.y:1.20-22: warning: symbol BAR is used, but is not defined as a token and has no rules [-Wother] 1 | %destructor {} QUX BAR | ^~~ foo.y:1.16-18: warning: symbol QUX is used, but is not defined as a token and has no rules [-Wother] 1 | %destructor {} QUX BAR | ^~~ Let's sort them by location instead: foo.y:1.16-18: warning: symbol 'QUX' is used, but is not defined as a token and has no rules [-Wother] 1 | %destructor {} QUX BAR | ^~~ foo.y:1.20-22: warning: symbol 'BAR' is used, but is not defined as a token and has no rules [-Wother] 1 | %destructor {} QUX BAR | ^~~ * src/location.h (location_cmp): Be robust to empty file names. * src/symtab.c (symbol_cmp): Sort by location. * tests/input.at: Adjust expectations. 2019-10-06 Akim Demaille diagnostics: suggest fixes for undeclared symbols From input.y:1.17-19: warning: symbol baz is used, but is not defined as a token and has no rules [-Wother] 1 | %printer {} foo baz | ^~~ to input.y:1.17-19: warning: symbol 'baz' is used, but is not defined as a token and has no rules; did you mean 'bar'? [-Wother] 1 | %printer {} foo baz | ^~~ | bar * bootstrap.conf: We need fstrcmp. * src/symtab.c (symbol_from_uniqstr_fuzzy): New. (complain_symbol_undeclared): Use it. * tests/diagnostics.at (Suggestions): New. * data/bison-default.css (insertion): Rename as... (fixit-insert): this, as this is what GCC uses. 2019-10-06 Akim Demaille style: isolate complain_symbol_undeclared * src/symtab.c (complain_symbol_undeclared): New. Use it. Use quote on the guilty symbol (like GCC does, and we also do elsewhere). * tests/input.at: Adjust. 2019-10-06 Akim Demaille style: simplify the handling of symbol and semantic_type tables Both are stored in a hash, and back in the days, we used to iterate over these tables using hash_do_for_each. However, the order of traversal was not deterministic, which was a nuisance for deterministic output (and therefore also a problem for tests). So at some point (83b60c97ee1f98bb1f15ffa38acdc4cc765515f5) we generated a sorted list of these symbols, and symbols_do actually iterated on that list. But we kept the constraints of using hash_do_for_each, which requires a lot of ceremonial code, and makes it hard/unnatural to preserve data between iterations (see the next commit). Alas, this is C, not C++. Let's remove this abstraction, and directly iterate on the sorted tables. * src/symtab.c (symbols_do): Remove. Adjust callers to use a simple for-loop instead. (table_sort): New. (symbols_check_defined): Use it. (symbol_check_defined_processor, symbol_pack_processor) (semantic_type_check_defined_processor, symbol_translation_processor): Remove. Simplify the corresponding functions (that no longer need to return a bool). 2019-10-06 Akim Demaille diagnostics: display suggested update after the caret-info This commit adds the suggestion in green, on the line below the caret-and-tildes. foo.y:1.1-14: warning: deprecated directive: '%error-verbose', use '%define parse.error verbose' [-Wdeprecated] 1 | %error-verbose | ^~~~~~~~~~~~~~ | %define parse.error verbose The current approach, with location_caret_suggestion, is fragile: there's a protocol of calls to the complain functions which is strict. We should rather have a richer structure describing the diagnostics, including with submessages such as the suggestions, passed in the end to the routines in charge of formatting and printing them. * src/location.h, src/location.c (location_caret_suggestion): New. * src/complain.c (deprecated_directive): Use it. * tests/diagnostics.at, tests/input.at: Adjust expectations. 2019-10-06 Akim Demaille diagnostics: isolate caret_set_column * src/location.c (caret_info): Add width and skip members. (caret_set_column): New. Use it. 2019-10-06 Akim Demaille diagnostics: isolate caret_set_file * src/location.c (caret_set_file): New. Store the current line's length in caret_info.line_len. Pay attention to fseek's return value. Extracted from... (location_caret): here. 2019-10-06 Akim Demaille tests: use tput to get the number of columns * tests/bison.in: here. 2019-10-06 Akim Demaille TODO: update I no longer agree with that item, there are indeed two things to report: lack of definition, and being useless. We could have either one without the other, they are not directly related. 2019-10-06 Akim Demaille yacc.c: work around warnings from G++ 4.8 input.c: In function 'int yyparse()': input.c: error: conversion to 'long int' from 'long unsigned int' may change the sign of the result [-Werror=sign-conversion] yyes_capacity = sizeof yyesa / sizeof *yyes; ^ cc1plus: all warnings being treated as errors * data/skeletons/yacc.c: here. 2019-10-06 Akim Demaille yacc.c: work around warnings from Clang++ 3.3 and 3.4 When we run the test suite with these C++ compilers to compile C code, we get: 239. synclines.at:440: testing syncline escapes: yacc.c ... ../../tests/synclines.at:440: $CC $CFLAGS $CPPFLAGS \"\\\"\".c -o \"\\\"\" || exit 77 stderr: stdout: ../../tests/synclines.at:440: COLUMNS=1000; export COLUMNS; bison --color=no -fno-caret -o \"\\\"\".c \"\\\"\".y ../../tests/synclines.at:440: $CC $CFLAGS $CPPFLAGS $LDFLAGS -o \"\\\"\" \"\\\"\".c $LIBS stderr: "\"".c:1102:41: error: implicit conversion loses integer precision: 'long' to 'int' [-Werror,-Wshorten-64-to-32] YYPTRDIFF_T yysize = yyssp - yyss + 1; ~~~~~~ ~~~~~~~~~~~~~^~~ 1 error generated. 193. conflicts.at:545: testing parse.error=verbose and consistent errors: lr.type=canonical-lr parse.lac=full ... input.c:737:75: error: implicit conversion loses integer precision: 'long' to 'int' [-Werror,-Wshorten-64-to-32] YYPTRDIFF_T yysize_old = *yytop == yytop_empty ? 0 : *yytop - *yybottom + 1; ~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~~^~~ input.c:901:48: error: implicit conversion loses integer precision: 'long' to 'int' [-Werror,-Wshorten-64-to-32] YYPTRDIFF_T yysize = yyesp - *yyes + 1; ~~~~~~ ~~~~~~~~~~~~~~^~~ * data/skeletons/yacc.c: Add more casts. 2019-10-05 Akim Demaille tests: avoid a GCC 4.8 warning GCC 4.8 reports: input.y:57:33: error: conversion to 'int' from 'long unsigned int' may alter its value [-Werror=conversion] int input_elts = sizeof input / sizeof input[0]; ^ * tests/local.at (AT_YYLEX_DEFINE(c)): Add a cast (sorry, Paul!). 2019-10-05 Paul Eggert * data/skeletons/glr.c (yysplitStack): Pacify Clang 8. 2019-10-05 Paul Eggert Avoid quiet conversion of pointer to bool * src/location.c (caret_set_file): * src/scan-code.l (contains_dot_or_dash): Do not quietly convert pointer to bool, as Oracle Developer Studio 12.6 complains and it is arguably confusing style anyway. 2019-10-05 Paul Eggert Port ARGMATCH_DEFINE_GROUP calls to C99 * src/complain.c, src/getargs.c: Omit ‘;’ after call to ARGMATCH_DEFINE_GROUP, as C99 does not allow ‘;’ there. 2019-10-05 Paul Eggert Port lexcalc scan.l to Solaris 10 * examples/c/lexcalc/scan.l: Include errno.h. 2019-10-05 Akim Demaille yacc.c: use casts instead of pragmas when losing integer width For instance with Clang 4, 8, etc.: input.c:1166:12: error: implicit conversion loses integer precision: 'int' to 'yy_state_num' (aka 'signed char') [-Werror,-Wconversion] *yyssp = yystate; ~ ^~~~~~~ And GCC 8: input.c:1166:12: error: implicit conversion loses integer precision: 'int' to 'yy_state_num' (aka 'signed char') [-Werror,-Wimplicit-int-conversion] *yyssp = yystate; ~ ^~~~~~~ * data/skeletons/yacc.c (YY_CONVERT_INT_BEGIN): Remove. Adjust callers. 2019-10-04 Akim Demaille yacc.c: fix warnings about undefined macros For instance with GCC 4.9 and --enable-gcc-warnings: 25. input.at:1201: testing Torturing the Scanner ... ../../tests/input.at:1344: $CC $CFLAGS $CPPFLAGS -c -o input.o input.c stderr: input.c:239:18: error: "__STDC_VERSION__" is not defined [-Werror=undef] # elif 199901 <= __STDC_VERSION__ ^ input.c:256:18: error: "__STDC_VERSION__" is not defined [-Werror=undef] # elif 199901 <= __STDC_VERSION__ ^ * data/skeletons/yacc.c: Check that __STDC_VERSION__ is defined before using it. 2019-10-04 Akim Demaille tests: check more state numbers * tests/torture.at (State number type): Also check 128, 129 and 32768. 2019-10-03 Paul Eggert * doc/bison.texi (Table of Symbols): Mention memory exhaustion. 2019-10-03 Paul Eggert Simplify mfcalc error handling * doc/bison.texi (Mfcalc Symbol Table, Mfcalc Lexer): Don’t abort on memory allocation failure or integer overflow. Instead, comment that these things aren’t checked for. 2019-10-03 Akim Demaille c++: fix comments suggesting to use %require * data/skeletons/location.cc, data/skeletons/stack.hh: Here. 2019-10-03 Akim Demaille lalr1.cc: simplify uses of size_t * data/skeletons/stack.hh (stack::index_type): New type. (stack::size, stack::operator[]): Be about an index_type rather than a size_type and an int. 2019-10-03 Akim Demaille c++: fixes for old compilers On the CI with GCC 6: examples/c++/calc++/parser.cc:845:5: error: 'ptrdiff_t' was not declared in this scope ptrdiff_t yycount = 0; ^~~~~~~~~ examples/c++/calc++/parser.cc:845:5: note: suggested alternatives: /usr/include/x86_64-linux-gnu/c++/6/bits/c++config.h:202:28: note: 'std::ptrdiff_t' typedef __PTRDIFF_TYPE__ ptrdiff_t; ^~~~~~~~~ * data/skeletons/lalr1.cc: Qualify ptrdiff_t and size_t with std::. 2019-10-03 Akim Demaille tests: be robust to -DNDEBUG input.y: In function 'yylex': input.y:67:7: error: unused variable 'input_elts' [-Werror=unused-variable] int input_elts = sizeof input / sizeof input[0]; ^~~~~~~~~~ cc1: all warnings being treated as errors * tests/input.at, tests/local.at: Avoid that. 2019-10-03 Akim Demaille CI: remove the symlink before creating it Currently we fail if we rerun a job that succeeded to push the tarball. 2019-10-03 Paul Eggert Adjust ‘Big horizontal’ test case * tests/torture.at (Big horizontal): Adjust to recent changes with integers. If there are states 0..256, Bison now uses a signed rather than an unsigned 16-bit integer. 2019-10-03 Paul Eggert regen 2019-10-03 Paul Eggert Prefer signed to unsigned integers This patch contains more fixes to prefer signed to unsigned integer types, as modern tools like 'gcc -fsanitize=undefined' can check for signed integer overflow but not unsigned overflow. * NEWS: Document the API change. * boostrap.conf (gnulib_modules): Add intprops. * data/skeletons/glr.c: Include stddef.h and stdint.h, since this skeleton can assume C99 or later. (YYSIZEMAX): Now signed, and the minimum of SIZE_MAX and PTRDIFF_MAX. (yybool) [!__cplusplus]: Now signed (which is how bool behaves). (YYTRANSLATE): Avoid use of unsigned, and make the macro safe even for values greater than UINT_MAX. (yytnamerr, struct yyGLRState, struct yyGLRStateSet, struct yyGLRStack) (yyaddDeferredAction, yyinitStateSet, yyinitGLRStack) (yyexpandGLRStack, yymarkStackDeleted, yyremoveDeletes) (yyglrShift, yyglrShiftDefer, yy_reduce_print, yydoAction) (yyglrReduce, yysplitStack, yyreportTree, yycompressStack) (yyprocessOneStack, yyreportSyntaxError, yyrecoverSyntaxError) (yyparse, yy_yypstack, yypstack, yypdumpstack): * tests/input.at (Torturing the Scanner): Prefer ptrdiff_t to size_t. * data/skeletons/c++.m4 (b4_yytranslate_define): * src/AnnotationList.c (AnnotationList__computePredecessorAnnotations): * src/AnnotationList.h (AnnotationIndex): * src/InadequacyList.h (InadequacyListNodeCount): * src/closure.c (closure_new): * src/complain.c (error_message, complains, complain_indent) (complain_args, duplicate_directive, duplicate_rule_directive): * src/gram.c (nritems, ritem_print, grammar_dump): * src/ielr.c (ielr_compute_ritem_sees_lookahead_set) (ielr_item_has_lookahead, ielr_compute_annotation_lists) (ielr_compute_lookaheads): * src/location.c (columns, boundary_print, location_print): * src/muscle-tab.c (muscle_percent_define_insert) (muscle_percent_define_check_values): * src/output.c (prepare_rules, prepare_actions): * src/parse-gram.y (id, handle_require): * src/reader.c (record_merge_function_type, packgram): * src/reduce.c (nuseless_productions, nuseless_nonterminals) (inaccessable_symbols): * src/relation.c (relation_print): * src/scan-code.l (variant, variant_table_size, variant_count) (variant_add, get_at_spec, show_sub_message, show_sub_messages) (parse_ref): * src/scan-gram.l () (scan_integer, convert_ucn_to_byte, handle_syncline): * src/scan-skel.l (at_complain): * src/symtab.c (complain_symbol_redeclared) (complain_semantic_type_redeclared, complain_class_redeclared) (symbol_class_set, complain_user_token_number_redeclared): * src/tables.c (conflict_tos, conflrow, conflict_table) (conflict_list, save_row, pack_vector): * tests/local.at (AT_YYLEX_DEFINE(c)): Prefer signed to unsigned integer. * data/skeletons/lalr1.cc (yy_lac_check_): * tests/actions.at (_AT_CHECK_PRINTER_AND_DESTRUCTOR): * tests/local.at (AT_YYLEX_DEFINE(c)): Omit now-unnecessary casts. * data/skeletons/location.cc (b4_location_define): * doc/bison.texi (Mfcalc Lexer, C++ position, C++ location): Prefer int to unsigned for line and column numbers. Change example to abort explicitly on memory exhaustion, and fix an off-by-one bug that led to undefined behavior. * data/skeletons/stack.hh (stack::operator[]): Also allow ptrdiff_t indexes. (stack::pop, slice::slice, slice::operator[]): Index arg is now ptrdiff_t, not int. (stack::ssize): New method. (slice::range_): Now ptrdiff_t, not int. * data/skeletons/yacc.c (b4_state_num_type): Remove. All uses replaced by b4_int_type. (YY_CONVERT_INT_BEGIN, YY_CONVERT_INT_END): New macros. (yylac, yyparse): Use them around conversions that -Wconversion would give false alarms about. Omit unnecessary casts. (yy_stack_print): Use int rather than unsigned, and omit a cast that doesn’t seem to be needed here any more. * examples/c++/variant.yy (yylex): * examples/c++/variant-11.yy (yylex): Omit no-longer-needed conversions to unsigned. * src/InadequacyList.c (InadequacyList__new_conflict): Don’t assume *node_count is unsigned. * src/output.c (muscle_insert_unsigned_table): Remove; no longer used. 2019-10-02 Paul Eggert Prefer signed types for indexes in skeletons * NEWS: Mention this. * data/skeletons/c.m4 (b4_int_type): Prefer char if it will do, and prefer signed types to unsigned if either will do. * data/skeletons/glr.c (yy_reduce_print): No need to convert rule line to unsigned long. (yyrecoverSyntaxError): Put action into an int to avoid GCC warning of using a char subscript. * data/skeletons/lalr1.cc (yy_lac_check_, yysyntax_error_): Prefer ptrdiff_t to size_t. * data/skeletons/yacc.c (b4_int_type): Prefer signed types to unsigned if either will do. * data/skeletons/yacc.c (b4_declare_parser_state_variables): (YYSTACK_RELOCATE, YYCOPY, yy_lac_stack_realloc, yy_lac) (yytnamerr, yysyntax_error, yyparse): Prefer ptrdiff_t to size_t. (YYPTRDIFF_T, YYPTRDIFF_MAXIMUM): New macros. (YYSIZE_T): Fix "! defined YYSIZE_T" typo. (YYSIZE_MAXIMUM): Take the minimum of PTRDIFF_MAX and SIZE_MAX. (YYSIZEOF): New macro. (YYSTACK_GAP_MAXIMUM, YYSTACK_BYTES, YYSTACK_RELOCATE) (yy_lac_stack_realloc, yyparse): Use it. (YYCOPY, yy_lac_stack_realloc): Cast to YYSIZE_T to pacify GCC. (yy_reduce_print): Use int instead of unsigned long when int will do. (yy_lac_stack_realloc): Prefer long to unsigned long when either will do. * tests/regression.at: Adjust to these changes. 2019-09-30 Akim Demaille yacc: use the most appropriate integral type for state numbers Currently we properly use the "best" integral type for tables, including those storing state numbers. However the variables for state numbers used in yyparse (and its dependencies such as yy_stack_print) still use int16_t invariably. As a consequence, very large models overflow these variables. Let's use the "best" type for these variables too. It turns out that we can still use 16 bits for twice larger automata: stick to unsigned types. However using 'unsigned' when 16 bits are not enough is troublesome and generates tons of warnings about signedness issues. Instead, let's use 'int'. Reported by Tom Kramer. https://lists.gnu.org/archive/html/bug-bison/2019-09/msg00018.html * data/skeletons/yacc.c (b4_state_num_type): New. (yy_state_num): Be computed from YYNSTATES. * tests/linear: New. * tests/torture.at (State number type): New. Use it. 2019-09-30 Akim Demaille yacc: introduce a type for states * data/skeletons/yacc.c (yy_state_num): New. Use it for arrays of states. 2019-09-30 Akim Demaille style: prefer symbolic values rather than litterals Instead of #define YYPACT_NINF -130 #define yypact_value_is_default(Yystate) \ (!!((Yystate) == (-130))) generate #define YYPACT_NINF (-130) #define yypact_value_is_default(Yyn) \ ((Yyn) == YYPACT_NINF) * data/skeletons/c.m4 (b4_table_value_equals): Add support for $4. * data/skeletons/glr.c, data/skeletons/yacc.c: Use it. Also, use shorter macro argument names, the name of the macro is clear enough. 2019-09-30 Akim Demaille style: change misleading macro argument name * data/skeletons/glr.c, data/skeletons/yacc.c (yypact_value_is_default): It does not take a rule number as argument. 2019-09-28 Akim Demaille Merge remote-tracking branch 'upstream/maint' * upstream/maint: c++: add copy ctors for compatibility with the IAR compiler CI: show git status CI: disable ICC tests: pass -jN from Make to the test suite quotearg: avoid leaks maint: post-release administrivia 2019-09-27 Akim Demaille c++: add copy ctors for compatibility with the IAR compiler Reported by Andreas Damm. https://savannah.gnu.org/support/?110032 * data/skeletons/lalr1.cc (stack_symbol_type::operator=): New overload, const, to please the IAR C++ compiler (version ca 2013). 2019-09-23 Akim Demaille CI: show git status 2019-09-23 Akim Demaille CI: disable ICC It seems that Intel changed something in their license management. https://github.com/nemequ/icc-travis/issues/15 2019-09-23 Akim Demaille tests: pass -jN from Make to the test suite I am sooooo tired of typing "make -j5 TESTSUITEFLAGS=-j5"... Should have done this years ago. * cfg.mk (TESTSUITEFLAGS): here. 2019-09-22 Akim Demaille quotearg: avoid leaks Reported by Tomasz Kłoczko. https://lists.gnu.org/archive/html/bug-bison/2019-09/msg00008.html * src/main.c (main): Free quotearg's memory later. 2019-09-22 Akim Demaille diagnostics: get the screen width from the terminal * bootstrap.conf: We need winsz-ioctl and winsz-termios. * src/location.c (columns): Use winsize to get the number of columns. Code taken from the GNU Coreutils. * src/location.h, src/location.c (caret_init): New. * src/complain.c (complain_init): Call it. * tests/bison.in: Export COLUMNS so that users of tests/bison can enjoy proper line truncation. 2019-09-22 Akim Demaille diagnostics: don't print ellipsis on the caret line From 9 | ...TUVWXYZ ABCDEFGHIJKLMNOPQRSTUVWXYZ ABCDEFGHIJKL | ... ^~~~~~~~~~~~~~~~~~~~~~~~~~ to 9 | ...TUVWXYZ ABCDEFGHIJKLMNOPQRSTUVWXYZ ABCDEFGHI... | ^~~~~~~~~~~~~~~~~~~~~~~~~~ * src/location.c (location_caret): here. * tests/diagnostics.at: Adjust expectations. 2019-09-22 Akim Demaille diagnostics: also show truncation at the end of line with "..." From 9 | ...TUVWXYZ ABCDEFGHIJKLMNOPQRSTUVWXYZ ABCDEFGHIJKL | ... ^~~~~~~~~~~~~~~~~~~~~~~~~~ to 9 | ...TUVWXYZ ABCDEFGHIJKLMNOPQRSTUVWXYZ ABCDEFGHI... | ... ^~~~~~~~~~~~~~~~~~~~~~~~~~ * src/location.c (location_caret): here. * tests/diagnostics.at: Adjust expectations. 2019-09-22 Akim Demaille diagnostics: check that quoted lines are truncated * tests/diagnostics.at (Screen width: 60 columns, Screen width: 80 columns, Screen width: 200 columns): New tests. 2019-09-22 Akim Demaille diagnostics: truncate quoted sources to fit the screen * src/location.c (min_int, columns): New. (location_caret): Compute the line width. Based on it, compute how many columns must be skipped before the quoted location and truncated after, to fit the sceen width. * tests/local.at (AT_QUELL_VALGRIND): Transform into... (AT_SET_ENV_IF, AT_SET_ENV): these. Define COLUMNS to protect the test suite from the user's environment. 2019-09-22 Akim Demaille diagnostics: learn how to count column number with multibyte chars So far diagnostics were cheating: in addition to the 'column' field of locations (based on actual screen width per multibyte characters and on tabulation expansion), the scanner sets the 'byte' field. Diagnostics used this byte count to decide where to insert (color) style. We want to be able to truncate the quoted lines when there are too wide to fit the screen. This requires that the diagnostics learn how to count columns, the byte-in-boundary trick no longer works. Bytes are still used for fix-its. * bootstrap.conf: We need mbfile for mbf_getc. * src/location.c (caret_info): We need an mbfile. (caret_set_file): Initialize it. (caret_getc): Convert to mbfile. (location_caret): Instead of relying on the byte position to decide where to insert the color style, count the current column using boundary_compute. 2019-09-22 Akim Demaille diagnostics: style: rename member for clariy * src/location.c (caret_info): Now that we no longer have a 'file' member (see previous commit), rename 'source' as 'file'. 2019-09-22 Akim Demaille diagnostics: style: use a boundary to track the caret_info * src/location.c (caret_info): Replace file and line with pos, a boundary. This will allow us to use features of the boundary type, such as boundary_compute. 2019-09-22 Akim Demaille diagnostics: extract boundary_compute from location_compute The handling of the contributions of the tabulations in the columns is burried inside location_compute. We will soon be willing to use the boundary part of the computation (to compute the current column number each time we read a multibyte char). * src/location.c (boundary_compute): New, extracted from... (location_compute): here. 2019-09-22 Akim Demaille diagnostics: style: add caret_set_file To make the following commits easier to read. * src/location.c (caret_set_file): New. 2019-09-22 Akim Demaille diagnostics: style: minor changes * src/location.c (location_caret): Factor two branches of an if. 2019-09-22 Akim Demaille CI: show git status 2019-09-22 Akim Demaille git: update ignores 2019-09-22 Akim Demaille git: update ignores 2019-09-21 Akim Demaille quotearg: avoid leaks Reported by Tomasz Kłoczko. https://lists.gnu.org/archive/html/bug-bison/2019-09/msg00008.html * src/main.c (main): Free quotearg's memory later. 2019-09-14 Akim Demaille tests: pass -jN from Make to the test suite I am sooooo tired of typing "make -j5 TESTSUITEFLAGS=-j5"... Should have done this years ago. * cfg.mk (TESTSUITEFLAGS): here. 2019-09-14 Akim Demaille java: handle eof in yytranslate * data/skeletons/lalr1.java (yytranslate_): Handle eof here, as is done in lalr1.cc. * tests/javapush.at: Adjust. 2019-09-14 Akim Demaille d: handle eof in yytranslate This changes the traces from Reading a token: Now at end of input. to Reading a token: Next token is token $end (7FFEE56E6474) which is ok. Actually it is even better, as it gives the location when locations are enabled, and is clearer when rules explicitly use the EOF token. * data/skeletons/lalr1.d (yytranslate_): Handle eof here, as is done in lalr1.cc. 2019-09-14 Akim Demaille regen 2019-09-14 Akim Demaille parser: use api.token.raw * src/parse-gram.y: Here. 2019-09-14 Akim Demaille api.token.raw: document it * doc/bison.texi: here. 2019-09-14 Akim Demaille api.token.raw: cannot be used with character literals * src/parse-gram.y (CHAR): api.token.raw and character literals are mutually exclusive. * tests/input.at (Character literals and api.token.raw): New. 2019-09-14 Akim Demaille api.token.raw: apply to the other skeletons * data/skeletons/c++.m4, data/skeletons/glr.c, * data/skeletons/lalr1.c, data/skeletons/lalr1.java: Add support for api.token.raw. * tests/scanner.at: Check them. 2019-09-14 Akim Demaille api.token.raw: check it * tests/local.at (AT_TOKEN_RAW_IF): New. * tests/local.mk: New. Use it. 2019-09-14 Akim Demaille api.token.raw: implement Bison used to feature %raw, documented as follows: @item %raw The output file @file{@var{name}.h} normally defines the tokens with Yacc-compatible token numbers. If this option is specified, the internal Bison numbers are used instead. (Yacc-compatible numbers start at 257 except for single character tokens; Bison assigns token numbers sequentially for all tokens starting at 3.) Unfortunately, as far as I can tell, it never worked: token numbers are indeed changed in the generated tables (from external token number to internal), yet the code was still applying the mapping from external token numbers to internal token numbers. This commit reintroduces the feature as it was expected to be. * data/skeletons/bison.m4 (b4_token_format): When api.token.raw is enabled, use the internal token number. * data/skeletons/yacc.c (yytranslate): Don't emit if api.token.raw is enabled. (YYTRANSLATE): Adjust. 2019-09-14 Akim Demaille style: tidy yacc.c * data/skeletons/yacc.c: Include 'c.m4' first. Then sort the handling of %define variables. * tests/input.at: Adjust. 2019-09-14 Akim Demaille CI: disable ICC It seems that Intel changed something in their license management. https://github.com/nemequ/icc-travis/issues/15 2019-09-14 Akim Demaille diagnostics: fix use of complain_indent * src/symtab.c (symbol_class_set): Here. * tests/diagnostics.at, tests/input.at, tests/regression.at: Adjust expectations. 2019-09-14 Akim Demaille input: stop treating lone CRs as end-of-lines We used to treat lone CRs (\r, aka ^M) as regular NLs (\n), probably to please Classic MacOS. As of today, it makes more sense to treat \r like a plain white space character. https://lists.gnu.org/archive/html/bison-patches/2019-09/msg00027.html * src/scan-gram.l (no_cr_read): Remove. Instead, use... (eol): this new abbreviation denoting end-of-line. * src/location.c (caret_getc): New. (location_caret): Use it. * tests/diagnostics.at (Carriage return): Adjust expectations. (CR NL): New. 2019-09-12 Akim Demaille Merge tag 'v3.4.2' into HEAD bison 3.4.2 * tag 'v3.4.2': (24 commits) version 3.4.2 CI: always uninstall icc news: more bug fixes thanks to Marc Schönefeld diagnostics: beware of unexpected EOF when quoting the source file gnulib: update build: fix distcheck tests: add noexcept to please GCC 9 news: update fix: don't die when EOF token is defined twice tests: check token redeclaration yacc.c: beware of GCC's -Wmaybe-uninitialized glr.c: initialize vector of bools gnulib: update check for memory exhaustion diagnostics: avoid global variables diagnostics: fix invalid error message indentation git: ignore files generated in gnulib-po c++: avoid duplicate definition of YYUSE gnulib: update CI: more compilers ... 2019-09-12 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2019-09-12 Akim Demaille version 3.4.2 * NEWS: Record release date. 2019-09-12 Akim Demaille CI: always uninstall icc 2019-09-12 Akim Demaille news: more bug fixes thanks to Marc Schönefeld 2019-09-12 Akim Demaille diagnostics: beware of unexpected EOF when quoting the source file When the input file contains lone CRs (aka, ^M, \r), the locations see a new line. Diagnostics look only at \n as end-of-line, so sometimes there is an offset in diagnostics. Worse yet: sometimes we loop endlessly waiting for \n to come from a continuous stream of EOF. Fix that: - check for EOF - beware not to call end_use_class if begin_use_class was not called (which would abort). This could happen if the actual line is shorter that the expected one. Prompted by a (private) report from Marc Schönefeld. * src/location.c (location_caret): here. * tests/diagnostics.at (Carriage return): New. 2019-09-11 Akim Demaille gnulib: update Contains the creation of the xhash module. https://lists.gnu.org/archive/html/bug-gnulib/2019-09/msg00046.html * src/muscle-tab.c, src/state.c, src/symtab.c, src/uniqstr.c: Use hash_xinitialize. 2019-09-11 Akim Demaille build: fix distcheck * configure.ac (gl_LIBOBJS): Adjust so that the generated files are indeed the expected ones. 2019-09-10 Akim Demaille diagnostics: beware of unexpected EOF when quoting the source file When the input file contains lone CRs (aka, ^M, \r), the locations see a new line. Diagnostics look only at \n as end-of-line, so sometimes there is an offset in diagnostics. Worse yet: sometimes we loop endlessly waiting for \n to come from a continuous stream of EOF. Fix that: - check for EOF - beware not to call end_use_class if begin_use_class was not called (which would abort). This could happen if the actual line is shorter that the expected one. Prompted by a (private) report from Marc Schönefeld. * src/location.c (location_caret): here. * tests/diagnostics.at (Carriage return): New. 2019-09-10 Akim Demaille gnulib: update Contains the creation of the xhash module. https://lists.gnu.org/archive/html/bug-gnulib/2019-09/msg00046.html * src/muscle-tab.c, src/state.c, src/symtab.c, src/uniqstr.c: Use hash_xinitialize. 2019-09-10 Akim Demaille build: fix distcheck * configure.ac (gl_LIBOBJS): Adjust so that the generated files are indeed the expected ones. 2019-09-08 Akim Demaille tests: add noexcept to please GCC 9 bison/tests/c++.at:552: bison --color=no -fno-caret -o list.cc list.y bison/tests/c++.at:552: $CXX $CXXFLAGS $CPPFLAGS $LDFLAGS -o list list.cc $LIBS stderr: gcc9/c++/ext/new_allocator.h: In instantiation of 'void __gnu_cxx::new_allocator<_Tp>::construct(_Up*, _Args&& ...) [with _Up = string; _Args = {string}; _Tp = string]': gcc9/c++/bits/alloc_traits.h:482:2: required from 'static void std::allocator_traits >::construct(std::allocator_traits >::allocator_type&, _Up*, _Args&& ...) [with _Up = string; _Args = {string}; _Tp = string; std::allocator_traits >::allocator_type = std::allocator]' gcc9/c++/bits/stl_uninitialized.h:888:67: required from 'void std::__relocate_object_a(_Tp*, _Up*, _Allocator&) [with _Tp = string; _Up = string; _Allocator = std::allocator]' gcc9/c++/bits/stl_uninitialized.h:920:47: required from '_ForwardIterator std::__relocate_a_1(_InputIterator, _InputIterator, _ForwardIterator, _Allocator&) [with _InputIterator = string*; _ForwardIterator = string*; _Allocator = std::allocator]' gcc9/c++/bits/stl_uninitialized.h:942:37: required from '_ForwardIterator std::__relocate_a(_InputIterator, _InputIterator, _ForwardIterator, _Allocator&) [with _InputIterator = string*; _ForwardIterator = string*; _Allocator = std::allocator]' gcc9/c++/bits/stl_vector.h:430:35: required from 'static constexpr bool std::vector<_Tp, _Alloc>::_S_nothrow_relocate(std::true_type) [with _Tp = string; _Alloc = std::allocator; std::true_type = std::integral_constant]' gcc9/c++/bits/stl_vector.h:446:28: required from 'void std::vector<_Tp, _Alloc>::_M_realloc_insert(std::vector<_Tp, _Alloc>::iterator, _Args&& ...) [with _Args = {const string&}; _Tp = string; _Alloc = std::allocator; std::vector<_Tp, _Alloc>::iterator = __gnu_cxx::__normal_iterator >; typename std::_Vector_base<_Tp, _Alloc>::pointer = string*]' gcc9/c++/bits/stl_vector.h:1195:4: required from 'void std::vector<_Tp, _Alloc>::push_back(const value_type&) [with _Tp = string; _Alloc = std::allocator; std::vector<_Tp, _Alloc>::value_type = string]' list.y:126:110: required from here gcc9/c++/bits/vector.tcc:459:44: in 'constexpr' expansion of 'std::vector::_S_use_relocate()' list.y:41:7: error: but 'string::string(string&&)' does not throw; perhaps it should be declared 'noexcept' [-Werror=noexcept] 41 | string (string&& s) | ^~~~~~ * tests/c++.at (Variants): Add noexcept where appropriate. 2019-09-08 Akim Demaille tests: add noexcept to please GCC 9 bison/tests/c++.at:552: bison --color=no -fno-caret -o list.cc list.y bison/tests/c++.at:552: $CXX $CXXFLAGS $CPPFLAGS $LDFLAGS -o list list.cc $LIBS stderr: gcc9/c++/ext/new_allocator.h: In instantiation of 'void __gnu_cxx::new_allocator<_Tp>::construct(_Up*, _Args&& ...) [with _Up = string; _Args = {string}; _Tp = string]': gcc9/c++/bits/alloc_traits.h:482:2: required from 'static void std::allocator_traits >::construct(std::allocator_traits >::allocator_type&, _Up*, _Args&& ...) [with _Up = string; _Args = {string}; _Tp = string; std::allocator_traits >::allocator_type = std::allocator]' gcc9/c++/bits/stl_uninitialized.h:888:67: required from 'void std::__relocate_object_a(_Tp*, _Up*, _Allocator&) [with _Tp = string; _Up = string; _Allocator = std::allocator]' gcc9/c++/bits/stl_uninitialized.h:920:47: required from '_ForwardIterator std::__relocate_a_1(_InputIterator, _InputIterator, _ForwardIterator, _Allocator&) [with _InputIterator = string*; _ForwardIterator = string*; _Allocator = std::allocator]' gcc9/c++/bits/stl_uninitialized.h:942:37: required from '_ForwardIterator std::__relocate_a(_InputIterator, _InputIterator, _ForwardIterator, _Allocator&) [with _InputIterator = string*; _ForwardIterator = string*; _Allocator = std::allocator]' gcc9/c++/bits/stl_vector.h:430:35: required from 'static constexpr bool std::vector<_Tp, _Alloc>::_S_nothrow_relocate(std::true_type) [with _Tp = string; _Alloc = std::allocator; std::true_type = std::integral_constant]' gcc9/c++/bits/stl_vector.h:446:28: required from 'void std::vector<_Tp, _Alloc>::_M_realloc_insert(std::vector<_Tp, _Alloc>::iterator, _Args&& ...) [with _Args = {const string&}; _Tp = string; _Alloc = std::allocator; std::vector<_Tp, _Alloc>::iterator = __gnu_cxx::__normal_iterator >; typename std::_Vector_base<_Tp, _Alloc>::pointer = string*]' gcc9/c++/bits/stl_vector.h:1195:4: required from 'void std::vector<_Tp, _Alloc>::push_back(const value_type&) [with _Tp = string; _Alloc = std::allocator; std::vector<_Tp, _Alloc>::value_type = string]' list.y:126:110: required from here gcc9/c++/bits/vector.tcc:459:44: in 'constexpr' expansion of 'std::vector::_S_use_relocate()' list.y:41:7: error: but 'string::string(string&&)' does not throw; perhaps it should be declared 'noexcept' [-Werror=noexcept] 41 | string (string&& s) | ^~~~~~ * tests/c++.at (Variants): Add noexcept where appropriate. 2019-09-08 Akim Demaille news: update 2019-09-08 Akim Demaille fix: don't die when EOF token is defined twice With %token EOF 0 EOF 0 we get input.y:3.14-16: warning: symbol EOF redeclared [-Wother] 3 | %token EOF 0 EOF 0 | ^~~ input.y:3.8-10: previous declaration 3 | %token EOF 0 EOF 0 | ^~~ Assertion failed: (nsyms == ntokens + nvars), function check_and_convert_grammar, file /Users/akim/src/gnu/bison/src/reader.c, line 839. Reported by Marc Schönefeld. * src/symtab.c (symbol_user_token_number_set): Register only the first definition of the end of input token. * tests/input.at (Symbol redeclared): Check that case. 2019-09-08 Akim Demaille tests: check token redeclaration * src/symtab.c (symbol_class_set): Report previous definitions when redeclared. * tests/input.at (Symbol redeclared): New. 2019-09-08 Akim Demaille yacc.c: beware of GCC's -Wmaybe-uninitialized Test 400 (calc.at:773: testing Calculator api.push-pull=both api.pure=full parse.error=verbose %debug %locations %defines api.prefix={calc} %verbose %yacc) fails on the CI with GCC 8 on Bionic: 400. calc.at:773: testing Calculator api.push-pull=both api.pure=full parse.error=verbose %debug %locations %defines api.prefix={calc} %verbose %yacc ... ../../tests/calc.at:773: bison --color=no -fno-caret -Wno-deprecated -o calc.c calc.y ../../tests/calc.at:773: $CC $CFLAGS $CPPFLAGS $LDFLAGS -o calc calc.c calc-lex.c calc-main.c $LIBS stderr: calc.y: In function 'int calcpush_parse(calcpstate*, int, const CALCSTYPE*, CALCLTYPE*)': calc.y:26:20: error: 'yylval.CALCSTYPE::ival' may be used uninitialized in this function [-Werror=maybe-uninitialized] %printer { fprintf (yyo, "%d", $$); } ; ^ calc.c:1272:9: note: 'yylval.CALCSTYPE::ival' was declared here YYSTYPE yylval YY_INITIAL_VALUE (= yyval_default); ^~~~~~ cc1plus: all warnings being treated as errors stdout: ../../tests/calc.at:773: exit code was 1, expected 0 400. calc.at:773: 400. Calculator api.push-pull=both api.pure=full parse.error=verbose %debug %locations %defines api.prefix={calc} %verbose %yacc (calc.at:773): FAILED (calc.at:773) * data/skeletons/c.m4 (yy_symbol_value_print): Disable the warning locally. 2019-09-08 Akim Demaille glr.c: initialize vector of bools The CI, with CC='gcc-7 -fsanitize=undefined,address -fno-omit-frame-pointer', reports: calc.cc:1652:50: runtime error: load of value 190, which is not a valid value for type 'bool' ../../tests/calc.at:867: cat stderr --- expout 2019-09-05 20:30:37.887257545 +0000 +++ /home/travis/build/bison-3.4.1.72-79a1-dirty/_build/tests/testsuite.dir/at-groups/438/stdout 2019-09-05 20:30:37.887257545 +0000 @@ -1 +1,2 @@ syntax error +calc.cc:1652:50: runtime error: load of value 190, which is not a valid value for type 'bool' 438. calc.at:867: 438. Calculator glr.cc (calc.at:867): FAILED (calc.at:867) The problem is that yylookaheadNeeds is not initialized in yyinitStateSet, and when it is copied, the value is not 0 or 1. * data/skeletons/glr.c (yylookaheadNeeds): Initialize yylookaheadNeeds. 2019-09-08 Akim Demaille gnulib: update Contains a fix for https://lists.gnu.org/archive/html/bug-bison/2019-08/msg00016.html. See https://lists.gnu.org/archive/html/bug-gnulib/2019-09/msg00005.html. Reported by 江 祖铭 (Zu-Ming Jiang). 2019-09-08 Akim Demaille check for memory exhaustion hash_initialize returns NULL when out of memory. Check for it, and die cleanly instead of crashing. Reported by 江 祖铭 (Zu-Ming Jiang). https://lists.gnu.org/archive/html/bug-bison/2019-08/msg00015.html * src/muscle-tab.c, src/state.c, src/symtab.c, src/uniqstr.c: Check the value returned by hash_initialize. 2019-09-08 László Várady diagnostics: avoid global variables * src/complain.c (indent_ptr): Remove. (error_message, complains): Take indent as an argument. Adjust callers. 2019-09-08 László Várady diagnostics: fix invalid error message indentation https://lists.gnu.org/archive/html/bison-patches/2019-08/msg00007.html When Bison is started with a flag that suppresses warning messages, the error_message() function can produce a few gigabytes of indentation because of a dangling pointer. * src/complain.c (error_message): Don't reset indent_ptr here, but... (complain_indent): here. * tests/diagnostics.at (Indentation with message suppression): Check this case. 2019-09-08 Akim Demaille git: ignore files generated in gnulib-po Because of them, the CI generates "-dirty" tarballs. 2019-09-08 Akim Demaille c++: avoid duplicate definition of YYUSE Reported by Frank Heckenbach. https://lists.gnu.org/archive/html/bug-bison/2019-06/msg00009.html * data/skeletons/lalr1.cc (b4_shared_declarations): Remove the duplicate definition of YYUSE, the other one coming from b4_attribute_define. 2019-09-08 Akim Demaille gnulib: update This update brings file from Gettext 0.20, which is not available on the CI yet. .travis.yml: Adjust. Use Bionic now that it's available. 2019-09-08 Akim Demaille CI: more compilers * .travis.yml: Bionic is now available, with GCC8. GCC7 sanitizers work, but they are too longer: cover only part 1. Redefine part 1 and part 2 so that part 1 is really the core of the tests: not playing with POSIX and C++ compiler for C code. 2019-09-08 Akim Demaille CI: fail fast 2019-09-08 Akim Demaille CI: propagate sftp failures * .travis.yml (stage: "compile"): here. 2019-09-08 Akim Demaille CI: avoid useless git costs Travis answered favorably to my suggestion to provide a means to disable git clone on some jobs (issue 7542). See https://docs.travis-ci.com/user/customizing-the-build/#disabling-git-clone. * .travis.yml: Disable git globally, enable it for i. the compile job, and ii. the test job on ICC which needs the install-icc.sh script. 2019-09-08 Akim Demaille CI: factor * .travis.yml (Clang 7 libc++ and ASAN part 2): Reuse bits from "Clang 7 libc++ and ASAN part 1". 2019-09-08 Akim Demaille regen 2019-09-08 Akim Demaille gnulib: update Contains a fix for https://lists.gnu.org/archive/html/bug-bison/2019-08/msg00016.html. See https://lists.gnu.org/archive/html/bug-gnulib/2019-09/msg00005.html. Reported by 江 祖铭 (Zu-Ming Jiang). 2019-09-07 Akim Demaille fix: don't die when EOF token is defined twice With %token EOF 0 EOF 0 we get input.y:3.14-16: warning: symbol EOF redeclared [-Wother] 3 | %token EOF 0 EOF 0 | ^~~ input.y:3.8-10: previous declaration 3 | %token EOF 0 EOF 0 | ^~~ Assertion failed: (nsyms == ntokens + nvars), function check_and_convert_grammar, file /Users/akim/src/gnu/bison/src/reader.c, line 839. Reported by Marc Schönefeld. * src/symtab.c (symbol_user_token_number_set): Register only the first definition of the end of input token. * tests/input.at (Symbol redeclared): Check that case. 2019-09-07 Akim Demaille tests: check token redeclaration * src/symtab.c (symbol_class_set): Report previous definitions when redeclared. * tests/input.at (Symbol redeclared): New. 2019-09-06 Akim Demaille glr.c: initialize vector of bools The CI, with CC='gcc-7 -fsanitize=undefined,address -fno-omit-frame-pointer', reports: calc.cc:1652:50: runtime error: load of value 190, which is not a valid value for type 'bool' ../../tests/calc.at:867: cat stderr --- expout 2019-09-05 20:30:37.887257545 +0000 +++ /home/travis/build/bison-3.4.1.72-79a1-dirty/_build/tests/testsuite.dir/at-groups/438/stdout 2019-09-05 20:30:37.887257545 +0000 @@ -1 +1,2 @@ syntax error +calc.cc:1652:50: runtime error: load of value 190, which is not a valid value for type 'bool' 438. calc.at:867: 438. Calculator glr.cc (calc.at:867): FAILED (calc.at:867) The problem is that yylookaheadNeeds is not initialized in yyinitStateSet, and when it is copied, the value is not 0 or 1. * data/skeletons/glr.c (yylookaheadNeeds): Initialize yylookaheadNeeds. 2019-09-06 Akim Demaille yacc.c: beware of GCC's -Wmaybe-uninitialized Test 400 (calc.at:773: testing Calculator api.push-pull=both api.pure=full parse.error=verbose %debug %locations %defines api.prefix={calc} %verbose %yacc) fails on the CI with GCC 8 on Bionic: 400. calc.at:773: testing Calculator api.push-pull=both api.pure=full parse.error=verbose %debug %locations %defines api.prefix={calc} %verbose %yacc ... ../../tests/calc.at:773: bison --color=no -fno-caret -Wno-deprecated -o calc.c calc.y ../../tests/calc.at:773: $CC $CFLAGS $CPPFLAGS $LDFLAGS -o calc calc.c calc-lex.c calc-main.c $LIBS stderr: calc.y: In function 'int calcpush_parse(calcpstate*, int, const CALCSTYPE*, CALCLTYPE*)': calc.y:26:20: error: 'yylval.CALCSTYPE::ival' may be used uninitialized in this function [-Werror=maybe-uninitialized] %printer { fprintf (yyo, "%d", $$); } ; ^ calc.c:1272:9: note: 'yylval.CALCSTYPE::ival' was declared here YYSTYPE yylval YY_INITIAL_VALUE (= yyval_default); ^~~~~~ cc1plus: all warnings being treated as errors stdout: ../../tests/calc.at:773: exit code was 1, expected 0 400. calc.at:773: 400. Calculator api.push-pull=both api.pure=full parse.error=verbose %debug %locations %defines api.prefix={calc} %verbose %yacc (calc.at:773): FAILED (calc.at:773) * data/skeletons/c.m4 (yy_symbol_value_print): Disable the warning locally. 2019-09-06 Akim Demaille lalr1.cc: fix LAC support * data/skeletons/lalr1.cc (ctor): Initialize yy_lac_established_. This is quite painful to write, and ugly to read. 2019-09-06 Akim Demaille style: fix comment * tests/atlocal.in: here. 2019-09-06 Akim Demaille CI: more compilers * .travis.yml: Bionic is now available, with GCC8. GCC7 sanitizers work, but they are too longer: cover only part 1. Redefine part 1 and part 2 so that part 1 is really the core of the tests: not playing with POSIX and C++ compiler for C code. 2019-09-06 Akim Demaille CI: fail fast 2019-09-04 Akim Demaille gnulib: update This update brings file from Gettext 0.20, which is not available on the CI yet. .travis.yml: Adjust. Use Bionic now that it's available. 2019-09-01 Akim Demaille check for memory exhaustion hash_initialize returns NULL when out of memory. Check for it, and die cleanly instead of crashing. Reported by 江 祖铭 (Zu-Ming Jiang). https://lists.gnu.org/archive/html/bug-bison/2019-08/msg00015.html * src/muscle-tab.c, src/state.c, src/symtab.c, src/uniqstr.c: Check the value returned by hash_initialize. 2019-08-30 Akim Demaille news: LAC for C++ 2019-08-29 Akim Demaille d: remove useless imports * examples/d/calc.y, tests/calc.at: here. 2019-08-18 László Várady diagnostics: avoid global variables * src/complain.c (indent_ptr): Remove. (error_message, complains): Take indent as an argument. Adjust callers. 2019-08-18 László Várady diagnostics: fix invalid error message indentation https://lists.gnu.org/archive/html/bison-patches/2019-08/msg00007.html When Bison is started with a flag that suppresses warning messages, the error_message() function can produce a few gigabytes of indentation because of a dangling pointer. * src/complain.c (error_message): Don't reset indent_ptr here, but... (complain_indent): here. * tests/diagnostics.at (Indentation with message suppression): Check this case. 2019-08-18 Akim Demaille c++: use resize to shrink a vector Suggested by Adrian Vogelsgesang. https://lists.gnu.org/archive/html/bison-patches/2019-08/msg00009.html * data/skeletons/lalr1.cc (yy_lac_check_): here. 2019-08-09 Akim Demaille lalr1.cc: check LAC support * tests/conflicts.at, tests/input.at, tests/regression.at: here. 2019-08-09 Adrian Vogelsgesang lalr1.cc: add LAC support Implement lookahead correction (LAC) for the C++ skeleton. LAC is a mechanism to make sure that we report the correct list of expected tokens if a syntax error occurs. So far, LAC was only supported for the C skeleton "yacc.c". * data/skeletons/lalr1.cc: Add LAC support. * doc/bison.texi: Update. 2019-08-09 Adrian Vogelsgesang style: readability improvements to yacc.c * data/skeletons/yacc.c (yysyntax_error): Change the nesting of `m4` conditions slightly to make it more readable. The generated C code stays unchanged. 2019-08-09 Adrian Vogelsgesang lalr1.cc: reduce "scope" * data/skeletons/lalr1.cc (yy_lr_goto_state_): Make it static. 2019-08-09 Adrian Vogelsgesang lalr1.cc: fix indentation of table declarations in the header * data/skeletons/lalr1.cc: Fix indentation of table declarations in the generated header. 2019-08-09 Akim Demaille tests: prepare LAC tests for more languages * tests/regression.at: Use %expect to avoid warnings. Set the keywords to facilitate running specific tests. Use macros such as AT_YYLEX_DECLARE to facilitate tests for other languages. Likewise for AT_FULL_COMPILE. 2019-08-09 Akim Demaille git: ignore files generated in gnulib-po Because of them, the CI generates "-dirty" tarballs. 2019-07-26 Akim Demaille diagnostics: use the modern argmatch interface * src/complain.h (warnings): Remove Werror. Adjust dependencies. Sort. Remove useless comments (see the doc in argmatch group). * src/complain.c (warnings_args, warnings_types): Remove. (warning_argmatch): Use argmatch_warning_value. (warnings_print_categories): Use argmatch_warning_argument. 2019-07-19 Akim Demaille doc: avoid spurious empty lines in the option table In Texinfo. empty lines in multitable rows generate empty lines in the output. Avoid them altogether. With help from Gavin Smith. https://lists.gnu.org/archive/html/bug-texinfo/2019-07/msg00000.html * build-aux/cross-options.pl: Separate rows with empty lines. So, to be more readable, generate a single line for each row. Use Perl format to this end. 2019-07-07 Akim Demaille --fixed-output-files: detach from --yacc See the previous commit. This option should be removed, -o suffices. * src/getargs.c (FIXED_OUTPUT_FILES): New. Add support for it. (getargs): Define loc, and use it. This is safer when we need to pass a pointer to a location. 2019-07-07 Akim Demaille %fixed-output-files: detach from %yacc The name fixed-output-files is pretty clear: generate y.tab.c, as Yacc does. So let's detach this from %yacc which does more: it requires POSIX Yacc behavior. This directive is obsolete since December 29th 2001 8c9a50bee13474c1491df8f79f075f5214dda0d1. It does not show in the doc. I don't want to spend more time on improving its diagnostics, it could be removed just as well as far as I'm concerned. * src/scan-gram.l, src/parse-gram.y (%fixed-output-files): Detach from %yacc. 2019-07-07 Akim Demaille style: clarify control flow * src/getargs.c (language_argmatch): Initialize msg. Check it instead of relying on a return. 2019-07-07 Akim Demaille remove MS-DOS support DJGPP support was dropped in Bison 3.3 (c239e53bab8e87b30889ac446d3b513da9951a35). AS_FILE_NAME was introduced in ae4048011562c250d2f3c96687422d99a38745ce. * src/getargs.c (AS_FILE_NAME): Remove. 2019-07-07 Akim Demaille style: declare options in the same order as in --help * src/getargs.c (long_options): here. 2019-07-07 Akim Demaille regen 2019-07-07 Akim Demaille gnulib: update Contains a fix for argmatch to get proper man pages. See https://lists.gnu.org/archive/html/bug-gnulib/2019-07/msg00038.html 2019-07-07 Akim Demaille style: comment change * src/getargs.c: here. 2019-07-07 Akim Demaille doc: remove the --report=look-aheads alias Years ago we moved from 'look-ahead' to 'lookahead', and that alias was kept for backward compatibility. But now that we use argmatch to generate the documentation, that value clutters the doc. * src/getargs.c (argmatch_report_args): Remove the --report=look-aheads alias. 2019-07-07 Akim Demaille doc: fix inaccuracies wrt --define and --force-define The doc says that -Dfoo=bar is the same as %define foo "bar". It is not: the quotes are not added (and it makes a difference). * doc/bison.texi (Tuning the Parser): Fix the definition of -D/-F * src/getargs.c (usage): Likewise. 2019-07-07 Akim Demaille doc: put diagnostics related options together * doc/bison.texi (Diagnostics): New section. Move --warning, --color and --style there. * src/getargs.c (usage): Likewise. 2019-07-07 Akim Demaille doc: move -y's documentation into "Tuning the Parser" Let's clarify --help: use clearer "section" names, as in the doc. Move --yacc to where it belongs. * src/getargs.c (usage): Rename "Parser" as "Tuning the Parser", as in the doc. Rename "Output" as "Output Files" Move --yacc to "Tuning the Parser". * doc/bison.texi: Likewise. 2019-07-07 Akim Demaille doc: document colorized diagnostics * src/getargs.c (argmatch_color_group): New. (usage): Document --color and --style. * doc/bison.texi (Bison Options): Split into three subsections. Document --color and --style. 2019-07-03 Akim Demaille gnulib: use new features of the argmatch module It can now generate the usage message. * src/complain.h (feature_fixit_parsable): Rename as... (feature_fixit): this, for column economy. Adjust dependencies. (warning_usage): New. Use it. * src/complain.h, src/complain.c, src/getargs.h, src/getargs.c: Use ARGMATCH_DEFINE_GROUP instead of the older interface. 2019-07-02 Akim Demaille preserve the indentation in the ouput Preserve the actions' initial indentation. For instance, on | %define api.value.type {int} | %% | exp: exp '/' exp { if ($3) | $$ = $1 + $3; | else | $$ = 0; } we used to generate | { if (yyvsp[0]) | yyval = yyvsp[-2] + yyvsp[0]; | else | yyval = 0; } now we produce | { if (yyvsp[0]) | yyval = yyvsp[-2] + yyvsp[0]; | else | yyval = 0; } See https://lists.gnu.org/archive/html/bison-patches/2019-06/msg00012.html. * data/skeletons/bison.m4 (b4_symbol_action): Output the code in column 0, leave indentation matters to the C code. * src/output.c (user_actions_output): Preserve the incoming indentation in the output. (prepare_symbol_definitions): Likewise for %printer/%destructor. * tests/synclines.at (Output columns): New. 2019-07-01 Akim Demaille style: prefer passing locations by pointer The code is inconsistent: sometimes we pass by value, sometimes by reference. Let's stick to the last, more conventional for large values in C. * src/scan-code.l: Pass locations by reference. 2019-06-30 Akim Demaille c++: avoid duplicate definition of YYUSE Reported by Frank Heckenbach. https://lists.gnu.org/archive/html/bug-bison/2019-06/msg00009.html * data/skeletons/lalr1.cc (b4_shared_declarations): Remove the duplicate definition of YYUSE, the other one coming from b4_attribute_define. 2019-06-27 Akim Demaille style: comment changes * examples/c/lexcalc/local.mk, examples/c/reccalc/local.mk: Here. 2019-06-23 Akim Demaille tests: restructure for clarity * tests/calc.at (AT_CALC_MAIN, AT_CALC_LEX): Rewrite on top of AT_LANG_DISPATCH. 2019-06-23 Akim Demaille d: track locations * configure.ac (DCFLAGS): Pass -g. * data/skeletons/d.m4 (b4_locations_if): Remove, let bison.m4's one do its job. * data/skeletons/lalr1.d (position): Leave filename empty by default. (position::toString): Don't print empty file names. (location::this): New ctor. (location::toString): Match the implementations of C/C++. (yy_semantic_null): Leave undefined, the previous implementation does not compile. * tests/calc.at: Improve the implementation for D. Enable more checks, in particular using locations. * tests/local.at (AT_YYERROR_DEFINE(d)): Fix its implementation. 2019-06-23 Akim Demaille d: style changes * data/skeletons/lalr1.d: Use a more traditional quotation scheme. Formatting changes. 2019-06-23 Akim Demaille d: put internal details inside the parser Avoid name clashes, etc. * data/skeletons/lalr1.d (YYStackElement, YYStack): Move inside the parser. 2019-06-22 Akim Demaille gnulib: update 2019-06-22 Akim Demaille remove "experimental" warnings Sadly enough, AFAIK, there were never answers to the "More user feedback will help to stabilize it" sentences. Remove them. * src/getargs.c: IELR, canonical LR and XML output are here to stay, and they are no more experimental than some other features. * doc/bison.texi: Likewise. Also remove "experimental" warning for Java, LAC, LR tuning options, and named references. 2019-06-22 Akim Demaille CI: propagate sftp failures * .travis.yml (stage: "compile"): here. 2019-06-20 Akim Demaille d: honor %define parse.trace * data/skeletons/lalr1.d: Don't generate debug code if parse.trace is not enabled. 2019-06-20 Akim Demaille d: style changes * data/skeletons/lalr1.d: here. 2019-06-20 Akim Demaille d: prefer delegation to duplication * data/skeletons/lalr1.d: Delegate the construction of the scanner. 2019-06-20 Akim Demaille d: enable #line output * data/skeletons/d.m4 (b4_sync_start): New. 2019-06-20 Akim Demaille d: style changes * data/skeletons/lalr1.d: here. * examples/d/calc.y: Remove incorrect support for decimal numbers. Formatting changes. 2019-06-20 Akim Demaille style: reduce scopes in glr.c * data/skeletons/glr.c: here. 2019-06-20 Akim Demaille java: honor %define parse.trace * data/skeletons/lalr1.java: Don't generate debug code if parse.trace is not enabled. 2019-06-19 Akim Demaille java: fix support for api.prefix * data/skeletons/java.m4: here. * tests/java.at: Check it. 2019-06-19 Akim Demaille java: style changes * data/skeletons/lalr1.java: Use more conventional function names for Java. Prefer < and <= to => and >. Use the same approach for m4 quotation as in the other skeletons. Fix indentation issues. * tests/calc.at, tests/java.at, tests/javapush.at: Fix quotation style. (main): Use 'args', not 'argv', the former seems more conventional and is used elsewhere in Bison. Prefer character literals to integers to denote characters. * examples/java/Calc.y: Likewise. 2019-06-15 Akim Demaille CI: avoid useless git costs Travis answered favorably to my suggestion to provide a means to disable git clone on some jobs (issue 7542). See https://docs.travis-ci.com/user/customizing-the-build/#disabling-git-clone. * .travis.yml: Disable git globally, enable it for i. the compile job, and ii. the test job on ICC which needs the install-icc.sh script. 2019-06-12 Akim Demaille style: simplify strings to translate * src/conflicts.c (log_resolution): Don't translate indentation. 2019-06-12 Akim Demaille style: reduce scopes, propagate const * src/conflicts.c (conflicts_output): here. 2019-06-12 Akim Demaille style: use clearer types * src/conflicts.c (conflicts): Array of Booleans. 2019-06-11 Akim Demaille tests: prefer %empty * tests/regression.at: here. 2019-06-09 Akim Demaille CI: factor * .travis.yml (Clang 7 libc++ and ASAN part 2): Reuse bits from "Clang 7 libc++ and ASAN part 1". 2019-06-09 Akim Demaille lr0: more debug traces * src/lr0.c (kernel_check): New. (new_itemsets, save_reductions): Add traces. 2019-06-09 Akim Demaille traces: add some colors This is an experiment. Maybe more styles will be used (in which case a short-hand function will be useful), maybe it will be just reverted. * data/bison-default.css (.traces0): New. * src/lalr.c (lalr): Use it. 2019-06-09 Akim Demaille tests: make sure the default action properly works in C++ See e3fdc370495ffdedadd6ac621e32e34a0e1a9de0: in C++ we generate explicitly the code for the default action instead of simply copying blindly the semantic value buffer. This is important when copying raw memory is not enough, as exemplified by move-only types. This is currently tested by examples/c++/variant.yy and variant-11.yy. But it is safer to also have a test in the main test suite. * tests/local.at (AT_REQUIRE_CXX_STD): Fix. (AT_BISON_OPTION_PUSHDEFS, AT_BISON_OPTION_POPDEFS): Define/undefine AT_BISON_OPTIONS. * tests/c++.at (Default action): New. 2019-06-09 Akim Demaille tests: main: support -s and -p * tests/local.at (AT_MAIN_DEFINE(c), AT_MAIN_DEFINE(c++)): here. 2019-06-04 Akim Demaille tests: remove useless support of '.' in integers * tests/calc.at: here. * doc/bison.texi: Avoid uninitialized variables. 2019-05-29 Akim Demaille tests: refactor checks on sets It will be convenient to check sets elsewhere. * tests/sets.at (AT_EXTRACT_SETS): Transform into... * tests/local.at (AT_SETS_CHECK): this. * tests/sets.at: Adjust. 2019-05-29 Akim Demaille update-test: some file names have dashes in them * build-aux/update-test (log): Rename as... (trace): this, to avoid clashes with the log variable. (getargs): Clarify the type of the arguments. 2019-05-26 Akim Demaille tests: take SHELL into account Reported by Dennis Clarke. http://lists.gnu.org/archive/html/bug-bison/2019-05/msg00053.html * examples/local.mk, tests/local.mk: here. 2019-05-26 Akim Demaille gnulib: update to get gnulib translations This update contains a fix needed for gnulib-po to work properly. https://lists.gnu.org/archive/html/bug-gnulib/2019-05/msg00146.html 2019-05-25 Akim Demaille doc: clarify the purpose of symbol_type constructors Reported by Frank Heckenbach. http://lists.gnu.org/archive/html/bug-bison/2019-02/msg00006.html * doc/bison.texi (Complete Symbols): Here. 2019-05-22 Akim Demaille thanks: fix an address 2019-05-22 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2019-05-22 Akim Demaille version 3.4.1 * NEWS: Record release date. 2019-05-22 Akim Demaille NEWS: update 2019-05-20 Akim Demaille CI: remove useless apt-get update The apt addons already ran it for us, it is not needed. * .travis.yml: here. 2019-05-20 Akim Demaille c++: beware of to_string portability issues Reported by Bruno Haible. http://lists.gnu.org/archive/html/bug-bison/2019-05/msg00033.html * m4/bison-cxx-std.m4 (_BISON_CXXSTD_11_snippet): Check it. 2019-05-20 Akim Demaille doc: avoid Texinfo portability issues Reported by Bruno Haible. http://lists.gnu.org/archive/html/bug-bison/2019-05/msg00024.html Fixed by Karl Berry. http://lists.gnu.org/archive/html/bug-bison/2019-05/msg00034.html * doc/bison.texi: Don't specify the langage, rely on the default. Avoid blank pages. 2019-05-19 Akim Demaille examples: don't run those that require f?lex when it's not available Reported by Bruno Haible. http://lists.gnu.org/archive/html/bug-bison/2019-05/msg00026.html * configure.ac (FLEX_WORKS): New. * examples/c/lexcalc/local.mk, examples/c/reccalc/local.mk: Use it. 2019-05-19 Akim Demaille diagnostics: don't crash when libtextstyle is installed Reported by neok m4700. https://lists.gnu.org/archive/html/bison-patches/2019-05/msg00025.html https://github.com/akimd/bison/pull/11 * src/complain.c (complain_init_color): style_file_prepare _needs_ a string as second argument. 2019-05-19 Akim Demaille CI: avoid useless git costs The final gain is small: 2h2min instead 2h9min. But that is still an improvement. * .travis.yml (git.depth): Make the clone very shallow. (git.submodules): Don't clone gnulib in test jobs. (jobs.include.compile.script): Do it here. 2019-05-19 Akim Demaille fix: copyable instead of copiable Reported by Frank Heckenbach. http://lists.gnu.org/archive/html/bug-bison/2019-05/msg00020.html * data/skeletons/lalr1.cc, doc/bison.texi: here. 2019-05-19 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2019-05-19 Akim Demaille version 3.4 * NEWS: Record release date. 2019-05-19 Akim Demaille fix: use copiable, not copyable Reported by Hans Åberg. http://lists.gnu.org/archive/html/bug-bison/2019-05/msg00017.html * data/skeletons/lalr1.cc, doc/bison.texi: here. 2019-05-19 Akim Demaille NEWS: update for 3.4 2019-05-19 Akim Demaille tests: adjust to GCC9 diagnostics with a margin * tests/synclines.at (_AT_SYNCLINES_COMPILE): Remove the margin. 2019-05-19 Akim Demaille regen 2019-05-19 Akim Demaille diagnostics: %pure-parser is obsolete Reported by Uxio Prego. http://lists.gnu.org/archive/html/bug-bison/2018-12/msg00029.html * src/scan-gram.l, src/parse-gram.y (PERCENT_PURE_PARSER) (handle_pure_parser): New. Issue a deprecation/update notice for %pure-parser. * doc/bison.texi (Java Bison Interface): Don't mention %pure-parser. * tests/actions.at, tests/input.at: Adjust. 2019-05-19 Akim Demaille diagnostics: clean up convention for colored diagnostics * data/diagnostics.css: Rename as... * data/bison-default.css: this. Add the GPL header. This is the convention followed by Bruno Haible in gettext. Adjust dependencies. * src/complain.c (complain_init_color): Use BISON_STYLE instead of BISON_DIAGNOSTICS_STYLE. 2019-05-19 Akim Demaille CI: use a pipeline: first build the tarball, then check it Build the tarball in one job, check it in many. Unfortunately no real gain in overall duration. With help from Clément Démoulins. * .travis.yml: here. Remove all the tricks that were used to be able to boostrap on old distros. (before_install): Merge into 'script', because before_install applies to all the jobs, and we don't want to run it for the 'compile' job. 2019-05-18 Akim Demaille examples: fix srcdir/builddir issues * examples/d/local.mk, examples/java/local.mk: here. 2019-05-18 Akim Demaille gnulib: update In preparation for Bison 3.4, revert to the version before this commit: commit 03752516b21091cf3c4beea7e8b9bcad462d50ed Author: John Darrington Date: Sun May 12 00:42:36 2019 +0200 version-etc: Ease translation. * lib/version-etc.c (version_etc_arn, emit_bug_reporting_address): Move URLs and formatting newlines out of translatable string. because it changes the messages for --version, translated in gnulib.po. These changes are not yet available on the translation project, so we would have a regression in the set of translated strings. 2019-05-13 Akim Demaille build: do not use $< in plain rules It works only in implicit rules (or with GNU Make, but not with Solaris Make). Reported by Bruno Haible. http://lists.gnu.org/archive/html/bug-bison/2019-05/msg00009.html Diagnosed thanks to Kiyoshi Kanazawa. * examples/c/reccalc/local.mk, examples/d/local.mk, * examples/java/local.mk: Don't use $< in non implicit rules. 2019-05-12 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2019-05-12 Akim Demaille version 3.3.91 * NEWS: Record release date. 2019-05-12 Akim Demaille NEWS: update 2019-05-12 Akim Demaille gnulib: update 2019-05-11 Akim Demaille style: remove incorrect comment * src/getargs.c: here. It's documented in getargs.h anyway. 2019-05-09 Akim Demaille doc: use colors for diagnostics in TeX too Thanks to Gavin Smith and Patrice Dumas. http://lists.gnu.org/archive/html/help-texinfo/2019-04/msg00015.html * doc/bison.texi (@colorWarning, @colorError, @colorNotice) (@colorOff): Define for TeX and HTML. (@dwarning, @derror, @dnotice): Use them. 2019-05-08 Akim Demaille gnulib: update to fix location tracking in UTF-8 on Solaris This update contains Bruno Haible's fix for the location tracking issue reported by Kiyoshi Kanazawa. https://lists.gnu.org/archive/html/bug-gnulib/2019-05/msg00020.html https://lists.gnu.org/archive/html/bug-bison/2019-04/msg00020.html 2019-05-08 Akim Demaille diagnostics: rename --style=debug as --color=debug It is more consistent with --color=html, --color=test, etc. * src/getargs.h, src/getargs.c (style_debug): Rename as... (color_debug): this. (getargs_colors): Rename --style=debug as --color=debug. Adjust dependencies. 2019-05-08 Akim Demaille diagnostics: support --color=html Based on a message from Bruno Haible. https://git.savannah.gnu.org/gitweb/?p=gettext.git;a=commitdiff;h=fe18e92743b7226791a5f28d7c786941a1bf8cc9 This does not generate proper HTML: special characters are not escaped for instance. This is a hidden feature meant for Bison developers, not end users. * src/complain.c (complain_init_color): Support --color=html. 2019-05-08 Akim Demaille tests: use %empty instead of comments * tests/c++.at, tests/glr-regression.at: here. 2019-05-08 Akim Demaille fixits: sort them before applying them An experimental commit introduced a fix-it hint that changes comments such as "/* empty */" into %empty. But in some case, because diagnostics are not necessarily emitted in order, the fixits also come in disorder, which must never happen, as the fixes are installed in one pass. * src/fixits.c (fixits_register): Insert them in order. 2019-05-04 Akim Demaille style: use warning_is_enabled instead of duplicating it * src/complain.c (deprecated_directive): Here. 2019-05-03 Akim Demaille fixits: be sure to preserve the action when adding %empty Currently we remove the rhs to install %empty instead. * src/reader.c (grammar_rule_check_and_complete): Insert the missing %empty in front of the rhs, not in replacement thereof. * tests/actions.at (Add missing %empty): Check that. 2019-05-03 Akim Demaille tests: don't duplicate the portability prologue * tests/actions.at, tests/input.at: Don't repeat the prologue, skip it. * tests/diagnostics.at, tests/local.at: Comment changes. 2019-05-03 Akim Demaille style: use consistently *_loc for locations Some members are called foo_location, others are foo_loc. Stick to the latter. * src/gram.h, src/location.h, src/location.c, src/output.c, * src/parse-gram.y, src/reader.h, src/reader.c, src/reduce.c, * src/scan-gram.l, src/symlist.h, src/symlist.c, src/symtab.h, * src/symtab.c: Use _loc consistently, not _location. 2019-05-03 Akim Demaille style: clarify the use of symbol_lists' locations symbol_list features a 'location' and a 'sym_loc' member. The former is expected to be set only for symbol_lists that denote a symbol (not a type name), and the latter should only denote the location of the symbol/type name. Yet both are set, and the name "location" is too unprecise. * src/symlist.h, src/symlist.c (symbol_list::location): Rename as rhs_loc for clarity. Move it to the "section" of data valid only for rules. * src/reader.c, src/scan-code.l: Adjust. 2019-05-03 Akim Demaille maint: update gnulib-po/.gitignore 2019-04-29 Akim Demaille tests: don't require a D compiler Reported by Kiyoshi Kanazawa. http://lists.gnu.org/archive/html/bug-bison/2019-04/msg00018.html * tests/atlocal.in (BISON_DC_WORKS): New. * tests/local.at (AT_COMPILE_D): Use it. 2019-04-29 Akim Demaille doc: use svg instead of png * doc/bison.texi, doc/local.mk: here. 2019-04-29 Akim Demaille doc: use colors * doc/bison.texi (dwarning, derror, dnotice): New. Use them in the diagnostics. * doc/local.mk (AM_MAKEINFOFLAGS): Pass customization variables. 2019-04-28 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2019-04-28 Akim Demaille version 3.3.90 * NEWS: Record release date. 2019-04-28 Akim Demaille package: add missing CLEANFILES * examples: here. 2019-04-28 Akim Demaille build: don't generate the graph reports Revert "build: also generate the graph reports" (4ec413da32760defe1bf382c048d1d2f67e0b58a). The problem is Automake's ylwrap which does not rename y.dot with the appropriate name. We should completely stop using Automake's support for Yacc, which is not something I will do right now. So step back. * Makefile.am (AM_YFLAGS_WITH_LINES): Don't pass --graph. 2019-04-28 Akim Demaille package: don't regen the parser during dist if unneeded * Makefile.am (gen-synclines): New. 2019-04-28 Akim Demaille package: don't ship the sources generated from the parser Because some of our examples use %C%_reccalc_SOURCES = %D%/parse.y Automake ships parse.y and parse.c, and possibly parse.h when it "understands" that there is one. This is not what we want: ship only parser.y. Yet we still want to use Automake to compile the sources from parser.y. The easiest seems to use nodist_%C%_reccalc_SOURCES = %D%/parse.y together with dist_reccalc_DATA = %D%/parse.y %D%/scan.l %D%/Makefile %D%/README.md which guarantees that parse.y is indeed shipped. * examples/c/calc/local.mk, examples/c/lexcalc/local.mk, * examples/c/reccalc/local.mk: Always use nodist_*SOURCES for parsers, let the dist_*_DATA rules do their job. 2019-04-28 Akim Demaille package: various fixes for syntax-check * cfg.mk: Disable checks where needed (e.g., we do want to check the behavior with tabs). (sc_at_parser_check): Remove. Unfortunately since a11c144609255bc6e42c2aff83548e91cbd05425 we no longer use the './' prefix to run programs in the current directory. That was so that we could run Java programs like the other, although they are no run with the `./` prefix (see 967a59d2c08a33f24708450561e2f8010b604523). As a consequence this sc check no longer makes sense. However, since now AT_PARSER_CHECK passes the `./` prefix itself, this sc-check was superfluous. * examples/c/reccalc/scan.l: Use memcpy, not strncpy. * src/ielr.c, src/reader.c: Obfuscate "lr(0)" so that the sc-check for "space before paren" does not fire. * tests/diagnostics.at: Avoid space-tab, use tab-tab. 2019-04-27 Akim Demaille doc: clarify -fsyntax-error * NEWS, doc/bison.texi: here. 2019-04-27 Akim Demaille regen 2019-04-27 Akim Demaille traces: use colors for the semantic values This makes reading the trace slightly easier. It would be very nice to highlight the "big steps", especially reductions. But this is a private experiment: do not use it. * data/diagnostics.css (value): New. * src/parse-gram.y: Use no delimiters and no c quotation for strings to facilitate debugging. (tron, troff, TRACE): New. Not very elegant, but until there is support for printf-formats in libtextstyle, it shall be enough. 2019-04-27 Akim Demaille diagnostics: give m4 precise locations Currently we pass only the columns based on the screen-width, which is important for the carets. But we don't pass the bytes-based columns, which is important for the colors. Pass both. * src/muscle-tab.c (muscle_boundary_grow): Also pass the byte-based column. * src/location.c (location_caret): Clarify. (boundary_set_from_string): Adjust to the new format. * tests/diagnostics.at (Tabulations and multibyte characters from M4): New. 2019-04-27 Akim Demaille diagnostics: fix locations coming from M4 Locations issued from M4 need the byte-based column for the diagnostics to work properly. Currently they were unassigned, which typically resulted in partially non-colored diagnostics. * src/location.c (boundary_set_from_string): Fix the parsed location. * src/muscle-tab.c (muscle_percent_define_default): Set the byte values. * tests/diagnostics.at (Locations from M4): New. 2019-04-27 Akim Demaille diagnostics: show locations in full when debugging This is meant for developers, not end users, that's why I attached it to --trace. * src/getargs.h, src/getargs.c (trace_locations): New. * src/location.c (location_print): Use it. 2019-04-27 Akim Demaille diagnostics: use flush, not fflush * src/complain.c: here. 2019-04-25 Akim Demaille build: use gettext-h We were using the gnulib's gettext module with tricks in bootstrap.conf to avoid useless files. Instead, use gnulib's gettext-h module. * .travis.yml: Force Gettext 0.18.3 on Trusty. * bootstrap.conf: Use gettext-h instead of gettext. (excluded_files): Remove. * configure.ac (AM_GNU_GETTEXT_VERSION): Bump to 0.19. 2019-04-25 Akim Demaille NEWS: update 2019-04-25 Akim Demaille api.location.type: support it in C Reported by Balázs Scheidler. * data/skeletons/c.m4 (b4_location_type_define): Use api.location.type if defined. * doc/bison.texi: Document it. * tests/local.at (AT_C_IF, AT_LANG_CASE): New. Support Span in C. * tests/calc.at (Span): Convert it to be usable in C and C++. Check api.location.type with yacc.c and glr.c. 2019-04-24 Akim Demaille updates: insert/remove %empty * src/reader.c (grammar_rule_check_and_complete): Generate fixits for adding/removing %empty. * tests/actions.at, tests/diagnostics.at, tests/existing.at: Adjust. 2019-04-24 Akim Demaille regen 2019-04-24 Akim Demaille diagnostics: better rule locations The "identifier and colon" of a rule is implemented as a single token, but whose location is only that of the identifier (so that messages about the lhs of a rule are accurate). When reducing empty rules, the default location is the single point location on the end of the previous symbol. As a consequence, when Bison parses a grammar, the location of the right-hand side of an empty rule is based on the lhs, *independently of the position of the colon*. And the colon can be way farther, separated by comments, white spaces, including empty lines. As a result, some messages look really bad. For instance: $ cat foo.y %% foo : /* empty */ bar : /* empty */ gives $ bison -Wall foo.y foo.y:2.4: warning: empty rule without %empty [-Wempty-rule] 2 | foo : /* empty */ | ^ foo.y:3.4: warning: empty rule without %empty [-Wempty-rule] 3 | bar | ^ The carets are not at the right column, not even the right line. This commit passes the colon "again" after the "id colon" token, which gives more accurate locations for these messages: $ bison -Wall foo.y foo.y:2.10: warning: empty rule without %empty [-Wempty-rule] 2 | foo : /* empty */ | ^ foo.y:4.2: warning: empty rule without %empty [-Wempty-rule] 4 | : /* empty */ | ^ * src/scan-gram.l (SC_AFTER_IDENTIFIER): Rollback the colon, so that we scan it again afterwards. (INITIAL): Scan colons. * src/parse-gram.y (COLON): New. (rules): Parse the colon after the rule's id_colon (and possible named reference). * tests/actions.at, tests/conflicts.at, tests/diagnostics.at, * tests/existing.at: Adjust. 2019-04-24 Akim Demaille fixits: track byte-columns, not character-columns Because the fix-its were ready the character-based columns, but were applied on byte-based columns, the result with multibyte characters or tabs could be "interesting". For instance %fixed-output_files %fixed_output-files %fixed-output-files %define api.prefix {foo} %no-default-prec would give %fixed-%fixed-output-files %fixed_output-files %fixed-orefix= "foo" o_default-prec * src/fixits.c (fixit_print, fixits_run): Work on byte-base columns. * tests/input.at: Check it. 2019-04-24 Akim Demaille diagnostics: expose a means to know whether a warning is enabled * src/complain.h, src/complain.c (warning_is_enabled): New. 2019-04-23 Akim Demaille gnulib: let it use its own PO domain See https://www.gnu.org/software/gnulib/manual/html_node/Localization.html. * bootstrap.conf: Create gnulib-po. * Makefile.am, configure.ac: Use it. * po/POTFILES.in: Remove files now in gnulib. * src/main.c: Open the bison-gnulib domain. 2019-04-23 Akim Demaille diagnostics: don't try to quote special files Based on a report by Todd Freed. http://lists.gnu.org/archive/html/bug-bison/2019-04/msg00000.html See also https://gcc.gnu.org/bugzilla/show_bug.cgi?id=90034 * src/location.c (caret_info): Also track the file name. (location_caret): Don't quote special files. 2019-04-23 Akim Demaille diagnostics: document the change of format * doc/bison.texiL Adjust output. Also, Graphviz has no uppercsae V. * NEWS: Explain the format change. 2019-04-23 Akim Demaille diagnostics: copy GCC9's format Currently, when we quote the source file, we indent it with one space, and preserve tabulations, so there is a discrepancy and the visual rendering is bad. One way out is to indent with a tab instead of a space, but then this space can be used for more information. This is what GCC9 does. Let's play copy cats. See https://lists.gnu.org/archive/html/bison-patches/2019-04/msg00025.html https://developers.redhat.com/blog/2019/03/08/usability-improvements-in-gcc-9/ https://gcc.gnu.org/onlinedocs/gccint/Guidelines-for-Diagnostics.html#Guidelines-for-Diagnostics * src/location.c (location_caret): Prefix quoted lines with the line number and a pipe, fitting 8 columns. * tests/actions.at, tests/c++.at, tests/conflicts.at, * tests/diagnostics.at, tests/input.at, tests/java.at, * tests/named-refs.at, tests/reduce.at, tests/regression.at, * tests/sets.at: Adjust expectations. Partly by "./build-aux/update-test tests/testsuite.dir/*/testsuite.log" repeatedly, and partly by hand. 2019-04-23 Akim Demaille diagnostics: fix the handling of multibyte characters This is a pity: efforts were invested in computing correctly the number of screen columns consumed by multibyte characters, but the routines that do that were fed by single-byte inputs... As a consequence Bison never displayed correctly locations when there are multibyte characters. * src/scan-gram.l (mbchar): New. Use it instead of . in the catch-all clause. * tests/diagnostics.at (Tabulations): Enhance into... (Tabulations and multibyte characters): this. 2019-04-23 Akim Demaille diagnostics: check the handling of tabulations * tests/diagnostics.at (Tabulations): here. 2019-04-23 Akim Demaille diagnostics: fix styling issues Single point locations (equal boundaries) are troublesome, and we were incorrectly ending the style in their case. Which results in an abort in libtextstyle. There is also a confusion between columns as displayed on the screen (which take into account multibyte characters and tabulations), and the number of bytes. Counting the screen-column incrementally (character by character) is uneasy (because of multibyte characters), and I don't want to maintain a buffer of the current line when displaying the diagnostic. So I believe the simplest solution is to track the byte number in addition to the screen column. * src/location.h, src/location.c (boundary): Add the byte-column. Adjust dependencies. * src/getargs.c, src/scan-gram.l: Adjust. * tests/diagnostics.at: Check zero-width locations. 2019-04-23 Akim Demaille diagnostics: check the styling Enable checking of styles even when libtextstyle is not installed. * src/getargs.h, src/getargs.c (style_debug): New. (getargs_colors): Set it when --style=debug. * src/complain.c (begin_use_class, end_use_class): Use it. * tests/diagnostics.at: New. 2019-04-23 Akim Demaille TODO: update Let's prepare 3.4 with more or less what we have. Schedule some features for 3.5 and 3.6. Remove obsolete stuff. 2019-04-19 Akim Demaille doc: sort the warning categories * doc/bison.texi, src/getargs.c: here. 2019-04-19 Akim Demaille style: formatting changes * tests/actions.at, tests/calc.at, tests/input.at: here. 2019-04-19 Akim Demaille graphviz: move constant computation out of a loop * src/graphviz.c (output_red): here. 2019-04-18 Akim Demaille diagnostics: fix memory leak in libtextstyle * src/complain.h, src/complain.c (complain_free): New. * src/main.c: Use it. 2019-04-17 Akim Demaille tests: remove useless feature * tests/calc.at (read_signed_integer): Rename as... (read_integer): this. We never read signs here. 2019-04-14 Akim Demaille traces: make closure() less verbose * src/getargs.h, src/getargs.c (trace_closure): New. * src/closure.c (closure): Use it. 2019-04-14 Akim Demaille build: also generate the graph reports * Makefile.am (AM_YFLAGS_WITH_LINES): here. 2019-04-12 Akim Demaille yacc.c: minor style change * data/skeletons/yacc.c: To improve consistency with other similar pieces of code. 2019-04-12 Akim Demaille style: scope reduction in lalr.c * src/lalr.c (initialize_goto_follows): here. 2019-04-12 Akim Demaille style: comment changes * src/closure.h, src/closure.c, src/lalr.c: here. 2019-04-12 Akim Demaille traces: improve logs * src/lalr.c: Move logs to a better place to understand the chronology of events. * src/symlist.c (symbol_list_syms_print): Don't dump core on type elements. 2019-04-07 Akim Demaille doc: minor fixes * doc/bison.texi: Use consistently $ and @kbd in shell examples. Prefer sticking to English words: output and file instead of outfile and infile. 2019-04-03 Akim Demaille regen 2019-04-03 Akim Demaille bison: use no-lines The 'regen' commit in Bison's history are a nuisance. They are especially big because of the #lines. Let's generate our parse without these lines in the repository, but generate them in the tarball. * Makefile.am (AM_YFLAGS_WITH_LINES): New. (AM_YFLAGS): Use it. (dist-hook): Regenerate the parser with #lines. 2019-04-03 Akim Demaille no-lines: avoid leaving an empty line instead of the syncline Currently, with --no-lines, instead of "#line file line\n", we emit "\n". Let's emit nothing. * data/skeletons/bison.m4 (b4_syncline): Emit at end-of-line when enabled. * data/skeletons/bison.m4, data/skeletons/c.m4, data/skeletons/glr.cc, * data/skeletons/lalr1.cc, src/output.c: Use dnl after b4_syncline to avoid spurious empty lines. * tests/synclines.at (Sync Lines): Make sure that --no-lines is like grep -v #line. * tests/calc.at: Make sure that a rich grammar file behaves properly with %no-lines. 2019-04-03 Akim Demaille java: use full locations for diagnostics about destructors Currently we use the syncline to report errors about a symbol's destructor/printer. This is not accurate (only file and line), and this is incorrect: the file name is double quotes (a recent change, needed to make sure we escape properly double quotes in it). And worst of all: with --no-line, b4_syncline expands to nothing. Rather, push the locations into the backend, and use them. * src/muscle-tab.h, src/muscle-tab.c (muscle_location_grow): Make it public. * src/output.c (prepare_symbol_definitions): Use it to pubish the location of the printer and destructor. * data/skeletons/lalr1.java: Use complain_at instead of complain. * tests/java.at (Java invalid directives): Adjust expectations. * data/skeletons/bison.m4 (b4_symbol_action_location): Remove. We should not use b4_syncline this way. 2019-04-03 Akim Demaille java: prefer errors to fatal errors Fatal errors are inconvenient, and should be reserved to cases where we cannot continue. Here, it could even be warnings actually: these directives will simply be ignored. * data/skeletons/lalr1.java: Prefer error (b4_complain) to fatal errors (b4_fatal). * tests/java.at (Java invalid directives): New. 2019-04-03 Akim Demaille tests: formatting changes * tests/javapush.at: here. 2019-04-03 Akim Demaille lalr: offer more flexibility in debugging routines * src/state.h, src/state.c (state_transitions_print): New, extracted from... (state_transitions_set): here. 2019-03-31 Akim Demaille lalr: don't overbook memory I never understood why we book ngotos+1 slots for relations between gotos: there are at most ngotos images, not ngotos+1 (and "includes" does have cases where a goto is in relation with itself, so it's not ngotos-1). Maybe bbf37f2534a8e5a6b4e28047f0a10903e6dc73f9 explains the +1: a bug left us register a goto several times on occasion, and the +1 might have been a means to avoid this problem in most cases. Now that this bug is addressed, we should no longer overbook memory, if only for the clarity of the code ("why ngotos+1 instead of ngotos?"). * src/lalr.c: A goto has at most ngotos images, not ngotos+1. While at it, avoid useless repeated call to map_goto introduced in bbf37f2534a8e5a6b4e28047f0a10903e6dc73f9. 2019-03-30 Akim Demaille lalr: show lookback for debug * src/lalr.c (lookback_print): New. (build_relations): Use it. Also show edges. 2019-03-30 Akim Demaille diagnostics: don't crash when declaring the error token as an nterm Reported by wcventure. http://lists.gnu.org/archive/html/bug-bison/2019-03/msg00008.html * src/symtab.c (complain_class_redeclared): Don't print empty locations. There can only be empty locations for predefined symbols. And the only symbol that is lexically available is the error token. So this appears to be the only possible way to have an error involving an empty location. * tests/input.at (Symbol class redefinition): Check it. 2019-03-30 Akim Demaille lalr: fix segmentation violation The "includes" relation [DeRemer 1982] is between gotos, so of course, for a given goto, there cannot be more that ngotos (number of gotos) images. But we manipulate the set of images of a goto as a list, without checking that an image was not already introduced. So we can "register" way more images than ngotos, leading to a crash (heap buffer overflow). Reported by wcventure. http://lists.gnu.org/archive/html/bug-bison/2019-03/msg00007.html For the records, this bug is present in the first committed version of Bison. * src/lalr.c (build_relations): Don't insert the same goto several times. * tests/sets.at (Build Relations): New. 2019-03-30 Akim Demaille state: more debug traces * src/state.c (state_transitions_set): Show the transitions. 2019-03-30 Akim Demaille style: rename variables for consistency * src/lalr.c: Use trans for transitions, and reds for reductions, as elsewhere in the code. * src/state.h: Comment changes. 2019-03-30 Akim Demaille gram: fix and improve log message It seems that not many people read these logs: the error was introduced in 2001 (3067fbef531832df1e43bbd28787655808361eed), * src/gram.c (grammar_dump): Fix the headers of the table: remove duplicate display of "Ritem Range". While at it, remove duplicate display of the rule number (and remove an incorrect comment about it: these numbers _are_ equal). * tests/sets.at (Reduced Grammar): Use useless rule, nterm and token in the example. 2019-03-30 Akim Demaille tests: add a tool for mass updates When we update some output format, too many adjustements must be made by hand. This script updates most tests based on the actual output made during the tests. * build-aux/update-test: New. 2019-03-25 Akim Demaille style: remove now useless _GL_UNUSED * src/getargs.c (getargs_colors): Here. Useless since 4d34b06fb3a38eb050439084476a6b3e292c5680. 2019-03-24 Theophile Ranquet tables: use bitsets for a performance boost Suggested by Yuri at . The improvement is marginal for most grammars, but notable for large grammars (e.g., PosgreSQL's postgre.y), and very large for the sample.y grammar submitted by Yuri in http://lists.gnu.org/archive/html/bison-patches/2012-01/msg00012.html. Measured with --trace=time -fsyntax-only. parser action tables postgre.y sample.y Before 0,129 (44%) 37,095 (99%) After 0,117 (42%) 5,046 (93%) * src/tables.c (pos): Replace this set of integer coded as an unsorted array or integers with... (pos_set): this bitset. 2019-03-24 Akim Demaille yacc.c: don't suggest api.header.include when --defines is not used See 4e19ab9fcd28c9361ff08f46e5e353effb0a0520: the suggestion to include the header file should not be emitted when the header is not generated. * data/skeletons/yacc.c: Here. 2019-03-24 Akim Demaille reader: clarify variable names * src/reader.c (grammar_rule_check_and_complete): When 'p' and 'lhs' are aliases, prefer the latter, for clarity and consistency. (grammar_current_rule_begin): Avoid 'p', current_rule suffices. * src/gram.h, src/gram.c: Comment changes. ptdr# calc.tab.c 2019-03-24 Akim Demaille diagnostics: style changes * src/location.c (location_caret): Clarify a bit. 2019-03-24 Akim Demaille diagnostics: use gnulib's libtextstyle-optional Bruno Haible just added a default implementation of libtextstyle's interface when the library is not available. https://lists.gnu.org/archive/html/bison-patches/2019-03/msg00025.html * gnulib: Update. * bootstrap.conf: Replace libtextstyle with libtextstyle-optional. * src/complain.c, src/getargs.c: Remove now useless cpp guards. 2019-03-23 Akim Demaille diagnostics: fix handling of style in limit cases * src/location.c (location_caret): Beware of the cases where the start and end columns are the same, or when the location is multilines. 2019-03-23 Akim Demaille warnings: don't use _Noreturn with G++ 4.7 in C++98 mode The timevar and bitset modules now use the c99 module which causes $CXX to now include -std=gnu++11 when possible. Unfortunately, G++ 4.7 does not implement [[noreturn]] in C++11 mode, so our tests of glr.cc (which uses _Noreturn) fail with input.cc:954:1: error: expected unqualified-id before '[' token right before [[noreturn]]. 4.8 works fine. * data/skeletons/c.m4 (b4_attribute_define): Do not use [[noreturn]] with GCC 4.7. 2019-03-17 Akim Demaille d: tests: use more a natural approach for the scanner See f8408562f8439654261418406296e4108d2a995f. * tests/calc.at: Stop imitating the C API. Prepare more tests to run in the future. %verbose works as expected (what a surprise, it's unrelated to the skeleton...). 2019-03-17 Akim Demaille regen 2019-03-17 Akim Demaille style: rename spec_defines_file as spec_header_file The variable spec_defines_file denotes the name of the generated header. Its name is derived from --defines/%defines, whose name in turn is derived from the fact that the header, in Yacc, contained the Not only does the header now contain a lot more than just the token definitions, but we no longer even generate macros, but an enum... Let's modernize our vocabulary. * src/files.h, src/files.c (spec_defines_file): Rename as... (spec_header_file): this. 2019-03-17 Akim Demaille yacc.c: provide a means to include the header in the implementation Currently when --defines is used, we generate a header, and paste an exact copy of it into the generated parser implementation file. Let's provide a means to #include it instead. We don't do it by default because of the Autotools' ylwrap. This program wraps invocations of yacc (that uses a fixed output name: y.tab.c, y.tab.h, y.output) to support a more modern naming scheme (dir/foo.y -> dir/foo.tab.c, dir/foo.tab.h, etc.). It does that by renaming the generated files, and then by running sed to propagate these renamings inside the files themselves. Unfortunately Automake's Makefiles uses Bison as if it were Yacc (with --yacc or with -o y.tab.c) and invoke bison via ylwrap. As a consequence, as far as Bison is concerned, the output files are y.tab.c and y.tab.h, so it emits '#include "y.tab.h"'. So far, so good. But now ylwrap processes this '#include "y.tab.h"' into '#include "dir/foo.tab.h"', which is not guaranteed to always work. So, let's do the Right Thing when the output file is not y.tab.c, in which case the user should %define api.header.include. Binding this behavior to --yacc is tempting, but we recently told people to stop using --yacc (as it also enables the Yacc warnings), but rather to use -o y.tab.c. Yacc.c is the only skeleton concerned: all the others do include their header. * data/skeletons/yacc.c (b4_header_include_if): New. (api.header.include): Provide a default value when the output is not y.tab.c. * src/parse-gram.y (api.header.include): Define. 2019-03-17 Akim Demaille d: don't link against LIBS * tests/local.at (AT_COMPILE_D): Don't pass LIBS, dmd does not like being given -lintl. 2019-03-17 Akim Demaille address warnings from GCC's UB sanitizer Running with CC='gcc-mp-8 -fsanitize=undefined' revealed Undefined Behaviors. https://lists.gnu.org/archive/html/bison-patches/2019-03/msg00008.html * src/state.c (errs_new): Don't call memcpy with NULL as source. * src/location.c (add_column_width): Don't assume that the column argument is nonnegative: the scanner sometimes "backtracks" (e.g., see ROLLBACK_CURRENT_TOKEN and DEPRECATED) in which case we can have negative column numbers (temporarily). Found in test 3 (Invalid inputs). 2019-03-16 Akim Demaille diagnostics: use libtextstyle for colored output Bruno Haible released libtextstyle, a library for colored output based on CSS. Let's use it to generate colored diagnostics, provided libtextstyle is available. See https://lists.gnu.org/archive/html/bug-gnulib/2019-01/msg00176.html https://lists.gnu.org/archive/html/bison-patches/2019-02/msg00073.html https://lists.gnu.org/archive/html/bison-patches/2019-02/msg00084.html https://lists.gnu.org/archive/html/bison-patches/2019-03/msg00007.html * bootstrap.conf (gnulib_modules): Use libtextstyle when possible. * data/diagnostics.css: New. * src/complain.c (begin_use_class, end_use_class, flush) (severity_style, complain_init_color): New. Use them. * src/getargs.c (getargs_colors): New. (getargs): Use it. Skip --color and --style. * src/location.h, src/location.c (location_print): Use a style. * tests/bison.in: Force --color=yes when stderr is a tty. * tests/local.at: Disable colors during the test suite. * tests/input.at: Adjust expectations to the extra options passed on the command line. 2019-03-16 Akim Demaille style: clean up complain.c * src/complain.c (severity_prefix): New. (error_message): Take the severity as argument, instead of the prefix. 2019-03-16 Akim Demaille yacc.c: emit the header before the implementation file * data/skeletons/yacc.c: here. This is more logical for the time stamps, but it's also required by following patches: the shared declarations are also in charge of handling api.value.type=union. So far, they are run in the implementation file in both cases (with or without header). But if we run them only in the header, then the implementation file is emited with incorrect support for api.value.type=union. Arguably we should not have such dependencies. This is because we have side-effects in our backend (redefining the symbols' type and type_tag). In the future we should find a better solution for this, without sacrificing the independence of the backend from bison itself (i.e., I don't think we should handle api.value.type=union in bison, leave it to m4). 2019-03-16 Akim Demaille simplify the generated #line Currently we generate things like: #line 683 "src/parse-gram.y" /* yacc.c:316 */ The first part is of course very important: compilers point the users to their grammar file rather than into the generated parser. The second part points to the place in the skeletons that generated this piece of code. This dependency on the Bison skeletons generates lots of useless 'git diff'. This location is useless for the regular user (who does not care about the skeletons) and is actually not useful for Bison developpers too (I never used this to locate the code in skeletons that generated output). So disable it completely. If someone thinks this was actually useful, a %define variable should be provided to control the level of verbosity of '#line', in replacement of --no-lines. So now, generate: #line 683 "src/parse-gram.y" * data/skeletons/bison.m4 (b4_sync_end): Emit nothing. 2019-03-13 Akim Demaille gnulib: update 2019-03-13 Akim Demaille tests: remove duplicates * tests/regression.at (Invalid inputs, Invalid inputs with {}): Remove, there are exact copies of them in input.at. 2019-03-02 Akim Demaille d: simplify the API to build the scanner of the example * examples/d/calc.y (calcLexer): Add an overload for File. Use it. 2019-03-01 H. S. Teoh d: modernize the scanner of the example https://lists.gnu.org/archive/html/bison-patches/2019-02/msg00121.html * examples/d/calc.y (CalcLexer): Stop shoehorning C's API into D: use a range based approach in the scanner, rather than some imitation of getc/ungetc. (main): Adjust. 2019-03-01 Akim Demaille d: tests: use fewer global variables * tests/calc.at: Move 'input' into the scanner. 2019-02-28 Akim Demaille lalr: clarify the count of lookaheads * src/lalr.c (state_lookahead_tokens_count): Remove wierd `+=` that is actually an `=`. 2019-02-28 Akim Demaille lalr: clarify the API * src/state.h, src/state.c (state_reduction_find): Clarify. Die on errors. * src/lalr.c (goto_list_new): New. Use it. 2019-02-28 Akim Demaille lalr: improve traces * src/lalr.c (follows_print): Just print the symbol tag. Take and print a title. Indent the output. Use it to print the various steps of the computation. (lookahead_tokens_print): Fix a lie: the number displayed is not the number of tokens. Don't display states that don't even have reductions. 2019-02-27 Akim Demaille lalr: print the 'reads' relation * src/relation.h, src/relation.c (relation_print): Accept and use a title. Don't print empty rows. Indent the output. Adjust dependencies. * src/lalr.c (initialize_goto_follows): Print 'reads' in traces. 2019-02-27 Akim Demaille style: comment changes * src/lr0.c: here. 2019-02-26 Akim Demaille dlang: initial changes to run the calc tests on it * configure.ac (DCFLAGS): Define. * tests/atlocal.in: Receive it. * data/skeletons/d.m4 (api.parser.class): Remove spurious YY. * data/skeletons/lalr1.d (yylex): Return an int instead of a YYTokenType, so that we can use characters as tokens. * examples/d/calc.y: Adjust. * tests/local.at: Initial support for D. (AT_D_IF, AT_DATA_GRAMMAR(D), AT_YYERROR_DECLARE(d)) (AT_YYERROR_DECLARE_EXTERN(d), AT_YYERROR_DEFINE(d)) (AT_MAIN_DEFINE(d), AT_COMPILE_D, AT_LANG_COMPILE(d), AT_LANG_EXT(d)): New. * tests/calc.at: Initial support for D. * tests/headers.at 2019-02-26 Akim Demaille d: improve the example * examples/d/calc.y: Exit with failure on errors. Remove useless operators (=, !) meant for the test suite. Add unary + for symmetry. * examples/d/calc.test: Adjust expectations. 2019-02-26 Akim Demaille tests: style changes * tests/local.at AT_YYERROR_DEFINE(java): Use more consistent names. 2019-02-25 Akim Demaille style: eliminate useless indirection * src/relation.h, src/relation.c (relation_digraph): Don't take the biteetv as a pointer, it is already a pointer (as it's an array). 2019-02-25 Akim Demaille style: rename function for clarity Commit db34f7988941444bdc5f2b6adcf7fb83648f9a18 renames the variable F as goto_follows, but forgot to rename this function. * src/lalr.c (initialize_F): Rename as... (initialize_goto_follows): this. 2019-02-25 Akim Demaille lalr: more debug traces I need to be able to read includes and goto_follows. * src/relation.h, src/relation.c (relation_print): Provide a means to pretty-print the nodes of the relation. * src/lalr.c (goto_print, follows_print): New. (set_goto_map): Use goto_print. (build_relations): Show INCLUDES. (compute_FOLLOWS): Rename as... (compute_follows): this. Show FOLLOWS. 2019-02-24 Akim Demaille style: minor changes * examples/c/calc/calc.y, src/lalr.c: Reduce scope. * src/gram.c: Prefer < to >. 2019-02-24 Akim Demaille style: clarify the computation of the lookback edges * src/lalr.c (build_relations): Reduce the scopes. Instead of keeping rp alive in two different loops, clarify the second one by having an index on the path we traverse (i.e., use that index to compute the source state _and_ the symbol that labels the transition). This allows to turn an obscure 'while'-loop in a clearer (IMHO) 'for'-loop. We also consume more variables (by introducing p instead of making more side effects on length), but we're in 2019, I don't think this matters. What does matter is that (IMHO again), this is now clearer. Also, use clearer names. 2019-02-24 Akim Demaille style: scope reduction in tables.c * src/tables.c: here. * src/lalr.c: Prefer < to >. 2019-02-24 Akim Demaille d: formatting changes * data/skeletons/d.m4, data/skeletons/lalr1.d: Avoid trailing spaces. 2019-02-23 Akim Demaille examples: remove stray examples * examples/c/reentrant-calc: Remove. I did not mean to include this example, it was replaced by examples/c/reccalc. 2019-02-21 Akim Demaille tests: factor the execution of Java parsers * tests/local.at (AT_MAIN_DEFINE(java)): Exit failure on failure. (AT_PARSER_CHECK): If in Java, run AT_JAVA_PARSER_CHECK. * tests/conflicts.at (AT_CONSISTENT_ERRORS_CHECK): Simplify. 2019-02-21 Akim Demaille tests: fix a Java tests * tests/conflicts.at (AT_CONSISTENT_ERRORS_CHECK): Fix quotation error. 2019-02-21 Akim Demaille tests: simplify AT_PARSER_CHECK usage Currently the caller must specify the ./ prefix to its command. Let's avoid that: it will be nicer to read, make it easier to have a version that works for Java and C/C++. * tests/local.at (AT_PARSER_CHECK): Prefix the command with ./. Adjust callers. 2019-02-21 Akim Demaille tests: java: factor the definition of Position * tests/local.at (AT_JAVA_POSITION_DEFINE): New. * tests/java.at, tests/javapush.at: Use it. 2019-02-21 Akim Demaille tests: dispatch per lang on AT_DATA_GRAMMAR * tests/java.at: Do that. * tests/conflicts.at: Simplify. * tests/actions.at, tests/c++.at, tests/input.at, tests/local.at, * tests/named-refs.at: Use AT_BISON_OPTION_PUSHDEFS/AT_BISON_OPTION_POPDEFS. 2019-02-21 Akim Demaille tests: dispatch per lang on the definition of yylex * tests/local.at (AT_YYLEX_DEFINE): Dispatch on the language. (AT_YYLEX_DEFINE(java)): New. * tests/conflicts.at, tests/java.at: Use it. 2019-02-21 Akim Demaille tests: de-duplicate * tests/conflicts.at (AT_YYLEX_PROTOTYPE): Don't define it, leave that task to AT_DATA_GRAMMAR. But be honest: tell AT_BISON_OPTION_PUSHDEFS all the options we use. 2019-02-21 Akim Demaille tests: formatting changes * tests/local.at: here. 2019-02-21 Akim Demaille graph: prefer *.gv to *.dot Reported by Hans Åberg. https://lists.gnu.org/archive/html/help-bison/2019-02/msg00064.html * src/files.c (spec_graph_file): Use `*.gv` when 3.4 or better, otherwise `*.dot`. * src/parse-gram.y (handle_require): Pretend we are already 3.4. * doc/bison.texi: Adjust. * tests/local.at, tests/output.at: Exercise this. 2019-02-21 Akim Demaille doc: fixes to please older versions of Texinfo Currently, we have errors: ./doc/bison.texi:13005: node `History' lacks menu item for `Yacc' despite being its Up target * doc/bison.texi (History): Add the local menu. While at it, add a few index entries. 2019-02-17 Akim Demaille doc: style changes * doc/bison.texi (Bibliography): Add [Corbett 1984] and [Johnson 1978]. (History): Use them. And other minor changes. 2019-02-17 Eric S. Raymond doc: a history section * bison.texi (A Brief History of the Greater Ungulates): New section. 2019-02-17 Akim Demaille examples: add an example with a reentrant parser in Flex+Bison Suggested by Eric S. Raymond. https://lists.gnu.org/archive/html/bison-patches/2019-02/msg00066.html * examples/c/reentrant-calc/Makefile, examples/c/reentrant-calc/README.md, * examples/c/reentrant-calc/parse.y, examples/c/reentrant-calc/scan.l * examples/c/reentrant-calc/lexcalc.test, * examples/c/reentrant-calc/local.mk: New. 2019-02-16 Akim Demaille examples: fixes in lexcalc * examples/c/lexcalc/parse.y: Formatting/comment changes. (line): Don't return a value. Print the result here, which avoids printing a value for lines with an error. (yyerror): Be sure to increment the pointed, not the pointer... * examples/c/lexcalc/lexcalc.test: Check errors. * examples/c/lexcalc/local.mk: Fix a dependency. 2019-02-16 Akim Demaille build: fix distcheck Regression introduced in 05a80977798abf472bfc11c477751303ad604733. * doc/local.mk (bison.help): Don't depend on the path of the bison executable. 2019-02-16 Akim Demaille style: move pkgdatadir to files.* Let's move it to a more logical place. * src/output.h, src/output.c (pkgdatadir): Move to... * src/files.h, src/files.c: here. 2019-02-14 Akim Demaille style: rename cleanup_caret as caret_free * src/location.c, src/location.h, src/main.c: here. 2019-02-14 Akim Demaille style: avoid default in switch on enums * src/assoc.c (assoc_to_string): here. 2019-02-14 Akim Demaille doc: run tests/bison, not src/bison * doc/local.mk: Don't run src/bison, as it expects to find all its files installed. Instead, run tests/bison which is ready to run in builddir. 2019-02-13 Akim Demaille gnulib: update 2019-02-12 Akim Demaille style: comment and names changes in map_goto * src/lalr.h, src/lalr.c: Use clearer names. 2019-02-12 Akim Demaille yacc: support parse.assert While hacking on the computation of the automaton, I had yystate being equal to -1, and the parser loops. Let's catch this when parser.assert is enabled. * data/skeletons/yacc.c (YY_ASSERT): New. Use it. Not using the name YYASSERT, to make it clear that this is private. glr.c should probably move to YY_ASSERT too. Also, while at it, report 'Entering state...' even before growing the stacks. 2019-02-12 Akim Demaille examples: depend on Bison's sources * examples/c/calc/local.mk, examples/c/lexcalc/local.mk, * examples/c/mfcalc/local.mk, examples/c/rpcalc/local.mk: Regenerate the files if dependencies have changed. 2019-02-12 Eric S. Raymond README: point to README-hacking * README (Build from git): New. * README-hacking: Describe easier submodule update. 2019-02-10 Akim Demaille doc: a single space before closing comments I don't think this style: /* If buffer is full, make it bigger. */ if (i == length) { length *= 2; symbuf = (char *) realloc (symbuf, length + 1); } /* Add this character to the buffer. */ symbuf[i++] = c; /* Get another character. */ c = getchar (); or more readable than /* If buffer is full, make it bigger. */ if (i == length) { length *= 2; symbuf = (char *) realloc (symbuf, length + 1); } /* Add this character to the buffer. */ symbuf[i++] = c; /* Get another character. */ c = getchar (); Actually, I think the latter is more readable, and helps with width issues in the PDF. As a matter of fact, I would happily move to // only for single line comments. * doc/bison.texi: A single space before closing comments. 2019-02-10 Akim Demaille doc: modernize the examples * doc/bison.texi: Prefer 'fun' to 'fnct'. Reduce local variable scopes. Prefer strdup to malloc + strcpy. Avoid gratuitous casts. Use simpler names (e.g., 'name' instead of 'fname'). Avoid uses of 0 for NULL. Avoid using NULL when possible (e.g., 'p' instead of 'p != NULL'). Prefer union names to casts (e.g. 'yylval.VAR = s' instead of '*((symrec**) &yylval) = s'). Give arguments a name in fun declarations. Use our typedefs instead of duplicating them (func_t). Stop promoting an explicit $$ = $1;, it should be implicit (Bison might be able to eliminate useless chain rules). Help a bit Texinfo by making smaller groups. Rely on the C compiler to call function pointers (prefer '$1->value.fun ($3)' to (*($1->value.fnctptr))($3)'). 2019-02-10 Akim Demaille examples: add a simple infix calculator in C Currently we have no simple example: rpcalc in reverse Polish, mfcalc has functions, and lexcalc is using lex. * examples/c/calc/Makefile, examples/c/calc/calc.y, * examples/c/calc/calc.test, examples/c/calc/local.mk: New. 2019-02-10 Akim Demaille examples: fix annoying off-by-one errors * examples/extexi: Since we issue #lines only at the beginning of @example, leave empty line when removing content (such as @comment lines), otherwise the lines that follow have incorrect source line location. This leaves ugly empty lines, but they are removed when you tidy the output for the end user: sequences of \n are mapped to at most two sucessive \n. 2019-02-09 Akim Demaille style: factor printing of rules * src/gram.h, src/gram.c (rule_print): New. Use it. 2019-02-09 Akim Demaille style: use lower case for variable names * src/relation.c (INDEX, VERTICES): Rename as... (indexes, vertices): these. 2019-02-09 Akim Demaille style: scope reduction in relation.c 2019-02-09 Akim Demaille report: stop counting uselessly * src/print.c (print_nonterminal_symbols): Replace left_count and right_count with on_left and on_right. 2019-02-09 Akim Demaille report: clean up its format The format is inconsistent. For instance most sections are indented (including "Terminals unused in grammar" for instance), but the sections "Terminals, with rules where they appear" and "Nonterminals, with rules where they appear" are not. Let's indent them. Also, these two sections try to wrap the output to avoid lines too long. Yet we don't do that in the rest of the file, for instance when listing the lookaheads of an item. For instance in the case of Bison's parse-gram.output we go from: Terminals, with rules where they appear "end of file" (0) 0 error (256) 28 88 "string" (258) 9 13 16 17 20 23 24 109 116 [...] Nonterminals, with rules where they appear $accept (58) on left: 0 input (59) on left: 1, on right: 0 prologue_declarations (60) on left: 2 3, on right: 1 3 prologue_declaration (61) on left: 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 22 23 24 25 26 27 28 29, on right: 3 [...] to Terminals, with rules where they appear "end of file" (0) 0 error (256) 28 88 "string" (258) 9 13 16 17 20 23 24 109 116 [...] Nonterminals, with rules where they appear $accept (58) on left: 0 input (59) on left: 1 on right: 0 prologue_declarations (60) on left: 2 3 on right: 1 3 prologue_declaration (61) on left: 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 22 23 24 25 26 27 28 29 on right: 3 [...] * src/print.c (END_TEST): Remove. (print_terminal_symbols): Don't try to wrap the output. (print_nonterminal_symbols): Likewise. Make two different lines for occurrences on the left, and occurrence on the rhs of the rules. Indent by 4 and 8, not 3. * src/reduce.c (reduce_output): Indent by 4, not 3. * tests/conflicts.at, tests/existing.at, tests/reduce.at, * tests/regression.at, tests/report.at: Adjust. 2019-02-05 Akim Demaille add LR(0) output This should not be used to generate parsers. My point is actually to facilitate debugging (when tweaking the generation of the LR(0) automaton for instance, not carying -yet- about lookaheads). * src/reader.c (prepare_percent_define_front_end_variables): Add lr(0). * src/conflicts.c (set_conflicts): Be robust to reds not having lookaheads at all. * src/ielr.c (LrType, lr_type_get): Adjust. (ielr): Implement support for LR(0). * src/lalr.c (lalr_free): Don't free LA when it's not computed. 2019-02-05 Akim Demaille style: scope reduction in derives.c * src/derives.c: here. 2019-02-05 Akim Demaille style: comment changes and refactoring in state.c * src/state.h, src/state.c: Comment changes. (transitions_to): Take a state* as argument. * src/lalr.h, src/lalr.c: Comment changes. (initialize_F): Use clear variable names. 2019-02-05 Akim Demaille tests: fix typos * tests/reduce.at: here. 2019-02-03 Akim Demaille Merge branch maint * maint: maint: post-release administrivia version 3.3.2 style: minor fixes NEWS: named constructors are preferable to symbol_type ctors gram: fix handling of nterms in actions when some are unused style: rename local variable CI: update the ICC serial number for travis-ci.org 2019-02-03 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2019-02-03 Akim Demaille version 3.3.2 * NEWS: Record release date. 2019-02-03 Akim Demaille style: minor fixes * NEWS, src/reduce.c, src/reduce.h: Use 'nonterminal'. Fix comments. 2019-02-03 Akim Demaille NEWS: named constructors are preferable to symbol_type ctors Reported by Frank Heckenbach. http://lists.gnu.org/archive/html/bug-bison/2019-01/msg00043.html 2019-02-03 Akim Demaille gram: fix handling of nterms in actions when some are unused Since Bison 3.3, semantic values in rule actions (i.e., '$...') are passed to the m4 backend as the symbol number. Unfortunately, when there are unused symbols, the symbols are renumbered _after_ the numbers were used in the rule actions. As a result, the evaluation of the skeleton failed because it used non existing symbol numbers. Which is the happy scenario: we could use numbers of other existing symbols... Reported by Balázs Scheidler. http://lists.gnu.org/archive/html/bug-bison/2019-01/msg00044.html Translating the rule actions after the symbol renumbering moves too many parts in bison. Relying on the symbol identifiers is more troublesome than it might first seem: some don't have an identifier (tokens with only a literal string), some might have a complex one (tokens with a literal string with characters special for M4). Well, these are tokens, but nterms also have issues: "dummy" nterms (for midrule actions) are named $@32 etc. which is risky for M4. Instead, let's simply give M4 the mapping between the old numbers and the new ones. To avoid confusion between old and new numbers, always emit pre-renumbering numbers as "orig NUM". * data/README: Give details about "orig NUM". * data/skeletons/bison.m4 (__b4_symbol, _b4_symbol): Resolve the "orig NUM". * src/output.c (prepare_symbol_definitions): Pass nterm_map to m4. * src/reduce.h, src/reduce.c (nterm_map): Extract it from nonterminals_reduce, to make it public. (reduce_free): Free it. * src/scan-code.l (handle_action_dollar): When referring to a nterm, use "orig NUM". * tests/reduce.at (Useless Parts): New, based Balázs Scheidler's report. 2019-02-03 Akim Demaille tests: strengthen some of them * tests/reduce.at: Check that the generated parsers are proper C. 2019-02-03 Akim Demaille package: rename data/README as data/README.md So that it is properly rendered by online git services. 2019-02-03 Akim Demaille style: reduce scopes * src/symlist.c (symbol_list_free): New. 2019-02-03 Akim Demaille style: prefer snprintf to sprintf * src/symtab.c (dummy_symbol_get): There's no need for the buffer to be so big and static. Use snprintf for safety. 2019-02-02 Akim Demaille style: comment and name changes * src/output.c (prepare_symbol_names): here. * src/reader.c: Remove obsolete comment. * src/scan-code.l: Use || for Boolean or. 2019-02-02 Akim Demaille style: comment changes * src/reader.c, src/scan-code.l: here. 2019-02-02 Akim Demaille make: regenerate the example parsers when bison changes * Makefile.am (dependencies): Also depend on Bison's sources. 2019-02-02 Akim Demaille style: rename local variable * src/reduce.c (nonterminals_reduce): Rename nontermmap as nterm_map. We will expose it. 2019-01-30 Akim Demaille gram: detect and report (in debug traces) useless chain rules A rule is a useless chain iff it's a chain (aka unit, or injection) rule (i.e., the RHS has length 1), and it's useless (it has no used defined semantic action). * src/gram.h, src/gram.c (rule_useless_chain_p): New. (grammar_dump): Report useless chain rules. * tests/sets.at: Check the traces. 2019-01-30 Akim Demaille lr(0): more debug traces * src/lr0.c (core_print, kernel_print): New. Use them. 2019-01-30 Akim Demaille lr(0): remove useless conditional * src/lr0.c (new_itemsets): There's no harm in setting a Boolean several times. 2019-01-30 Akim Demaille style: sort includes and avoid assignments * src/symtab.c: Sort includes. * src/gram.c (grammar_rules_print_xml): Avoid assignments to define 'usefulness'. 2019-01-30 Akim Demaille style: use item_rule * src/print-graph.c, src/print-xml.c: here. 2019-01-30 Akim Demaille gram: factor the printing of items and the computation of their rule There are several places where we need to recover the rule from an item, let's factor that into item_rule. We also want to print items in a nice way: we do it when generating the *output file, but it is also useful in debug messages. * src/gram.h, src/gram.c (item_rule, item_print): New. * src/print.c (print_core): Use them. * src/state.h, src/state.c: Propagate constness. 2019-01-30 Akim Demaille style: scope reduction in print-xml * src/print-xml.c: here. 2019-01-30 Akim Demaille tests: check XML and dot reports * tests/report.at: Here. 2019-01-28 Akim Demaille CI: update the ICC serial number for travis-ci.org On travis-ci.org, there are five concurrent slaves, instead of three on travis-ci.com. 2019-01-28 Akim Demaille CI: update the ICC serial number for travis-ci.org On travis-ci.org, there are five concurrent slaves, instead of three on travis-ci.com. 2019-01-28 Akim Demaille style: comment changes * src/lr0.c, src/state.c, src/state.h: here. 2019-01-28 Akim Demaille closure: initialize it once for all The memory allocated by 'closure' (and some data such as 'fderives') is used to computed a state's full itemset from its core. This is needed during the construction of the LR(0) automaton, and the memory is reclaimed immediately afterwards. Unfortunately the reports (graph, text, xml) also need this information when describing the states with their full itemsets. As a consequence the memory was allocated again, fderives computed again too, and more --trace reports are generated which only duplicate what was already reported. Stop that. It does mean that we release the memory later (hence the peak memory usage is higher now), but I don't think that's a problem today. * src/lr0.c (generate_states): Don't call closure_free. * src/state.c (states_free): Do it here. (for symmetry with closure_new which is called in generate_states). * src/print-graph.c, src/print-xml.c, src/print.c: You can now expect the closure module to be functional. 2019-01-28 Akim Demaille style: rename closure_* functions as closure_* This is more consistent with the other files. * closure.h, closure.c (new_closure, free_closure): Rename as... (closure_new, closure_free): this. Adjust dependencies. 2019-01-28 Akim Demaille lr0: use a bitset for the set of "shiftable symbols" This will make it easier to add new elements (that might already be part of shift_symbol) without having to worry about the size of shift_symbol (which is currently a fixed size vector). I could not measure any significant differences in performances in the generation of LR(0) automaton (benched on gramamrs of Ruby, C, and C++). * src/lr0.c (shift_symbol): Make it a bitset. 2019-01-28 Akim Demaille add -fsyntax-only When debugging Bison itself, this is very handy, especially when tweaking the frontend badly enough to break the backends. It can also be used to check a grammar. * src/getargs.h, src/getargs.c (feature_syntax_only): New. (feature_args, feature_types): Adjust. * src/main.c (main): Use it. 2019-01-27 Akim Demaille style: beware of collisions on status * src/symtab.h (status): Rename as... (declaration_status): this, to avoid colliding with status, the argument of 'usage'. 'status' seems a tad too general to be used only here. 2019-01-27 Akim Demaille gnulib: update 2019-01-27 Akim Demaille usage: document -ffixit * src/getargs.c (usage): Document -ffixit. Document the aliases of -f. 2019-01-27 Akim Demaille style: reduce scopes in state.c and ielr.c 2019-01-27 Akim Demaille Merge branch 'maint' * maint: maint: post-release administrivia version 3.3.1 yacc: issue warnings, not errors, for Bison extensions style: formatting changes in NEWS and complain.c tests: don't depend on the user's definition of SHELL 2019-01-27 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2019-01-27 Akim Demaille version 3.3.1 * NEWS: Record release date. 2019-01-27 Akim Demaille yacc: issue warnings, not errors, for Bison extensions Reported by Kiyoshi Kanazawa. http://lists.gnu.org/archive/html/bug-bison/2019-01/msg00029.html * src/getargs.c (getargs): Let --yacc imply -Wyacc, not -Werror=yacc. * tests/input.at: Adjust. * doc/bison.tex (Bison Options): Document. 2019-01-27 Akim Demaille style: formatting changes in NEWS and complain.c 2019-01-27 Kiyoshi Kanazawa tests: don't depend on the user's definition of SHELL http://lists.gnu.org/archive/html/bug-bison/2019-01/msg00031.html * examples/test (SHELL): Set it to /bin/sh. 2019-01-26 Akim Demaille traces: always print the reduced grammar and fix it * src/gram.c (grammar_dump): Print the effective number first instead of last. And fix it (remove the incorrect "+1"). Use t/f for Booleans. * src/reduce.c: When asked, always print the reduced grammar, even if there was nothing useless. * tests/sets.at (Reduced Grammar): Check that. 2019-01-26 Akim Demaille style: rename LR0.* as lr0.* Let's stick to lower case for file names. * src/LR0.h, src/LR0.c: Rename as... * src/lr0.h, src/lr0.c: these. 2019-01-26 Akim Demaille style: rename print_graph.* as print-graph.* These are the only files with _. * src/print_graph.h, src/print_graph.c: Rename as... * src/print-graph.h, src/print-graph.c: these. 2019-01-26 Akim Demaille style: various fixes * src/gram.c: Use consistent variable names. Prefix prefix unary operators. (grammar_dump): Use rule_rhs_length instead of duplicating it. * src/reduce.c: Avoid useless variables. 2019-01-26 Akim Demaille style: comment changes in gram.h * src/gram.h: Shorten comments. 2019-01-26 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2019-01-26 Akim Demaille version 3.3 * NEWS: Record release date. 2019-01-26 Akim Demaille NEWS: update 2019-01-26 Akim Demaille c++: fix comment * data/skeletons/c++.m4: here. 2019-01-26 Akim Demaille style: formatting changes * data/skeletons/lalr1.cc: Add dnl. * data/skeletons/bison.m4: Comment the use of dnl. 2019-01-26 Akim Demaille tests: run the printer/destructor test on glr.cc * tests/actions.at (_AT_CHECK_PRINTER_AND_DESTRUCTOR): Adjust for glr.cc, and use it. 2019-01-24 Akim Demaille gnulib: update 2019-01-22 Akim Demaille --update: when used, do not generate the output files It is inconvenient that we also generate the output files when we update the grammar file, and it's somewhat unexpected. Let's not do that. * src/main.c (main): Skip generation when --update is passed. * src/getargs.c (usage): Update the help message. * doc/bison.texi (Bison Options): Likewise. * tests/input.at: Check that we don't generate the output. 2019-01-22 Akim Demaille diagnostics: let redundant definitions be only warnings After all, this is clearly harmless. * src/muscle-tab.c (muscle_percent_define_insert): Let equal definitions of a %define variable be only a warning. Adjust test cases. 2019-01-22 Akim Demaille tests: improve check for updated variable names * tests/input.at ("%define" backward compatibility): Don't define twice "api.namespace", so that we don't get an error, which stops the process too soon to see an error about the value given to 'lr.keep-unreachable-state'. 2019-01-21 Akim Demaille diagnostics: remove redundancy Don't repeat the name of the warning in the sub messages. E.g., remove the second "[-Wother]" in the following message foo.y:2.1-27: warning: %define variable 'parse.error' redefined [-Wother] %define parse.error verbose ^~~~~~~~~~~~~~~~~~~~~~~~~~~ foo.y:1.1-27: previous definition [-Wother] %define parse.error verbose ^~~~~~~~~~~~~~~~~~~~~~~~~~~ * src/complain.c (error_message): Don't print the warning type when it's indented. Adjust test cases. 2019-01-20 Akim Demaille c++: better "scope" a workaround for GCC * data/skeletons/lalr1.cc: Enable it only for GCC 4.8 and before. 2019-01-20 Akim Demaille c++: address -Wweak-vtables warnings Reported by Derek Clegg http://lists.gnu.org/archive/html/bug-bison/2019-01/msg00021.html aux/parser-internal.h:429:12: error: 'syntax_error' has no out-of-line virtual method definitions; its vtable will be emitted in every translation unit [-Werror,-Wweak-vtables] struct syntax_error : std::runtime_error To avoid this warning, we need syntax_error to have a virtual function defined in a compilation unit. Let it be the destructor. To comply with C++98, this dtor should be 'throw()'. Merely making YY_NOEXCEPT be 'throw()' in C++98 triggers errors (http://lists.gnu.org/archive/html/bug-bison/2019-01/msg00022.html), so let's introduce YY_NOTHROW and flag only ~syntax_error with it. Also, since we now have an explicit dtor, we need to provide an copy ctor. * configure.ac (warn_cxx): Add -Wweak-vtables. * data/skeletons/c++.m4 (YY_NOTHROW): New. (syntax_error): Declare the dtor, and define the copy ctor. * data/skeletons/glr.cc, data/skeletons/lalr1.cc (~syntax_error): Define. 2019-01-20 Akim Demaille style: prefer bool to char * src/state.h, src/state.c (state::consistent): Make it a bool. Adjust dependencies. 2019-01-20 Akim Demaille glr.cc: be more alike lalr1.cc 2019-01-20 Akim Demaille style: formatting changes * data/skeletons/c++.m4: Un-remove an end-of-line. 2019-01-20 Akim Demaille NEWS: fixes 2019-01-19 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2019-01-19 Akim Demaille version 3.2.91 * NEWS: Record release date. 2019-01-18 Akim Demaille style: various fixes Some reported by syntax-check. * po/POTFILES.in: Add fixits.cc. * src/muscle-tab.c: Don't cast for free. * src/files.c: Reduce scopes. * cfg.mk: We need the cast for free in muscle_percent_define_insert. 2019-01-18 Akim Demaille doc: document -ffixit and --update * doc/bison.texi (Bison Options): here. 2019-01-18 Akim Demaille doc: style fixes * doc/bison.texi: Use @kbd where appropriate. Update ^~~~ marks for caret-errors. * build-aux/cross-options.pl: Do not add quotes to %define's argument. 2019-01-18 Akim Demaille NEWS: update for fixits and --update 2019-01-17 Akim Demaille fixits: handle duplicates of %name-prefix The test case "Deprecated directives" (currently 56) no longer emits warnings after 'bison -u'! * src/files.h, src/files.c (spec_name_prefix_loc): New. * src/parse-gram.y (handle_name_prefix): Emit fixits for duplicate %name-prefix. * tests/input.at (Deprecated directives): Adjust. 2019-01-17 Akim Demaille regen 2019-01-17 Akim Demaille fixits: handle %file-prefix * src/files.h, src/files.c (spec_file_prefix_loc): New. * src/scan-gram.l (%file-prefix): Delegate diagnostics to... * src/parse-gram.y (handle_file_prefix): here. * src/complain.c (duplicate_directive): Quote the directive. * tests/input.at: Adjust. 2019-01-17 Akim Demaille fixits: handle per-rule duplicates * src/complain.c (duplicate_rule_directive): Here. * tests/actions.at (Invalid uses of %empty): Check it. 2019-01-17 Akim Demaille fixits: fix warnings about duplicates * src/complain.c (duplicate_directive): Fix the complaint level. * tests/input.at: Adjust. 2019-01-16 Akim Demaille diagnostics: properly indent the "previous declaration" message * src/complain.c (duplicate_directive, duplicate_rule_directive): Here. 2019-01-16 Akim Demaille regen 2019-01-16 Akim Demaille style: rename some functions for consistency "handle_" is the prefix used in scan-code.l for instance. * src/parse-gram.y (do_error_verbose, do_name_prefix, do_require) (do_skeleton, do_yacc): Rename as... (handle_error_verbose, handle_name_prefix, handle_require) (handle_skeleton, handle_yacc): these. 2019-01-16 Akim Demaille fixits: report duplicate %yacc directives We should use -ffixit and --update to clean files with duplicate directives. And we should complain only once about duplicate obsolete directives: keep only the "duplicate" warning. Let's start with %yacc. For instance on: %fixed-output_files %fixed-output-files %yacc %% exp: This run of bison: $ bison /tmp/foo.y -u foo.y:1.1-19: warning: deprecated directive, use '%fixed-output-files' [-Wdeprecated] %fixed-output_files ^~~~~~~~~~~~~~~~~~~ foo.y:2.1-19: warning: duplicate directive [-Wother] %fixed-output-files ^~~~~~~~~~~~~~~~~~~ foo.y:1.1-19: previous declaration %fixed-output_files ^~~~~~~~~~~~~~~~~~~ foo.y:3.1-5: warning: duplicate directive [-Wother] %yacc ^~~~~ foo.y:1.1-19: previous declaration %fixed-output_files ^~~~~~~~~~~~~~~~~~~ bison: file 'foo.y' was updated (backup: 'foo.y~') gives: %fixed-output-files %% exp: * src/location.h, src/location.c (location_empty): New. * src/complain.h, src/complain.c (duplicate_directive): New. * src/getargs.h, src/getargs.c (yacc_flag): Instead of a Boolean, be the location of the definition. Update dependencies. * src/scan-gram.l (%yacc, %fixed-output-files): Move the handling of its warnings to... * src/parse-gram.y (do_yacc): This new function. * tests/input.at (Deprecated Directives): Adjust expectations. 2019-01-16 Akim Demaille style: rename duplicate_directive as duplicate_rule_directive * src/complain.h, src/complain.c: here. Adjust callers. 2019-01-16 Akim Demaille fixits: suggest running --update if there are fixits * src/fixits.h, src/fixits.c (fixits_empty): New. * src/complain.c (deprecated_directive): Register the Wdeprecated fixits only if -Wdeprecated was enabled, so that we don't apply updates if the user didn't ask for them. * src/main.c (main): If there were fixits, issue a warning suggesting running with --update. Free uniqstrs after the fixits, since the latter use the former. * tests/headers.at, tests/input.at: Update expectations. 2019-01-16 Akim Demaille fixits: avoid generating empty lines * src/fixits.c (fixits_run): If erase the content of a line, also erase the following \n. * tests/input.at (Deprecated directives): Update expectations. 2019-01-15 Akim Demaille configure: don't try to run C++ warnings on C Reported by Derek Clegg. https://lists.gnu.org/archive/html/bison-patches/2019-01/msg00066.html * configure.ac: here. 2019-01-15 Akim Demaille c, c++: avoid implicit fall-throw Reported by Derek Clegg. http://lists.gnu.org/archive/html/bug-bison/2019-01/msg00004.html * configure.ac (warn_common): Add -Wimplicit-fallthrough. This does trigger failures in the test suite. * data/skeletons/glr.c, data/skeletons/lalr1.cc, * data/skeletons/yacc.c, tests/c++.at: Make fall-throws explicit. 2019-01-15 Akim Demaille c++: avoid -Wundefined-func-template warnings from clang Reported by Derek Clegg. http://lists.gnu.org/archive/html/bug-bison/2019-01/msg00006.html Clang does not like this: template struct basic_symbol : D { basic_symbol(); }; struct by_type {}; struct symbol_type : basic_symbol { symbol_type(){} }; It gives: $ clang++-mp-7.0 -Wundefined-func-template foo.cc -c foo.cc:11:3: warning: instantiation of function 'basic_symbol::basic_symbol' required here, but no definition is available [-Wundefined-func-template] symbol_type(){} ^ foo.cc:4:3: note: forward declaration of template entity is here basic_symbol(); ^ foo.cc:11:3: note: add an explicit instantiation declaration to suppress this warning if 'basic_symbol::basic_symbol' is explicitly instantiated in another translation unit symbol_type(){} ^ 1 warning generated. The same applies for the basic_symbol's destructor and `clear()`. * configure.ac (warn_cxx): Add -Wundefined-func-template. This triggered one failure in the test suite: * tests/headers.at (Sane headers): here, where we check that we can compile the generated headers in other compilation units than the parser's. Add a variant type to make sure that basic_symbol and symbol_type are properly generated in this case. * data/skeletons/c++.m4 (basic_symbol): Inline the definitions of the destructor and of `clear` in the class definition. 2019-01-14 Akim Demaille Revert the last two commits They should not have been pushed, sorry about that. This reverts - commit 8575bd06ae6e65f3a30b21a3e022a968e4c7ae7a. - commit 55bf52860eac5c1394dc344a691220272df32b09. 2019-01-14 Akim Demaille c++: avoid warnings about extraneous semi-colons Reported by Derek Clegg. * configure.ac (warn_common): Add -Wextra-semi. * data/skeletons/c++.m4: Remove extraneous semi-colon. 2019-01-14 Akim Demaille WIP 2019-01-14 Akim Demaille diagnostics: add fixit support for duplicate removal * src/muscle-tab.c (muscle_percent_define_insert): Register a fixit for duplicate removal. * tests/input.at: Adjust expectations. 2019-01-14 Akim Demaille regen 2019-01-14 Akim Demaille diagnostics: improve the accuracy for %error-verbose Avoid duplicate warnings about %error-verbose, once for deprecation, another for duplicate. Keep only the duplicate warning for the second occurrence of %error-verbose. This will help removal fixits. * src/scan-gram.l (%error-verbose): Return as a PERCENT_ERROR_VERBOSE token. * src/parse-gram.y (do_error_verbose): New. Use it. * src/muscle-tab.c (muscle_percent_variable_update): Handle pseudo variables such as %error-verbose. 2019-01-14 Akim Demaille diagnostics: avoid duplicate warnings for deprecated directives Currently, on %define parser_class_name "Parser" %define parser_class_name "Parser" %% exp:; we issue: foo.y:1.9-25: warning: deprecated directive, use '%define api.parser.class {Parser}' [-Wdeprecated] %define parser_class_name "Parser" ^~~~~~~~~~~~~~~~~ foo.y:2.9-25: warning: deprecated directive, use '%define api.parser.class {Parser}' [-Wdeprecated] %define parser_class_name "Parser" ^~~~~~~~~~~~~~~~~ foo.y:2.9-25: error: %define variable 'api.parser.class' redefined %define parser_class_name "Parser" ^~~~~~~~~~~~~~~~~ foo.y:1.9-25: previous definition %define parser_class_name "Parser" ^~~~~~~~~~~~~~~~~ Let's get rid of the second warning about the deprecated variable parser_class_name. This is noise, but it will also be a problem with fixits for removing duplicates, as we will first generate the update, and then it's too late to remove it: fixits do not edit the result of previous fixits. So generate this instead: foo.y:1.1-34: warning: deprecated directive, use '%define api.parser.class {Parser}' [-Wdeprecated] %define parser_class_name "Parser" ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ foo.y:2.1-34: error: %define variable 'api.parser.class' redefined %define parser_class_name "Parser" ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ foo.y:1.1-34: previous definition %define parser_class_name "Parser" ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ * src/muscle-tab.c (muscle_percent_variable_update): Pass the warning to the caller, instead of issuing it. (muscle_percent_define_insert): Issue this warning only if we don't have to complain about a duplicate definition. * tests/input.at: Adjust expectations. 2019-01-14 Akim Demaille diagnostics: update the grammar file Let's use the fixits to actually update the grammar files. * src/getargs.h, src/getargs.c (update_flag): New. * src/fixits.h, src/fixits.c (fixits_run): New. * src/main.c (main): Invoke fixits_run when --update is passed. * tests/input.at (Deprecated directives): Check --update. 2019-01-14 Akim Demaille diagnostics: improve accuracy for deprecated %define variables * src/parse-gram.y: Use the location of the whole definition to record the location of a %define variable, instead of just the name of the variable. Adjust tests. 2019-01-14 Akim Demaille diagnostics: keep the fixits Introduce proper support for fixits, instead of just printing them on demand. * bootstrap.conf: We need gnulib's xlists. * src/fixits.h, src/fixits.c: New. * src/complain.c (deprecated_directive): Use fixits_register. * src/main.c (main): Use fixits_free. 2019-01-14 Akim Demaille diagnostics: add -ffixit support for deprecated features Issue directives for IDE/editors to fix the source file. http://clang.llvm.org/docs/UsersManual.html#cmdoption-fdiagnostics-parseable-fixits Do it for deprecated features. For instance: $ cat foo.y %error-verbose %name-prefix = "foo" %name-prefix="bar" %define parser_class_name "Parser" %% exp:; $ LC_ALL=C ./_build/8d/tests/bison -ffixit /tmp/foo.yy /tmp/foo.yy:1.1-14: warning: deprecated directive, use '%define parse.error verbose' [-Wdeprecated] %error-verbose ^^^^^^^^^^^^^^ fix-it:"/tmp/foo.yy":{1:1-1:15}:"%define parse.error verbose" /tmp/foo.yy:3.1-20: warning: deprecated directive, use '%define api.prefix {foo}' [-Wdeprecated] %name-prefix = "foo" ^^^^^^^^^^^^^^^^^^^^ fix-it:"/tmp/foo.yy":{3:1-3:21}:"%define api.prefix {foo}" /tmp/foo.yy:4.1-18: warning: deprecated directive, use '%define api.prefix {bar}' [-Wdeprecated] %name-prefix="bar" ^^^^^^^^^^^^^^^^^^ fix-it:"/tmp/foo.yy":{4:1-4:19}:"%define api.prefix {bar}" /tmp/foo.yy:5.9-25: warning: deprecated directive, use '%define api.parser.class {Parser}' [-Wdeprecated] %define parser_class_name "Parser" ^^^^^^^^^^^^^^^^^ fix-it:"/tmp/foo.yy":{5:9-5:26}:"%define api.parser.class {Parser}" /tmp/foo.yy:5.9-25: error: %define variable 'api.parser.class' is not used %define parser_class_name "Parser" ^^^^^^^^^^^^^^^^^ * src/getargs.h, src/getargs.c (feature_fixit_parsable): New. (feature_types, feature_args): Use it. * src/complain.c (deprecated_directive): Use it. * tests/input.at: Check it. 2019-01-14 Akim Demaille diagnostics: prefer ^~~~ to ^^^^ to underline code That's what both GCC and Clang do, and it is indeed much nicer to read. From: foo.y:1.1-14: warning: deprecated directive, use '%define parse.error verbose' [-Wdeprecated] %error-verbose ^^^^^^^^^^^^^^ foo.y:4.1-20: warning: deprecated directive, use '%define api.prefix {foo}' [-Wdeprecated] %name-prefix = "foo" ^^^^^^^^^^^^^^^^^^^^ to: foo.y:1.1-14: warning: deprecated directive, use '%define parse.error verbose' [-Wdeprecated] %error-verbose ^~~~~~~~~~~~~~ foo.y:4.1-20: warning: deprecated directive, use '%define api.prefix {foo}' [-Wdeprecated] %name-prefix = "foo" ^~~~~~~~~~~~~~~~~~~~ * src/location.c (location_caret): Use ^~~~. Adjust tests expectations. 2019-01-14 Akim Demaille regen 2019-01-14 Akim Demaille diagnostics: improve them for %name-prefix Currently the diagnostics for %name-prefix are not precise enough. In particular, they does not show that braces must be used instead of quotes. Before: foo.y:3.1-14: warning: deprecated directive, use '%define api.prefix' [-Wdeprecated] %name-prefix = "foo" ^^^^^^^^^^^^^^ After: foo.y:3.1-20: warning: deprecated directive, use '%define api.prefix {foo}' [-Wdeprecated] %name-prefix = "foo" ^^^^^^^^^^^^^^^^^^^^ To do this we need the value passed to %name-prefix, so move the warning from the scanner to the parser. Accuracy will be very important for the forthcoming changes. * src/parse-gram.y (do_name_prefix): New. (PERCENT_NAME_PREFIX): Have a semantic value: the raw source, with possibly underscores, equal sign, and spaces. This is used to provide a more accurate message. It does not take comments into account, but... * src/scan-gram.l (%name-prefix): Delegate the warnings to the parser. * tests/headers.at, tests/input.at: Adjust expectations. 2019-01-14 Akim Demaille diagnostics: style: avoid allocating memory when not needed * src/muscle-tab.c (muscle_percent_variable_update): Avoid allocating memory when it is not needed, which should be most of the time (when there's no update to perform). Adjust callers. 2019-01-14 Akim Demaille c++: avoid warnings about extraneous semi-colons Reported by Derek Clegg. http://lists.gnu.org/archive/html/bug-bison/2019-01/msg00005.html * configure.ac (warn_cxx): Add -Wextra-semi. * data/skeletons/c++.m4: Remove extraneous semi-colon. 2019-01-13 Akim Demaille style: minor changes * src/muscle-tab.c: Sort alphabetically. * src/scan-gram.l: Reduce scopes. Initialize variables. 2019-01-13 Akim Demaille c++: beware of -Wshadow This line: slice slice (yystack_, yylen); triggers warnings: parse.h:1790:11: note: shadowed declaration is here Reported by Frank Heckenbach. http://lists.gnu.org/archive/html/bug-bison/2019-01/msg00002.html * configure.ac (warn_c): Move -Wshadow to... (warn_common): here. * data/skeletons/stack.hh (slice): Define as an inner class of stack. * data/skeletons/lalr1.cc: Adjust. Rename the variable as 'range' instead of 'slice'. 2019-01-12 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2019-01-12 Akim Demaille version 3.2.90 * NEWS: Record release date. 2019-01-12 Akim Demaille yacc: fix relocatability * src/yacc.in (prefix): Define it, as it's typically needed for exec_prefix. 2019-01-12 Akim Demaille syntax-check: adjust paths * cfg.mk: here. (require_config_h): New. 2019-01-12 Akim Demaille style: formatting clean up * data/skeletons/d.m4, examples/d/calc.y, src/output.c, * src/parse-gram.y: No tab, no trailing spaces. Reported by syntax-check. 2019-01-12 Akim Demaille po: remove bitset/stats.c * po/POTFILES.in: here. 2019-01-12 Akim Demaille tests: fix usage of AT_PARSER_CHECK The parser must be the first command. Caught by syntax-check. * tests/c++.at: here. 2019-01-12 Akim Demaille gnulib: update 2019-01-12 Akim Demaille NEWS: update 2019-01-12 Akim Demaille style: isolate the creation of tname in a function * src/output.c (prepare_symbol_names): New. Use it. 2019-01-12 Akim Demaille tests: formatting changes * tests/input.at, tests/regression.at: here. 2019-01-05 Akim Demaille doc: use @option consistently * doc/bison.texi: Use @option, not @code, for options. 2019-01-05 Akim Demaille yacc.c: avoid negated if * data/skeletons/yacc.c: Prefer a "direct" conditional. 2019-01-05 Akim Demaille package: bump copyrights to 2019 2019-01-05 Akim Demaille java/d: rename some %define variables for consistency See 890ee8a1fd288b3cc1c21c49ea0ece696ef40565 and https://lists.gnu.org/archive/html/bison-patches/2019-01/msg00024.html. * data/skeletons/d.m4, data/skeletons/java.m4 (abstract, annotations, extends, final, implements, public, strictfp): Rename as... (api.parser.abstract, api.parser.annotations, api.parser.extends) (api.parser.final, api.parser.implements, api.parser.public) (api.parser.strictfp): these. * src/muscle-tab.c (muscle_percent_variable_update): Ensure backward compatibility. * doc/bison.texi, examples/d/calc.y, examples/java/Calc.y, tests/input.at: Adjust. 2019-01-05 Akim Demaille java/d: remove useless macros There are many macros that are defined and used just once (b4_public_if, b4_abstract_if, etc.). That's overkill. Rather, let's define a macro to build the "public class YYParser" line. It appears that the same syntax with "extends", "abstract", etc. is implemented in the D parser, which looks very fishy... * data/skeletons/d.m4, data/skeletons/java.m4 (b4_public_if) (b4_abstract_if, b4_final_if, b4_strictfp_if): Replace with (b4_parser_class_declaration): this. * data/skeletons/lalr1.d, data/skeletons/lalr1.java: Adjust. 2019-01-05 Akim Demaille style: clean tests * tests/named-refs.at: here. 2019-01-05 Akim Demaille gnulib: update In particular, report uninitialized submodules instead of breaking symlinks. For reports, see https://lists.gnu.org/archive/html/bug-bison/2011-05/msg00012.html https://lists.gnu.org/archive/html/help-bison/2018-12/msg00034.html For a fix, see https://lists.gnu.org/archive/html/bug-gnulib/2019-01/msg00024.html 2019-01-05 Akim Demaille glr.cc: fix the handling of syntax_error from the scanner Commit 90a8537e6287f92fb3d5be0258a69247a742f12e was right, but issued two error messages. Commit 80ef7e7639f99618bee490b2dea02b5fd9ab28e5 tried to address that by mapping yychar and yytoken to empty, but that completely breaks the invariants of glr.c. In particular, yygetToken can be called repeatedly and is expected to return the latest result, unless yytoken is YYEMPTY. Since the previous attempt was "recording" that the token was coming from an exception by setting it to YYEMPTY, instead of getting again the faulty token, we fetched another one. Rather, revert to the first approach: map yytoken to "invalid token", but record in yychar the fact that we come from an exception thrown in the scanner. * data/skeletons/glr.c (YYFAULTYTOK): New. (yygetToken): Use it to record syntax errors from the scanner. * tests/c++.at (Syntax error as exception): In addition to checking syntax_error with error recovery, make sure it also behaves as expected without. 2019-01-03 Akim Demaille clearly deprecate %name-prefix * src/scan-gram.l (%name-prefix): Issue a deprecation warning. * tests/calc.at, tests/headers.at, tests/input.at, tests/java.at, * tests/javapush.at, tests/local.at: Adjust expectations. Or disable -Wdeprecated. * doc/bison.texi: Document that %name-prefix is replaced by %define api.prefix. 2019-01-03 Akim Demaille examples: clean up the Java/D examples * examples/java/Calc.y: Fix indentation. Sort. Don't use %name-prefix, since api.parser.class is already defined. * examples/d/calc.y: Likewise. 2019-01-03 Akim Demaille doc: formatting changes * doc/bison.texi: here. 2019-01-02 Akim Demaille rename parser_class_name as api.parser.class The previous name was historical and inconsistent. * src/muscle-tab.c (define_directive): Use the proper value passing syntax, based on the muscle kind. (muscle_percent_variable_update): Use the right value passing syntax. Migrate from parser_class_name to api.parser.class. * data/skeletons: Migrate from parser_class_name to api.parser.class. * doc/bison.texi (%define Summary): Document both parser_class_name and api.parser.class. Promote the latter over the former. 2019-01-02 Akim Demaille style: formatting changes * src/scan-gram.l: Here. 2019-01-02 Akim Demaille style: glr.c: prefer returning a value rather than passing pointers This is very debatable. This function is not pure at all, so it could stick to returning void: that's a common coding style to tell the difference between "real" (pure) functions and side-effecting subroutines. However, we already have this style elsewhere (e.g., yylex), and I feel the callers are somewhat nice to read this way. * data/skeletons/glr.c (yygetLRActions): Return the action rather than passing by pointer. While at it, fix type of yytoken. Adjust callers. 2019-01-02 Akim Demaille glr.cc: don't issue two error messages when syntax_error is thrown Reported by Askar Safin. https://lists.gnu.org/archive/html/bison-patches/2019-01/msg00000.html * data/skeletons/glr.c (yygetToken): Return YYEMPTY when an exception is thrown. * data/skeletons/lalr1.cc: Log when an exception is caught. * tests/c++.at (Syntax error as exception): Be sure to recover from error before triggering another error. 2019-01-02 Akim Demaille skeletons: shorten b4_parser_class_name to b4_parser_class * skeletons/c++.m4, skeletons/d.m4, skeletons/glr.c, skeletons/glr.cc, * skeletons/java.m4, skeletons/lalr1.cc, skeletons/lalr1.d, * skeletons/lalr1.java: Here. 2019-01-02 Akim Demaille glr.cc: remove duplicate definition of YYLLOC_DEFAULT It's already provided by glr.c. * data/skeletons/glr.cc (b4_post_prologue): Here. 2019-01-02 Akim Demaille style: sort includes in scanners * src/scan-code.l, src/scan-gram.l, src/scan-skel.l: Reorder includes. 2019-01-02 Akim Demaille style: formatting changes in the doc * doc/bison.texi: here. 2019-01-02 Akim Demaille style: remove stray empty lines * data/skeletons/glr.c, data/skeletons/glr.cc: here. * data/skeletons/bison.m4 (b4_glr_cc_if): Move it here. 2018-12-31 Akim Demaille glr.cc: support syntax_error exceptions Kindly requested by Аскар Сафин (Askar Safin). http://lists.gnu.org/archive/html/bug-bison/2018-12/msg00033.html * data/skeletons/glr.c (b4_glr_cc_if): New. Use it. (yygetToken): Catch syntax_errors. * data/skeletons/glr.cc (YY_EXCEPTIONS): New. * tests/c++.at: Check it. 2018-12-31 Akim Demaille glr.c: factor the calls to yylex The call protocol of yylex is quite complex, and repeated three times. Let's factor it. * data/skeletons/glr.c (yygetToken): New. Use it. 2018-12-31 Akim Demaille style: reduce scopes in glr.c * data/skeletons/glr.c (yyrecoverSyntaxError): here. 2018-12-30 Akim Demaille tests: refactor 2018-12-30 Akim Demaille c++: inline the implementation of syntax_error in its definition This way, it is easier to make sure its implementation is available in glr.cc too, which is not the case currently. * data/skeletons/c++.m4 (b4_public_types_define): Move the implementation of syntax_error... (b4_public_types_declare): here. 2018-12-29 Akim Demaille style: formatting changes * tests/input.at: here. 2018-12-29 Akim Demaille symbol: don't crash on symbol without content When running with --trace=parse, we may crash. * src/symtab.c (symbol_print): Avoid that. 2018-12-28 Akim Demaille README-hacking: update 2018-12-28 Akim Demaille reader: get rid of a useless function Useless since 58d7a1a1c7497ba51a35fcf991c5b047f692fe18 (2006). * src/parse-gram.y, src/reader.h (token_name): Remove, unused. 2018-12-27 Akim Demaille parsers: fix minor stylistic issues * data/skeletons/variant.hh (b4_token_constructor_declare): Remove, unused since the previous commit. Fix indentation issues. * data/skeletons/c++.m4: Fix indentation issues. 2018-12-27 Akim Demaille c++: check several parsers in the same program * tests/local.at (AT_LOCATION_TYPE_IF): Turn into... (AT_LOCATION_TYPE_SPAN_IF): this. Adjust dependencies. * tests/headers.at (Several parsers): Add another C++ parser, which uses the first C++ parser's locations. 2018-12-26 Akim Demaille c++: variants: fuse declarations and definitions We used to create a short definition of yy::parser with all the implementations of its member functions outside. But yy::parser is no longer short and simple to read. Maintaining each function twice is painful: a lot of redundancy but different indentation levels, output which depends on whether we are in a header or not (see d132c2d5455135f63a7497c38358eadd6e3a6af8), etc. Let's simplify this and put the implementations into the class definition itself. Discussed in this monologue: https://lists.gnu.org/archive/html/bison-patches/2018-12/msg00058.html. * data/skeletons/c++.m4, data/skeletons/lalr1.cc, * data/skeletons/variant.hh (b4_basic_symbol_constructor_define) (_b4_token_constructor_declare, b4_token_constructor_declare) Merge into... (b4_basic_symbol_constructor_define, _b4_token_constructor_define) (b4_token_constructor_define): these. 2018-12-26 Akim Demaille examples: fix dependencies Commit 112ccb5ed73ba5c64b0b5300d8b9b686f02f094c moved the skeletons from dist_pkgdata_DATA to dist_skeletons_DATA, hence broke the dependencies. * Makefile.am (dependencies): New. Use it where appropriate. 2018-12-26 Akim Demaille c++: move stack inside yy::parser We used to define such auxiliary structures outside the class, mainly as a matter of style to keep the definition of yy::parser short and simple. However, now there's a lot more code generated inside the class definition (e.g., all the token constructors), so the readability no longer applies. However, if we move stack (and slice) inside yy::parser, then it should no longer be needed to change the namespace to have multiple parsers: changing the class name should suffice. One common argument against inner classes is that they code bloat. It hardly applies here, since typically different parsers will have different semantic value types, hence different actual stack types. * data/skeletons/lalr1.cc: Invoke b4_stack_define inside yy::parser. 2018-12-25 Akim Demaille package: make bison a relocatable package Suggested by David Barto https://lists.gnu.org/archive/html/help-bison/2015-02/msg00004.html and Victor Zverovich. https://lists.gnu.org/archive/html/bison-patches/2018-10/msg00121.html This is very easy to do, thanks to work by Bruno Haible in gnulib. See "Supporting Relocation" in gnulib's documentation. * bootstrap.conf: We need relocatable-prog and relocatable-script (for yacc). * src/yacc.in: New. * configure.ac, src/local.mk: Instantiate it. * src/main.c, src/output.c (main, pkgdatadir): Use relocatable2. * doc/bison.texi (FAQ): Document it. 2018-12-25 Akim Demaille README: wrap paragraphs 2018-12-25 Akim Demaille gnulib: update 2018-12-25 Akim Demaille package: move skeletons into data/skeletons * data/bison.m4, data/c++-skel.m4, data/c++.m4, data/c-like.m4, * data/c-skel.m4, data/c.m4, data/d-skel.m4, data/d.m4, data/glr.c, * data/glr.cc, data/java-skel.m4, data/java.m4, data/lalr1.cc, * data/lalr1.d, data/lalr1.java, data/location.cc, data/stack.hh, * data/variant.hh, data/yacc.c: Move to... * data/skeletons: here. Use b4_skeletonsdir instead of b4_pkgdatadir. * data/local.mk, src/output.c: Adjust. 2018-12-24 Akim Demaille c++: style: use consistently this/that instead of this/other * data/lalr1.cc, data/variant.hh: here. 2018-12-24 Akim Demaille c++: also provide a copy constructor for symbol_type Suggested by Wolfgang Thaller. http://lists.gnu.org/archive/html/bug-bison/2018-12/msg00081.html * data/c++.m4 (basic_symbol, by_type): Instead of provide either move or copy constructor, always provide the copy one. * tests/c++.at (C++ Variant-based Symbols Unit Tests): Check it. 2018-12-24 Akim Demaille c++: fix double free when a symbol_type was moved Currently the following piece of code crashes (with parse.assert), because we don't record that s was moved-from, and we invoke its dtor. { auto s = parser::make_INT (42); auto s2 = std::move (s); } Reported by Wolfgang Thaller. http://lists.gnu.org/archive/html/bug-bison/2018-12/msg00077.html * data/c++.m4 (by_type): Provide a move-ctor. (basic_symbol): Be sure not to read a moved-from value. * tests/c++.at (C++ Variant-based Symbols Unit Tests): Check this case. 2018-12-24 Akim Demaille c++: style: improve tests * tests/c++.at (C++ Variant-based Symbols Unit Tests): Provide better assertions. Use them. Avoid useless Bison invocations. 2018-12-24 Akim Demaille c++: style: use consistently this/that instead of this/other * data/c++.m4: here. 2018-12-23 Akim Demaille tests: fixes * tests/c++.at: Use AT_YYLEX_PROTOTYPE etc. Which requires that we use the same argument names (lvalp, etc.). * tests/local.at (AT_NAME_PREFIX): Fix regex. 2018-12-22 Akim Demaille c++: style: rename a few macros for clarity * data/c++.m4, data/lalr1.cc, data/variant.hh: s/b4_symbol_constructor/b4_token_constructor/g, as this is really what is being defined. 2018-12-22 Akim Demaille c++: exhibit a safe symbol_type Instead of introducing make_symbol (whose name, btw, somewhat infringes on the user's "name space", if she defines a token named "symbol"), let's make the construction of symbol_type safer, using assertions. For instance with: %token ':' ID INT; generate: symbol_type (int token, const std::string&); symbol_type (int token, const int&); symbol_type (int token); It does mean that now named token constructors (make_ID, make_INT, etc.) go through a useless assert, but I think we can ignore this: I assume any decent compiler will inline the symbol_type ctor inside the make_TOKEN functions, which will show that the assert is trivially verified, hence I expect no code will be emitted for it. And anyway, that's an assert, NDEBUG controls it. * data/c++.m4 (symbol_type): Turn into a subclass of basic_symbol. Declare symbol constructors when variants are enabled. * data/variant.hh (_b4_type_constructor_declare) (_b4_type_constructor_define): Replace with... (_b4_symbol_constructor_declare, _b4_symbol_constructor_def): these. Generate symbol_type constructors. * doc/bison.texi (Complete Symbols): Document. * tests/types.at: Check. 2018-12-22 Akim Demaille c++: provide symbol constructors per type On %token FOO BAR we currently generate make_FOO(int) and make_BAR(int). However, in order to factor their scanners, some users would also like to have make_symbol(tok, int), where tok is FOO or BAR. To ensure type safety, add assertions that do check that value type and token type match. Bind this assertion to the parse.assert %define variable. Suggested by Frank Heckenbach. http://lists.gnu.org/archive/html/bug-bison/2018-12/msg00034.html Should also match expectations from Аскар Сафин. http://lists.gnu.org/archive/html/bug-bison/2018-12/msg00023.html * data/variant.hh: Use b4_token_visible_if where applicable. (_b4_type_constructor_declare, _b4_type_constructor_define): New. Use them. 2018-12-22 Akim Demaille c++: style changes * data/c++.m4, data/variant.hh: Improve layout of the generated code. Avoid casts. (_b4_symbol_constructor_declare, _b4_symbol_constructor_define): Rename as... (_b4_token_maker_declare, _b4_token_maker_define): these. * tests/types.at: Improve pair printing. 2018-12-22 Akim Demaille style: simplify tests * tests/types.at: Simplify C++ instantiations. 2018-12-19 Akim Demaille style: use b4_token_visible_if And other formatting/comment changes. * data/variant.hh: Here. 2018-12-19 Akim Demaille NEWS: update 2018-12-19 Akim Demaille c++: fix token constructors for types with commas Bitten by macros, again. See 680b715518795c8648360fcf05f3772c04d2eed2. * data/variant.hh (_b4_symbol_constructor_declare) (_b4_symbol_constructor_define): Do not use user types, which can include commas as in `std::pair`, to macros. * tests/local.at: Adjust the lex related macros to support the case of token constructors. * tests/types.at: Also check token constructors on types with commas. 2018-12-19 Akim Demaille gnulib: update 2018-12-17 Akim Demaille symbols: document the overhaul of symbol declarations * doc/bison.texi (Symbol Decls): New. 2018-12-16 Akim Demaille symbols: check more invalid declarations * tests/input.at (Invalid %nterm uses): Rename as... (Invalid symbol declarations): this. Extend. 2018-12-16 Akim Demaille symbols: check the previous commit * tests/input.at (Symbol declarations): New. 2018-12-16 Akim Demaille regen 2018-12-16 Akim Demaille symbols: clean up their parsing Prompted by Rici Lake. http://lists.gnu.org/archive/html/bug-bison/2018-10/msg00000.html We have four classes of directives that declare symbols: %nterm, %type, %token, and the family of %left etc. Currently not all of them support the possibility to have several type tags (``), and not all of them support the fact of not having any type tag at all (%type). Let's unify this. - %type POSIX Yacc specifies that %type is for nonterminals only. However, some Bison users want to use it for both tokens and nterms (actually, Bison's own grammar does this in several places, e.g., CHAR). So it should accept char/string literals. As a consequence cannot be used to declare tokens with their alias: `%type foo "foo"` would be ambiguous (are we defining foo = "foo", or are these two different symbols?) POSIX specifies that it is OK to use %type without a type tag. I'm not sure what it means, but we support it. - %token Accept token declarations with number and string literal: (ID|CHAR) NUM? STRING?. - %left, etc. They cannot be the same as %token, because we accept to declare the symbol with %token, and to then qualify its precedence with %left. Then `%left foo "foo"` would also be ambiguous: foo="foo", or two symbols. They cannot be simply a list of identifiers, but POSIX Yacc says we can declare token numbers here. I personally think this is a bad idea, precedence management is tricky in itself and should not be cluttered with token declaration issues. We used to accept declaring a token number on a string literal here (e.g., `%left "token" 1`). This is abnormal. Either the feature is useful, and then it should be supported in %token, or it's useless and we should not support it in corner cases. - %nterm Obviously cannot accept tokens, nor char/string literals. Does not exist in POSIX Yacc, but since %type also works for terminals, it is a nice option to have. * src/parse-gram.y: Avoid relying on side effects. For instance, get rid of current_type, rather, build the list of symbols and iterate over it to assign the type. It's not always possible/convenient. For instance, we still use current_class. Prefer "decl" to "def", since in the rest of the implementation we actually "declare" symbols, we don't "define" them. (token_decls, token_decls_for_prec, symbol_decls, nterm_decls): New. Use them for %token, %left, %type and %nterm. * src/symlist.h, src/symlist.c (symbol_list_type_set): New. * tests/regression.at b/tests/regression.at (Token number in precedence declaration): We no longer accept to give a number to string literals. 2018-12-15 Akim Demaille symbols: set tag_seen when assigning a type to symbols * src/reader.h, src/reader.c (tag_seen): Move to... * src/symtab.h, src/symtab.c: here. (symbol_type_set): Set it to true. * src/parse-gram.y: Don't. 2018-12-14 Akim Demaille tests: isolate test about Yacc warnings * tests/input.at (Yacc warnings): New. (AT_CHECK_UNUSED_VALUES): Remove checks about yacc. 2018-12-14 Akim Demaille parser: warn about string literals in Yacc mode * src/scan-gram.l (scan_integer): Warn. * tests/input.at (Yacc warnings on symbols): Check. 2018-12-14 Akim Demaille parser: warn about hexadecimal token numbers in Yacc mode * src/scan-gram.l (scan_integer): Warn. * tests/input.at (Yacc warnings on symbols): Check. 2018-12-14 Akim Demaille parser: reprecate %nterm back After having spent quite some time on cleaning the handling of symbol declarations in the grammar files, I believe we should keep it. It looks like it's a duplicate of %type, but it is not. While POSIX Yacc requires %type to apply only to nonterminal symbols, it appears that both byacc and bison accept it for tokens too. And some experienced users do actually expect this feature to group symbols (terminal or not) by type ("On the other hand, it is generally more useful IMHO to group terminals and non-terminals with the same type tag together", http://lists.gnu.org/archive/html/bug-bison/2018-10/msg00000.html). Even Bison's own parser does this today (see CHAR). Basically reverts 7928c3e6fbdf47ff81184966cee937e6aa694b94. * src/scan-gram.l (%nterm): Dedeprecate, but issue a Wyacc warning. * tests/input.at: Adjust expectations. (Yacc warnings on symbols): New. * src/symtab.c (symbol_class_set): Fix error introduced in 20b07467938cf880a1d30eb30d6e191843a21fec. 2018-12-11 Eduard Staniloiu CI: add dmd support * .travis.yml: here. 2018-12-11 Akim Demaille style: s/non-terminal/nonterminal/ I personally prefer 'non terminal', or 'non-terminal', but 'nonterminal' is the common spelling. * data/glr.c, src/parse-gram.y, src/symtab.c, src/symtab.h, * tests/input.at, doc/refcard.tex: here. 2018-12-11 Akim Demaille style: rename error functions for clarity * src/symtab.c (symbol_redeclaration, semantic_type_redeclaration) (user_token_number_redeclaration): Rename as... (complain_symbol_redeclared, complain_semantic_type_redeclared) (complain_user_token_number_redeclared): this. 2018-12-11 Akim Demaille parser: improve the error message for symbol class redefinition Currently our error messages include both "symbol redeclared" and "symbol redefined", and they mean something different. This is obscure, let's make this clearer. I think the idea between 'definition' vs. 'declaration' is that in the case of the nonterminals, the actual definition is its set of rules, so %nterm would be about declaration. The case of %token is less clear. * src/symtab.c (complain_class_redefined): New. (symbol_class_set): Use it. Simplify the logic of this function to clearly skip its body when the preconditions are not met. * tests/input.at (Symbol class redefinition): New. 2018-12-11 Akim Demaille examples: simplify computation of yydebug * examples/c/lexcalc/parse.y: here. 2018-12-10 Akim Demaille C++: support variadic emplace Suggested by Askar Safin. http://lists.gnu.org/archive/html/bug-bison/2018-12/msg00006.html * data/variant.hh: Implement. * tests/types.at: Check. * doc/bison.texi: Document. 2018-12-09 Akim Demaille examples: add a simple Flex+Bison example in C Suggested by Askar Safin. http://lists.gnu.org/archive/html/bug-bison/2018-12/msg00003.html * examples/c/lexcalc/Makefile, examples/c/lexcalc/README.md, * examples/c/lexcalc/lexcalc.test, examples/c/lexcalc/local.mk, * examples/c/lexcalc/parse.y, examples/c/lexcalc/scan.l: New. 2018-12-09 Akim Demaille regen 2018-12-09 Akim Demaille examples: sort them per language and complete them Convert some of the READMEs to Markdown, which is now more common, and nicely displayed in some git hosting services. Add missing READMEs and Makefiles. Generate XML, HTML and Dot files. Be sure to ship the test files. Complete CLEANFILES to remove all generated files. * examples/calc++: Move into... * examples/c++: here. * examples/mfcalc, examples/rpcalc: Move into... * examples/c: here. * examples/README.md, examples/c++/calc++/Makefile, examples/c/local.mk, * examples/c/mfcalc/Makefile, examples/c/rpcalc/Makefile, * examples/d/README.md, examples/java/README.md: New files. * examples/test (medir): Be robust to deeper directory nesting. 2018-12-09 Akim Demaille regen 2018-12-09 Akim Demaille parser: minor refactoring * src/parse-gram.y (symbol.prec): Reuse int.opt. 2018-12-09 Akim Demaille parser: move checks inside the called functions Revamping the handling of the symbols is the grammar is much more delicate than I anticipated. Let's first move things around for clarity. * src/symtab.c (symbol_make_alias): Don't accept to alias non-terminals. (symbol_user_token_number_set): Don't accept user token numbers for non-terminals. Don't do anything in case of redefinition, instead of trying to update. The flow is eaier to follow this way. 2018-12-08 Akim Demaille d: fix double definition of YYSemanticType * data/lalr1.d: When moving to b4_user_union_members, it also defines b4_tag_seen_flag, so we had two definitions. 2018-12-08 Akim Demaille gnulib: update 2018-12-06 Akim Demaille parser: fix incorrect condition to raise a syntax error * src/parse-gram.y (symbol_def): Fix test. 2018-12-06 Akim Demaille d: fix use of b4_union_members * data/lalr1.d: Use b4_user_union_members instead. 2018-12-06 Akim Demaille style: comment changes * data/variant.hh: here. 2018-12-06 Akim Demaille java, d: add a Makefile for the example * examples/java/Makefile, examples/d/Makefile: New. 2018-12-05 Akim Demaille style: scope reduction in ielr.c * src/ielr.c: here. 2018-12-05 Akim Demaille style: scope reduction in lalr.c * src/lalr.c: here. 2018-12-05 Akim Demaille d, java: compute static subtractions * data/d.m4, data/java.m4: Use b4_subtract where appropriate. 2018-12-04 Akim Demaille d: add an example * examples/d/calc.test, examples/d/calc.y, examples/d/local.mk: 2018-12-04 Akim Demaille d: update the skeleton * data/d.m4, data/lalr1.d: Catch up with Bison. And actually, also catch up with D. 2018-12-04 Akim Demaille d: add experimental support for the D language * configure.ac (ENABLE_D): New. * src/getargs.c (valid_languages): Add d. 2018-12-04 Akim Demaille d: add skeleton for the D language Contributed by Oliver Mangold. https://lists.gnu.org/archive/html/help-bison/2012-01/msg00000.html * README-D.txt, d-skel.m4, d.m4, lalr1.d: New. 2018-12-04 Akim Demaille examples: regenerate them when version.texi changes When we extract the examples from the documentation, %require "@value{VERSION}" is replaced with the current version. If we change the git branch, without changing the documentation, the generated examples will %require a version of Bison that differs from the actual version. * examples/local.mk (extracted.stamp): Depend on doc/version.texi. 2018-12-04 Akim Demaille skeletons: start some technical documentation * data/README: Convert to Markdown. Start documenting some of the macros used in all our skeletons. Simplify and fix the documentation of the macros in the skeletons. 2018-12-03 Akim Demaille regen 2018-12-03 Akim Demaille backend: revamp the handling of symbol types Currently it is the front end that passes the symbol types to the backend. For instance: %token NUM %type exp1 exp2 exp1: NUM { $$ = $1; } exp2: NUM { $$ = $1; } In both cases, $$ and $1 are passed to the backend as having type 'ival' resulting in code like `val.ival`. This is troublesome in the case of api.value.type=union, since in that the case the code this: %define api.value.type union %token NUM %type exp1 exp2 exp1: NUM { $$ = $1; } exp2: NUM { $$ = $1; } because in this case, since the backend does not know the symbol being processed, it is forced to generate casts in both cases: *(int*)(&val)`. This is unfortunate in the first case (exp1) where there is no reason at all to use a cast instead of `val.NUM` and `val.exp1`. So instead delegate the computation of the actual value type to the backend: pass $$ as `symbol-number, ival` and $$ as `symbol-number, MULL`, instead of passing `ival` before. * src/scan-code.l (handle_action_dollar): Find the symbol the action is about, not just its tyye. Pass both symbol-number, and explicit type tag ($n when there is one) to b4_lhs_value and b4_rhs_value. * data/bison.m4 (b4_symbol_action): adjust to the new signature to b4_dollar_pushdef. * data/c-like.m4 (_b4_dollar_dollar, b4_dollar_pushdef): Accept the symbol-number as new argument. * data/c.m4 (b4_symbol_value): Accept the symbol-number as new argument, and use it. (b4_symbol_value_union): Accept the symbol-number as new argument, and use it to prefer ready a union member rather than casting the union. * data/yacc.c (b4_lhs_value, b4_rhs_value): Accept the new symbol-number argument. Adjust uses of b4_dollar_pushdef. * data/glr.c (b4_lhs_value, b4_rhs_value): Adjust. * data/lalr1.cc (b4_symbol_value_template, b4_lhs_value): Adjust to the new symbol-number argument. * data/variant.hh (b4_symbol_value, b4_symbol_value_template): Accept the new symbol-number argument. * data/java.m4 (b4_symbol_value, b4_rhs_data): New. (b4_rhs_value): Use them. * data/lalr1.java: Adjust to b4_dollar_pushdef, and use b4_rhs_data. 2018-12-03 Akim Demaille style: comment and formatting changes * data/bison.m4, data/c++.m4, data/glr.c, data/java.m4, data/lalr1.cc, * data/yacc.c, src/scan-code.l: Fix comments. Prefer POS to denote the position of a symbol in a rule, since NUM is also used to denote symbol numbers. 2018-12-03 Akim Demaille NEWS: update 2018-12-03 Akim Demaille java: make sure the build dir exists * examples/java/local.mk (%D%/Calc.java): here. 2018-12-03 Akim Demaille c++: don't define variant, directly define semantic_type Instead of defining yy::variant and then alias yy::parser::semantic_type to variant, directly define yy::parser::semantic_type. This model is more appropriate if we want to sit the storage on top of unions in C++11. * data/variant.hh (b4_variant_define): Specialize and inline the definition into... (b4_value_type_declare): Here. Define union_type here. * data/lalr1.cc: Adjust. 2018-12-01 Akim Demaille NEWS: update 2018-12-01 Akim Demaille C++: use noexcept and constexpr There are probably more opportunities for them. So far, I observed no performance improvements. * data/c++.m4, data/lalr1.cc, data/stack.hh: here. 2018-12-01 Akim Demaille CI: also display the examples' test suite log * .travis.yml: here. 2018-12-01 Akim Demaille java: add an example * examples/java/Calc.y: New, based on test 495: "Calculator parse.error=verbose %locations". * examples/java/Calc.test, examples/java/local.mk: New. * configure.ac (ENABLE_JAVA): New. * examples/test (prog): Be ready to run Java programs. 2018-12-01 Akim Demaille style: unsigned int -> unsigned See https://lists.gnu.org/archive/html/bison-patches/2018-08/msg00027.html * src/output.c (muscle_insert_unsigned_int_table): Rename as... (muscle_insert_unsigned_table): this. 2018-12-01 Akim Demaille output: restore yyrhs and yyprhs This was demanded several times. See for instance: - David M. Warme https://lists.gnu.org/archive/html/help-bison/2011-04/msg00003.html - box12009 http://lists.gnu.org/archive/html/bug-bison/2016-10/msg00001.html Basically, this reverts: - commit 3d3bc1fe30f356cf674a979409e86ea0f88de4a0 Get rid of (yy)rhs and (yy)prhs - commit d333175f63f402dbadb647175e40ad88bf1defb5 Avoid compiler warning. Note that since these tables are not needed in the generated parsers, no skeleton requests them. This change only brings back their definition to M4, making it possible to user-defined skeletons to use these tables. * src/output.c (muscle_insert_item_number_table): Define. (prepare_rules): Generate the rhs and prhs tables. 2018-11-30 Akim Demaille regen 2018-11-30 Akim Demaille parser: shorten side-effects on current_type * src/parse-gram.y (tag.opt): Don't change current_type. Rather, return its value. Adjust dependencies. 2018-11-30 Akim Demaille style: reduce scopes * src/symlist.c: here. 2018-11-29 Akim Demaille tests: don't name C++ files *.c * tests/synclines.at (syncline escapes): Here. Otherwise, Clang generates an error and skips the test. 2018-11-29 Akim Demaille gnulib: update 2018-11-29 Akim Demaille regen 2018-11-29 Akim Demaille parser: factor the symbol definition * src/parse-gram.y (int.opt, string_as_id.opt): New. (symbol_def): Use it. 2018-11-29 Akim Demaille parser: improve location of string alias errors * src/parse-gram.y (symbol_def): Pass the right location for symbol_make_alias. * tests/regression.at (Duplicate string): Move to... * tests/input.at: here. (Token collisions): New. 2018-11-29 Akim Demaille diagnostics: complain about Bison directives when -Wyacc * src/complain.h, src/complain.c (bison_directive): New. * src/scan-gram.l (BISON_DIRECTIVE): New. Use it for Bison extensions. 2018-11-28 Akim Demaille doc: formatting changes * doc/bison.texi: here. 2018-11-27 Akim Demaille style: fix quotation in the test suite * tests/input.at: here. 2018-11-27 Akim Demaille regen 2018-11-27 Akim Demaille %nterm: do not accept character literals Reported by Rici Lake. http://lists.gnu.org/archive/html/bug-bison/2018-10/msg00000.html * src/complain.h: Formatting change. * src/parse-gram.y (id): Reject character literals used in a context for non-terminals. * tests/input.at (Invalid %nterm uses): Check that. 2018-11-27 Akim Demaille %nterm: do not accept numbers nor string alias Reported by Rici Lake. http://lists.gnu.org/archive/html/bug-bison/2018-10/msg00000.html * src/parse-gram.y (symbol_def): Refuse string aliases and numbers for non-terminals. (prologue_declaration): Recover from errors ended with ';'. * tests/input.at (Invalid %nterm uses): New. 2018-11-27 Akim Demaille TODO: update 2018-11-26 Akim Demaille doc: formatting changes * doc/bison.texi: Here. 2018-11-26 Akim Demaille style: comment changes * tests/testsuite.at: here. 2018-11-26 Akim Demaille gnulib: update ignores 2018-11-26 Akim Demaille gnulib: update to use its bitsets Bison's bitset were moved to gnulib. * lib/abitset.c, lib/abitset.h, lib/bbitset.h, lib/bitset.c, * lib/bitset.h, lib/ebitset.c, lib/ebitset.h, lib/lbitset.c, * lib/bitset_stats.c, lib/bitset_stats.h, lib/bitsetv-print.c, * lib/bitsetv-print.h, lib/bitsetv.c, lib/bitsetv.h, * lib/lbitset.h, lib/vbitset.c, lib/vbitset.h: Remove. * gnulib: Update. * bootstrap.conf, lib/local.mk: Adjust. 2018-11-25 Akim Demaille gnulib: use conditional dependencies * bootstrap.conf: here. 2018-11-25 Akim Demaille CI: run on xenial Xenial (Ubuntu 16.04) is finally available on Travis. Let's use it. Among the changes: - Automake 1.14.1 -> 1.15.0 - Doxygen 1.8.6 -> 1.8.11 - Flex 2.5.35 -> 2.6.0, with plenty of new compiler warnings - Gettext 0.18.3 -> 0.19.7 - Graphviz 2.36.0 -> 2.38.0 - Texinfo 5.2 -> 6.1 * .travis.yml: here. 2018-11-25 Akim Demaille CI: we don't need git show * .travis.yml: Don't run it. 2018-11-25 Akim Demaille regen 2018-11-25 Akim Demaille warning: avoid warnings about unreachable code Reported by Uxio Prego. https://lists.gnu.org/archive/html/help-bison/2018-11/msg00031.html We also need to move the unreachable 'goto' to a reachable place, otherwise clang complains about the code being unreachable anyway. See also https://bugs.llvm.org/show_bug.cgi?id=39736. Interestingly, we don't have to apply that trick to `#define YYCDEBUG if (false) std::cerr`, clang does not warn when the code comes from macro expansion. * configure.ac: Use -Wunreachable-code when supported. * data/lalr1.cc, data/yacc.c: Pacify clang's warning about `if (0)` by using a macro. Another possibility was to move this statement to a reachable place. * tests/actions.at, tests/c++.at: Avoid generating unreachable code. 2018-11-24 Akim Demaille yacc.c: avoid generating dead code We should probably introduce some struct and functions to deal with stack management, rather than open coding it. yyparse would be much nicer to read, and a better model for possible other skeletons. * data/yacc.c (yyparse::yysetstate): Avoid generating code when neither yyoverflow nor YYSTACK_RELOCATE is defined. 2018-11-22 Akim Demaille %expect-rr: tune the number of conflicts per rule Currently on a grammar such as exp : a '1' | a '2' | a '3' | b '1' | b '2' | b '3' a: b: we count only one rr-conflict on the `b:` rule, i.e., we expect: b: %expect-rr 1 although there are 3 conflicts in total. That's because in the conflicted state we count only a single conflict, not three (one for each of the lookaheads: '1', '2', '3'). State 0 0 $accept: . exp $end 1 exp: . a '1' 2 | . a '2' 3 | . a '3' 4 | . b '1' 5 | . b '2' 6 | . b '3' 7 a: . %empty ['1', '2', '3'] 8 b: . %empty ['1', '2', '3'] '1' reduce using rule 7 (a) '1' [reduce using rule 8 (b)] '2' reduce using rule 7 (a) '2' [reduce using rule 8 (b)] '3' reduce using rule 7 (a) '3' [reduce using rule 8 (b)] $default reduce using rule 7 (a) exp go to state 1 a go to state 2 b go to state 3 See https://lists.gnu.org/archive/html/bison-patches/2013-02/msg00106.html. * src/conflicts.c (rule_has_state_rr_conflicts): Rename as... (count_rule_state_sr_conflicts): this. DWIM. (count_rule_rr_conflicts): Adjust. * tests/conflicts.at (%expect-rr in grammar rules) (%expect-rr too much in grammar rules) (%expect-rr not enough in grammar rules): New. 2018-11-22 Akim Demaille %expect-rr: fix the computation of the overall number of conflicts On a grammar such as exp: "num" | "num" | "num" we currently report only one RR conflict, instead of two. This bug is present since the origins of Bison commit 08089d5d35ece0c7d41659cc1bc09638e2abb151 Author: David MacKenzie Date: Tue Apr 20 05:42:52 1993 +0000 Initial revision and was preserved in commit 676385e29c4aedfc05d20daf1ef20cd4ccc84856 Author: Paul Hilfinger Date: Fri Jun 28 02:26:44 2002 +0000 Initial check-in introducing experimental GLR parsing. See entry in ChangeLog dated 2002-06-27 from Paul Hilfinger for details. See https://lists.gnu.org/archive/html/bison-patches/2018-11/msg00011.html * src/conflicts.h, src/conflicts.c (count_state_rr_conflicts) (count_rr_conflicts): Use only the correct count of conflicts. * tests/glr-regression.at: Fix expectations. 2018-11-21 Akim Demaille tests: generate *.output files * tests/glr-regression.at: here. 2018-11-21 Akim Demaille %expect: tune the number of conflicts per rule Currently on a grammar such as exp: "number" | exp "+" exp | exp "*" exp we count only one sr-conflict for both binary rules, i.e., we expect: exp: "number" | exp "+" exp %expect 1 | exp "*" exp %expect 1 although there are 4 conflicts in total. That's because in the states in conflict, for instance that for the "+" rule: State 6 2 exp: exp . "+" exp 2 | exp "+" exp . [$end, "+", "*"] 3 | exp . "*" exp "+" shift, and go to state 4 "*" shift, and go to state 5 "+" [reduce using rule 2 (exp)] "*" [reduce using rule 2 (exp)] $default reduce using rule 2 (exp) we count only a single conflict, although there are two (one on "+" and another with "*"). See https://lists.gnu.org/archive/html/bison-patches/2013-02/msg00106.html. * src/conflicts.c (rule_has_state_sr_conflicts): Rename as... (count_rule_state_sr_conflicts): this. DWIM. (count_rule_sr_conflicts): Adjust. * tests/conflicts.at (%expect in grammar rules): New. 2018-11-21 Akim Demaille regen 2018-11-21 Akim Demaille style: reduce scopes * src/conflicts.c, src/reader.c: Minor style changes. 2018-11-21 Paul Hilfinger allow %expect and %expect-rr modifiers on individual rules This change allows one to document (and check) which rules participate in shift/reduce and reduce/reduce conflicts. This is particularly important GLR parsers, where conflicts are a normal occurrence. For example, %glr-parser %expect 1 %% ... argument_list: arguments %expect 1 | arguments ',' | %empty ; arguments: expression | argument_list ',' expression ; ... Looking at the output from -v, one can see that the shift-reduce conflict here is due to the fact that the parser does not know whether to reduce arguments to argument_list until it sees the token AFTER the following ','. By marking the rule with %expect 1 (because there is a conflict in one state), we document the source of the 1 overall shift- reduce conflict. In GLR parsers, we can use %expect-rr in a rule for reduce/reduce conflicts. In this case, we mark each of the conflicting rules. For example, %glr-parser %expect-rr 1 %% stmt: target_list '=' expr ';' | expr_list ';' ; target_list: target | target ',' target_list ; target: ID %expect-rr 1 ; expr_list: expr | expr ',' expr_list ; expr: ID %expect-rr 1 | ... ; In a statement such as x, y = 3, 4; the parser must reduce x to a target or an expr, but does not know which until it sees the '='. So we notate the two possible reductions to indicate that each conflicts in one rule. See https://lists.gnu.org/archive/html/bison-patches/2013-02/msg00105.html. * doc/bison.texi (Suppressing Conflict Warnings): Document %expect, %expect-rr in grammar rules. * src/conflicts.c (count_state_rr_conflicts): Adjust comment. (rule_has_state_sr_conflicts): New static function. (count_rule_sr_conflicts): New static function. (rule_nast_state_rr_conflicts): New static function. (count_rule_rr_conflicts): New static function. (rule_conflicts_print): New static function. (conflicts_print): Also use rule_conflicts_print to report on individual rules. * src/gram.h (struct rule): Add new fields expected_sr_conflicts, expected_rr_conflicts. * src/reader.c (grammar_midrule_action): Transfer expected_sr_conflicts, expected_rr_conflicts to new rule, and turn off in current_rule. (grammar_current_rule_expect_sr): New function. (grammar_current_rule_expect_rr): New function. (packgram): Transfer expected_sr_conflicts, expected_rr_conflicts to new rule. * src/reader.h (grammar_current_rule_expect_sr): New function. (grammar_current_rule_expect_rr): New function. * src/symlist.c (symbol_list_sym_new): Initialize expected_sr_conflicts, expected_rr_conflicts. * src/symlist.h (struct symbol_list): Add new fields expected_sr_conflicts, expected_rr_conflicts. * tests/conflicts.at: Add tests "%expect in grammar rule not enough", "%expect in grammar rule right.", "%expect in grammar rule too much." 2018-11-21 Akim Demaille NEWS: update 2018-11-21 Akim Demaille gnulib: update 2018-11-21 Akim Demaille remove ancient lint directives * data/c++.m4, data/yacc.c: Remove surprising remains of lint directives. 2018-11-21 Jannick doc: calc++: remove ancient fix for flex * doc/bison.texi (Calc++ Scanner): Remove fix for Flex 2.5.31-2.5.33. 2018-11-21 Jannick doc: calc++: ignore \r in the scaner * doc/bison.texi (Calc++ Scanner): Ignore \r. 2018-11-20 Akim Demaille style: harmonize the labels of yyparse * data/glr.c, data/lalr1.cc, data/yacc.c: Fix indentation and other formatting issues. 2018-11-20 Akim Demaille regen 2018-11-20 Akim Demaille style: avoid lengthy actions We also lack a consistent naming for directive implementations. `directive_skeleton` is too long, `percent_skeleton` is not very nice looking, `process_skeleton` looks ambiguous, `do_skeleton` is somewhat ambiguous too, but seems a better track. * src/parse-gram.y (version_check): Rename as... (do_require): this. (do_skeleton): New. Use it. 2018-11-20 Akim Demaille c++: using macros around user types breaks when they include comma We may generate code such as basic_symbol (typename Base::kind_type t, YY_RVREF (std::pair) v); which, of course, breaks, because YY_RVREF sees two arguments. Let's not play tricks with _VA_ARGS__, I'm unsure about it portability. Anyway, I plan to change more things in this area. Reported by Sébastien Villemot. http://lists.gnu.org/archive/html/bug-bison/2018-11/msg00014.html * data/variant.hh (b4_basic_symbol_constructor_declare) (b4_basic_symbol_constructor_define): Don't use macro on user types. * tests/types.at: Check that we support pairs. 2018-11-16 Akim Demaille README: update 2018-11-16 Akim Demaille tests: remove duplicate definition Probably imported by 6d58c632025cb6928a90e4176577982bfb9c3981, a merge commit. * tests/atlocal.in (POSIXLY_CORRECT_IS_EXPORTED): Define it once. 2018-11-16 Akim Demaille glr.c: fix use of _Noreturn In C++, [[noreturn]] must not be between "static" and the rest of the function signature, it must precede it. C's _Noreturn does not seem to have such a constraint, but it is therefore compatible with the C++ constraint. Since we #define _Noreturn as [[noreturn]] is modern C++, be sure to push the _Noreturn first. Unfortunately this was not caught by the test suite, because it always loads config.h first, and config.h contains another definition of _Noreturn that does not use [[noreturn]], and hides ours. That's probably a sign we should avoid always loading config.h. * data/glr.c (yyFail, yyMemoryExhausted): here. 2018-11-16 Akim Demaille tests: run the api.value.type tests for all C++ standards * tests/local.at (AT_LANG_FOR_EACH_STD): New. (AT_REQUIRE_CXX_VERSION): Rename as... (AT_REQUIRE_CXX_STD): this. Accept an argument for what to do when the requirement is not met. * tests/types.at (api.value.type): Check all the C++ stds. 2018-11-16 Akim Demaille CI: split the ASAN job in two The following commit introduce even more compilations/runs than before, and with ASAN on, we go beyond to 50min credit from Travis. * .travis.yml (Clang 7 libc++ and ASAN): Split in two. 2018-11-14 Akim Demaille c++: use YY_CPLUSPLUS * data/c++.m4: here. 2018-11-13 Akim Demaille c++: factor the handling of __cplusplus into YY_CPLUSPLUS * data/c++.m4 (b4_cxx_portability): Define it. Use it. * data/lalr1.cc, data/variant.hh: Use it. 2018-11-13 Akim Demaille style: formatting changes * src/scan-gram.l: here. 2018-11-13 Akim Demaille tests: clarify some magic constant * tests/c++.at (C++ Variant-based Symbols Unit Tests): here. 2018-11-13 Akim Demaille examples: remove useless includes * examples/c++/variant-11.yy, examples/c++/variant.yy: here. Fix warning when storing a long into an int. 2018-11-12 Akim Demaille tests: compile the C++ examples with warnings * examples/c++/local.mk: Pass $(WARN_CXXFLAGS_TEST). 2018-11-12 Akim Demaille regen 2018-11-12 Akim Demaille parser: deprecate %error-verbose It is unfortunate that %error_verbose was properly diagnosed as obsoleted by "%define parse.error verbose", but %error-verbose was not. * src/parse-gram.y (%error-verbose): Remove support. * src/scan-gram.l: Do it here instead, with a warning. * tests/input.at (Deprecated directives): Check it. 2018-11-12 Akim Demaille tests: migrate from %error-verbose to %define parse.error verbose * tests/actions.at, tests/c++.at, tests/conflicts.at, * tests/cxx-type.at, tests/existing.at, tests/glr-regression.at, * tests/headers.at, tests/input.at, tests/java.at, tests/javapush.at, * tests/local.at, tests/regression.at, tests/skeletons.at, * tests/torture.at: Here. 2018-11-12 Akim Demaille parser: deprecate %nterm It has several weaknesses. Reported by Rici Lake. http://lists.gnu.org/archive/html/bug-bison/2018-10/msg00000.html * src/scan-gram.l: here. 2018-11-12 Akim Demaille tests: fix syncline tests These tests are skipped with GCC: "\"".c:1:5: error: function declaration isn't a prototype [-Werror=strict-prototypes] int main() { return 0; } ^~~~ * tests/synclines.at: Stop writing C++ in C. * tests/local.at: Formatting changes. 2018-11-11 Akim Demaille NEWS: expected features of Bison 3.3 2018-11-11 Akim Demaille yacc: reduce scope in push mode * data/yacc.c (yypull_parse): Here. 2018-11-11 Akim Demaille style: comment changes * data/c++.m4, data/glr.c, data/lalr1.java, data/yacc.c (yytranslate, YYTRANSLATE): Harmonize comments. 2018-11-10 Akim Demaille c++: simplify a default construction * data/lalr1.cc (yytnamerr_): here. 2018-11-10 Akim Demaille regen 2018-11-10 Akim Demaille reader: no longer accept %define variable names in quotes It was never documented. * src/parse-gram.y (variable): Here. 2018-11-10 Akim Demaille dogfooding: use api.value.type union * src/parse-gram.y (api.value.type): Set to union. Replace occurrences of %union with explicit %types. * src/scan-gram.l: Adjust yylval's field names. (RETURN_VALUE): No longer needs the Field argument. Use it more. 2018-11-10 Akim Demaille djgpp: remove Support for DJGPP was announced to be removed in the NEWS of Bison 3.1 (2018-08-27) unless someone expressed interest. There was no answer. * djgpp: Remove. * NEWS, Makefile.am, cfg.mk, po/POTFILES.in: Adjust. 2018-11-10 Akim Demaille scanner: simplify use of gettext * src/scan-gram.l (unexpected_end): Leave the actual call to gettext to the caller. 2018-11-10 Akim Demaille style: clean up the scanner and parser * src/scan-gram.l: Formatting changes. Add "missing" assertion for symmetry. * src/parse-gram.y: Formatting changes. 2018-11-09 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2018-11-09 Akim Demaille version 3.2.1 * NEWS: Record release date. 2018-11-08 Akim Demaille build: fix issues in the generated tarball Reported by Andre da Costa Barros. https://savannah.gnu.org/patch/?9716 * examples/calc++/local.mk: We no longer generate position.hh and stack.hh. Leaving them here triggers their concurrent generation, which fails. (%C%_calc___CPPFLAGS): Fix the extracted headers in the source tree. * examples/mfcalc/local.mk (%C%_mfcalc_CPPFLAGS): Ditto. 2018-11-07 Akim Demaille build: fix typo Reported by Horst Von Brand. https://savannah.gnu.org/support/?109580 * examples/local.mk (.PHOMY): Rename as... (.PHONY): this. 2018-11-06 Akim Demaille NEWS: update 2018-11-06 Akim Demaille examples: ship them Currently, the examples are extracted on the user's side. Unfortunately, that requires that the user has Perl, which is otherwise not needed for Bison. Let's ship the examples instead. The examples were handled this way so that we could depend on configure flags: if --enable-gcc-warnings is passed, it is understood as "I'm a maintainer", so the examples are generated with `#line`s. Regular users should not see them, so they are now unconditionally removed when rolling a tarball. Reported by Mike Frysinger. https://lists.gnu.org/archive/html/bison-patches/2015-04/msg00000.html * examples/local.mk: Ship all the extracted files. (examples-unline): New. Make sure that the generated tarballs do not contain the #lines. 2018-11-06 Akim Demaille build: minor fixes in doc/ * doc/local.mk: Consistently use *.tmp for temporary files. Fix comments. 2018-11-05 Akim Demaille CI: maximize chances of errors sooner * .travis.yml: Try clang and icc soon, so that we don't have to wait for the end of the run to know that they fail. 2018-11-04 Akim Demaille c++: improve the generated documentation * data/lalr1.cc, data/location.cc: Improve documenting comments. * tests/c++.at (Doxygen Documentation): Fix AT_BISON_OPTION_PUSHDEFS, so that the generated yyerror is correct. * tests/c++.at, tests/headers.at: Prefer %empty. 2018-11-04 Akim Demaille tests: don't fail if the C++ compiler does not work Also, make sure that `make dist` generates a correct tarball even if the C++ compiler does not work. Reported by Nelson H. F. Beebe. * m4/cxx.m4 (BISON_CXX_WORKS): Define to true/false instead of true/exit 77. The latter is too dangerous to use (it directly quits). (ENABLE_CXX): New name for the Automake conditional, for consistency with ENABLE_CXX11 etc. * tests/local.at (AT_COMPILE, AT_COMPILE_CXX): Adjust to the new semantics of BISON_CXX_WORKS. * examples/c++/local.mk: Skip the variant test if C++ does not work. * examples/calc++/local.mk: Likewise. 2018-11-04 Akim Demaille tests: don't disable C++ warnings in C files This triggers error with GCC. See eff6739124c61bb5660d78453210d1d6a17d30e7. * tests/testsuite.h: Disable -Wzero-as-null-pointer-constant only for C++. 2018-11-04 Akim Demaille c++: workaround portability issue On some systems (x86_64-pc-solaris2.11), with Developer Studio 12.5's CC, we get: ".../include/CC/Cstd/vector.cc", line 127: Error: Cannot assign const yy::parser::stack_symbol_type to yy::parser::stack_symbol_type without "yy::parser::stack_symbol_type::operator=(const yy::parser::stack_symbol_type&)";. ".../include/CC/Cstd/vector", line 475: Where: While instantiating "std::vector::__insert_aux(yy::parser::stack_symbol_type*, const yy::parser::stack_symbol_type&)". ".../include/CC/Cstd/vector", line 475: Where: Instantiated from non-template code. 1 Error(s) detected. Don't expect __cplusplus to be always defined. If it's not, consider this is C++98. Reported by Nelson H. F. Beebe. * data/c++.m4, data/lalr1.cc, examples/c++/variant.yy, tests/local.at, * tests/testsuite.h: An undefined __cplusplus means pre C++11. 2018-11-03 Akim Demaille tests: work around getopt portability issues On some systems, we don't use our getopt. As a consequence the error messages vary: $ bison --skeleton bison: option requires an argument -- skeleton Try 'bison --help' for more information. instead of bison: option '--skeleton' requires an argument Try 'bison --help' for more information. Reported by Jannick and Nelson H. F. Beebe. https://lists.gnu.org/archive/html/bison-patches/2018-10/msg00140.html * tests/input.at (Invalid number of arguments): work around getopt portability issues. 2018-11-03 Akim Demaille doc: -Wzero-as-null-pointer-constant was added to GCC 4.7 It is not supported by previous versions. See https://www.gnu.org/software/gcc/gcc-4.7/changes.html Reported by Nelson H. F. Beebe. * doc/bison.texi (Calc++ Scanner): here. 2018-11-02 Adam Sampson examples: #include in calc++ strerror is defined by , and recent versions of GNU libstdc++ no longer include this automatically from . 2018-10-31 Akim Demaille c: provide a definition of _Noreturn that works for C++ On Solaris, GCC 7.3 defines: -std=c++14 -std=c++17 __cplusplus 201402L 201703L __STDC_VERSION__ 199901L 201112L So the current #definition of _Noreturn sees that 201112 <= __STDC_VERSION__, i.e., that C11 is supported, so it expects _Noreturn to be supported. Apparently it is not. This is only for C++, the test suite works for C. However, the test suite does not try several C standards, maybe we should... http://lists.gnu.org/archive/html/bug-bison/2018-10/msg00064.html * data/c.m4 (b4_attribute_define): Define _Noreturn as [[noreturn]] in modern C++. 2018-10-30 Akim Demaille c: update the definition of _Noreturn Does not work on Solaris 11.3 x86/64: 479. c++.at:1293: testing C++ GLR parser identifier shadowing ... ======== Testing with C++ standard flags: '-std=c++17' ./c++.at:1332: $BISON_CXX_WORKS stderr: stdout: ./c++.at:1332: $CXX $CXXFLAGS $CPPFLAGS $LDFLAGS -o input input.cc $LIBS stderr: input.cc:837:8: error: '_Noreturn' does not name a type static _Noreturn void ^~~~~~~~~ input.cc:845:8: error: '_Noreturn' does not name a type static _Noreturn void ^~~~~~~~~ Reported by Kiyoshi Kanazawa. http://lists.gnu.org/archive/html/bug-bison/2018-10/msg00051.html * data/c.m4 (b4_attribute_define): Use the snippet which is currently in gnulib's m4/gnulib-common.m4 (which seems a little more advanced than lib/_Noreturn.h). 2018-10-30 Akim Demaille tests: don't expect the shell to support 'local' It doesn't work on Solaris 11.3 x86/64. Reported by Kiyoshi Kanazawa. http://lists.gnu.org/archive/html/bug-bison/2018-10/msg00051.html * examples/test: Don't use 'local'. 2018-10-30 Akim Demaille bitset: fix warning Reported by Hans Åberg. http://lists.gnu.org/archive/html/bug-bison/2018-10/msg00047.html * lib/bitset.c (bitset_count_): here. 2018-10-30 Akim Demaille build: fix use of gnulib Make variables Reported by Kiyoshi Kanazawa. http://lists.gnu.org/archive/html/bug-bison/2018-10/msg00048.html * lib/local.mk (lib_libbison_a_LIBADD): Merge into... * src/local.mk (src_bison_LDADD): here. 2018-10-29 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2018-10-29 Akim Demaille version 3.2 * NEWS: Record release date. 2018-10-29 Akim Demaille build: don't depend on the libc to generate bison.help The "Report translation bugs to..." part of --help is issued only on glibc systems. So if the tarball is not wrapped on such a system, and used on such a system (or the converse), then bison.help will differ on the user's system, and help2man will be called to update bison.1. But help2man should not be a requirement. Reported by Alexandre Duret-Lutz. * doc/local.mk (doc/bison.help): Remove the possible doc about translation bugs. Pass LC_ALL=C, as reported in src/getargs.c's usage(). (doc/cross-options.texi): Use bison.help instead of calling bison --help. 2018-10-29 Akim Demaille c++: always issue the "generated by" message Some users rely on this sentence to know that the file can be ignored. Reported by Alexandre Duret-Lutz. * data/bison.m4 (b4_generated_by): New. (b4_copyright): Use it. * data/location.cc, data/stack.hh: Use it too, for the stub files (position.hh and stack.hh). 2018-10-28 Akim Demaille cfg.mk: remove exceptions for timevar They appear to be no longer needed. * cfg.mk: here. 2018-10-28 Akim Demaille style: clean up src/AnnotationList.c * src/AnnotationList.c: Reduce scopes. 2018-10-28 Akim Demaille style: clean up print.c * src/print.c: Reduce scopes. 2018-10-27 Akim Demaille bitset: clean up bbitset.h * lib/libiberty.h: Inline in... * lib/bbitset.h: here. * lib/local.mk: Adjust. 2018-10-27 Akim Demaille bitset: clean up bitset.h * lib/bitset.h: Fix include order. 2018-10-27 Akim Demaille bitset: clean up vbitset.c * lib/vbitset.c: Reduce scopes, etc. 2018-10-27 Akim Demaille bitset: clean up lbitset.c * lib/lbitset.c: Reduce scopes, etc. 2018-10-27 Akim Demaille bitset: clean up ebitset.c * lib/ebitset.c: Reduce scopes, etc. 2018-10-27 Akim Demaille bitset: clean up bitset_stats.c * lib/bitset_stats.c: Reduce scopes, etc. 2018-10-27 Akim Demaille bitset: clean up bitset.c * lib/bitset.c: Reduce scopes, etc. 2018-10-27 Akim Demaille bitset: clean up abitset.c * lib/abitset.c: Reduce scopes, etc. 2018-10-27 Jannick xml2dot.xsl: fix typos in comments 2018-10-27 Akim Demaille doc: fix distcheck The extracted example, simple.yy, does not use %require "3.2", so it generates a stack.hh, which breaks distcheck. * doc/bison.texi: Fix it. 2018-10-27 Akim Demaille NEWS: prepare for 3.2 2018-10-26 Akim Demaille tests: beware of Windows file name constraints Don't expect to be able to build a file named '"\"".y' (6 characters) on Windows. Reported by Jannick. http://lists.gnu.org/archive/html/bug-bison/2018-10/msg00042.html * tests/synclines.at (syncline escapes): Skip if we failed to create the file. 2018-10-26 Akim Demaille tests: fix invocation of m4_map * tests/actions.at, tests/synclines.at: m4_map takes a list of arguments in $2, m4_map_args takes arguments in $2, $3, etc. 2018-10-26 Akim Demaille doc: spell check * README, doc/bison.texi, examples/README, examples/c++/README: here. 2018-10-26 Akim Demaille examples: add a Makefile for C++ short examples * examples/c++/Makefile: New. * examples/c++/local.mk, examples/c++/README: Adjust. 2018-10-26 Akim Demaille doc: some improvements * doc/bison.texi (Calc++ Scanner): Show how exception can be thrown from auxiliary functions. Clarify the meaning of the various flex %options we use. Get rid of a warning. (Calc++ Parsing Driver): Use the parser as a functor. 2018-10-26 Akim Demaille examples: check the errors * examples/test (run): Check stderr, unless -noerr is passed. * examples/calc++/calc++.test, examples/mfcalc/mfcalc.test: Check errors. 2018-10-25 Akim Demaille doc: minor fixes * doc/bison.texi: Simplify wording. Fix Texinfo error. (Complete Symbols): Handle the token EOF. (Calc++ Parser): In the modern C++ world, prefer assignment to swap. (Strings are Destroyed): Prefer an explicit 'continue' to a comment. 2018-10-24 Akim Demaille configure: quit on trying to get ICC and Flex be friends The CI is using Flex 2.5.35. And ICC is too picky for it. Let's stop making these warnings errors. I wish I could disable them in the source files using the ICC version and the Flex version, but ICC's pragma support is unclear, and I'm tired of fighting it. * configure.ac (FLEX_SCANNER_CXXFLAGS): Make warnings warnings. * examples/c++/local.mk: Comment changes. 2018-10-24 Akim Demaille doc: mention earlier how to disable the generation of location.hh Suggested by Victor Khomenko. * doc/bison.texi (C++ Bison Interface): Here. 2018-10-24 Akim Demaille c++: std::to_string is available in C++11 Reported by Victor Khomenko. http://lists.gnu.org/archive/html/bug-bison/2018-10/msg00033.html * doc/bison.texi, examples/c++/variant-11.yy: Use std::to_string instead of ours. 2018-10-24 Akim Demaille examples: move the variant examples into the C++ directory * examples/variant-11.test examples/variant-11.yy, * examples/variant.test examples/variant.yy: Move into examples/c++/. * examples/c++/README: New. * examples/README, examples/c++/local.mk, examples/local.mk: Adjust. 2018-10-24 Akim Demaille doc: an introductory example for C++ Suggested by Victor Khomenko. http://lists.gnu.org/archive/html/bug-bison/2018-08/msg00037.html * doc/bison.texi (A Simple C++ Example): New. * examples/c++/local.mk, examples/c++/simple.test: New. Extract, check, and install this new example. * examples/local.mk: Adjust. * examples/test: Adjust to the case where the dirname differs from the test name. 2018-10-24 Akim Demaille build: remove a few copies of the Copyright from the generated Makefile * build-aux/local.mk, cfg.mk, examples/calc++/local.mk, * examples/local.mk, examples/mfcalc/local.mk, * examples/rpcalc/local.mk, lib/local.mk, src/local.mk, * tests/local.mk: Use Automake comments so that we don't get a copy of each in the generated Makefile. 2018-10-24 Akim Demaille c++: make operator() an alias to the parse function * data/glr.cc, data/lalr1.cc (operator()): New. * doc/bison.texi: Update. 2018-10-23 Akim Demaille build: enable more warnings during tests Prompted by Derek Clegg. http://lists.gnu.org/archive/html/bug-bison/2018-10/msg00018.html * configure.ac: here. 2018-10-23 Akim Demaille regen 2018-10-23 Akim Demaille yacc.c: work around strange typing issues On the CI, both GCC and Clang report: src/parse-gram.c: In function 'yy_lac': src/parse-gram.c:1479:29: error: format '%hd' expects argument of type 'int', but argument 3 has type 'yytype_int16 {aka long int}' [-Werror=format=] YYDPRINTF ((stderr, " G%hd", yystate)); ^ Although yytype_int16 is supposed to be a short int, not a long int. This must be explored. * data/yacc.c (yy_lac): Work around typing issue. 2018-10-23 Akim Demaille yacc.c: don't define _Noreturn uselessly Clang warns: aux/x.h:97:11: error: macro name is a reserved identifier [-Werror,-Wreserved-id-macro] # define _Noreturn YY_ATTRIBUTE ((__noreturn__)) Reported by Derek Clegg. http://lists.gnu.org/archive/html/bug-bison/2018-10/msg00024.html * data/c.m4 (b4_attribute_define): Don't define _Noreturn unconditionally. * data/glr.c: Ask for _Noreturn. 2018-10-23 Akim Demaille pacify ICC 16.0.3 20160415 Found on the CI. yacc.c: error #2259: non-pointer conversion from "int" to "yytype_int16={short}" may lose significant bits yystate = (0 <= yyi && yyi <= YYLAST && yycheck[yyi] == *yyesp ^ glr.c: error #2259: non-pointer conversion from "int" to "yybool={unsigned char}" may lose significant bits yybool yynormal YY_ATTRIBUTE_UNUSED = (yystackp->yysplitPoint == YY_NULLPTR); ^ error #2259: non-pointer conversion from "int" to "yybool={unsigned char}" may lose significant bits return yypact_value_is_default (yypact[yystate]); ^ error #2259: non-pointer conversion from "int" to "yybool={unsigned char}" may lose significant bits return 0 < yyaction; ^ error #2259: non-pointer conversion from "int" to "yybool={unsigned char}" may lose significant bits return yyaction == 0; ^ error #2259: non-pointer conversion from "int" to "yybool={unsigned char}" may lose significant bits yystackp->yytops.yylookaheadNeeds[yys] = yychar != YYEMPTY; ^ * data/glr.c, data/yacc.c: Avoid these warnings. 2018-10-23 Akim Demaille flex: work around more warnings * doc/bison.texi: here. 2018-10-23 Akim Demaille examples: clang 5 defines nullptr as a macro * examples/variant.yy: So don't do it. * examples/variant-11.yy: Fix comment. 2018-10-23 Akim Demaille glr.c: be strict about types * data/glr.c: Don't use `foo |= bar` with foo and bar being yybool: the result appears to be an int, not a yybool. Use yybool where appropriate. Add casts where needed. 2018-10-23 Akim Demaille yacc.c: fix warnings about integral types Reported by Derek Clegg. http://lists.gnu.org/archive/html/bug-bison/2018-10/msg00018.html Rather than adding casts, we should be more careful with types. For instance yystate should be a yytype_int16. But currently we can't: it is also used sometimes for storing other things that state numbers. * data/yacc.c (yyparse): Add missing casts. 2018-10-22 Akim Demaille yacc.c: clarify the computation of yystate The yacc.c skeleton is old, and was using many tricks to save registers. Today's register allocators can do this themselves. Let's keep the code simpler to read and let compilers do their job. * data/yacc.c: Avoid using yystate for different types of content. An inline function would be better, but doing this portably will be a problem. 2018-10-22 Akim Demaille printf returns a signed int * tests/local.at: Adjust location_print's signature. 2018-10-22 Akim Demaille tests: be strict about types * tests/actions.at, tests/c++.at, tests/cxx-type.at, * tests/glr-regression.at, tests/local.at, tests/torture.at, * tests/types.at: Pay stricter attention to types to avoid warnings. 2018-10-22 Akim Demaille c++: fix signedness issues * data/lalr1.cc, data/stack.hh: The callers of stack use int, while stack is based on size_type. Add overloads to avoid warnings. 2018-10-22 Akim Demaille c++: minor changes * data/lalr1.cc: Fix oldish comment. * data/stack.hh: Prefer typename for type names. Use size() instead of duplicating it. * examples/variant-11.yy, examples/variant.yy (yylex): Use int, as this is the type of the semantic value. 2018-10-22 Akim Demaille all: display a clear warning about private macros * data/bison.m4 (b4_disclaimer): New. * data/glr.c, data/glr.cc, data/lalr1.cc, data/yacc.c: Use it. 2018-10-21 Akim Demaille c++: minor simplification * data/stack.hh: Prefer a default argument value to two constructors. 2018-10-21 Akim Demaille regen 2018-10-21 Akim Demaille all: avoid useless comments and #lines Currently we emit useless code for places where we might issue user content, but there is none. This commit avoids this. Besides, some of the comments looked like implementation details ("Copy the first part of user declarations"), rather than made for the reader of the result ("First part of user prologue"). On Bison's parse-gram.c we get: @@ -76,10 +76,6 @@ #define yynerrs gram_nerrs -/* Copy the first part of user declarations. */ - -#line 82 "src/parse-gram.c" /* yacc.c:339 */ - * data/bison.m4 (b4_define_user_code): Accept a comment to document the section. Do not emit any code if the content is empty. Adjust callers to not emit the comment. Do not * data/glr.c, data/glr.cc, data/lalr1.cc, data/lalr1.java, data/yacc.c: Adjust. 2018-10-21 Akim Demaille tests: refactor * tests/actions.at, tests/synclines.at: Prefer iteration to copy-paste. 2018-10-21 Akim Demaille tests: rename AT_SKEL_CC_IF/AT_SKEL_JAVA_IF as AT_CXX_IF/AT_JAVA_IF The previous name is too obscure, and the other macros for C++ use CXX, not CC. * tests/local.at (AT_SKEL_CC_IF, AT_SKEL_JAVA_IF): Rename as... (AT_CXX_IF, AT_JAVA_IF): these. Adjust callers. 2018-10-21 Akim Demaille c++: check that emplace for rvalues works See the previous commit. * tests/local.at (AT_REQUIRE_CXX_VERSION): New. * tests/types.at (api.value.type): Check emplace in C++14. 2018-10-20 Akim Demaille c++: prefer a perfect forwarding version of variant's emplace * data/variant.hh (emplace): In modern C++, use only a perfect forwarding version. And use it. * doc/bison.texi: Document it. 2018-10-20 Akim Demaille c++: prefer 'emplace' to 'build' When we introduced variants in Bison, C++ did not have the 'emplace' functions, and we chose 'build'. Let's align with modern C++ and promote 'emplace' rather than 'build'. * data/lalr1.cc, data/variant.hh (emplace): New. (build): Deprecate in favor of emplace. * doc/bison.texi: Adjust. 2018-10-20 Akim Demaille %printer: promote yyo rather than yyoutput * doc/bison.texi: Promote yyo rather than yyoutput. * data/c.m4, data/glr.cc, tests/types.at, tests/calc.at, tests/regression.at: Adjust. 2018-10-20 Akim Demaille doc: improve the C++ section * doc/bison.texi (C++ Parser): file.hh and location.hh are no longer mandatory. Various minor fixes. 2018-10-20 Akim Demaille doc: reorder C++ sections * doc/bison.texi (C++ Parser Interface): Document before semantic_type and location_type. 2018-10-20 Akim Demaille git: don't ignore auxiliary Texinfo files As a matter of fact, I think it is wrong to gitignore generated files that belong to the build tree. There should be the strict minimum, and it's up to people that build in place to adjust their own ~/.gitignore. * doc/.gitignore: here. Remove files we no longer produce (thanks to texi2dvi). 2018-10-20 Akim Demaille c++: do not exhibit private macros * examples/variant-11.yy: here. 2018-10-20 Akim Demaille c++: don't obfuscate std::move when not needed * data/lalr1.cc, data/variant.hh: Avoid macros that depend on the version of C++ when not needed. 2018-10-20 Akim Demaille build: add missing gnulib libs Reported by Denis Excoffier. * lib/local.mk, src/local.mk: here. 2018-10-18 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2018-10-18 Akim Demaille version 3.1.91 * NEWS: Record release date. 2018-10-18 Akim Demaille NEWS: update 2018-10-18 Akim Demaille examples: force stack resizing with unique_ptr In the previous commit we fixed a problem when the C++ stack was resized. The test was using ints. Let's add a test with someone quite touchy: unique_ptr * examples/variant-11.yy: Accept an argument, which is the number of numbers to send to the parser. * examples/variant-11.test: Check with many numbers. 2018-10-18 Akim Demaille lalr1.cc: fix stack symbol move In some casing, once we moved a stack symbol, we forget to mark the source stack symbol as emptied. As a consequence, it may be destroyed a second time. This happens when the stack has to be resized. * data/lalr1.cc (stack_symbol_type::stack_symbol_type): Record that the source was emptied. (stack_symbol_type::operator=): Likewise. * tests/c++.at (C++ Variant-based Symbols Unit Tests): Force the stack to be resized. Check its content. 2018-10-17 Akim Demaille doc: improve the doc of the examples * examples/README: here. 2018-10-17 Akim Demaille reader: recognize C++ even when it's not lalr1.cc or glr.cc * src/reader.c (grammar_rule_check_and_complete): If a user uses her own skeleton but sets the language to C++, recognize it as C++. 2018-10-17 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2018-10-17 Akim Demaille version 3.1.90 * NEWS: Record release date. 2018-10-16 Akim Demaille examples: don't generate the position/stack files * examples/variant-11.yy, examples/variant.yy: Require 3.2. 2018-10-16 Akim Demaille gnulib: update 2018-10-16 Akim Demaille pacify syntax-checks * lib/lbitset.c, tests/c++.at: here. * cfg.mk: Add exceptions. 2018-10-16 Akim Demaille regen 2018-10-16 Akim Demaille generate the default action only for C++ This commit adds restrictions to what was done in 01898726e27c8cf64f8fcea7f26f8ce62f3f5cf2 [1]. Rici Lake [2] has shown that it's risky to disable the pre-action, at least now. Also, generating the default $$ = $1 action can have bad effects in some cases [3]. The original change [1] was prompted for C++. Let's try it there only, for a start. We could restrict it further to lalr1.cc with variants, but we need to see in the wild how this change behaves. And it is not unreasonable to expect grammar files in C++ to behave better wrt types. See [1] https://lists.gnu.org/archive/html/bison-patches/2018-10/msg00050.html [2] https://lists.gnu.org/archive/html/bison-patches/2018-10/msg00061.html [3] https://lists.gnu.org/archive/html/bison-patches/2018-10/msg00066.html * src/getargs.c: Style changes. * src/reader.c (grammar_rule_check_and_complete): Complete only for C++. 2018-10-16 Akim Demaille regen 2018-10-16 Akim Demaille C++: let %require "3.2" disable the generation of obsolete files The files stack.hh and position.hh are deprecated. Rather than devoting specify %define variables to discard them (api.position.file and api.stack.file), and rather than having to use special rules when api.location.file is used, let's simply decide that from %require "3.2" onwards, these files will not be generated. The only noticeable thing here is that, in order to be able to check the behavior of %require "3.2", to have this version (which is still 3.1-*) to accept %require "3.2". * src/gram.h, src/gram.c (required_version): New. * src/parse-gram.y (version_check): Set it. * src/output.c (prepare): Pass it m4. * data/bison.m4 (b4_required_version_if): Receive it and use it. * data/location.cc, data/stack.hh: Replace the api.*.file with only required version comparison. * tests/input.at: No longer check api.stack.file and api.position.file. * NEWS, doc/bison.texi: Don't mention them. Document the %require 3.2 behavior. * tests/output.at: Use %require 3.2 instead. 2018-10-15 Akim Demaille java: bump to Java SE 7 macOS 10.14 no longer supports versions of Java earlier than 5. And Java 6 will be deprecated by the end of this year. So let's move our requirement to Java 7. Reported by Yu Yijun. https://lists.gnu.org/archive/html/bug-bison/2018-09/msg00060.html Suggested by Paul Eggert and Bruno Haible. http://lists.gnu.org/archive/html/bug-gnulib/2018-10/msg00094.html * configure.ac: Require Java 7, both compiler and runtime. 2018-10-15 Akim Demaille doc: do not advertise %nterm Reported by Rici Lake. http://lists.gnu.org/archive/html/bug-bison/2018-10/msg00000.html * NEWS, doc/bison.texi: here. 2018-10-14 Akim Demaille generate the default semantic action Currently, in C, the default semantic action is implemented by being always run before running the actual user semantic action. As a consequence, when the user action is run, $$ is already set as $1. In C++ with variants, we don't do that, since we cannot manipulate the semantic value without knowing its exact type. When variants are enabled, the only guarantee is that $$ is default contructed and ready to the used. Some users still would like the default action to be run with variants. Frank Heckenbach's parser in C++17 (http://lists.gnu.org/archive/html/bug-bison/2018-04/msg00011.html) provides this feature, but relying on std::variant's dynamic typing, which we forbid in lalr1.cc. The simplest seems to be actually generating the default semantic action (in all languages/skeletons). This makes the pre-action (that sets $$ to $1) useless. But... maybe some users depend on this, in spite of the comments that clearly warn againt this. So let's not turn this off just yet. * src/reader.c (grammar_rule_check_and_complete): Rename as... (grammar_rule_check_and_complete): this. Install the default semantic action when applicable. * examples/variant-11.yy, examples/variant.yy, tests/calc.at: Exercise the default semantic action, even with variants. 2018-10-14 Jannick xml2xhtml.xsl: add UTF-8 encoding To make arrows appear nicely in the browser. Currently it is shown as some garbled something in mine (Firefox). * data/xslt/xml2xhtml.xsl: here. 2018-10-14 Akim Demaille reader: reorder some calls to separate checks from assignments * src/reader.c (packgram): Move assignments to rules[ruleno] after the checks on the rule. 2018-10-14 Akim Demaille C++: style: add missing space before parens * data/c++.m4, data/lalr1.cc, data/stack.hh, data/variant.hh, * examples/variant-11.yy: here. 2018-10-14 Akim Demaille gnulib: update timevar See https://lists.gnu.org/archive/html/bug-gnulib/2018-10/msg00005.html. 2018-10-11 Akim Demaille style: modernize lib/vbitset.h 2018-10-11 Akim Demaille style: modernize lib/vbitset.c 2018-10-11 Akim Demaille style: modernize lib/lbitset.c 2018-10-11 Akim Demaille style: modernize lib/lbitset.h 2018-10-11 Akim Demaille style: modernize lib/ebitset.c 2018-10-11 Akim Demaille style: modernize lib/ebitset.h 2018-10-11 Akim Demaille style: modernize lib/bitsetv.c 2018-10-11 Akim Demaille style: modernize lib/bitsetv.h 2018-10-11 Akim Demaille style: modernize lib/bitsetv-print.c 2018-10-11 Akim Demaille style: modernize lib/bitsetv-print.h 2018-10-11 Akim Demaille style: modernize lib/bitset_stats.c 2018-10-11 Akim Demaille style: modernize lib/bitset_stats.h 2018-10-11 Akim Demaille style: modernize lib/bitset.c 2018-10-11 Akim Demaille style: modernize lib/bitset.h 2018-10-11 Akim Demaille style: modernize lib/bbitset.h 2018-10-11 Akim Demaille style: modernize lib/abitset.c 2018-10-10 Akim Demaille style: modernize lib/abitset.h 2018-10-09 Akim Demaille C++: issue a better CPP guard and Doxygen file command Currently we use "" as \file argument, and as base for the CPP guard. This is not nice when is absolute, in which case it is expected that the user will use api.location.include to get something nicer. If defined, use that name instead. * data/location.cc (b4_location_path): New. Use it. * tests/c++.at (Shared locations): Check the guard and Doxygen doc. 2018-10-09 Akim Demaille C++: remove stray empty line * data/stack.hh: here. 2018-10-09 Akim Demaille doc: spell check 2018-10-09 Akim Demaille doc: document api.*.file and the like * doc/bison.texi (Exposing the Location Classes): New. (%define Summary): Document api.location.file, api.location.include, api.stack.file and api.position.file. (C++ Bison Interface): stack.hh and position.hh are deprecated. 2018-10-07 Akim Demaille build: add missing gnulib libs * src/local.mk (LDADD): Here. 2018-10-07 Akim Demaille build: let timevar be dealt with by gnulib * lib/local.mk (lib_libbison_a_SOURCES): Remove timevar. 2018-10-07 Akim Demaille build: fix distcheck Now that distcheck no longer fails (see previous commit), let's address the shortcomings. * Makefile.am (CLEANDIRS, clean-local): New. * doc/local.mk, examples/calc++/local.mk, examples/local.mk, * examples/mfcalc/local.mk, examples/rpcalc/local.mk, * src/local.mk (CLEANDIRS): Get rid of Apple's *.dSYM directories. (CLEANFILES): Get rid of *.output files. * examples/variant-11.yy, examples/variant.yy: Don't generate any of the auxiliary files (location.hh and the like). 2018-10-07 Akim Demaille README: work around a nasty behavior of gettext `make update-po` runs: package_gnu="$(PACKAGE_GNU)"; \ test -n "$$package_gnu" || { \ if { if (LC_ALL=C find --version) 2>/dev/null | grep GNU >/dev/null; then \ LC_ALL=C find -L $(top_srcdir) -maxdepth 1 -type f \ -size -10000000c -exec grep 'GNU @PACKAGE@' \ /dev/null '{}' ';' 2>/dev/null; \ else \ LC_ALL=C grep 'GNU @PACKAGE@' $(top_srcdir)/* 2>/dev/null; \ fi; \ } | grep -v 'libtool:' >/dev/null; then \ package_gnu=yes; \ else \ package_gnu=no; \ fi; \ }; \ and based on the result, put GNU or not in the following line from bison.pot: # This file is distributed under the same license as the GNU bison package. It turns out that in my environment some log files had the 'GNU bison' string (note the lower case), but in distcheck, these files are no longer visible, so the generate bison.pot was different, and distcheck failed because we try to update bison.pot, which is read only in distcheck. The heuristics should look accept 'GNU Bison', not just 'GNU bison'. But let's please it to make sure we have our 'GNU'. * README: Mention 'GNU bison'. 2018-10-07 Akim Demaille NEWS: document api.*.file changes 2018-10-06 Akim Demaille c++: provide a means to control how location.hh is included Users may want to generate the location file elsewhere, say $top_srcdir/include/ast/location.hh. Yet, we should not generate `#include "$top_srcdir/include/ast/location.hh"` but probably something like `#include `, or `#include "ast/location.hh", or `#include `. It entirely depends on the compiler flags (-I/-isystem) that are used. Bison cannot guess what is expected, so let's give the user a means to tell how the location file should be included. * data/location.cc (b4_location_file): New. * data/glr.cc, data/lalr1.cc: Use it. 2018-10-06 Akim Demaille c++: support absolute api.location.file names In the case a user wants to create location.hh elsewhere, it can be helpful to define api.location.file to some possibly absolute path such as -Dapi.location.file='"$(top_srcdir)/include/ast/location.hh"'. Currently this does not work with `-o foo/parser.cc`, as we join foo/ and $(top_srcdir) together, the latter starting with slash. We should not try to do that in m4, manipulating file names is quite complex when you through Windows file name in. Let m4 delegate this to gnulib. * src/scan-skel.l (at_output): Accept up to two arguments. * data/bison.m4 (b4_output): Adjust. * tests/skeletons.at (Fatal errors but M4 continues producing output): Adjust to keep the error. * data/location.cc, data/stack.hh: Leave the concatenation to @output. * tests/output.at: Exercise api.location.file with an absolute path. 2018-10-06 Akim Demaille c++: when api.location.file is defined, don't generate stack.hh Make it easier to have fewer files. * data/stack.hh: Don't generate stack.hh when api.location.file is specified. * tests/calc++.at, tests/output.at: Adjust tests. 2018-10-06 Akim Demaille c++: make position.hh completely useless Let's put the definition of position into location.hh, there's no real value in keeping them separate: they are small, and share the same requirements. To help users transition to this new model, still generate position.hh by default, but as a simple include to location.hh. * data/location.cc (api.position.file): Accept only 'none' as possible value. (position.hh): Make it a stub. (location.hh): Adjust. (b4_position_define): Merge into... (b4_location_define): this. * data/glr.cc, data/lalr1.cc, tests/input.at, tests/output.at: Adjust. 2018-10-06 Akim Demaille c++: make stack.hh completely useless Let's completely deprecate stack.hh. Don't provide a means to give it a new name, allow only its removal. See https://lists.gnu.org/archive/html/bison-patches/2018-09/msg00151.html and https://lists.gnu.org/archive/html/bison-patches/2018-09/msg00182.html. * data/stack.hh: Reduce stack.hh to a simple comment explaining how to get rid of it. * data/lalr1.cc: Adjust * tests/input.at (%define file variables): Adjust. * tests/output.at: Remove cases where stack.hh was removed. 2018-10-06 Akim Demaille c++: add support for api.position.file and api.location.file * data/location.cc: Sort includes. (b4_position_file, b4_location_file): New. When there's a file for locations but not for positions, include the definition of position in the location file. * data/lalr1.cc (b4_shared_declarations): Include the position/location file when it exists. Otherwise, define the class. * data/glr.cc: Likewise. * tests/input.at (%define file variables): Check them. * tests/output.at (C++ output): Check various cases with api.position.file and api.location.file. 2018-10-06 Akim Demaille c++: provide control over the stack.hh file name It was not a good idea to generate the file stack.hh. It never was. But now we have to deal with backward compatibility: if we stop generating it, the build system of some build system will probably break. So offer the user a means to (i) decide what the name of the output file should be, and (ii) not generate this file at all (its content will be inline where the parser is defined). * data/lalr1.cc (b4_percent_define_check_file_complain) (b4_percent_define_check_file): New. * data/stack.hh: Generate the file only if api.stack.file is not empty. In that case, use it as file name. * data/lalr1.cc: Adjust to include the right file, or to include the definition of stack. * tests/calc.at, tests/output.at: Exercise api.stack.file. 2018-10-06 Akim Demaille tests: c++: don't fuse prefix and namespace They are not the same concept. It appears that we still consider that api.prefix is the default for api.namespace. We should stop that. * tests/local.at (AT_BISON_OPTION_PUSHDEFS, AT_BISON_OPTION_POPDEFS): Separate namespace and prefix. Adjust dependencies. 2018-10-06 Akim Demaille style: tests: factor file extension computations * tests/local.at (AT_LANG_HDR): New. * tests/calc.at, tests/headers.at, tests/synclines.at: Use it, and AT_LANG_EXT. 2018-10-06 Akim Demaille c++: style changes * data/lalr1.cc: Formatting changes. Remove duplicate definition of YY_NULLPTR. Add quotes to help Emacs track balanced parens. 2018-10-06 Akim Demaille lib: introduce xpath_join * lib/path-join.h, lib/path-join.c: New. * lib/local.mk: Adjust. * src/output.c: Use it. 2018-10-06 Akim Demaille CI: travis finally knows about llvm-toolchain-trusty-7 See https://github.com/travis-ci/apt-source-safelist/pull/392. 2018-10-06 Akim Demaille gnulib: update 2018-10-06 Akim Demaille doc: restore spello made on purpose 2018-10-05 Akim Demaille THANKS: add Josh Soref 2018-10-05 Josh Soref spelling: whether 2018-10-05 Josh Soref spelling: typedef 2018-10-05 Josh Soref spelling: troubleshooting 2018-10-05 Josh Soref spelling: transparent 2018-10-05 Josh Soref spelling: translated 2018-10-05 Josh Soref spelling: transitions 2018-10-05 Josh Soref spelling: transition 2018-10-05 Josh Soref spelling: tracking 2018-10-05 Josh Soref spelling: suppress 2018-10-05 Josh Soref spelling: succesful 2018-10-05 Josh Soref spelling: storage 2018-10-05 Josh Soref spelling: safely 2018-10-05 Josh Soref spelling: responsibility 2018-10-05 Josh Soref spelling: resources 2018-10-05 Josh Soref spelling: releases 2018-10-05 Josh Soref spelling: reduction 2018-10-05 Josh Soref spelling: reachable 2018-10-05 Josh Soref spelling: rawtoknumflag 2018-10-05 Josh Soref spelling: possibly 2018-10-05 Josh Soref spelling: persistent 2018-10-05 Josh Soref spelling: outputting 2018-10-05 Josh Soref spelling: otherwise 2018-10-05 Josh Soref spelling: occurrence 2018-10-05 Josh Soref spelling: namespace 2018-10-05 Josh Soref spelling: minimal 2018-10-05 Josh Soref spelling: invocations 2018-10-05 Josh Soref spelling: initialize 2018-10-05 Josh Soref spelling: incorrectly 2018-10-05 Josh Soref spelling: included 2018-10-05 Josh Soref spelling: illicit 2018-10-05 Josh Soref spelling: handling 2018-10-05 Josh Soref spelling: gratuitously 2018-10-05 Josh Soref spelling: grammar 2018-10-05 Josh Soref spelling: generation 2018-10-05 Josh Soref spelling: generate 2018-10-05 Josh Soref spelling: forewarn 2018-10-05 Josh Soref spelling: fnchange 2018-10-05 Josh Soref spelling: family 2018-10-05 Josh Soref spelling: extensions 2018-10-05 Josh Soref spelling: enum 2018-10-05 Josh Soref spelling: diagnostics 2018-10-05 Josh Soref spelling: determined 2018-10-05 Josh Soref spelling: detailed 2018-10-05 Josh Soref spelling: descriptive 2018-10-05 Josh Soref spelling: definitions 2018-10-05 Josh Soref spelling: declaration 2018-10-05 Josh Soref spelling: corrupted 2018-10-05 Josh Soref spelling: consuming 2018-10-05 Josh Soref spelling: consistently 2018-10-05 Josh Soref spelling: conflicts 2018-10-05 Josh Soref spelling: concatenation 2018-10-05 Josh Soref spelling: complete 2018-10-05 Josh Soref spelling: compatibility 2018-10-05 Josh Soref spelling: comparison 2018-10-05 Josh Soref spelling: combination 2018-10-05 Josh Soref spelling: characters 2018-10-05 Josh Soref spelling: builddir 2018-10-05 Josh Soref spelling: assoc 2018-10-05 Josh Soref spelling: appropriate 2018-10-05 Josh Soref spelling: alignment 2018-10-05 Josh Soref spelling: aggregate 2018-10-05 Josh Soref spelling: adjust 2018-10-05 Josh Soref spelling: additional 2018-10-05 Josh Soref spelling: accurately 2018-10-05 Akim Demaille README-hacking: details about make check-local 2018-10-05 Akim Demaille gnulib: update 2018-10-04 Akim Demaille main: fix error message for missing argument * src/getargs.c (getargs): Don't display any argv other that argv[0] when reporting a missing argument. * tests/bison.in: Neutralize path differences in stderr. * tests/input.at (Invalid number of arguments): New. 2018-09-30 Akim Demaille gnulib: move timevar to it * lib/timevar.c, lib/timevar.h, m4/timevar.m4: Remove. * gnulib: Update. * configure.ac: Adjust. * lib/timevar.def: Use lower case for the timevvars. Adjust dependencies. 2018-09-29 Akim Demaille style: comment changes * data/glr.cc, data/lalr1.cc: here. 2018-09-29 Akim Demaille gnulib: update 2018-09-29 Paul Eggert getargs: use LC_MESSAGES trick only on glibc * src/getargs.c (usage): Rely on setlocale (LC_MESSAGES, NULL) trick only on glibc, as POSIX does not specify the output of setlocale in this case, and the Gnulib localename module source code indicates that the trick works only on glibc. 2018-09-29 Paul Eggert uniqstr: avoid need for VLAs C11 no longer requires support for variable-length arrays, and VS2015 does not have them. Redo UNIQSTR_CONCAT to use a method that is simpler and better anyway. * src/uniqstr.c (uniqstr_vsprintf): Remove; no longer needed. * src/uniqstr.h (UNIQSTR_GEN_FORMAT, UNIQSTR_GEN_FORMAT_): * src/uniqstr.c (uniqstr_concat): New function. * src/uniqstr.h (UNIQSTR_CONCAT): Use it instead of using uniqstr_vsprintf. 2018-09-26 Akim Demaille doc: clean up the C++ section * doc/bison.texi: Minor fixes in typography. It is no longer require to pass --defines for C++ (it was addressed long ago). No longer refer to the `variant` define variable, it was replaced by `api.value.type variant`. Prefer nullptr to 0 for the null pointer. Use deftypeop for constructors. (Complete Symbols): Give the expected signature of yylex. Don't document the symbol_type constructors, as we want users to focus on make_TOKEN. Also show the case without locations. 2018-09-26 Akim Demaille CI: fixes for clang and asan Bison's test 464 (Syntax error as exception) fails on the CI. Do not use clang with asan on Ubuntu's libc++. https://bugs.llvm.org/show_bug.cgi?id=17379 * .travis.yml (Clang 7 libc++ and ASAN): New. (Clang 6 -O3 and libc++): Really use libc++. (Clang 5): Don't use libc++, nor asan (does not work either, same reason). 2018-09-24 Akim Demaille style: reduce scopes in muscle-tab.c 2018-09-24 Akim Demaille style: remove useless parens * data/bison.m4, data/glr.c, data/glr.cc, data/lalr1.cc, * data/lalr1.java, data/location.cc, data/yacc.c: Call b4_output_end without parens. 2018-09-24 Akim Demaille c++: fix warning message for automove * src/scan-code.l: Remove 'enabled'. Use only $k (numeric), even for named references, for clarity. * tests/c++.at: Adjust expectations. 2018-09-24 Akim Demaille style: minor refactoring * data/bison.m4: Formatting changes. * src/scan-code.l: Avoid loops, prefer standard string functions. (find_prefix_end): Be const correct. Avoid useless intermediate variables. (variant_add): Be const correct. (parse_ref): Prefer variable definitions to assignments. 2018-09-24 Akim Demaille CI: don't exit * .travis.yml: Prefer `false` to `exit`, as it completely ends the script (so we don't get the logs). 2018-09-24 Akim Demaille CI: really use Clang 3.3 and 3.4, not 5.0 * .travis.yml: Don't define CC/CXX, it does not work. Use `[[...]]` instead of `[...]`. Show the compiler versions. (Clang 3.3, Clang 3.4): Specify the path to avoid using /usr/local/clang-5.0.0/bin's clang. 2018-09-23 Akim Demaille CI: more compiler configurations * .travis.yml (GCC 8): Use sanitizers. (Clang 5 -O3): Remove, replaced by... (Clang 7 ASAN and libc++, Clang 6 -O3 and libc++): New. 2018-09-23 Akim Demaille build: rename and simplify the -std checks for C++ Too much code duplication. * m4/bison-cxx-std.m4: s/BISON_CXX_COMPILE_STDCXX/BISON_CXXSTD/. (BISON_CXXSTD): New. * configure.ac: Use it. 2018-09-23 Akim Demaille build: check for C++98 and 03 like the others * m4/bison-cxx-std.m4 (BISON_CXX_COMPILE_STDCXX_98) (BISON_CXX_COMPILE_STDCXX_03): New. * configure.ac: Use them. 2018-09-23 Akim Demaille build: use our own version of ax_check_link_flag The message on configure is misleading: checking whether the linker accepts -std=c++11... yes checking whether the linker accepts -std=c++14... yes checking whether the linker accepts -std=c++17... no It is the compiler that we check, not just the linker. * m4/ax_check_link_flag.m4: Remove. * m4/bison-check-compiler-flag.m4: New. * m4/bison-cxx-std.m4: Use it. 2018-09-23 Akim Demaille build: fix Autoconf macros to check for C++ standard flags * m4/bison-cxx-std.m4: Since now we link the program, we need a program: main was missing and linking was failing. 2018-09-23 Akim Demaille tests: fix a memory leak This has been bugging me for while. I was hard to reproduce: it worked only on GNU/Linux, probably because libc++ implements the small string optimization, while libstdc++ did not and actually allocated on the heap for this small string. See https://lists.gnu.org/archive/html/bison-patches/2018-09/msg00110.html. * tests/types.at (api.value.type): Do not provide a semantic value to EOF. 2018-09-22 Akim Demaille doc: work around Flex's use of 'register' The CI uses an old version of Flex. See 65fa634cdcfc5cf59b8b074670f488bba4df57cd. * doc/bison.texi (calc++/scanner.ll): Here. 2018-09-22 Akim Demaille timevar: don't declare getrusage if we don't use it This fails on MinGW. Reported by Simon Sobisch. http://lists.gnu.org/archive/html/bug-bison/2018-09/msg00058.html * lib/timevar.c: Don't provide default prototypes for functions we don't use. 2018-09-22 Akim Demaille timevar: get rid of a useless macro * lib/timevar.h (timevar_report): Rename as... (timevar_enabled): this. * lib/timevar.c (TIMEVAR_ENABLE): Remove. 2018-09-22 Akim Demaille timevar: introduce and use get_current_time * lib/timevar.c: here. Remove useless prototypes. (timevar_accumulate): Be const correct. 2018-09-22 Akim Demaille timevar: rename get_time as set_to_current_time * lib/timevar.c: here. 2018-09-22 Akim Demaille timevar: reduce scopes * lib/timevar.c: here. 2018-09-22 Akim Demaille timevar: document in the header, not in the implementation * lib/timevar.c: Move documentation from here... * lib/timevar.h: to there. 2018-09-22 Akim Demaille timevar: remove useless 'extern' for prototypes * lib/timevar.h, lib/timevar.c: here. 2018-09-22 Akim Demaille timevar: rename init_timevar as timevar_init * lib/timevar.h, lib/timevar.c: here. * src/main.c: Adjust. 2018-09-22 Akim Demaille timevar: we don't care about backward compatibility * lib/timevar.h, lib/timevar.c (get_run_time, print_time): Remove. 2018-09-22 Akim Demaille timevar: prefer #elif * lib/timevar.c: Use #if/#elif to be clearer about mutually exclusive cases. Indent CPP nested directives. 2018-09-22 Akim Demaille timevar: assume ANSI C Suggested by Bruno Haible. https://lists.gnu.org/archive/html/bug-gnulib/2018-09/msg00102.html * lib/timevar.c: Wow... This was still KnR C! 2018-09-22 Akim Demaille timevar: remove remains of GCC * lib/timevar.h, lib/timevar.c: Rename the header guard. Get rid of parts meant for GCC only. 2018-09-22 Akim Demaille news: c++: move semantics 2018-09-22 Akim Demaille c++: issue a warning with a value is moved several times Suggested by Frank Heckenbach. http://lists.gnu.org/archive/html/bug-bison/2018-09/msg00022.html * src/scan-code.l (parse_ref): Check multiple occurrences of rhs values. * tests/c++.at (Multiple occurrences of $n and api.value.automove): New. 2018-09-22 Akim Demaille c++: introduce api.value.automove Based on work by Frank Heckenbach. See http://lists.gnu.org/archive/html/bug-bison/2018-04/msg00000.html and http://lists.gnu.org/archive/html/bug-bison/2018-09/msg00019.html. * data/lalr1.cc (b4_rhs_value): Use YY_MOVE api.rhs.automove is set. * doc/bison.texi (%define Summary): Document api.rhs.automove. * examples/variant-11.yy: Use it. * tests/local.at (AT_AUTOMOVE_IF): New. * tests/c++.at (Variants): Check move semantics. 2018-09-22 Akim Demaille tests: c++: use a custom string type The forthcoming automove feature, to be properly checked, will require that we can rely on the value of a moved-from string, which is not something the C++ standard guarantees. So introduce our own wrapper. Suggested by Frank Heckenbach. https://lists.gnu.org/archive/html/bison-patches/2018-09/msg00111.html * tests/c++.at (Variants): Introduce and use a new 'string' class. 2018-09-22 Akim Demaille tests: prepare a test for automove The 'Variants' tests are well suited to check support for move, and in particular for the forthcoming automove feature. But the tests were written to show the best practice in C++98, using swap: list "," item { std::swap ($$, $1); $$.push_back ($3); } This cannot work with std::move. So, make this example simpler, based on regular assignment instead of swap, which is a regression for C++98 (as the new traces show), but will be an improvement for modern C++ with automove. * tests/c++.at (Variants): Stop using swap. We don't generate a header file, so remove the 'require' code section. Adjust expectations. 2018-09-20 Akim Demaille style: reduce scopes in gram.c * src/gram.c: here. 2018-09-20 Akim Demaille style: reduce scopes in reduce.c * src/reduce.c: Here. 2018-09-20 Akim Demaille gnulib: update 2018-09-20 Akim Demaille build: work around ICC's limitations Several types of failures. First, unable to pass the file name properly to the linker. ./synclines.at:416: $CC $CFLAGS $CPPFLAGS $LDFLAGS -o \"\\\"\" \"\\\"\".c $LIBS stderr: ld: cannot open output file "/"": No such file or directory stdout: Unable to save under such a file name. ./synclines.at:421: $CXX $CXXFLAGS $CPPFLAGS -c $LDFLAGS -o \"\\\"\" \"\\\"\".cc $LIBS stderr: error: can't open file "/"" for write compilation aborted for "\"".cc (code 1) Spurious output because of warning flags is failed to reject as an error during configure: ./headers.at:343: $CXX $CXXFLAGS $CPPFLAGS $LDFLAGS c-only.o cxx-only.o -o c-and-cxx || exit 77 --- /dev/null 2018-09-18 21:21:37.745649000 +0000 +++ /home/travis/build/akimd/bison/tests/testsuite.dir/at-groups/222/stderr 2018-09-18 21:28:17.291919519 +0000 @@ -0,0 +1,7 @@ +icpc: command line warning #10006: ignoring unknown option '-Wcast-align' +icpc: command line warning #10006: ignoring unknown option '-fparse-all-comments' +icpc: command line warning #10006: ignoring unknown option '-Wdocumentation' +icpc: command line warning #10006: ignoring unknown option '-Wnull-dereference' +icpc: command line warning #10006: ignoring unknown option '-Wnoexcept' +icpc: command line warning #10006: ignoring unknown option '-fno-color-diagnostics' +icpc: command line warning #10006: ignoring unknown option '-Wno-keyword-macro' stdout: * tests/local.at (AT_SKIP_IF_CANNOT_LINK_C_AND_CXX): Also ignore stderr, as with ICC we get * tests/synclines.at (syncline escapes): Don't link the output. 2018-09-19 Akim Demaille doc: fix typo Introduced in the previous commit. * doc/bison.texi: here. 2018-09-19 Akim Demaille style: use midrule only, not mid-rule The code was already using midrule only, never mid_rule. This is simpler to remember, and matches a similar change we made from look-ahead to lookahead. * NEWS, doc/bison.texi, src/reader.c, src/scan-code.h, src/scan-code.l * tests/actions.at, tests/c++.at, tests/existing.at: here. 2018-09-19 Akim Demaille style: use _foo for private macros, not foo_ We use both styles, let's stick to a single one. Autoconf uses the prefix one, let's do the same. * data/bison.m4, data/c++.m4, data/c-like.m4, data/lalr1.cc, * data/variant.hh, data/yacc.c: Rename all the b4_*_ macros as _b4_*. 2018-09-19 Akim Demaille build: don't accept a broken standard lib for C++ On the CI, we had failures such as: ./c++.at:401: $PREPARSER ./list stderr: ./list: error while loading shared libraries: libc++.so.1: cannot open shared object file: No such file or directory because we accepted `-std=c++ -stdlib=libc++` although libc++ is not installed on the machine. * m4/ax_check_compile_flag.m4 (AX_CHECK_COMPILE_FLAG): Rewrite as... * m4/bison-check-compile-flag.m4 (BISON_CHECK_COMPILE_FLAG): this, so that we use AC_LINK_IFELSE to check the compiler (and its std lib) instead of AC_COMPILE_IFELSE. 2018-09-18 Paul Eggert doc: document older compiler issues * doc/bison.texi (Compiler Requirements for GLR): Rename from Compiler Requirements. (I can't build Bison): Add FAQ for older compilers. 2018-09-18 Akim Demaille glr.c: work around ICC limitations The CI is littered with # -*- compilation -*- 423. regression.at:907: testing Dancer %glr-parser ... ./regression.at:907: bison -fno-caret -o dancer.c dancer.y ./regression.at:907: $BISON_C_WORKS stderr: stdout: ./regression.at:907: $CC $CFLAGS $CPPFLAGS $LDFLAGS -o dancer dancer.c $LIBS stderr: icc: command line warning #10006: ignoring unknown option '-Wcast-align' icc: command line warning #10006: ignoring unknown option '-fparse-all-comments' icc: command line warning #10006: ignoring unknown option '-Wdocumentation' icc: command line warning #10006: ignoring unknown option '-Wnull-dereference' icc: command line warning #10006: ignoring unknown option '-Wbad-function-cast' icc: command line warning #10006: ignoring unknown option '-fno-color-diagnostics' icc: command line warning #10006: ignoring unknown option '-Wno-keyword-macro' dancer.c(755): error #1628: function declared with "noreturn" does return } ^ dancer.c(761): error #1628: function declared with "noreturn" does return } ^ compilation aborted for dancer.c (code 2) ICC sees that `longjmp(buf, 1);` does not return, it sees that `abort();` does not either, but fails to see it for `longjmp(buf, 1); abort();` * data/glr.c (YYLONGJMP): Be even clearer on the fact this does not return. 2018-09-18 Akim Demaille TODO: more 2018-09-18 Akim Demaille CI: change strategy to pass CXXFLAGS and the like Putting them in the env is useless. We don't want to pass `CPPFLAGS="$CPPFLAGS"` to configure, as it means "set it to nothing" when $CPPFLAGS is not set, which is not what we want. This correctly started to use libc++, but it is not installed on the Ubuntu. We will see later if we can use it. * .travis.yml: Define CONFIGUREFLAGS, and pass it to configure. 2018-09-18 Akim Demaille CI: also use GCC 4.7 and 4.8 * .travis.yml (matrix): here. 2018-09-18 Akim Demaille CI: name the items of the matrix * .travis.yml: here. 2018-09-18 Akim Demaille CI: also check with ICC * build-aux/install-icc.sh: New. * .travis.yml (icc): New. Use -k to get as many errors as possible from the start. * src/complain.c (warnings_types): Use a more precise type. 2018-09-18 Akim Demaille CI: be sure to exit on failures a807cfa6eb1a5362ead0b7c99bdc8fd2f4f896da completely broke the whole point of having a CI: we always exit with success! 2018-09-18 Akim Demaille build: strengthen the C++ standard flag test On the CI, we have this spurious failure with clang 3.9 with -std=c++17: In file included from list.y:23: In file included from /usr/include/c++/4.8/iostream:39: In file included from /usr/include/c++/4.8/ostream:38: In file included from /usr/include/c++/4.8/ios:42: In file included from /usr/include/c++/4.8/bits/ios_base.h:41: In file included from /usr/include/c++/4.8/bits/locale_classes.h:40: In file included from /usr/include/c++/4.8/string:52: In file included from /usr/include/c++/4.8/bits/basic_string.h:2815: In file included from /usr/include/c++/4.8/ext/string_conversions.h:43: /usr/include/c++/4.8/cstdio:120:11: error: no member named 'gets' in the global namespace using ::gets; ~~^ This shows that our test, based on gl_WARN_ADD, is a joke. We have to really check for at least a bit of C++. * m4/ax_check_compile_flag.m4, m4/bison-cxx-std.m4: New. * configure.ac: Use them to make sure the compiler actually works. 2018-09-18 Akim Demaille tests: fix memory leak This was reported by ASAN on the CI. * tests/types.at (api.value.type): Don't set a semantic value to EOF. 2018-09-18 Akim Demaille glr.c: prefer true/false to 1/0 in C++ * data/glr.c: here. 2018-09-18 Akim Demaille doc: work around Flex's use of 'register' The CI uses an old version of Flex. * doc/bison.texi (calc++/scanner.ll): Here. 2018-09-18 Akim Demaille tests: fight G++ warnings about zero as null pointer constant In C++ pre C++11 it is standard practice to use 0 for the null pointer. But GCC pre 8 -std=c++98 with -Wzero-as-null-pointer-constant warns about this. So disable -Wzero-as-null-pointer-constant when compiling C++ pre 11. Let's do this in AT_DATA_SOURCE_PROLOGUE (which is pasted on top of all the test grammar files). Unfortunately, that shifts all the locations in the expected error messages, which would be too noisy. Instead, let's introduce testsuite.h, which can vary in length, and include it in AT_DATA_SOURCE_PROLOGUE. * tests/testsuite.h: New. Disable -Wzero-as-null-pointer-constant's warning with GCC pre 8, C++ pre 11. * tests/local.at (AT_DATA_SOURCE_PROLOGUE): Use it. * tests/atlocal.in (CPPFLAGS): Find it. * tests/local.mk: Ship it. * data/c.m4 (YY_NULLPTR): Prefer ((void*)0) to 0 in C. 2018-09-18 Akim Demaille CI: make sure `git describe` works For some reasons, the checkout on travis may not have any tags, so `git describe` fails, so bootstrap fails. * .travis.yml: If git describe fails, install some tag. 2018-09-18 Akim Demaille CI: install Doxygen * .travis.yml: here, so that its tests are not skipped. Remove valgrind: it's too expensive on the CI, and asan does the job. 2018-09-16 Akim Demaille style: prefer %D% in Automake files * tests/local.mk: Prefer %D%/ to tests/. 2018-09-15 Akim Demaille style: reduce scopes in complain.c 2018-09-15 Akim Demaille style: reduce scopes in tables.c * src/tables.c: here. * src/state.h: Formatting changes. 2018-09-15 Akim Demaille style: reduce scopes in graphviz.c 2018-09-15 Akim Demaille style: reduce scopes in LR0.c 2018-09-15 Akim Demaille style: reduce scopes in print_graph.c * src/print_graph.c: here. 2018-09-15 Akim Demaille doc: formatting changes * doc/bison.texi: No changes in the output. 2018-09-13 Akim Demaille tests: run the C++ tests on all the available standards This is much of course more efficient than in the matrix of the CI (or on our own machines), but a bit more tedious. * configure.ac (CXX03_CXXFLAGS, CXX11_CXXFLAGS, CXX14_CXXFLAGS) (CXX17_CXXFLAGS, CXX2A_CXXFLAGS, STDCXX_FLAGS): New. * tests/atlocal.in: Receive them. * tests/local.at (AT_FOR_EACH_CXX): New. * tests/c++.at: Use AT_FOR_EACH_CXX. 2018-09-13 Akim Demaille tests: allow to override variables with envvars * tests/atlocal.in: Allow the user to change interesting variables (CFLAGS, CXXFLAGS, etc.). 2018-09-13 Akim Demaille lalr1.cc: modern C++ no longer needs an assignment for symbols Reported by Frank Heckenbach. http://lists.gnu.org/archive/html/bug-bison/2018-03/msg00002.html Actually the assignment operator should never be needed: the C++98 requirements for vector::push_back is CopyInsertable, which does not require an assignment operator. However, libstdc++ shipped with GCC up to (and including) 6 uses the assignment operator (which affects Clang on top of libstdc++, but also ICC). So let's keep it for legacy C++. See https://gcc.godbolt.org/z/q0XXmC. * data/lalr1.cc (stack_symbol_type::operator=): Remove. * data/c++.m4 (basic_symbol::operator=): Ditto. * tests/c++.at (C++ Variant-based Symbols Unit Tests): Adjust. 2018-09-13 Akim Demaille lalr1.cc: support move semantics Modern C++ (i.e., C++11 and later) introduced "move only" types: types such as std::unique_ptr that can never be duplicated. They must never be copied (by assignments and constructors), they must be "moved". The implementation of lalr1.cc used to copy symbols (including their semantic values). This commit ensures that values are only moved in modern C++, yet remain compatible with C++98/C++03. Suggested by Frank Heckenbach, who provided a full implementation on top of C++17's std::variant. See http://lists.gnu.org/archive/html/bug-bison/2018-03/msg00002.html, and https://lists.gnu.org/archive/html/bison-patches/2018-04/msg00002.html. Symbols (terminal/non terminal) are handled by several functions that used to take const-refs, which resulted eventually in a copy pushed on the stack. With modern C++ (C++11 and later) the callers must use std::move, and the callees must take their arguments as rvalue refs (foo&&). In order to avoid duplicating these functions to support both legacy C++ and modern C++, let's introduce macros (YY_MOVE, YY_RVREF, etc.) that rely on copy-semantics for C++98/03, and move-semantics for modern C++. That's easy for inner types, when the parser's functions pass arguments to each other. Functions facing the user (make_NUMBER, make_STRING, etc.) should support both rvalue-refs (for instance to support move-only types: make_INT (std::make_unique (1))), and lvalue-refs (so that we can pass a variable: make_INT (my_int)). To avoid the multiplication of the signatures (there is also the location), let's take the argument by value. See: https://lists.gnu.org/archive/html/bison-patches/2018-09/msg00024.html. * data/c++.m4 (b4_cxx_portability): New. (basic_symbol): In C++11, replace copy-ctors with move-ctors. In C++11, replace copies with moves. * data/lalr1.cc (stack_symbol_type, yypush_): Likewise. Use YY_MOVE to avoid useless copies. * data/variant.hh (variant): Support move-semantics. (make_SYMBOL): In C++11, in order to support both read-only lvalues, and rvalues, take the argument as a copy. * data/stack.hh (yypush_): Use rvalue-refs in C++11. * tests/c++.at: Use move semantics. * tests/headers.at: Adjust to the new macros (YY_MOVE, etc.). * configure.ac (CXX98_CXXFLAGS, CXX11_CXXFLAGS, CXX14_CXXFLAGS) (CXX17_CXXFLAGS, ENABLE_CXX11): New. * tests/atlocal.in: Receive them. * examples/variant.yy: Don't define things in std. * examples/variant-11.test, examples/variant-11.yy: New. Check the support of move-only types. * examples/README, examples/local.mk: Adjust. 2018-09-12 Akim Demaille tests: factor the definition of full compilation * tests/local.at (AT_LANG_EXT): New. (AT_FULL_COMPILE): Simplify. 2018-09-10 Akim Demaille CI: use clang with libc++ GCC uses libstdc++. Let's also check libc++. * .travis.yml: here. 2018-09-10 Akim Demaille CI: use address sanitizer * .travis.yml (matrix): Use the latest (available) clang with asan. 2018-09-10 Akim Demaille CI: sort the matrix in reverse-chronological There are only three builds at a time: show the result of modern compilers first. * .travis.yml (matrix): Sort in reverse-chronological. 2018-09-10 Akim Demaille build: use -fparse-all-comments with -Wdocumentation Clang checks only /** ... */ comments without this flag. * configure.ac (warn_common): Also check -fparse-all-comments. 2018-09-09 Akim Demaille TODO: minor updates 2018-09-09 Akim Demaille build: fix support for --disable-dependency-tracking Reported by Juan Manuel Guerrero. https://lists.gnu.org/archive/html/bug-bison/2014-07/msg00000.html. * examples/local.mk (%D%/extracted.stamp): Make sure the output directory exists. * examples/extexi (process): Likewise. 2018-09-09 Akim Demaille configure.ac: fix definition of NO_EXCEPTIONS_CXXFLAGS * configure.ac: Always define it, not just when --enable-gcc-warnings is passed. 2018-09-09 Akim Demaille skeletons: style/comment changes * data/c++.m4, data/c.m4, data/glr.c: Here. 2018-09-09 Akim Demaille variant: indent better the generated code * data/variant.hh (b4_basic_symbol_constructor_declare) (b4_basic_symbol_constructor_define): here. 2018-09-09 Akim Demaille lalr1.cc: don't generate useless constructors when variant is used This generates less code, which is nicer to read, but also takes less chances with compilers such as G++ 4.8 that are too strict and check "dead code" (templated code that is not instantiated). * data/c++.m4 (b4_symbol_type_declare, b4_symbol_type_define): When variants are used, don't generate code meant for non variants. 2018-09-09 Akim Demaille CI: Clang 6.0 is not available But Clang 3.3 and 3.4 are. * .travis.yml (addons): Remove, it appears to be ignore if the matrix also defines it. (matrix): Update. 2018-09-08 Akim Demaille build: work around warnings in Flex See ea0db44fedc8d5cbdc5c3180bef0285d7ae83803. We also need to disable the warning in the examples (but don't want to clutter the documentation with such details). * doc/bison.texi (scanner.ll): Disable Clang's -Wdocumentation. While at it, hide the other kludges. 2018-09-08 Akim Demaille CI: more compiler configurations * .travis.yml: here. 2018-09-08 Akim Demaille configure: reveal the name of the Valgrind suppression file we use * configure.ac: here. * build-aux/Linux.valgrind (libstdcxx_init): New. 2018-09-08 Akim Demaille build: work around warnings in Flex 2.5.35 That's the version on Ubuntu Precise. See also 1dac131ec45ffa1e382319a94640c65bd10f6aa5. * src/flex-scanner.h: Disable -Wdocumentation. * doc/bison.texi: Turn off a warning triggered by Flex 2.6.4. 2018-09-08 Akim Demaille CI: show the version of the tools we use We have failures on Flex output, which are probably related to an old release. Let's check. In file included from src/scan-code-c.c:3: src/scan-code.c:2198:21: error: empty paragraph passed to '@param' command [-Werror,-Wdocumentation] * @param line_number ~~~~~~~~~~~~~~~~~^ * .travis.yml: here. 2018-09-08 Akim Demaille CI: run more maintainer tests and show the logs Running all these tests might be overkill: it is very long, and don't need full portability checks. Besides, some tests under Valgrind are too slow and get killed by the CI (timeout of 10min without output). * .travis.yml: here. 2018-09-08 Akim Demaille CI: enable compiler warnings * .travis.yml: here. * README-hacking: We no longer aim at K&R C. 2018-09-06 Akim Demaille build: work around GCC warnings on Flex code See ef98967ada3c1cd48c177d7349e65a709bb49b97. * src/flex-scanner.h: Disable -Wnull-dereference for GCC 6+. 2018-09-06 Akim Demaille tests: fix target naming convention We have some maintainer-check-foo and some maintainer-foo-check. Keep only the former. * tests/local.mk (maintainer-push-check, maintainer-xml-check) (maintainer-release-check): Rename as... (maintainer-check-push, maintainer-check-xml) (maintainer-check-release): these. 2018-09-06 Akim Demaille tests: fix variable naming convention Most of our variables for C++ flags are named FOO_CXXFLAGS, not CXXFLAGS_FOO. * configure.ac, tests/atlocal.in, tests/calc.at (NO_EXCEPTIONS_CXXFLAGS): Rename as... (CXXFLAGS_NO_EXCEPTIONS): this. 2018-09-06 Akim Demaille tests: fix maintainer-check-g++ make recipe Clang++ issues warnings when it's used to compile C. This make target is precisely checking whether we can do that. * configure.ac (NO_DEPRECATED_CXXFLAGS): New. * tests/atlocal.in: Use it. 2018-09-06 Akim Demaille tests: fix maintainer-check-valgrind make recipe * tests/local.mk (maintainer-check-valgrind): Run the with Valgrind when it's available, not the converse. 2018-09-06 Akim Demaille CI: prepare for travis * .travis.yml: New. 2018-09-06 Akim Demaille examples: beware of shell portability issues This completes 2d7e7438024e47650c3a0c9f5f313c6eb6acae2d. Some shells don't grok "local var=`cmd`" very well: they need the rhs to be quoted. ./examples/test: 72: local: you.,: bad variable name FAIL examples/variant.test (exit status: 2) Reported by Étienne Renault. * examples/test (run): Quote the values in 'local' assignments. 2018-09-04 Akim Demaille tests: style changes * tests/c++.at: Formatting changes. Use 'using' to shorten the code. 2018-09-02 Akim Demaille tests: disable GCC7 warnings for some tests With GCC7 we have warnings (false positive): x8.c: In function 'x8_parse': x8.c:1233:16: error: 'yylval' may be used uninitialized in this function [-Werror=maybe-uninitialized] yylval = *yypushed_val; ~~~~~~~^~~~~~~~~~~~~~~ x8.c: In function 'x8_pull_parse': x8.c:1233:16: error: 'yylval' may be used uninitialized in this function [-Werror=maybe-uninitialized] yylval = *yypushed_val; ~~~~~~~^~~~~~~~~~~~~~~ See also 9645a2b20ee7cbfa8bb4ac2237f87d598afe349c. * tests/local.at (AT_PUSH_IF): New. (AT_BISON_OPTION_POPDEFS): Pop it, and pop AT_PURE_IF. * tests/headers.at (Several parsers, Several parsers): Disable these warnings when in push parser. 2018-09-02 Akim Demaille C++: don't issue the definition of symbol_type when not used Currently, in glr.cc, we emit the definitions of basic_symbol and symbol_type, although there are not used. * data/c++.m4 (b4_public_types_declare): Extract these definitions from here, and move them... (b4_symbol_type_declare): here. (b4_public_types_declare): Also remove the definition of the symbol constructors. * data/lalr1.cc (b4_shared_declarations): Adjust: call b4_symbol_type_declare and b4_symbol_constructor_declare. 2018-09-02 Akim Demaille examples: beware of shell portability issues Some shells don't grok `local var=$val` very well: they need the rhs to be quoted. ./examples/test: 66: local: you.,: bad variable name FAIL examples/variant.test (exit status: 2) Reported by Étienne Renault. * examples/test (run): Quote the values in 'local' assignments. 2018-08-31 Akim Demaille C++: leave 'inline' on the definition, not the declaration This is for consistency with the other uses of 'inline' in the C++ skeletons. On examples/variant.yy, this change gives: --- examples/variant.hh 2018-08-31 07:16:57.214222580 +0200 +++ examples/variant.hh 2018-08-31 07:19:52.285431997 +0200 @@ -444,15 +444,15 @@ typedef basic_symbol symbol_type; // Symbol constructors declarations. - static inline + static symbol_type make_END_OF_FILE (const location_type& l); - static inline + static symbol_type make_TEXT (const ::std::string& v, const location_type& l); - static inline + static symbol_type make_NUMBER (const int& v, const location_type& l); @@ -945,19 +945,23 @@ }; return static_cast (yytoken_number_[type]); } + // Implementation of make_symbol for each symbol type. + inline parser::symbol_type parser::make_END_OF_FILE (const location_type& l) { return symbol_type (token::END_OF_FILE, l); } + inline parser::symbol_type parser::make_TEXT (const ::std::string& v, const location_type& l) { return symbol_type (token::TEXT, v, l); } + inline parser::symbol_type parser::make_NUMBER (const int& v, const location_type& l) { @@ -967,7 +971,7 @@ } // yy -#line 971 "examples/variant.hh" // lalr1.cc:380 +#line 975 "examples/variant.hh" // lalr1.cc:380 and no changes on variant.cc. * data/c++.m4 (b4_public_types_define): Formatting changes. * data/variant.hh (b4_symbol_value_template_, b4_symbol_constructor_declare_): Move the 'inline' from declaration to implementation. 2018-08-30 Akim Demaille c++: style changes Instead of parser::stack_symbol_type::stack_symbol_type (const stack_symbol_type& that) : super_type (that.state, that.location) { value = that.value; } generate parser::stack_symbol_type::stack_symbol_type (const stack_symbol_type& that) : super_type (that.state, that.value, that.location) {} * data/lalr1.cc (stack_symbol_type): Improve the copy ctor, when not using the variants. (yypush_): Rename arguments for clarity. 2018-08-30 Akim Demaille c++: the assignment operator does not have to be const * data/lalr1.cc (stack_symbol_type::operator=): Don't copy the argument, move it. 2018-08-30 Akim Demaille tests: style changes * tests/c++.at (C++ Variant-based Symbols): Rename as... (C++ Variant-based Symbols Unit Tests): this. Comment/style changes. 2018-08-27 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2018-08-27 Akim Demaille version 3.1 * NEWS: Record release date. 2018-08-27 Akim Demaille build: tabs are ok in a Makefile * cfg.mk: TABs are ok in examples/calc++/Makefile. 2018-08-26 Akim Demaille C++: make sure the generated header is self container See the previous commit. * data/lalr1.cc: Be sure to define YY_NULLPTR. * tests/headers.at: Check the case that was failing. 2018-08-26 Akim Demaille tests: check that headers are sane The header generated for variants with assertions but without locations, is not self-contained. Prepare a check for this. * tests/headers.at (Sane headers): New, extracted from... (Several parsers): here. 2018-08-25 Akim Demaille "C++: restore copy-constructor for stack_symbol_type Benchmarks show that it is more efficient to keep this copy constructor, rather than forcing the use of the default constructor and then assignment. This reverts commit 7ab25ad0208d00f509613e1e151aa3043cf2862f. 2018-08-25 Akim Demaille examples: calc++: a Makefile and a README * examples/calc++/Makefile, examples/calc++/README: New. * examples/calc++/local.mk: Ship and install them. * doc/bison.texi: Formatting changes. 2018-08-25 Akim Demaille gnulib: update 2018-08-25 Jiahao Li variant: fix uninitialized memory access in `variant<>` Currently, in bison's C++ parser template (`lalr.cc`), the `variant<>` struct's `build()` method uses placement-new in the form `new (...) T` to initialize a variant type. However, for POD variant types, this will leave the memory space uninitialized. If we subsequently tries to `::move` into a variant object in such state, the call can trigger clang's undefined behavior sanitizer due to accessing the uninitialized memory. https://lists.gnu.org/archive/html/bison-patches/2018-08/msg00098.html * data/variant.hh (build): Always initialize the stored value. 2018-08-24 Akim Demaille NEWS: update 2018-08-24 Akim Demaille examples: calc++: minor improvements * doc/bison.texi (A Complete C++ Example): Prefer throw exceptions from the scanner. Show the invalid characters. Since the scanner sends exceptions, it no longer needs to report errors, so we can get rid of the driver's routine to report error, do it in yyerror. Use @group/@end group to improve rendering. 2018-08-24 Akim Demaille examples: calc++: make sure the file name in location is set Reported by Hans Åberg. http://lists.gnu.org/archive/html/bug-bison/2018-08/msg00039.html * doc/bison.texi (A Complete C++ Example): Move the token's location from the scanner to the driver. 2018-08-24 Akim Demaille examples: calc++: remove prefixes This example uses the calcxx_ prefix for each class. That's uselessly heavy. * doc/bison.texi (A Complete C++ Example): Simplify the class names. Since now 'driver' denotes the class, use 'drv' for the values. Formatting changes. 2018-08-23 Akim Demaille examples: fix the leading empty line * examples/extexi: Really avoid the first empty line. Remove useless `next`. 2018-08-23 Akim Demaille examples: shorten the name of the calc++ files * doc/bison.texi: Turn the calc++- prefix into calc++/. * examples/extexi (%file_wanted): Replace with (&file_wanted): this. * examples/calc++/local.mk: Adjust. 2018-08-23 Akim Demaille tests: disable -Wmaybe-uninitialized in some tests On these tests, at -O2 and above, GCC 8 complains that yylval may be uninitialized. But it seems wrong: it is initialized. Rather than turning off the warning in the skeleton (hence possibility hiding relevant warnings of user parsers), let's turn it off in the tests only. 163: parse.error=verbose and consistent errors: FAILED (conflicts.at:625) 165: parse.error=verbose and consistent errors: lr.default-reduction=consistent FAILED (conflicts.at:635) 166: parse.error=verbose and consistent errors: lr.default-reduction=accepting FAILED (conflicts.at:641) 167: parse.error=verbose and consistent errors: lr.type=canonical-lr FAILED (conflicts.at:645) 168: parse.error=verbose and consistent errors: parse.lac=full FAILED (conflicts.at:650) 169: parse.error=verbose and consistent errors: parse.lac=full lr.default-reduction=accepting FAILED (conflicts.at:655) We get: input.c: In function 'yyparse': input.c:980:9: error: 'yylval' may be used uninitialized in this function [-Werror=maybe-uninitialized] YYSTYPE yylval YY_INITIAL_VALUE (= yyval_default); ^~~~~~ cc1: all warnings being treated as errors See https://lists.gnu.org/archive/html/bison-patches/2018-08/msg00063.html. * tests/conflicts.at (AT_CONSISTENT_ERRORS_CHECK): Disable -Wmaybe-uninitialized. 2018-08-19 Akim Demaille doc: clarify that the push parser object can be reused Suggested by Rici Lake. https://lists.gnu.org/archive/html/bug-bison/2018-08/msg00033.html * doc/bison.texi: Complete description of the first node in the main @menu. (Push Decl): Remove the 'experimental' warnings about push parser. Clarify that the push parser object can be reused in several parses. 2018-08-19 Akim Demaille lalr1.cc: support compilation with disabled support for exceptions Reported by Brooks Moses http://lists.gnu.org/archive/html/bison-patches/2018-02/msg00000.html * data/lalr1.cc (YY_EXCEPTIONS): New. Use it to disable try/catch clauses. * doc/bison.texi (C++ Parser Interface): Document it. * configure.ac (CXXFLAGS_NO_EXCEPTIONS): New. * tests/atlocal.in: Receive it. * tests/local.at (AT_FULL_COMPILE, AT_LANG_COMPILE): Accept a new argument, extra compiler flags. * tests/calc.at: Run the C++ calculator with exception support disabled. 2018-08-19 Akim Demaille examples: add empty lines Currently the examples are too dense, let's put empty lines where '#line' would be issued. And also remove some spurious empty lines (remains from @group, @end group, etc.). * examples/extexi: Do that. * examples/local.mk (extexiFLAGS): Rename as... (EXTEXIFLAGS): this. 2018-08-19 Akim Demaille examples: check the variant example * examples/mfcalc/local.mk, examples/rpcalc/local.mk: Define the programs in a more natural order, source, preproc, then linker. * examples/test: Be ready to work on programs that are not in a subdir. * examples/variant.test: New. * examples/local.mk: Use it. * examples/variant.yy: Don't use 0 for nullptr. Use a more natural output for a list of string. 2018-08-18 Akim Demaille C++: fix portability issue with MSVC 2017 Visual Studio issues a C4146 warning on '-static_cast(rhs)'. The code is weird, probably to cope with INT_MIN. Let's go back to using std::max (whose header is still included in position.hh...) like originally, but with the needed casts. Reported by 長田偉伸, and with help from Rici Lake. See also http://lists.gnu.org/archive/html/bug-bison/2013-02/msg00000.html and commit 75ae8299840bbd854fa2474d38402bbb933c6511. * data/location.cc (position::add_): Take min as an int. Use std::max. While here, get rid of a couple of useless inlines. 2018-08-18 Akim Demaille build: fix concurrent build failure Reported by Dengke Du and Robert Yang. https://lists.gnu.org/archive/html/bison-patches/2017-07/msg00000.html * src/local.mk (src/yacc): Make sure the directory exists. 2018-08-18 Akim Demaille lalr1.cc: remove debug comment * data/lalr1.cc: Remove a comment about indentation. I'm not sure it would be nice to indent even more, it's already quite of the right. 2018-08-18 Akim Demaille doc: fix the name of the GFDL section Reported by dine <2500418497@qq.com>. http://lists.gnu.org/archive/html/bug-bison/2016-10/msg00000.html I mirrored what the Coreutils do. * doc/bison.texi (Copying This Manual): Rename as... (GNU Free Documentation License): this, since that the name we used in the preamble. 2018-08-18 Akim Demaille doc: clarify the destructor selection example Reported by Gary L Peskin. http://lists.gnu.org/archive/html/help-bison/2016-02/msg00000.html * doc/bison.texi (Destructor Decl): here. 2018-08-18 Akim Demaille portability: don't use _Pragma with ICC ICC defines __GNUC__ [1], but does not support GCC's _Pragma for diagnostics. As a matter of fact, I believe it does not support _Pragma at all (only #pragma) [2]. Reported by Maxim Prohorenko. https://savannah.gnu.org/support/index.php?108339 [1] https://software.intel.com/en-us/cpp-compiler-18.0-developer-guide-and-reference-gcc-compatibility-and-interoperability [2] https://software.intel.com/en-us/cpp-compiler-18.0-developer-guide-and-reference-pragmas * data/c.m4 (YY_IGNORE_MAYBE_UNINITIALIZED_BEGIN): Exclude ICC from the club. 2018-08-18 Akim Demaille doc: typed mid-rule actions * doc/bison.texi (Mid-Rule Actions): Restructure to insert... (Typed Mid-Rule Actions): this new section. Move the manual translation of mid-rule actions into regular actions to... (Mid-Rule Action Translation): here. 2018-08-18 Akim Demaille escape properly the file names in #line for printer/destructor Reported by Jannick. http://lists.gnu.org/archive/html/bug-bison/2017-05/msg00001.html "Amusingly" enough, we have the same problem with %defines when the parser file name has backslashes or quotes: we generate #includes with an incorrect C string. * src/output.c (prepare_symbol_definitions): Escape properly the file names before passing them to M4. * data/bison.m4, data/lalr1.cc: Don't simply put the file name between two quotes (that should have been strong enough a smell...), expect the string to be properly quoted. * tests/synclines.at: New tests to check this. 2018-08-18 Akim Demaille tests: fix title and improve quoting * tests/synclines.at: here. Also, prefer '%code' to ;%{...%}' for yylex/yyerror prototypes. 2018-08-18 Akim Demaille style: reduce scopes * src/output.c: here. 2018-08-18 Akim Demaille style: reduce scopes * src/scan-code.l: here. 2018-08-18 Akim Demaille regen 2018-08-18 Akim Demaille style: reduce scopes * src/parse-gram.y: Declare iterator within the for-loop. 2018-08-18 Akim Demaille tests: style changes * tests/c++.at, tests/local.at: Formatting and title changes. 2018-08-17 Akim Demaille reader: simplify the search of the start symbol Suggested by Paul Eggert. * src/reader.c (find_start_symbol): Don't check 'res', we know it is not null. That suffices to avoid the GCC warnings. * bootstrap.conf: We don't need 'assume', which doesn't exist anyway. 2018-08-15 Akim Demaille c++: fix GCC8 warnings about uninitialized values In 0931d14728fb4a2272399f2c927ae78e2607b4fb I removed too many initializations from some ctors: some were not about base ctors, but about member variables. In fact, more of them were missing to please GCC 8. While at it, generate more natural code for C++ without variant: instead of template parser::basic_symbol::basic_symbol (const basic_symbol& other) : Base (other) , value () { value = other.value } generate template parser::basic_symbol::basic_symbol (const basic_symbol& other) : Base (other) , value (other.value) {} * data/c++.m4 (basic_symbol::basic_symbol): Always initialize 'value', it might be a POD without a ctor. * data/lalr1.cc (stack_symbol_type::stack_symbol_type): Likewise. * data/variant.hh (variant::variant): Default initialize the buffer too. 2018-08-15 Akim Demaille tests: style: use %empty * tests/conflicts.at: here. 2018-08-15 Akim Demaille tests: fix warnings in push mode Fix warning with GCC 8, -DNDEBUG. 422. push.at:83: testing Multiple impure instances ... input.y: In function 'main': input.c:1022:12: error: potential null pointer dereference [-Werror=null-dereference] if (!yyps->yynew && yyps->yyss != yyps->yyssa) ~~~~^~~~~~~ * data/yacc.c (pstate_delete): Do nothing if called on null pointer. 2018-08-15 Akim Demaille c++: avoid GCC 8 warnings GCC 8 issues warnings whose root cause was a bit hard to find. calc.cc: In member function 'virtual int yy::parser::parse()': calc.cc:810:18: warning: '*((void*)& +8)' may be used uninitialized in this function [-Wmaybe-uninitialized] , location (l) ^ calc.cc: In member function 'void yy::parser::yypush_(const char*, yy::parser::stack_symbol_type&)': calc.cc:810:18: warning: '*((void*)& +8)' may be used uninitialized in this function [-Wmaybe-uninitialized] , location (l) ^ calc.cc: In member function 'void yy::parser::yypush_(const char*, yy::parser::state_type, yy::parser::symbol_type&)': calc.cc:810:18: warning: '*((void*)& +8)' may be used uninitialized in this function [-Wmaybe-uninitialized] , location (l) ^ The problem is with locations that don't have a constructor, such as Span (in calc.cc) which is POD. It is POD on purpose: so that we can use that structure to test glr.cc which cannot use non POD in its (C) stacks. * data/c++.m4 (basic_symbol): Also ensure that 'location' is initialized. 2018-08-15 Akim Demaille tests: avoid compiler warnings * tests/calc.at (AT_CALC_MAIN): Declare yyparse and operator<< in an unnamed namespace to avoid "not declared" warnings (clang -Weverything). Remove useless prototypes. 2018-08-15 Akim Demaille build: work around GCC warnings on Flex code With GCC 7.3.0 and Flex 2.6.4, we get warnings on all the generated scanners: examples/calc++/calc++-scanner.cc: In function 'void yyrestart(FILE*)': examples/calc++/calc++-scanner.cc:1611:20: error: potential null pointer dereference [-Werror=null-dereference] /* %endif */ ~~~~~~~~~~~ ^ examples/calc++/calc++-scanner.cc:1607:19: error: potential null pointer dereference [-Werror=null-dereference] /* %if-c-only */ ~~~~~~~~~~~~~~~ ^ examples/calc++/calc++-scanner.cc:1611:20: error: potential null pointer dereference [-Werror=null-dereference] /* %endif */ ~~~~~~~~~~~ ^ examples/calc++/calc++-scanner.cc:1607:19: error: potential null pointer dereference [-Werror=null-dereference] /* %if-c-only */ ~~~~~~~~~~~~~~~ ^ cc1plus: all warnings being treated as errors Obviously the lines are incorrect, and the warnings are emitted twice. Still, let's get rid of these warnings. * doc/bison.texi, src/flex-scanner.h: Disable these warnings in code generated by Flex. 2018-08-15 Akim Demaille fix incorrect C code Commit 3df32101e7978eaafa63bce8908de3dcae4d9cda introduced invalid C code. Caught by GCC 7.3.0. * bootstrap.conf (gnulib_modules): We need assume. * src/reader.c (find_start_symbol): Fix the signature (too much C++, sorry...). Prefer 'assume' to 'assert', so that we don't have these warnings even when NDEBUG is defined. 2018-08-14 Akim Demaille examples: fix Englishoes * examples/README: Fix my mistakes. 2018-08-14 Akim Demaille examples: ship and install variant.yy This file was meant to be shown as an example. Install it. * README, data/README: Put Emacs metadata in the final section. * examples/README: New. * examples/variant.yy: Use %empty. * examples/local.mk: Install both these files. 2018-08-14 Akim Demaille C++: remove useless copy-constructor We currently generate copy constructors such as the following one (taken from examples/variant.yy): parser::stack_symbol_type::stack_symbol_type (const stack_symbol_type& that) : super_type (that.state, that.location) { switch (that.type_get ()) { case 3: // TEXT case 8: // item value.copy< ::std::string > (that.value); break; case 7: // list value.copy< ::std::vector > (that.value); break; case 4: // NUMBER value.copy< int > (that.value); break; default: break; } } they are actually useless: we never need it. * data/lalr1.cc: Don't generate the stack_symbol_type copy ctor. 2018-08-14 Akim Demaille C++: symbol constructors: add a missing reference Fix a typo so that instead of basic_symbol::basic_symbol (typename Base::kind_type t, const int v) we now generate basic_symbol::basic_symbol (typename Base::kind_type t, const int& v) * data/variant.hh (b4_basic_symbol_constructor_declare) (b4_basic_symbol_constructor_define): Add missing reference. 2018-08-14 Akim Demaille C++: remove useless calls to the base default constructor * data/c++.m4, data/stack.hh, data/variant.hh: here. 2018-08-14 Akim Demaille C++: prefer size_type to unsigned for indexes * data/stack.hh (size_type): New, based on the container type. 2018-08-14 Akim Demaille regen 2018-08-14 Akim Demaille style: m4: remove useless reference to 'int' in integral types * m4/cxx.m4: Prefer 'unsigned' to 'unsigned int'. 2018-08-14 Akim Demaille style: doc: remove useless reference to 'int' in integral types * doc/bison.texi: Prefer 'unsigned' to 'unsigned int'. Likewise for long and short. 2018-08-14 Akim Demaille style: lib: remove useless reference to 'int' in integral types * lib/abitset.c, lib/bbitset.h, lib/bitset.c, lib/bitset.h, * lib/bitset_stats.c, lib/bitsetv-print.c, lib/bitsetv.c, * lib/bitsetv.h, lib/ebitset.c, lib/lbitset.c, lib/timevar.c, * lib/vbitset.c: Prefer 'unsigned' to 'unsigned int'. Likewise for long and short. 2018-08-14 Akim Demaille style: src: remove useless reference to 'int' in integral types * src/AnnotationList.c, src/AnnotationList.h, src/InadequacyList.h, * src/closure.c, src/closure.h, src/gram.c, src/gram.h, src/ielr.c, * src/location.c, src/output.c, src/reader.c, src/relation.c, * src/scan-code.l, src/scan-gram.l, src/tables.c, src/tables.h: Prefer 'unsigned' to 'unsigned int'. Likewise for long and short. 2018-08-14 Akim Demaille style: tests: remove useless reference to 'int' in integral types * tests/actions.at, tests/cxx-type.at: Prefer 'unsigned' to 'unsigned int'. Likewise for long and short. 2018-08-14 Akim Demaille style: data: remove useless reference to 'int' in integral types * data/c.m4, data/glr.c, data/yacc.c: Prefer 'unsigned' to 'unsigned int'. Likewise for long and short. 2018-08-12 Akim Demaille gnulib: update * bootstrap.conf: gnulib_mk is defined again by bootstrap. 2018-08-12 Akim Demaille doc: avoid type aliases * doc/bison.texi (C++ Location Values): Use 'unsigned' instead of 'uint'. 2018-08-12 Akim Demaille doc: remove the "experimental" warning for some features Several features were flagged 'experimental' and waiting for user feedback to 'stabilize', but i. AFAIK, no user ever reported anything about them, ii. they'be been here long enough to prove they don't do harm. * doc/bison.texi: No longer experimental: default %printer and %destructor (typed: <*> and untyped: <>), %define api.value.type union and variant, Java parsers, XML output, LR family (lr, ielr, lalr), semantic predicates (%?). 2018-08-12 Akim Demaille tests: check variants and typed mid-rule actions See http://lists.gnu.org/archive/html/bison-patches/2018-08/msg00013.html * tests/c++.at (Variants and Typed Mid-rule Actions): New. 2018-08-12 Akim Demaille tests: fix minor issues * tests/actions.at: Fix some log messages. Prefer #error to fprintf: it fixes the invalid use of yyoutput in %destructor, and it is an even stronger check: that the code is not even emitted. The portability of #error is not really a problem here, since the point is anyway to have the compilation fail. 2018-08-11 Akim Demaille c++: variant: add more assertions * data/variant.hh (variant::as): Check yytypeid_ before checking *yytypeid_. 2018-08-11 Akim Demaille doc: -fcaret is enabled by default * doc/bison.texi (Mid-Rule Action Translation): So no need to pass it. 2018-08-11 Akim Demaille style: reduce scopes * src/closure.c, src/conflicts.c: here. 2018-08-11 Akim Demaille NEWS: update 2018-08-11 Akim Demaille rule actions cannot be typed Make sure that we cannot apply a type to the (main) action of a rule. * src/reader.c (grammar_rule_check): Issue the warning. * tests/input.at (Cannot type action): Check the warning. 2018-08-11 Akim Demaille warn about typed mid-rule actions in Yacc mode * src/reader.c (grammar_current_rule_action_append): Warn. * tests/input.at (AT_CHECK_UNUSED_VALUES): Check. 2018-08-11 Akim Demaille tests: check typed mid-rule actions * tests/input.at (_AT_UNUSED_VALUES_DECLARATIONS): Check typed mid-rule actions. * tests/report.at (Reports): Check that types of typed mid-rule actions are reported. * tests/actions.at (Typed mid-rule actions): Check that the values of typed mid-rule actions are correct. 2018-08-11 Akim Demaille regen 2018-08-11 Akim Demaille add support for typed mid-rule actions Prompted on Piotr Marcińczyk's message: http://lists.gnu.org/archive/html/bug-bison/2017-06/msg00000.html. See also http://lists.gnu.org/archive/html/bug-bison/2018-06/msg00001.html. Because their type is unknown to Bison, the values of midrule actions are not treated like the others: they don't have %printer and %destructor support. In addition, in C++, (Bison) variants cannot work properly. Typed midrule actions address these issues. Instead of: exp: { $$ = 1; } { $$ = 2; } { $$ = $1 + $2; } write: exp: { $$ = 1; } { $$ = 2; } { $$ = $1 + $2; } * src/scan-code.h, src/scan-code.l (code_props): Add a `type` field to record the declared type of an action. (code_props_rule_action_init): Add a type argument. * src/parse-gram.y: Accept an optional type tag for actions. * src/reader.h, src/reader.c (grammar_current_rule_action_append): Add a type argument. (grammar_midrule_action): When a mid-rule is typed, pass its type to the defined dummy non terminal symbol. 2018-08-05 Akim Demaille tests: make room for more cases * tests/input.at (AT_CHECK_UNUSED_VALUES): Add an empty line to allow more symbols, and adjust line numbers. Use a more consistent m4 quoting scheme. 2018-08-05 Akim Demaille warnings: address -Wnull-dereference in reader.c Based on a patch by David Michael. http://lists.gnu.org/archive/html/bison-patches/2018-07/msg00000.html * src/reader.c (find_start): New, extracted from... (check_and_convert_grammar): here. 2018-08-05 Akim Demaille style: ielr: reduce scopes * src/ielr.c: Use modern C to reduce the scopes of some variables. 2018-07-26 Akim Demaille style: move to C99 to reduce scopes * src/symtab.c, src/reader.c: Freely mix statements and variable definitions. And use for-loops with initializers. 2018-07-26 Akim Demaille style: split a function in two grammar_current_rule_action_append was used in two different places: for actual action (`{...}`), and for predicates (`%?{...}`). Let's split this in two different functions. * src/reader.h, src/reader.c (grammar_current_rule_predicate_append): New. Extracted from... (grammar_current_rule_action_append): here. Remove arguments that don't apply. Adjust dependencies. 2018-07-26 Akim Demaille tests: fix typo * tests/actions.at: Remove (harmless) stray character. 2018-07-26 Akim Demaille print: remove unused function This function was unused since 1991's original import by rms (e06f0c34427faedc7afbec9554adbffc4c87312e). * src/print.c (print_token): Remove. 2018-06-23 Akim Demaille doc: fix Texinfo syntax error * doc/bison.texi (Understanding): here. 2018-06-22 Akim Demaille doc: we now show the type of the symbols * doc/bison.texi (Understanding Your Parser): Update the output from Bison. Use types in the example, and show them in the report. * NEWS: Update. 2018-06-18 Akim Demaille tests: check the typed symbols in the reports * tests/report.at: New. * tests/local.mk, tests/testsuite.at: Use it. 2018-06-18 Akim Demaille report: display the type of the symbols * src/print.c (print_nonterminal_symbols, print_terminal_symbols): Also should the type of the symbols. 2018-06-18 Akim Demaille style: reduce scopes * src/print.c (print_terminal_symbols, print_nonterminal_symbols): Here. 2018-06-18 Akim Demaille style: split large function * src/print.c (print_grammar): Split into... (print_terminal_symbols, print_nonterminal_symbols): these. Adjust dependencies. 2018-06-18 Akim Demaille style: reduce scopes * src/print.c (print_grammar): Shorten scopes. 2018-06-18 Akim Demaille gnulib: update Fixes the `make install-pdf` problem reported by Hans Åberg in http://lists.gnu.org/archive/html/bug-bison/2018-06/msg00000.html that had already been fixed by Joel E. Denny in http://lists.gnu.org/archive/html/bug-bison/2012-04/msg00011.html Final fix in http://lists.gnu.org/archive/html/bug-gnulib/2018-06/msg00019.html 2018-06-17 Akim Demaille Merge maint into master * upstream/maint: (48 commits) THANKS: update an address tests: adjust syncline tests to GCC 7 glr: fix improperly placed synclines bison: be git grep friendly Replace ftp with https maint: post-release administrivia version 3.0.5 bison: style: indentation fixes regen bison: please address sanitizer C++: style: fix indentation NEWS: update C++: style: prefer `unsigned` to `unsigned int` C++: style: space before paren C++: fix -Wdeprecated warnings tests: fix -Wdeprecated warning maint: update syntax-check exclusions autoconf: update regen Update copyright years ... 2018-05-30 Akim Demaille THANKS: update an address 2018-05-30 Akim Demaille tests: adjust syncline tests to GCC 7 GCC 7 also underlines the error. syncline.c:4:2: error: #error "4" #error "4" ^~~~~ * tests/synclines.at (_AT_SYNCLINES_COMPILE): Remove tildas from GCC 7. 2018-05-29 Akim Demaille glr: fix improperly placed synclines Predicates with GLR are issued with synclines in the middle of C code: case 2: if (! (#line 6 "sempred.y" /* glr.c:816 */ new_syntax)) YYERROR; #line 793 "sempred.tab.c" /* glr.c:816 */ break; Reported by Rici Lake. http://lists.gnu.org/archive/html/bug-bison/2018-05/msg00033.html * data/c.m4 (b4_predicate_case): Be sure to start on column 0. It would be nicer if b4_syncline could ensure this by itself (that would avoid ugly code when synclines are disabled), but that's way more work. * tests/glr-regression.at (Predicates): Be a real end-to-end test. This would have caught this error years ago... 2018-05-29 Akim Demaille bison: be git grep friendly * src/output.c (user_actions_output): Make calls to b4_case and b4_predicate_case explicit. 2018-05-29 Akim Demaille Replace ftp with https Reported by Hans Åberg. * README, cfg.mk, doc/bison.texi: here. 2018-05-27 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2018-05-27 Akim Demaille version 3.0.5 * NEWS: Record release date. 2018-05-27 Akim Demaille bison: style: indentation fixes * src/parse-gram.y: here. 2018-05-27 Akim Demaille regen 2018-05-27 Akim Demaille bison: please address sanitizer * src/parse-gram.y (add_param): Asan does not like that the second argument of strspn is not 0-terminated. 2018-05-27 Akim Demaille C++: style: fix indentation * data/variant.hh (b4_symbol_variant): De-indent, as the callers are indented. 2018-05-27 Akim Demaille NEWS: update 2018-05-27 Akim Demaille C++: style: prefer `unsigned` to `unsigned int` * data/c++.m4: here. 2018-05-27 Akim Demaille C++: style: space before paren * data/c++.m4, data/lalr1.cc: here. 2018-05-27 Akim Demaille C++: fix -Wdeprecated warnings For instance on test 99: In file included from @@.cc:56: @@.hh:409:26: error: definition of implicit copy constructor for 'stack_symbol_type' is deprecated because it has a user-declared copy assignment operator [-Werror,-Wdeprecated] stack_symbol_type& operator= (const stack_symbol_type& that); ^ Reported by Derek Clegg. https://lists.gnu.org/archive/html/bison-patches/2018-05/msg00036.html * configure.ac (warn_tests): Add -Wdeprecated. * data/lalr1.cc (stack_symbol_type): Add an explicit copy ctor. We cannot rely on the explicit default implementation (`= default`) as we support C++ 98. 2018-05-27 Akim Demaille tests: fix -Wdeprecated warning With recent compilers: input.yy:49:5: error: definition of implicit copy assignment operator for 'Object' is deprecated because it has a user-declared destructor [-Werror,-Wdeprecated] ~Object () ^ input.yy:130:35: note: in implicit copy assignment operator for 'Object' first required here { yylhs.value.as< Object > () = yystack_[0].value.as< Object > (); } * tests/c++.at (Object): Add missing assignment operator. 2018-05-19 Akim Demaille maint: update syntax-check exclusions sc_two_space_separator_in_usage complains about bootstrap: two_space_separator_in_usage /Users/akim/src/gnu/bison/bootstrap:905: --aux-dir $build_aux\ /Users/akim/src/gnu/bison/bootstrap:906: --doc-base $doc_base\ /Users/akim/src/gnu/bison/bootstrap:907: --lib $gnulib_name\ /Users/akim/src/gnu/bison/bootstrap:908: --m4-base $m4_base/\ /Users/akim/src/gnu/bison/bootstrap:909: --source-base $source_base/\ /Users/akim/src/gnu/bison/bootstrap:910: --tests-base $tests_base\ /Users/akim/src/gnu/bison/bootstrap:911: --local-dir $local_gl_dir\ maint.mk: help2man requires at least two spaces between an option and its description * cfg.mk: Exclude bootstrap from this check. 2018-05-19 Akim Demaille autoconf: update * submodules/autoconf: Update to latest master. No difference on the M4 files we use. 2018-05-19 Akim Demaille regen 2018-05-12 Akim Demaille Update copyright years Run `make update-copyright`. 2018-05-12 Nate Guerin Add a missing word in the documentation Small patch adds the word 'to' to the documentation. 2018-05-12 Akim Demaille Examples: improve C++ style * examples/variant.yy: Prefer vector to list. Remove useless inline. 2018-05-12 Akim Demaille Avoid compiler warnings At least GCC 7.3, with -O1 or -O2 (but not -O0 or -O3) generates warnings with -Wnull-dereference when using yyformat: it fails to see yyformat cannot be null. Reported by Frank Heckenbach, https://savannah.gnu.org/patch/?9620. * configure.ac: Use -Wnull-dereference if supported. * data/glr.c, data/lalr1.cc, data/yacc.c: Define yyformat in such a way that GCC cannot not see that yyformat is defined. Using `default: abort();` also addresses the issue, but forces the inclusion of `stdlib.h`, which we avoid. 2018-05-10 Akim Demaille C++: fix uses of `inline` Sometimes `inline` would be used in *.cc files on symbols that are not exported (useless but harmless), and sometimes on exported symbols such as the constructor of syntax_error (harmful: linking fails). Reported several times, including: - by Dennis T http://lists.gnu.org/archive/html/bug-bison/2016-03/msg00002.html - by Frank Heckenbach https://savannah.gnu.org/patch/?9616 * data/c++.m4 (b4_inline): New: expands to `inline` or nothing. Use it where appropriate. * data/lalr1.cc: Use it where appropriate. * tests/c++.at (Syntax error as exception): Put the scanner in another compilation unit to exercise the constructor of syntax_error. 2018-05-10 Akim Demaille C++: remove useless `inline` in CC files * data/glr.cc, data/lalr1.cc: Remove `inline` from implementations that are not in headers. 2018-05-10 Akim Demaille C++: remove useless `inline` on templates Templates are implicitly `inline`. * data/c++.m4, data/lalr1.cc: Remove `inline` from templates. 2018-05-08 Akim Demaille style: don't use std::endl * data/lalr1.cc, doc/bison.texi, etc/bench.pl.in, examples/variant.yy, * tests/actions.at, tests/atlocal.in, tests/c++.at, tests/headers.at, * tests/local.at, tests/types.at: Don't use std::endl, it flushes uselessly, and is considered bad style. 2018-05-08 Akim Demaille doc: wrap * README-hacking: Refill paragraphs. 2018-05-08 Akim Demaille gnulib: update * README-hacking: Commit before bootstrapping. * bootstrap.conf: gnulib_mk is no longer defined by bootstrap. * bootstrap, gnulib, lib/.gitignore, m4/.gitignore: Update/regen. 2018-05-08 Akim Demaille tests: we might need to find gnulib headers 315. calc.at:596: testing Calculator ... ++ cat ++ test x = x1 ++ set +x bison/tests/calc.at:596: bison -fno-caret -o calc.c calc.y ++ bison -fno-caret -o calc.c calc.y ++ set +x bison/tests/calc.at:596: $BISON_C_WORKS stderr: stdout: ++ set +x bison/tests/calc.at:596: $CC $CFLAGS $CPPFLAGS $LDFLAGS -o calc calc.c $LIBS ++ ccache clang-mp-6.0 -Qunused-arguments -O3 -g -Wall -Wextra -Wno-sign-compare -Wcast-align -Wdocumentation -Wformat -Wpointer-arith -Wwrite-strings -Wbad-function-cast -Wshadow -Wstrict-prototypes -Wmissing-declarations -Wmissing-prototypes -Wmissing-declarations -Wmissing-prototypes -Wundef -pedantic -Wsign-compare -fno-color-diagnostics -Wno-keyword-macro -Werror -Ibison/_build/6s/lib -DNDEBUG -isystem /opt/local/include -I/opt/local/include -L/opt/local/lib -o calc calc.c bison/_build/6s/lib/libbison.a -lintl -Wl,-framework -Wl,CoreFoundation stderr: In file included from calc.y:198: bison/_build/6s/lib/unistd.h:592:11: fatal error: 'getopt-pfx-core.h' file not found # include ^~~~~~~~~~~~~~~~~~~ 1 error generated. stdout: bison/tests/calc.at:596: exit code was 1, expected 0 315. calc.at:596: 315. Calculator (calc.at:596): FAILED (calc.at:596) * tests/atlocal.in (CPPFLAGS): Find gnulib's headers. 2018-05-08 Akim Demaille getargs: rename argument to avoid gnulib's renaming With Clang 6.0: CC src/bison-getargs.o bison/src/getargs.c:67:12: error: parameter 'option' not found in the function declaration [-Werror,-Wdocumentation] * \param option option being decoded. ^~~~~~ bison/src/getargs.c:67:12: note: did you mean 'rpl_option'? * src/getargs.c: Don't use `option` as a documentation argument. 2017-09-22 Paul Eggert Capitalize "Polish" when it's a proper adjective 2017-09-17 Paul Eggert Adjust to recent Gnulib changes 2017-09-17 Paul Eggert autoconf: update 2017-09-17 Paul Eggert gnulib: update 2015-08-12 Akim Demaille gnulib: update 2015-08-12 Akim Demaille lalr1, yacc: use the default location as initial error location Currently lalr1.cc makes an out-of-bound access when trying to read @1 in rules with an empty rhs (i.e., when there is no @1) that raises an error (YYERROR). glr.c already gracefully handles this by using @$ as initial location for the errors. Let's do that in yacc.c and lalr1.cc. * data/lalr1.cc, data/yacc.c: Use @$ to initialize the error location. * tests/actions.at: Check that case. 2015-08-12 Akim Demaille style: formatting and comment changes * data/glr.c: Avoid empty lines. * data/lalr1.cc: Use the same comments as in glr.c and yacc.c. 2015-08-12 Akim Demaille c++: style: use "unsigned", not "unsigned int" This style appears to be more traditional, at least in C++. For instance in the standard, [facets.examples]. There are occurrences using "unsigned int" too though. * data/lalr1.cc, data/location.cc, data/stack.hh: here. 2015-08-12 Akim Demaille c++: style: remove useless "inline" and fix space issues * data/lalr1.cc, data/c++.m4: Formatting changes. * data/stack.hh: Remove useless "inline". Add documentation. * data/location.cc: Prefer {} for empty bodies. 2015-08-12 Akim Demaille tests: beware of additional warnings from GCC 5 * tests/synclines.at (AT_SYNCLINES_COMPILE): Avoid warnings about unused functions. 2015-08-12 Akim Demaille tests: beware that clang warns about "#define private public" We use this trick to write some test about internal details. But since we use -Werror, clang++ 3.6 dies issueing a warning about it. * configure.ac (warn_tests): Disable this warning. 2015-08-12 Akim Demaille tests: update our Valgrind suppression files * build-aux/linux-gnu.valgrind, build-aux/darwin11.4.0.valgrind: Rename as... * build-aux/Linux.valgrind, build-aux/Darwin.valgrind: these. * build-aux/Linux.valgrind: Add suppression clause. * configure.ac: Update. * tests/local.mk: Use it. 2015-03-03 Akim Demaille doc: improve html and pdf rendering * doc/bison.texi: Help html conversion to understand where the function names end. Beware of PDF width. 2015-03-03 Akim Demaille doc: fixes in the C++ part Reported by Askar Safin. http://lists.gnu.org/archive/html/bug-bison/2015-02/msg00018.html http://lists.gnu.org/archive/html/bug-bison/2015-02/msg00019.html * doc/bison.texi (Split Symbols): Fix access to token types. yylval is a pointer, so use ->. Fix coding style issues: space before paren. 2015-02-10 Akim Demaille tests: be robust to platforms that support UTF-8 even with LC_ALL=C Because musl supports UTF-8 with LC_ALL=C, gcc produces: input.y: In function ‘yyparse’: instead of: input.y: In function 'yyparse': Reported by Ferdinand Thiessen. http://lists.gnu.org/archive/html/bug-bison/2015-02/msg00001.html * tests/synclines.at (AT_SYNCLINES_COMPILE): Skip syncline tests when we can't trust error messages issued about a function body. 2015-02-10 Akim Demaille tests: java: avoid recent Java features Tests 463 and 464 fail with Java 1.4 compilers. Reported by Michael Felt. * tests/javapush.at: Use StringBuffer instead of StringBuilder. 2015-01-26 Akim Demaille tests: c++: fix symbol lookup issue Sun C 5.13 SunOS_sparc 2014/10/20 reports errors on tests 430-432. Reported by Dennis Clarke. * tests/c++.at (Variants): Be sure to emit operator<< before using it: use "%code top" rather than "%code". Prefer std::vector to std::list. Do not define anything in std::, to avoid undefined behavior. 2015-01-23 Akim Demaille Merge remote-tracking branch 'origin/maint' * origin/maint: maint: post-release administrivia version 3.0.4 gnulib: update build: re-enable compiler warnings, and fix them tests: c++: fix a C++03 conformance issue tests: fix a title c++: reserve 200 slots in the parser's stack tests: be more robust to unrecognized synclines, and try to recognize xlc tests: fix C++ conformance build: fix some warnings build: avoid infinite recursions on include_next 2015-01-23 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2015-01-23 Akim Demaille version 3.0.4 * NEWS: Record release date. 2015-01-23 Akim Demaille gnulib: update 2015-01-23 Akim Demaille build: re-enable compiler warnings, and fix them There are warnings (-Wextra) in generated C++ code: ltlparse.cc: In member function 'ltlyy::parser::symbol_number_type ltlyy::parser::by_state::type_get() const': ltlparse.cc:452:33: warning: enumeral and non-enumeral type in conditional expression return state == empty_state ? empty_symbol : yystos_[state]; Reported by Alexandre Duret-Lutz. It turns out that -Wall and -Wextra were disabled because of a stupid typo. * configure.ac: Fix the stupid typo. * data/lalr1.cc, src/AnnotationList.c, src/InadequacyList.c, * src/ielr.c, src/print.c, src/scan-code.l, src/symlist.c, * src/symlist.h, src/symtab.c, src/tables.c, tests/actions.at, * tests/calc.at, tests/cxx-type.at, tests/glr-regression.at, * tests/named-refs.at, tests/torture.at: Fix warnings, mostly issues about variables used only with assertions, which are disabled with -DNDEBUG. 2015-01-22 Akim Demaille tests: c++: fix a C++03 conformance issue This fixes test 241 on xLC: "input.y", line 42.11: 1540-0274 (S) The name lookup for "report" did not find a declaration. "input.y", line 42.11: 1540-1292 (I) Static declarations are not considered for a function call if the function is not qualified. where report is: static void report (std::ostream& yyo, int ival, float fval) { yyo << "ival: " << ival << ", fval: " << fval; } and line 42 is: %printer { report (yyo, $$, $$); } ; It turns out that indeed this function must not be declared static, . Let's put it into an anonymous namespace. Reported by Thomas Jahns. http://lists.gnu.org/archive/html/bug-bison/2015-01/msg00059.html * tests/actions.at (Qualified $$ in actions): Don't use "static", prefer anonymous namespace. 2015-01-20 Akim Demaille tests: fix a title * tests/conflicts.at: De-overquote. 2015-01-20 Akim Demaille c++: reserve 200 slots in the parser's stack This is consistent with what is done with yacc.c and glr.c. Because it also avoids that the stack needs to be resized very soon, it should help keeping tests about destructors more reliable. Indeed, if the stack is created too small, very soon the C++ library needs to enlarge it, which means creating a new one, copying the elements from the initial one onto it, and then destroy the elements of the initial stack: that would be a spurious call to a destructor. Reported by Thomas Jahns. http://lists.gnu.org/archive/html/bug-bison/2015-01/msg00059.html * data/stack.hh (stack::stack): Reserve 200 slots. * tests/c++.at: Remove traces of stack expansions. 2015-01-20 Akim Demaille tests: be more robust to unrecognized synclines, and try to recognize xlc Reported by Thomas Jahns. http://lists.gnu.org/archive/html/bug-bison/2015-01/msg00059.html * tests/synclines.at (AT_SYNCLINES_COMPILE): Rename as... (_AT_SYNCLINES_COMPILE): this. Try to recognize xlc locations. (AT_SYNCLINES_COMPILE): New. Skips the test if we can't read the synclines. 2015-01-20 Akim Demaille tests: fix C++ conformance Reported by Thomas Jahns. http://lists.gnu.org/archive/html/bug-bison/2015-01/msg00059.html * tests/c++.at (Exception safety): Add missing include. Don't use const_iterator for erase. 2015-01-18 Akim Demaille build: fix some warnings Reported by John Horigan. http://lists.gnu.org/archive/html/bug-bison/2015-01/msg00034.html * src/graphviz.c, src/symtab.h: Address compiler warnings. 2015-01-16 Akim Demaille Merge remote-tracking branch 'origin/maint' into origin/master * origin/maint: doc: minor fixes gnulib: strtoul is considered obsolete and now useless c++: avoid warnings when destructors don't use $$ maint: post-release administrivia version 3.0.3 gnulib: update 2015-01-16 Akim Demaille build: avoid infinite recursions on include_next On MacOS X 10.5 PPC with Apple's GCC 4.0.1: % uname -a Darwin aria.cielonegro.org 9.8.0 Darwin Kernel Version 9.8.0: Wed Jul 15 16:57:0 1 PDT 2009; root:xnu-1228.15.4~1/RELEASE_PPC Power Macintosh % gcc --version powerpc-apple-darwin9-gcc-4.0.1 (GCC) 4.0.1 (Apple Inc. build 5493) Copyright (C) 2005 Free Software Foundation, Inc. This is free software; see the source for copying conditions. There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. building in place enters into an infinite recursion on "#include_next": % gmake V=1 [snip] depbase=`echo lib/math.o | sed 's|[^/]*$|.deps/&|;s|\.o$||'`;\ gcc -std=gnu99 -I. -Ilib -I. -I./lib -g -O2 -MT lib/math.o -MD -MP -MF $depbase.Tpo -c -o lib/math.o lib/math.c &&\ mv -f $depbase.Tpo $depbase.Po In file included from lib/math.h:27, from lib/math.h:27, from lib/math.h:27, from lib/math.h:27, [snip] from lib/math.h:27, from lib/math.h:27, from lib/math.c:3: lib/math.h:27:23: error: #include nested too deeply Makefile:3414: recipe for target 'lib/math.o' failed gmake[2]: *** [lib/math.o] Error 1 Using -I./lib instead of -Ilib fixes the problem. Reported by Pho. * Makefile.am (AM_CPPFLAGS): Use -I./lib instead of -Ilib. 2015-01-16 Akim Demaille doc: minor fixes * doc/bison.texi: Fix warnings about colon in reference names. * data/bison.m4, src/files.h: Fix comments. * doc/Doxyfile.in: update. 2015-01-15 Akim Demaille gnulib: strtoul is considered obsolete and now useless * bootstrap.conf: here. 2015-01-15 Akim Demaille c++: avoid warnings when destructors don't use $$ * data/c++.m4: here. 2015-01-15 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2015-01-15 Akim Demaille version 3.0.3 * NEWS: Record release date. 2015-01-15 Akim Demaille gnulib: update 2015-01-14 Akim Demaille symbol: use the first occurrence as an LHS as defining location Currently on the following grammar: %type foo %% start: foo | bar | "baz" foo: foo bar: bar bison reports: warning: 2 nonterminals useless in grammar [-Wother] warning: 4 rules useless in grammar [-Wother] 1.13-15: warning: nonterminal useless in grammar: foo [-Wother] %type foo ^^^ 3.14-16: warning: nonterminal useless in grammar: bar [-Wother] start: foo | bar | "baz" ^^^ [...] i.e., the location of the first occurrence of a symbol is taken as its definition point. In the case of nonterminals, the first occurrence as a left-hand side of a rule makes more sense: warning: 2 nonterminals useless in grammar [-Wother] warning: 4 rules useless in grammar [-Wother] 4.1-3: warning: nonterminal useless in grammar: foo [-Wother] foo: foo ^^^ 5.1-3: warning: nonterminal useless in grammar: bar [-Wother] bar: bar ^^^ [...] * src/symtab.h, src/symtab.c (symbol::location_of_lhs): New. (symbol_location_as_lhs_set): New. * src/parse-gram.y (current_lhs): Use it. * tests/reduce.at: Update locations. 2015-01-14 Akim Demaille reduce: don't complain about rules whose lhs is useless In the following grammar, the 'exp' nonterminal is trivially useless. So, of course, its rules are useless too. %% input: '0' | exp exp: exp '+' exp | exp '-' exp | '(' exp ')' Previously all the useless rules were reported, including those whose left-hand side is the 'exp' nonterminal: warning: 1 nonterminal useless in grammar [-Wother] warning: 4 rules useless in grammar [-Wother] 2.14-16: warning: nonterminal useless in grammar: exp [-Wother] input: '0' | exp ^^^ 2.14-16: warning: rule useless in grammar [-Wother] input: '0' | exp ^^^ ! 3.6-16: warning: rule useless in grammar [-Wother] ! exp: exp '+' exp | exp '-' exp | '(' exp ')' ! ^^^^^^^^^^^ ! 3.20-30: warning: rule useless in grammar [-Wother] ! exp: exp '+' exp | exp '-' exp | '(' exp ')' ! ^^^^^^^^^^^ ! 3.34-44: warning: rule useless in grammar [-Wother] ! exp: exp '+' exp | exp '-' exp | '(' exp ')' ! ^^^^^^^^^^^ The interest of being so verbose is dubious. I suspect most of the time nonterminals are not expected to be useless, so the user wants to fix the nonterminal, not remove its rules. And even if the user wanted to get rid of its rules, the position of these rules probably does not help more that just having the name of the nonterminal. This commit discard these messages, marked with '!', and keep the others. In particular, we still report: 2.14-16: warning: rule useless in grammar [-Wother] input: '0' | exp ^^^ All the useless rules (including the '!' ones) are still reported in the reports (xml, text, etc.); only the diagnostics on stderr change. * src/gram.c (grammar_rules_useless_report): Don't complain about useless rules whose lhs is useless. * src/reduce.h, src/reduce.c (reduce_nonterminal_useless_in_grammar): Take a sym_content as argument. Adjust callers. * tests/reduce.at (Useless Rules, Underivable Rules, Reduced Automaton): Adjust. 2015-01-14 Akim Demaille style: reduce: use unsigned to count a number of objects * src/reduce.h, src/reduce.c (nuseful_productions, nuseless_productions) (nuseful_nonterminals, nuseless_nonterminals): Declare as unsigned. Simplify "0 <" tests into non-zero tests. 2015-01-14 Akim Demaille style: reduce: introduce and use a swap for bitset * src/reduce.c (bitset_swap): New. Use it. 2015-01-14 Akim Demaille style: reduce: reduce scopes and other stylistic changes * src/reduce.c: Various stylistic changes: Reduce scopes. Prefer ++i to i++. Prefer < to >. 2015-01-13 Akim Demaille Merge remote-tracking branch 'origin/maint' * origin/maint: tests: split a large test case into several smaller ones package: a bit of trouble shooting indications doc: liby's main arms the internationalization bison: avoid warnings from static code analysis c++: fix the use of destructors when variants are enabled style: tests: simplify the handling of some C++ tests c++: symbols can be empty, so use it c++: variants: don't leak the lookahead in error recovery c++: provide a means to clear symbols c++: clean up the handling of empty symbols c++: comment and style changes c++: variants: comparing addresses of typeid.name() is undefined c++: locations: complete the API and fix comments build: do not clean figure sources in make clean 2015-01-13 Akim Demaille tests: split a large test case into several smaller ones * tests/conflicts.at (AT_CONSISTENT_ERRORS_CHECK): Move AT_SETUP/AT_CLEANUP into it, so that we don't skip non Java tests following a test case in Java. 2015-01-12 Akim Demaille package: a bit of trouble shooting indications * README-hacking: here. 2015-01-12 Akim Demaille doc: liby's main arms the internationalization Reported by Nicolas Bedon. * doc/bison.texi (Yacc Library): Document the call the setlocale. 2015-01-09 Akim Demaille bison: avoid warnings from static code analysis A static analysis tool reports that some callers of symbol_list_n_get might get NULL and not handle it properly. This is not the case, yet we can suppress this pattern. Reported by Mike Sullivan. * src/symlist.c (symbol_list_n_get): Actually it is never called to return 0. Enforce this postcondition via aver. (symbol_list_n_type_name_get): Simplify accordingly. In particular, discards a (translated) useless error message. * src/symlist.h: Adjust documentation. * src/scan-code.l: Style change. 2015-01-09 Akim Demaille c++: fix the use of destructors when variants are enabled When using variants, destructors generate invalid code. Reported by Michael Catanzaro. * data/c++.m4 (~basic_symbol): b4_symbol_foreach works on yysym: define it. * tests/c++.at (Variants): Check it. 2015-01-08 Akim Demaille style: tests: simplify the handling of some C++ tests * tests/c++.at: here. (Doxygen): Pass %define, so that files such as position.hh etc. are generated, instead of putting everything into input.hh. 2015-01-08 Akim Demaille c++: symbols can be empty, so use it The previous patches ensure that symbols (symbol_type and stack_symbol_type) can be empty, cleared, and their emptiness can be checked. Therefore, yyempty, which codes whether yyla is empty or not, is now useless. In C skeletons (e.g., yacc.c), the fact that the lookahead is empty is coded by "yychar = YYEMPTY", which is exactly what this patch restores, since yychar/yytoken corresponds to yyla.type. * data/lalr1.cc (yyempty): Remove. Rather, depend on yyla.empty (). 2015-01-08 Akim Demaille c++: variants: don't leak the lookahead in error recovery During error recovery, when discarding the lookeahead, we don't destroy it, which is caught by parse.assert assertions. Reported by Antonio Silva Correia. With an analysis and suggested patch from Michel d'Hooge. * tests/c++.at (Variants): Strengthen the test to try syntax errors with discarded lookahead. 2015-01-08 Akim Demaille c++: provide a means to clear symbols The symbol destructor is currently the only means to clear a symbol. Unfortunately during error recovery we might have to clear the lookahead, which is a local variable (yyla) that has not yet reached its end of scope. Rather that duplicating the code to destroy a symbol, or rather than destroying and recreating yyla, let's provide a means to clear a symbol. Reported by Antonio Silva Correia, with an analysis from Michel d'Hooge. * data/c++.m4, data/lalr1.cc (basis_symbol::clear, by_state::clear) (by_type::clear): New. (basic_symbol::~basic_symbol): Use clear. 2015-01-08 Akim Demaille c++: clean up the handling of empty symbols * data/c++.m4, data/lalr1.cc (yyempty_): Remove, replaced by... (empty_symbol, by_state::empty_state): these. (basic_symbol::empty): New. 2015-01-08 Akim Demaille c++: comment and style changes * data/c++.m4, data/lalr1.cc: More documentation. Tidy. * tests/c++.at (string_cast): Rename as... (to_string): this C++11 name. 2015-01-07 Akim Demaille c++: variants: comparing addresses of typeid.name() is undefined Instead of storing and comparing pointers to names of types, store pointers to the typeids, and compares the typeids. Reported by Thomas Jahns. * data/variant.hh (yytname_): Replace with... (yytypeid_): this. 2015-01-05 Akim Demaille c++: locations: complete the API and fix comments There are no support for += between locations, and some comments are wrong. Reported by Alexandre Duret-Lutz. * data/location.cc: Fix. * doc/bison.texi: Document. * tests/c++.at: Check. 2015-01-05 Akim Demaille build: do not clean figure sources in make clean "make clean && make" fails in in-tree builds. * doc/local.mk (CLEANDIRS): Replace with... (CLEANFILES): this safer list of files to clean. 2015-01-05 Akim Demaille Merge remote-tracking branch 'origin/maint' * origin/maint: build: don't try to generate docs when cross-compiling package: fix a reporter's name %union: fix the support for named %union package: bump to 2015 flex: don't trust YY_USER_INIT yacc.c: fix broken union when api.value.type=union and %defines are used doc: fix missing xref gnulib: update location: remove some ugly debugging code traces build: use abort to pacify compiler errors package: bump to 2014 doc: specify documentation encoding 2015-01-05 Akim Demaille build: don't try to generate docs when cross-compiling When cross-compiling don't run the generated bison to update the docs. Reported by Aaro Koskinen. * configure.ac (CROSS_COMPILING): New. * doc/local.mk: Use it. 2015-01-04 Akim Demaille package: fix a reporter's name * THANKS, build-aux/git-log-fix: s/Bernd Edligner/Bernd Edlinger/. 2015-01-04 Akim Demaille %union: fix the support for named %union Bison supports a union tag, for obscure reasons. But it does a poor job at it, especially since Bison 3.0. Reported by Stephen Cameron and Tobias Frost. It did not ensure that the name was not given several times. An easy way to do this is to make the %union tag be handled as a %define variable, as they cannot be defined several times. Since Bison 3.0, the synclines were wrongly placed, resulting in invalid code. Addressing this issue, because of the way the union tag was stored (as a code muscle), would have been tedious. Unless we rather define the %union tag as a %percent variable, whose synclines are easier to manipulate. So replace the b4_union_name muscle by the api.value.union.name %define variable, document, and check. * data/bison.m4: Make sure that api.value.union.name has a keyword value. * data/c++.m4: Make sure that api.value.union.name is not defined. * data/c.m4 (b4_union_name): No longer use it, use api.value.union.name. * doc/bison.texi (%define Summary): Document it. * src/parse-gram.y (union_name): No longer define b4_uion_name, but api.value.union.name. * tests/input.at (Redefined %union name): New. * tests/synclines.at (%union name syncline): New. * tests/types.at: Check named %unions. 2015-01-04 Akim Demaille package: bump to 2015 Which also requires: * gnulib: Update. 2014-12-31 Akim Demaille flex: don't trust YY_USER_INIT Reported by Bernd Edlinger and others. * src/scan-gram.l: here. 2014-12-31 Akim Demaille yacc.c: fix broken union when api.value.type=union and %defines are used Reported by Rich Wilson. * data/c.m4 (b4_symbol_type_register): Append to b4_union_members, not b4_user_union_members. The latter invokes the former, but it is the former which is reinitialized to empty by b4_value_type_setup_union. * tests/types.at: Check it. This reveals another bug, this time in the case of glr.c parsers. * data/glr.c: Generate the header file before the implementation file, to be sure that the setup is run before what depends on it. 2014-12-31 Akim Demaille doc: fix missing xref Reported by xolodho. * doc/bison.texi (Printer Decl): here. 2014-12-29 Akim Demaille gnulib: update 2014-02-03 Akim Demaille location: remove some ugly debugging code traces * data/location.cc: here. 2014-02-03 Akim Demaille build: use abort to pacify compiler errors clang, with -DNDEBUG and -Werror fails on some functions that might lack a return. This is because aver is just another assert, discarded with -DNDEBUG. So use abort. * src/muscle-tab.c, src/scan-skel.l: here. 2014-02-03 Akim Demaille package: bump to 2014 * AUTHORS, ChangeLog-2012, Makefile.am, NEWS, PACKAGING, README, * README-alpha, README-hacking, THANKS, TODO, bootstrap.conf, * build-aux/darwin11.4.0.valgrind, build-aux/local.mk, * build-aux/update-b4-copyright, * build-aux/update-package-copyright-year, cfg.mk, configure.ac, * data/README, data/bison.m4, data/c++-skel.m4, data/c++.m4, * data/c-like.m4, data/c-skel.m4, data/c.m4, data/glr.c, data/glr.cc, * data/java-skel.m4, data/java.m4, data/lalr1.cc, data/lalr1.java, * data/local.mk, data/location.cc, data/stack.hh, data/variant.hh, * data/xslt/bison.xsl, data/xslt/xml2dot.xsl, data/xslt/xml2text.xsl, * data/xslt/xml2xhtml.xsl, data/yacc.c, djgpp/Makefile.maint, * djgpp/README.in, djgpp/config.bat, djgpp/config.sed, * djgpp/config.site, djgpp/config_h.sed, djgpp/djunpack.bat, * djgpp/local.mk, djgpp/subpipe.c, djgpp/subpipe.h, * djgpp/testsuite.sed, doc/bison.texi, doc/local.mk, doc/refcard.tex, * etc/README, etc/bench.pl.in, etc/local.mk, * examples/calc++/calc++.test, examples/calc++/local.mk, * examples/extexi, examples/local.mk, examples/mfcalc/local.mk, * examples/mfcalc/mfcalc.test, examples/rpcalc/local.mk, * examples/rpcalc/rpcalc.test, examples/test, examples/variant.yy, * lib/abitset.c, lib/abitset.h, lib/bbitset.h, lib/bitset.c, * lib/bitset.h, lib/bitset_stats.c, lib/bitset_stats.h, * lib/bitsetv-print.c, lib/bitsetv-print.h, lib/bitsetv.c, * lib/bitsetv.h, lib/ebitset.c, lib/ebitset.h, lib/get-errno.c, * lib/get-errno.h, lib/lbitset.c, lib/lbitset.h, lib/libiberty.h, * lib/local.mk, lib/main.c, lib/timevar.c, lib/timevar.def, * lib/timevar.h, lib/vbitset.c, lib/vbitset.h, lib/yyerror.c, * m4/bison-i18n.m4, m4/c-working.m4, m4/cxx.m4, m4/flex.m4, * m4/timevar.m4, src/AnnotationList.c, src/AnnotationList.h, * src/InadequacyList.c, src/InadequacyList.h, src/LR0.c, src/LR0.h, * src/Sbitset.c, src/Sbitset.h, src/assoc.c, src/assoc.h, * src/closure.c, src/closure.h, src/complain.c, src/complain.h, * src/conflicts.c, src/conflicts.h, src/derives.c, src/derives.h, * src/files.c, src/files.h, src/flex-scanner.h, src/getargs.c, * src/getargs.h, src/gram.c, src/gram.h, src/graphviz.c, * src/graphviz.h, src/ielr.c, src/ielr.h, src/lalr.c, src/lalr.h, * src/local.mk, src/location.c, src/location.h, src/main.c, * src/muscle-tab.c, src/muscle-tab.h, src/named-ref.c, * src/named-ref.h, src/nullable.c, src/nullable.h, src/output.c, * src/output.h, src/parse-gram.c, src/parse-gram.y, src/print-xml.c, * src/print-xml.h, src/print.c, src/print.h, src/print_graph.c, * src/print_graph.h, src/reader.c, src/reader.h, src/reduce.c, * src/reduce.h, src/relation.c, src/relation.h, src/scan-code.h, * src/scan-code.l, src/scan-gram.h, src/scan-gram.l, src/scan-skel.h, * src/scan-skel.l, src/state.c, src/state.h, src/symlist.c, * src/symlist.h, src/symtab.c, src/symtab.h, src/system.h, * src/tables.c, src/tables.h, src/uniqstr.c, src/uniqstr.h, * tests/actions.at, tests/atlocal.in, tests/bison.in, tests/c++.at, * tests/calc.at, tests/conflicts.at, tests/cxx-type.at, * tests/existing.at, tests/glr-regression.at, tests/headers.at, * tests/input.at, tests/java.at, tests/javapush.at, tests/local.at, * tests/local.mk, tests/named-refs.at, tests/output.at, tests/push.at, * tests/reduce.at, tests/regression.at, tests/sets.at, * tests/skeletons.at, tests/synclines.at, tests/testsuite.at, * tests/torture.at, tests/types.at: here. 2014-01-03 Paul Eggert doc: specify documentation encoding * doc/bison.texi: Add '@documentencoding UTF-8'; needed since the manual contains UTF-8 characters. This will cause the .info files to contain UTF-8 quotes and the like, which should be OK nowadays. Add @documentlanguage while we're at it. 2013-12-10 Akim Demaille symbols: properly fuse the properties of two symbol aliases This completes and fixes a7280757105b2909f6a58fdd1c582de8e278319a. Reported by Valentin Tolmer. Before it Bison used to put the properties of the symbols (associativity, printer, etc.) in the 'symbol' structure. An identifier-named token (FOO) and its string-named alias ("foo") duplicated these properties, and symbol_check_alias_consistency() checked that both had compatible properties and fused them, at the end of the parsing of the grammar. The commit a7280757105b2909f6a58fdd1c582de8e278319a introduces a sym_content structure that keeps all these properties, and ensures that both aliases point to the same sym_content (instead of duplicating). However, it removed symbol_check_alias_consistency, which resulted in the non-fusion of *existing* properties: %token FOO "foo" %left FOO %left "foo" was properly diagnosed as a redeclaration, but %left FOO %left "foo" %token FOO "foo" was not, as the properties of FOO and "foo" were not checked before fusion. It certainly also means that %left "foo" %token FOO "foo" did not transfer properly the associativity to FOO. The fix is simple: reintroduce symbol_check_alias_consistency (under a better name, symbol_merge_properties) and call it where appropriate. Also, that commit made USER_NUMBER_HAS_STRING_ALIAS useless, but left it. * src/symtab.h (USER_NUMBER_HAS_STRING_ALIAS): Remove, unused. Adjust dependencies. * src/symtab.c (symbol_merge_properties): New, based on the former symbol_check_alias_consistency. * tests/input.at: Re-enable tests that we now pass. 2013-12-10 Akim Demaille Merge remote-tracking branch 'origin/maint' * origin/maint: package: install the examples package: install README and the like in docdir diagnostics: fix the order of multiple declarations reports symbol: provide an easy means to compare them in source order 2013-12-09 Akim Demaille package: install the examples Currently, we do not install the various examples extracted from the documentation. Let's do it, as they are useful starting points. * configure.ac: When --enable-gcc-warnings is set, enable ENABLE_GCC_WARNINGS. * examples/extexi: No longer issue synclines by default. * examples/local.mk: Except if ENABLE_GCC_WARNINGS. * examples/calc++/local.mk, examples/mfcalc/local.mk, * examples/rpcalc/local.mk: Install the example files. 2013-12-09 Akim Demaille package: install README and the like in docdir * Makefile.am: here. 2013-12-09 Akim Demaille diagnostics: fix the order of multiple declarations reports On %token FOO "foo" %printer {} "foo" %printer {} FOO we report /tmp/foo.yy:2.10-11: error: %printer redeclaration for FOO %printer {} "foo" ^^ /tmp/foo.yy:3.10-11: previous declaration %printer {} FOO ^^ * src/symtab.c (locations_sort): New. Use it. * tests/input.at (Invalid Aliases): Stress the order of diagnostics. 2013-12-09 Akim Demaille symbol: provide an easy means to compare them in source order * src/symtab.c (symbols_sort): New. (user_token_number_redeclaration): Taken from here. 2013-12-09 Akim Demaille Merge remote-tracking branch 'origin/maint' * origin/maint: (43 commits) maint: post-release administrivia version 3.0.2 gnulib: update output: do not generate source files when late errors are caught output: record what generated files are source or report files output: do not generate source files when early errors are caught xml: also use "%empty" with html output style: formatting changes xml: also display %empty for empty right-hand sides reports: display %empty in the generated pointed-rules news: YYERROR vs variants style: scope reduction in lalr.cc lalr1.cc: formatting changes lalr1.cc: fix the support of YYERROR with variants tests: check $$'s destruction with variant, YYERROR, and no error recovery tests: simplify useless obfuscation skeletons: use better names when computing a "goto" maint: post-release administrivia version 3.0.1 aver: it is no longer "protected against NDEBUG" ... 2013-12-05 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2013-12-05 Akim Demaille version 3.0.2 * NEWS: Record release date. 2013-12-05 Akim Demaille gnulib: update * gnulib: here. 2013-12-04 Akim Demaille output: do not generate source files when late errors are caught Reported by Alexandre Duret-Lutz as "second problem" in: http://lists.gnu.org/archive/html/bug-bison/2013-09/msg00015.html * bootstrap.conf: We need the "unlink" module. * src/files.h, src/files.c (unlink_generated_sources): New. * src/output.c: Use it. * tests/output.at: Check the case of late errors. 2013-12-04 Akim Demaille output: record what generated files are source or report files * src/files.h, src/files.c (output_file_name_check): Take an additional argument to record whether a file is a source or report file. * src/files.c (generated_file): New. (file_names, file_names_count): Replace with... (generated_files, generated_files_size): these. * src/scan-skel.l: Adjust. 2013-12-04 Akim Demaille output: do not generate source files when early errors are caught Reported by Alexandre Duret-Lutz as "second problem" in: http://lists.gnu.org/archive/html/bug-bison/2013-09/msg00015.html One problem is that some errors are caught early, before the generation of output files, while others can only be detected afterwards (since, for instance, skeletons can raise errors themselves). This will be addressed in two steps: early errors do not generate source files at all, while later errors will remove the files that have already been generated. * src/scan-skel.l (yyout): Open to /dev/null when there are errors. * tests/output.at (AT_CHECK_FILES): Factored out of... (AT_CHECK_OUTPUT): this. Fuse the "SHELLIO" argument in the "FLAGS" one. Use $5 to denote the expected exit status. Add a test case for early errors. 2013-11-26 Akim Demaille xml: also use "%empty" with html output * data/xslt/xml2xhtml.xsl: No longer issue an Epsilon, display as in dot and text formats. 2013-11-26 Akim Demaille style: formatting changes * src/print-xml.c: here. 2013-11-26 Akim Demaille xml: also display %empty for empty right-hand sides * data/xslt/xml2dot.xsl, data/xslt/xml2text.xsl: Display %empty where needed. 2013-11-26 Akim Demaille reports: display %empty in the generated pointed-rules * src/print.c (print_core): Use %empty for empty rules. * src/print_graph.c (print_core): Ditto. * tests/conflicts.at, tests/output.at, tests/reduce.at: Adjust expectations. 2013-11-26 Akim Demaille news: YYERROR vs variants 2013-11-18 Akim Demaille style: scope reduction in lalr.cc * src/lalr.c: Shorten variable scopes. (lookahead_tokens_print): Use the same variable name in two loops iterating over the same structure. 2013-11-15 Akim Demaille lalr1.cc: formatting changes * data/lalr1.cc: Fix indentation. 2013-11-15 Akim Demaille lalr1.cc: fix the support of YYERROR with variants When variant are enabled, the yylhs variable (the left-hand side of the rule being reduced, i.e. $$ and @$) is explicitly destroyed when YYERROR is called. This is because before running the user code, $$ is initialized, so that the user can properly use it. However, when quitting yyparse, yylhs is also reclaimed by the C++ compiler: the variable goes out of scope. Instead of trying to be too smart, let the compiler do its job: reduce the scope of yylhs to exactly the reduction. This way, whatever the type of scope exit (regular, exception, return, goto...) this variable will be properly reclaimed. Reported by Paolo Simone Gasparello. * data/lalr1.cc (yyparse): Reduce the scope of yylhs. * tests/c++.at: We now pass this test. 2013-11-15 Akim Demaille tests: check $$'s destruction with variant, YYERROR, and no error recovery When variant are enabled, the yylhs variable (the left-hand side of the rule being reduced, i.e. $$ and @$) is explicitly destroyed when YYERROR is called. This is because before running the user code, $$ is initialized, so that the user can properly use it. However, when quitting yyparse, yylhs is also reclaimed by the C++ compiler: the variable goes out of scope. This was not detected by the test suite because (i) the Object tracker was too weak, and (ii) the problem does not show when there is error recovery. Reported by Paolo Simone Gasparello. * tests/c++.at (Exception safety): Improve the objects logger to make sure that we never destroy twice an object. Also track copy-constructors. Use a set instead of a list. Display the logs before running the function body, this is more useful in case of failure. Generalize to track with and without error recovery. 2013-11-15 Akim Demaille tests: simplify useless obfuscation * tests/c++.at: $$ is not special for M4, there is no need to "escape" it. 2013-11-14 Akim Demaille skeletons: use better names when computing a "goto" * data/glr.c (yyLRgotoState): Name the symbol argument yysym, instead of yylhs. * data/lalr1.cc (yy_lr_goto_state_): Likewise. * data/lalr1.java (yy_lr_goto_state_): New, modeled after the previous two routines. Use it. 2013-11-12 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2013-11-12 Akim Demaille version 3.0.1 * NEWS: Record release date. 2013-11-12 Akim Demaille aver: it is no longer "protected against NDEBUG" Apply the same rules for aver as for assert: no side effects, especially not important ones. * src/AnnotationList.c, src/muscle-tab.c: Adjust aver uses to resist to -DNDEBUG. 2013-11-08 Akim Demaille parsers: rename YY_NULL as YY_NULLPTR to avoid conflicts with Flex Flex also defines YY_NULL (to 0). Avoid gratuitous conflicts. * data/c.m4 (b4_null_define): Rename YY_NULL as YY_NULLPTR. * data/glr.c, data/lalr1.cc, data/location.cc, data/variant.hh, * data/yacc.c, src/parse-gram.c, tests/actions.at, tests/c++.at, * tests/cxx-type.at, tests/glr-regression.at, tests/headers.at, * tests/push.at, tests/regression.at: Adjust. 2013-11-05 Akim Demaille build: use Automake 1.14's non-recursive Makefile features * configure.ac: Require Automake 1.14. * examples/calc++/local.mk, examples/local.mk, examples/mfcalc/local.mk, * examples/rpcalc/local.mk, tests/local.mk: Use %D% and %C%. 2013-11-05 Akim Demaille build: restore maintainer-push-check * tests/local.mk: here. 2013-11-05 Akim Demaille c++: use __attribute__((__pure__)) to avoid warnings Building C++ parsers with -Wsuggest-attribute=const and -Wsuggest-attribute=noreturn triggers warning in generated code. * data/lalr1.cc: Call b4_attribute_define. (debug_stream, debug_level): Flag as pure. * tests/headers.at (Several parsers): There are now more YY macros that "leak". 2013-11-05 Akim Demaille skeletons: update the handling of compiler attributes * data/c.m4 (b4_attribute_define): Instead of defining __attribute__, define YY_ATTRIBUTE conditionally. (YY_ATTRIBUTE_PURE, YY_ATTRIBUTE_UNUSED, _Noreturn): New. Use them. * data/glr.c: Use them. 2013-11-05 Akim Demaille gnulib: update 2013-10-24 Akim Demaille style: use /* ... */ comments * src/complain.c: Here. 2013-10-24 Akim Demaille tests: skip C++ tests that are too demanding for some compilers Some tests now fail when compiled with G++ 4.3 or 4.4 on MacPorts. * tests/local.at (AT_SKIP_IF_EXCEPTION_SUPPORT_IS_POOR): New. * tests/c++.at (Exception safety): Use it. 2013-10-22 Akim Demaille install: do not install yacc.1 when --disable-yacc * configure.ac (ENABLE_YACC): New conditional. (YACC_SCRIPT, YACC_LIBRARY): Remove. * lib/local.mk, src/local.mk: Use the former instead of the latter. * doc/local.mk: Use ENABLE_YACC to avoid installing yacc.1. 2013-10-22 Akim Demaille style: avoid tabs * src/scan-code.l: here. 2013-10-22 Akim Demaille c++: fix generated doxygen comments * configure.ac: Enable -Wdocumentation if supported. * data/lalr1.cc: Fix comments. 2013-10-22 Akim Demaille fix: uniqstr are already pointers * src/uniqstr.c (uniqstr_assert): Remove incorrect double indirection, and now useless cast. 2013-10-22 Paul Eggert bison: pacify Sun C 5.12 * src/scan-code.l (show_sub_message): Redo initializations to work around a bogus Sun C 5.12 warning. (parse_ref): Remove unreachable code that Sun C 5.12 complains about. * src/uniqstr.h (uniqstr_vsprintf): Use _GL_ATTRIBUTE_FORMAT_PRINTF (...) instead of __attribute__ ((__format__ (__printf__, ...))). Otherwise, Sun C 5.12 complains about an unknown attribute. 2013-10-22 Paul Eggert maint: git now ignores rpcalc * examples/rpcalc/.gitignore: Ignore rpcalc. 2013-10-22 Paul Eggert build: examples/calc++/calc++ requires flex * configure.ac (FLEX_CXX_WORKS): New AM_CONDITIONAL. * examples/calc++/local.mk (examples/calc++/calc++): Build if FLEX_CXX_WORKS, not BISON_CXX_WORKS. 2013-10-22 Paul Eggert maint: mention help2man, texinfo, apt-get * README-hacking: Add help2man, texinfo. Describe how to add packages if you're using Debian. 2013-10-22 Paul Eggert maint: git now ignores .log and .trs files * .gitignore: Add *.log, *.trs. 2013-10-21 Akim Demaille tests: fix incorrect object construction Reported by Ken Moffat. http://lists.gnu.org/archive/html/bug-bison/2013-10/msg00009.html * tests/c++.at (Exception safety): Here. 2013-10-16 Akim Demaille glr: allow spaces between "%?" and "{" in predicates Reported by Rici Lake. http://lists.gnu.org/archive/html/bug-bison/2013-10/msg00004.html http://stackoverflow.com/questions/19330171/ * src/scan-gram.l: Do not try to be too smart when diagnosing invalid directives. * tests/glr-regression.at (Predicates): New test. 2013-10-16 Akim Demaille diagnostics: "-Werror -Wno-error=foo" must not emit errors Currently "-Werror -Wno-error=foo" still turns "foo" warnings into errors. Reported by Alexandre Duret-Lutz. See http://lists.gnu.org/archive/html/bug-bison/2013-09/msg00015.html. * src/complain.c (errority, errority_flag): New. (complain_init): Initialize the latter. (warning_argmatch): Extract the loop iterating on the flag's bits. Set and unset errority_flag here. (warnings_argmatch): -Wno-error is not the same as -Wno-error=everything: we must remember if category foo was explicitly turned in an error/warning via -W(no-)error=foo. (warning_severity): Use errority_flag. * tests/input.at (Symbols): Just check --yacc, not -Wyacc, that's the job of tests on -W. (-Werror is not affected by -Wnone and -Wall): Rename as... (-Werror combinations): this. Tests more combinations of -W, -W(no-)error, and -W(no-)error=foo. * tests/local.at (AT_BISON_CHECK_WARNINGS): Don't expect -Werror to turn runs that issue warnings into runs with errors, as the warnings might be enforced as warnings by -Wno-error=foo, in which case -Werror does not change anything. * doc/bison.texi (Bison Options): Try to be clearer about how -W(no-)error and -W(no-)error=foo interact. 2013-10-16 Akim Demaille comment changes * src/complain.h, src/complain.c: More documentation, more comments. 2013-10-04 Andreas Schwab location: fix EOF check * location.c (location_caret): Use int, not char, for values from getc. 2013-09-19 Akim Demaille style: variant: remove empty line * data/variant.hh (b4_symbol_constructor_define_): Remove stray eol. 2013-09-19 Akim Demaille Merge remote-tracking branch 'origin/maint' * origin/maint: glr: more assertions glr: shorten scopes glr: formatting changes glr: better use of tracing macros examples: improve the output of the "variant" example variant: remove useless assertion tests: remove stray debugging traces tests: do not use grep -q build: don't require flex for ordinary builds maint: update .gitignore build: port to pre-5.8.7 perl tests: minor change to make it easier to test other skeletons uniqstr: fix assertion 2013-09-19 Akim Demaille glr: simplify the invocation of YYLLOC_DEFAULT The commit which introduces yyresolveLocations (commit 8710fc41aaebc5d167a2783a4b8b60849a803869) saves and restores the look-ahead (type, value and location) for no clear reason. This appears to be useless. * data/glr.c (yyresolveLocations): Don't save/restore the current look-ahead to call YYLLOC_DEFAULT. Minor style changes. 2013-09-19 Akim Demaille glr: more assertions * data/glr.c (yyaddDeferredAction, yyglrShiftDefer, yypdumpstack): More assertions. 2013-09-19 Akim Demaille glr: shorten scopes * data/glr.c (yyglrReduce): Define yyflag with its value. 2013-09-19 Akim Demaille glr: formatting changes * data/glr.c: here. 2013-09-19 Akim Demaille glr: better use of tracing macros * data/glr.c (yydestroyGLRState): Use YY_SYMBOL_PRINT instead of yy_symbol_print. 2013-09-19 Akim Demaille examples: improve the output of the "variant" example * examples/variant.yy: Improve the printing of lists. 2013-09-19 Akim Demaille variant: remove useless assertion * data/variant.hh (move): Remove precondition assertion which is ensured by the first call of the body (this precondition is also one of "build"). 2013-09-19 Akim Demaille tests: remove stray debugging traces * tests/atlocal.in: Remove traces. Be ready to remove conftest.dSYM generated on OS X. 2013-09-04 Akim Demaille tests: do not use grep -q Reported by Daniel Galloway. http://lists.gnu.org/archive/html/bug-bison/2013-08/msg00020.html * tests/java.at: Ignore grep's output instead. 2013-08-25 Paul Eggert build: don't require flex for ordinary builds * configure.ac (LEX): Don't fail if this is lex, as flex is not required for ordinary builds. Instead, issue a warning and substitute a no-op LEX. Reported by Michael Felt in . 2013-08-25 Paul Eggert maint: update .gitignore * .gitignore: Add *.eps, *.o, *.pdf, *.png, *.stamp, *~, .deps, .dirstamp. Needed to suppress unwanted chatter from 'git status' after a bootstrap build. 2013-08-24 Paul Eggert build: port to pre-5.8.7 perl * examples/local.mk (extract): Omit -f from perl options. This doesn't work with perl versions before 5.8.7 that are configured without USE_SITECUSTOMIZE. Reported by Michael Felt in . 2013-08-01 Akim Demaille tests: minor change to make it easier to test other skeletons * tests/c++.at (Variants): Pass the skeleton as argument. 2013-08-01 Valentin Tolmer uniqstr: fix assertion * src/uniqstr.c (uniqstr_assert): Really make sure str is a uniqstr, not just whether some uniqstr with the same content was registered. 2013-08-01 Valentin Tolmer symbols: improve symbol aliasing Rather than having duplicate info in the symbol and the alias that has to be resolved later on, both the symbol and the alias have a common pointer to a separate structure containing this info. * src/symtab.h (sym_content): New structure. * src/symtab.c (sym_content_new, sym_content_free, symbol_free): New * src/AnnotationList.c, src/conflicts.c, src/gram.c, src/gram.h, * src/graphviz.c, src/ielr.c, src/output.c, src/parse-gram.y, src/print.c * src/print-xml.c, src/print_graph.c, src/reader.c, src/reduce.c, * src/state.h, src/symlist.c, src/symtab.c, src/symtab.h, src/tables.c: Adjust. * tests/input.at: Fix expectations (order changes). 2013-08-01 Akim Demaille build: ship the ASCII art figures We don't ship the *.txt files that are used to build the info file. Reported by Colin Daley. * doc/figs/example.txt: New. * doc/local.mk (bison.info): Depend on the txt files. And ship them. 2013-08-01 Akim Demaille doc: prefer the ".gv" extension to ".dot" See http://marc.info/?l=graphviz-devel&m=129418103126092 for the motivation (basically, some word processor now uses *.dot). * doc/figs/example-reduce.dot: Rename as... * doc/figs/example-reduce.gv: this. * doc/figs/example-shift.dot: Rename as... * doc/figs/example-shift.gv: this. * doc/figs/example.dot: Rename as... * doc/figs/example.gv: this. * doc/local.mk: Adjust. 2013-07-25 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2013-07-25 Akim Demaille version 3.0 * NEWS: Record release date. 2013-07-25 Akim Demaille regen 2013-07-25 Akim Demaille news: prepare 3.0 * NEWS (3.0): Reorder. 2013-07-25 Akim Demaille tests: fix invalid assignment when using variants in C++11 * tests/c++.at (Exception safety): In variant mode $$ is an instance of Object. Assigning YY_NULL in C++98 is incorrect, but behaves ok, as it assigns YY_NULL=0 using Object::operator= (char v). It is wrong in C++11 as there is operator for "$$ = nullptr". 2013-07-25 Akim Demaille yacc: beware of "uninitialized uses" warnings Again some issues with the fact that yylval is reported by GCC as possibly not initialized in some cases. Here, the case at hand is the %destructor. I am still not convinced that it is worth going all the trouble of using pragmas to disable temporarily some warnings, instead of just initializing the looking symbol once for all, but that's what Paul voted for, see . * data/c.m4 (b4_attribute_define): Define YY_IGNORE_MAYBE_UNINITIALIZED_BEGIN, YY_IGNORE_MAYBE_UNINITIALIZED_END, YY_INITIAL_VALUE here, as we will need them in the generation of the destructor function, which is defined in yacc.c before yyparse, which was in charge of defining these macros. * data/yacc.c (b4_declare_scanner_communication_variables): Simplify: trying to factor the definitions of the case pure and impure is too complex. Actually, it is not even clear that this macro should really exist, as even the calls are complex. Be careful not to issue a lone ";", as this is a statement, and C90 forbids declarations after statements ; so write "YY_INITIAL_VALUE(Decl;)", not "YY_INITIAL_VALUE(Decl);". 2013-07-25 Akim Demaille gnulib: update 2013-07-03 Akim Demaille tests: skip C++ tests if we can't compile a simple program There are possible conflicts between gnulib replacement functions (in ) and their C++ wrappers (in ). Trying to address these in configure seems too hard, and I don't know how to fix the issue in gnulib. Cowardly avoid the problem by skipping C++ tests when this happens. Reported by Stefano Lattarini. http://lists.gnu.org/archive/html/bug-bison/2013-06/msg00001.html * tests/atlocal.in (BISON_CXX_WORKS): Also set it to "skip" if we can't compile a simple program using . * tests/local.at: Comment changes. 2013-07-03 Akim Demaille tests: fix 'find' portability issues Reported by Stefano Lattarini. http://lists.gnu.org/archive/html/bug-bison/2013-06/msg00000.html * tests/output.at (AT_CHECK_OUTPUT): Use Perl instead. 2013-06-24 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2013-06-24 Akim Demaille version 2.7.91 * NEWS: Record release date. 2013-06-24 Akim Demaille NEWS: prepare for 2.7.91 * NEWS (2.7.91): Java push parsers. 2013-06-24 Akim Demaille java: rename YYMORE as YYPUSH_MORE for consistency with C http://lists.gnu.org/archive/html/bison-patches/2013-06/msg00008.html * data/lalr1.java, doc/bison.texi, tests/javapush.at: s/YYMORE/YYPUSH_MORE. 2013-06-21 Akim Demaille tests: fix Java push failure when running with BISON_USE_PUSH_FOR_PULL * tests/javapush.at (Trivial Push Parser with api.push-pull verification): When push for pull is enabled, there is one such function generated. 2013-06-21 Akim Demaille style: minor changes in the Java tests * tests/java.at (AT_CHECK_JAVA_GREP): Ignore the exit status. * tests/javapush.at (AT_CHECK_JAVA_GREP): Be more alike the previous one. Formating changes. Remove stray debugging "jj" file. 2013-06-21 Akim Demaille java: push: do not reset the error counter * data/lalr1.java (parse): here, when in push-pull is in "both" mode. This breaks the test suite, for instance make check TESTSUITEFLAGS='-d 388 BISON_USE_PUSH_FOR_PULL=1'. More generally make maintainer-push-check. 2013-06-14 Akim Demaille build: add Valgrind suppression file for GNU/Linux * build-aux/linux-gnu.valgrind: New. * build-aux/local.mk: Ship it. * configure.ac: Use it. 2013-06-13 Dennis Heimbigner java: add push-parser support * data/lalr1.java: Capture the declarations as m4 macros to avoid duplication. When push parsing, the declarations occur at the class instance level rather than within the parse() function. Change the way that the parser state is initialized. For push-parsing, the parse state declarations are moved to "push_parse_initialize()", which is called on the first invocation of "push_parse()". The %initial-action code is also inserted after the invocation of "push_parse_initialize()". The body of the parse loop is modified to return values at appropriate points when doing push parsing. In order to make push parsing work, it is necessary to divide YYNEWSTATE into two states: YYNEWSTATE and YYGETTOKEN. On the first call to push_parse(), the state is YYNEWSTATE. On all later entries, the state is set to YYGETTOKEN. The YYNEWSTATE switch arm falls through into YYGETTOKEN. YYGETTOKEN indicates that a new token is potentially needed. Normally, with a pull parser, this new token would be obtained by calling "yylex()". In the push parser, the value YYMORE is returned to the caller. On the next call to push_parse(), the parser will return to the YYGETTOKEN state and continue operation. * tests/javapush.at: New test file for java push parsing. * tests/testsuite.at: Use it. * tests/local.mk: Adjust. * doc/bison.texi (Java Push Parser Interface): New. 2013-06-11 Akim Demaille build: ship all the files, even if the C++ compiler is broken * examples/calc++/local.mk: Be sure to ship calc++.test even if the current C++ compiler is not sufficient to run the tests. 2013-06-05 Dennis Heimbigner style: comment changes in Java skeleton * data/lalr1.java: Here. 2013-06-03 Akim Demaille tests: fix a G++ warning * tests/c++.at: Use YY_NULL instead of 0 for the null pointer. And formatting changes. 2013-06-03 Akim Demaille build: fix a warning from clang * src/muscle-tab.c: Declare local functions static. 2013-05-30 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2013-05-30 Akim Demaille version 2.7.90 * NEWS: Record release date. 2013-05-30 Akim Demaille style: syntax-check fixes * data/yacc.c, src/Sbitset.c, src/Sbitset.h, src/muscle-tab.h, * src/output.c, src/parse-gram.y, src/reader.c, src/symtab.c, * src/uniqstr.c, src/uniqstr.h: Fix space before parens. * cfg.mk (_space_before_paren_exempt): Add needed exceptions. 2013-05-30 Akim Demaille xml: use %empty in the text output * data/xslt/xml2text.xsl: here. 2013-05-30 Akim Demaille build: locally disable new GCC warnings that fail on Flex generated code * configure.ac: here. 2013-05-30 Akim Demaille fix a memory leak * src/print-xml.c (num_escape_bufs): New. (print_xml): Be sure to release all the escape_bufs. 2013-05-30 Akim Demaille regen 2013-05-30 Akim Demaille build: be sure to include config.h first in the generated parser Using %code for config.h is wrong, as some headers will already have been included by Bison. In some cases, e.g., glibc's string.h, this results in some declaration not being made for lack of definition of _GNU_SOURCE, which is performed by config.h. * src/parse-gram.y: here. 2013-05-29 Akim Demaille Merge remote-tracking branch 'origin/maint' * origin/maint: maint: post-release administrivia version 2.7.1 regen 2013-05-29 Petr Machata drop unused options --raw, -n, -e, --include and -I * --raw appears to be ignored. It was marked as obsolete in the commit ec3bc39, and documented as no longer supported as of 1.29 (2001-09-07). Support for %raw appears to have been dropped in e9955c83 on 2002-06-11, but --raw was kept around. Maybe it's time to drop it as well? * Commit e9955c83 dropped support for %no-parser as well, and converted it to option. --no-parser was later dropped in 728c4be2 on 2007-08-12, but -n was kept around, probably as an omission. All three are documented as removed since 2.3b (2008-05-27). * -e existed for a single day in 2001. It was introduced in eeeb962b on 2001-11-27. The handling was removed in c7925b99 on 2001-11-28, but "e" was kept in the list of short options. Probably an omission. * --include appears to be dead code. The option sets a variable, but that variable is not used anywhere. It was added in f6bd5427 on 2001-11-26 as a %-directive, with comments that it's not yet implemented. It was converted to a command-line option later, but doesn't seem to ever have been actually implemented. 2013-05-28 Akim Demaille gnulib: update 2013-04-22 Akim Demaille diagnostics: always point to the first directive Some directives cannot be used several times (e.g., a given symbol may only have a single printer). In case of repeated definitions, an error is issued for the second definition, yet it is not discarded, and becomes the definition used for the rest of the file. This is not consistent with the idea that multiple definitions are not allowed: discard any repeated directive. * src/symtab.c (symbol_type_set, symbol_code_props_set) (semantic_type_code_props_set, symbol_class_set, symbol_translation): Discard repeated directives. * tests/input.at (Default %printer and %destructor redeclared) (Per-type %printer and %destructor redeclared): Update expectations. 2013-04-22 Akim Demaille tests: factor test for printer/desctructor redefined * tests/input.at (Default %printer and %destructor redeclared): Introduce AT_TEST to factor. 2013-04-22 Akim Demaille diagnostics: use appropriate location for useless precedence/associativity * src/symtab.c (symbol_precedence_set): Use prec_location, not location (which is the first occurrence of the symbol, possibly just %token). Also, as redefinitions are not allowed, keep the first values, not the subsequent ones. * tests/conflicts.at, tests/existing.at, tests/regression.at: Adjust. 2013-04-22 Akim Demaille tests: factor duplicate expected warnings * tests/existing.at: Instead of "t ? abc : aBc", write "a(t?b:B)c". 2013-04-19 Akim Demaille tests: enable -Wsign-compare and fix corresponding warnings -Wsign-compare was disabled for bison's own code, following gnulib's approach. However, the generated parsers should not trigger such warnings. Reported by Efi Fogel. http://lists.gnu.org/archive/html/help-bison/2013-04/msg00018.html See also http://stackoverflow.com/questions/16101062 for the weird "-(unsigned)i" piece of code. * configure.ac (warn_tests): Enable -Wsign-compare. * data/location.cc (position::add_): New. (position::lines, position::columns): Use it. * tests/actions.at (AT_CHECK_PRINTER_AND_DESTRUCTOR): Fix signedness issues. 2013-04-18 Akim Demaille muscle: check more cases of %define variables with code values * data/bison.m4 (b4_percent_define_check_kind): Fix overquotation. (api.location.type, api.position.type): Check they have code values here. * data/c++.m4 (api.location.type): No longer checked here. (parser_class_name): Check it here. * data/java.m4 (api.value.type, init_throws, lex_throws, parser_class_name) (throws, annotations, extends, implements): Check they have code values. * doc/bison.texi: Fix every incorrect occurrence of %define. Document the additional syntax for %define: code values. Document the additional syntax for -D/-F: string and code values. * tests/calc.at, tests/headers.at, tests/input.at, tests/java.at, * tests/local.at: Fix dependencies. 2013-04-18 Akim Demaille regen 2013-04-18 Akim Demaille parser: do not convert $ and @ in code values of %define variables * src/parse-gram.y (value: "{...}"): Just strip the braces, but pass the value as is. 2013-04-18 Akim Demaille parser: no longer use the "braceless" non-terminal The purpose of this symbol was only to factor function calls. As a result the actions were indeed simpler, but the grammar was somewhat uselessly obfuscated. Get rid of this symbol, but introduce functions to simplify dependencies. There is no (intended) changes of behavior here. * src/parse-gram.y (strip_braces, translate_code( (translate_code_braceless): New. (braceless): Remove, use "{...}" instead, and one of the previous functions depending on the context. (STRING, "%{...%}", EPILOGUE): Declare as , instead of , the difference between both is useless (well, I couldn't make sense of it, even after having read the initial commit that introduced them). (%union): Remove the now useless "chars" type. Adjust the printers. * src/scan-gram.l: Adjust. 2013-04-18 Akim Demaille regen 2013-04-18 Akim Demaille style: avoid %{...%} in our parser * src/parse-gram.y (%{...%}): Split in %code and %code requires. * src/location.h: Add missing includes for self containedness. 2013-04-18 Akim Demaille style: use %code for local function declarations in our parser * src/parse-gram.y (version_check, gram_error, char_name, lloc_default): Move their prototypes from %{...%} to %code. (YYLLOC_DEFAULT, YY_LOCATION_PRINT): Move from %{...%} to %code. (current_lhs): Move its implementation to the epilogue. 2013-04-16 Akim Demaille regen 2013-04-16 Akim Demaille muscle: check the kind of api.prefix, api.location.type * data/bison.m4: Check api.prefix. * data/c++.m4: Check api.location.type. * doc/bison.texi: Fix uses of api.value.type, api.prefix, api.location.type. Document {...} values for %define. * src/parse-gram.y: Fix use of api.prefix. * tests/calc.at: Fix uses of api.location.type. * tests/input.at: Check api.prefix, and api.location.type. 2013-04-15 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2013-04-15 Akim Demaille version 2.7.1 * NEWS: Record release date. 2013-04-15 Akim Demaille regen 2013-04-15 Akim Demaille muscle: enforce definition syntax for keyword variables * src/muscle-tab.c (muscle_percent_define_get_kind) (muscle_percent_define_check_kind): New. (muscle_percent_define_default): Variables with a default value are of "keyword" kind. (muscle_percent_define_flag_if, muscle_percent_define_check_values): Check that the variable is of keyword kind. * data/bison.m4: Likewise, but in M4. That is to say... (b4_percent_define_default): Define the kind when the variable is undefined. (b4_percent_define_check_kind): Use a better error message. (_b4_percent_define_check_values, _b4_percent_define_check_values): Former "enum" variables should be defined using the keyword syntax. * doc/bison.texi: Update. A couple of fixes. * tests/input.at (%define keyword variables): New. 2013-04-15 Akim Demaille muscle: let -D/-F support the three kinds of %define variable values See http://lists.gnu.org/archive/html/bison-patches/2013-04/msg00012.html * src/getargs.c (getargs): Recognize {value} and "value" for -D and -F. 2013-04-15 Akim Demaille muscle: minor refactoring * src/muscle-tab.c (muscle_percent_define_default): Reduce the scopes. 2013-04-15 Akim Demaille muscle: minor simplification which uncovers a missing warning * src/muscle-tab.c (muscle_percent_define_ensure): Discover the virtues of || to factor conditionals. * NEWS: As api.pure is no longer flagged as "used" by accident, we now have warnings for useless definitions. * tests/calc.at: So remove api.pure settings when running C++ tests, since C++ skeletons use a pure interface. 2013-04-15 Akim Demaille muscle: factor the field retrieval * src/muscle-tab.c (muscle_percent_define_get_raw): New. Use it where appropriate. (location_decode): No longer fetch the value from the table, take the value as argument. 2013-04-15 Akim Demaille muscle: factor the handling of used variables * src/muscle-tab.c (muscle_percent_define_use): New, corresponding to b4_percent_define_use. Use it where appropriate. 2013-04-15 Akim Demaille muscle: factor the computation of variable names * src/muscle-tab.c (muscle_name): New. Use it. Propagate "uniqstr" as value type instead of plain "char const *". 2013-04-15 Akim Demaille muscle: factor the kind check in M4 * data/bison.m4 (b4_percent_define_check_kind): New. Use it to check api.token.prefix. * data/c++.m4: Check the kind of api.namespace. * doc/bison.texi: Update a reference to former 'namespace' variable. * tests/input.at ("%define" code variables): Check api.namespace. 2013-04-15 Akim Demaille muscle: factor conditionals on defined %define variables * data/bison.m4 (b4_percent_define_ifdef_): New. Use it where appropriate. 2013-04-11 Akim Demaille api.token.prefix: use code values * data/bison.m4: Remove useless (and incorrect: m4_* instead of b4_*) default assignment to api.token.prefix. Check that api.token.prefix is assigned code. * tests/input.at (%define code variables): New test. * NEWS, doc/bison.texi, tests/c++.at, tests/calc.at, * tests/java.at, tests/local.at: Adjust to use braces. 2013-04-11 Akim Demaille c++: fix several issues with locations Reported by Daniel Frużyński. http://lists.gnu.org/archive/html/bug-bison/2013-02/msg00000.html * data/location.cc (position::columns, position::lines): Check for underflow. Fix some weird function signatures. (location): Accept signed integers as arguments where appropriate. Add operator- and operator+=. * doc/bison.texi (C++ position, C++ location): Various fixes and completion. * tests/c++.at (C++ Locations): New tests. 2013-04-11 Akim Demaille muscles: be sure that %code snippets are not glue together on a single line Recently "braceless" in the parser was changed so that an eol was no longer added to the value. This is not correct when a %code is used multiple times, because the syncline of the next snippet might be appended to the last (and not ended) line of the previous snippet. * src/muscle-tab.h (muscle_grow): Make it private. * src/muscle-tab.c (muscle_grow): Accept a fourth argument: a required terminator. Adjust callers. * tests/input.at (Multiple %code): New. 2013-04-11 Akim Demaille style: fix comments * tests/actions.at: Fix incorrect "prototype". 2013-04-10 Akim Demaille Merge remote-tracking branch 'origin/maint' * origin/maint: glr.cc: fix a clang warning maint: update copyright years build: fix VPATH issue build: avoid clang's colored diagnostics in the test suite tests: please clang and use ".cc", not ".c", for C++ input gnulib: update skeletons: avoid empty switch constructs lalr1.cc: fix compiler warnings yacc.c: do not use __attribute__ unprotected tests: style changes 2013-04-09 Akim Demaille api.value.type: use keyword/brace values Suggested by Joel E. Denny. http://lists.gnu.org/archive/html/bison-patches/2013-03/msg00016.html * data/bison.m4 (b4_percent_define_get_kind): New. (b4_variant_flag): Check that api.value.type is defined as the 'variant' keyword value. * data/c.m4 (_b4_value_type_setup_keyword): New. (b4_value_type_setup): Use it to simplify reading. Use b4_define_silent. Decode api.value.type, including its type. (b4_value_type_define): Likewise. * data/c++.m4 (b4_value_type_declare): Adjust the decoding of api.value.type, taking its kind into account. * doc/bison.texi: Adjust all the examples to the new syntax. * NEWS: Ditto. * tests/types.at: Adjust 2013-04-09 Akim Demaille api.value.type: diagnose guaranteed failure with --yacc Instead of generating invalid C code, generate an error when --yacc and '%define api.value.type union' are used together. * data/bison.m4: Issue an error in this case. * tests/types.at (%yacc vs. %define api.value.type union): New, check this error. * doc/bison.texi (Type Generation): Document it. * tests/output.at: Check that '-o y.tab.c' and '-y' behave equally wrt generated file names. * NEWS (Use of YACC='bison -y'): New. Promote the use of 'bison -o y.tab.c'. 2013-04-09 Akim Demaille doc: style changes * doc/bison.texi (Destructor Decl, Printer Decl): Group series of %token and %type together. 2013-04-09 Akim Demaille doc: display locations in error as recommended by GNU Coding Standards * doc/bison.texi (Actions and Locations): here. 2013-04-09 Akim Demaille doc: api.value.type union * doc/bison.texi (Type Generation): New section. (Multi-function Calc): Convert to use api.value.type=union. 2013-04-09 Akim Demaille doc: move the section about "%union" where types are discussed * doc/bison.texi (Union Decl): Move to... (Defining Language Semantics): here. 2013-04-09 Akim Demaille doc: deprecate #define YYSTYPE in favor of %define api.value.type * doc/bison.texi: Convert examples with YYSTYPE to use api.value.type. Deprecate YYSTYPE. 2013-04-09 Akim Demaille value type: accept "->" in type tags Provide a means to dereference pointers when defining tags. One example could be: %code requires { typedef struct ListElementType { union value { int intVal; float floatVal; char* charptrVal; } value; struct ListElementType* next; } ListElementType; } %union { ListElementType* list; } %token value.charptrVal> STRING %token value.intVal> INTEGER %token value.floatVal> REAL %type ElementList LiteralType * src/scan-code.l, src/scan-gram.l: Accept "->" in tags. * tests/types.at: Add more test cases to cover this case. 2013-04-09 Akim Demaille style: simplify the scanning of type tags * src/scan-gram.l: Remove the rule for simple tags: the "complex" case subsumes it. It was more efficient, but duplicated the code for a negligible benefit. 2013-04-09 Akim Demaille api.value.type: implement proper support, check, and document * data/c.m4 (b4_symbol_type_register, b4_type_define_tag) (b4_symbol_value_union, b4_value_type_setup_union) (b4_value_type_setup_variant, b4_value_type_setup): New. (b4_value_type_define): Use it to set up properly the type. Handle the various possible values of api.value.type. * data/c++.m4 (b4_value_type_declare): Likewise. * data/lalr1.cc (b4_value_type_setup_variant): Redefine. * tests/types.at: New. Exercise all the C/C++ skeletons with different types of api.value.type values. * tests/local.mk, tests/testsuite.at: Use it. * doc/bison.texi (%define Summary): Document api.value.type. * NEWS: Advertise it, together with api.token.constructor. 2013-04-09 Akim Demaille m4: allow the definition of side-effect only macros * data/bison.m4 (b4_divert_kill, b4_define_silent): New. * data/c.m4: Comment change. 2013-04-09 Akim Demaille variant: fix inconsistent quotation * data/variant.hh (b4_char_sizeof): De-overquote. (b4_value_type_declare): De-underquote. 2013-04-09 Akim Demaille m4: style changes in error messages * data/bison.m4: Use $0 to denote the current macro's name. 2013-04-08 Akim Demaille glr.cc: fix a clang warning * data/glr.cc (b4_epilogue): Be sure to end with an end-of-line, so that the file does end with one. 2013-04-08 Akim Demaille maint: update copyright years Run "make update-copyright". 2013-04-08 Akim Demaille build: fix VPATH issue * Makefile.am (update-b4-copyright, update-package-copyright-year): Fix path to build-aux. 2013-04-08 Akim Demaille build: avoid clang's colored diagnostics in the test suite The syncline tests, which try to recognize compiler diagnostics, are confused by escapes for colors. * configure.ac (warn_tests): New, to factor the warnings for both C and C++ tests. Add -fno-color-diagnostics to it. * tests/local.at (AT_TEST_TABLES_AND_PARSE): Do not remove glue together compiler flags. 2013-04-08 Akim Demaille tests: please clang and use ".cc", not ".c", for C++ input When fed with foo.c, clang++ 3.2 answers: clang: error: treating 'c' input as 'c++' when in C++ mode, this behavior is deprecated * tests/output.at (AT_CHECK_OUTPUT_FILE_NAME): Use *.cc and *.hh for C++. 2013-04-08 Akim Demaille gnulib: update 2013-04-08 Akim Demaille skeletons: avoid empty switch constructs Reported by Rob Conde. http://lists.gnu.org/archive/html/bug-bison/2013-03/msg00003.html * data/c.m4 (b4_symbol_actions): Rename as... (_b4_symbol_actions): this. (b4_symbol_actions): New wrapper. Do not emit empty switches. Adjust all b4_symbol_actions callers. 2013-04-08 Akim Demaille lalr1.cc: fix compiler warnings Reported by Rob Conde. http://lists.gnu.org/archive/html/bug-bison/2013-03/msg00003.html * data/stack.hh (operator=, stack(const stack&)): Make this class uncopyable, i.e., "undefine" these operators: make them private and don't implement them. (clear): New. * data/lalr1.cc: Use it instead of an assignment. (parser): Make this class uncopyable. 2013-04-08 Akim Demaille yacc.c: do not use __attribute__ unprotected Reported by Victor Khomenko. http://lists.gnu.org/archive/html/bug-bison/2013-04/msg00001.html * data/glr.c (YYUSE, __attribute__): Fuse their definition into... * data/c.m4 (b4_attribute_define): this new macro. * data/yacc.c, data/glr.c: Use it. 2013-04-05 Akim Demaille api.namespace: demonstrate and use {...} values instead of "..." values * tests/c++.at, tests/input.at: Use "%define api.namespace {foo}" instead of using quotes. * tests/local.at (AT_SETUP_STRIP, AT_NAME_PREFIX): Recognize uses of braces instead of quotes. * doc/bison.texi: Use braces for api.namespace's values. 2013-04-05 Akim Demaille grammar: do not add a \n at the end of blocks of code Now that we use "braceless" (which is {...} blocks of code with initial and final braces stripped) to denote "short" values (such as api.namespaces), the added end-of-line is a nuisance. As a matter of fact, this extra-safety was useless, as every expansion of "braceless" (aka, "user code") is followed by an end of line. * src/parse-gram.y, src/parse-gram.c (braceless): Instead of replacing the final brace by \n, just delete the brace. 2013-04-04 Akim Demaille regen 2013-04-04 Akim Demaille grammar: record the kind of %define variable values Provide a means to tell the difference between "keyword" values (e.g., %define api.pull both), "string" values (e.g., %define file.name "foo"), and "code" values (e.g., %define api.namespace {calc}). Suggested by Joel E. Denny. http://lists.gnu.org/archive/html/bison-patches/2013-03/msg00016.html * src/muscle-tab.h, src/muscle-tab.c (muscle_kind, muscle_kind_new) (muscle_kind_string): New. (muscle_percent_define_insert): Take the kind as new argument. Insert it in the muscle table. Adjust callers. * src/getargs.c: Adjust callers. * src/parse-gram.y: Ditto. (content.opt): Remove, replaced by... (value): this new non-terminal, whose semantics value is stored in the new "value" union member. Provide a printer. Support values in braces in additions to keyword and string values. fuse me 2013-04-04 Akim Demaille style: fix comments * src/muscle-tab.c (muscle_percent_define_ensure): Update obsolete comments. 2013-04-04 Akim Demaille regen 2013-04-04 Akim Demaille grammar: style changes * src/parse-gram.y (PARAM_TYPE): Remove useless typedef guard. There's a header guard. Use 'yyo' with %printer. Use a consistent style for %union one-liners. 2013-04-04 Akim Demaille grammar: split %union to group together related aspects * src/parse-gram.y (INT): Fuse the %type and %token declaration. Move its %union right before its introduction. (%union): Split in several %unions, right before their use. 2013-04-04 Akim Demaille muscle: refactor * src/muscle-tab.c (muscle_lookup, muscle_entry_new): New. (muscle_insert, muscle_grow, muscle_find_const, muscle_find): Use them. 2013-04-03 Akim Demaille style: comment changes * src/muscle-tab.c: Move the documentation of public functions to... * src/muscle-tab.h: here. Fix comment consistency issues. 2013-04-03 Akim Demaille muscle: minor refactoring * src/muscle-tab.h (MUSCLE_INSERT_C_STRING): Use MUSCLE_INSERT_STRING. 2013-03-06 Akim Demaille regen 2013-03-06 Valentin Tolmer gram: correct token numbering in precedence declarations In a precedence declaration, when tokens are declared with a litteral character (e.g., 'a') or with a identifier (e.g., B), Bison behaved differently: the litteral tokens would be numbered first, and then the other ones, leading to the following grammar: %right A B 'c' 'd' being numbered as such: 'c' 'd' A B. * src/parse-gram.y (symbol.prec): Set the symbol number when reading the symbols. * tests/conflicts.at (Token declaration order: literals vs. identifiers): New. 2013-03-04 Akim Demaille maint: update autoconf submodule * submodules/autoconf: Up to master. No significant changes in the files we use (m4sugar.m4 and foreach.m4). 2013-03-04 Akim Demaille diagnostics: no longer include the yacc category in -Wall It would be a pity to warn the users against Bison features... http://lists.gnu.org/archive/html/bison-patches/2013-02/msg00107.html * src/complain.h, src/complain.c (Wall): Disable Wyacc. (Weverything): New (hidden so far) category which really denotes all the categories (what used to be Wall). (warnings_args, warnings_types): Adjust. (warning_argmatch): Now !none = Weverything and conversely, no longer Wall. * NEWS, doc/bison.texi, src/getargs.c: Adjust the documentation. * tests/input.at (-Werror is not affected by -Wnone and -Wall): Adjust by not using a -Wyacc type of warning. 2013-03-04 Akim Demaille grammar: no longer detect and cure missing semicolon at end of actions Bison 3.0 is already breaking backward compatibility with other features. It is an appropriate time to drop this feature. Note that it was disabled when --yacc is passed. See http://lists.gnu.org/archive/html/bison-patches/2013-02/msg00102.html Basically, revert e8cd1ad655bcc704b06fb2f191dc3ac1df32b796. * src/scan-code.l (braces_level, need_semicolon, in_cpp): Remove. Remove every rule needed to detect and add missing semicolon. * tests/actions.at (Fix user actions without a trailing semicolon): Remove. * NEWS: Adjust. 2013-03-04 Akim Demaille build: stop using bison -y * Makefile.am (YACC): Pass -o y.tab.c, so that ylwrap is happy, and yet we don't pass --yacc to bison. (AM_YFLAGS): Disable Yacc warnings. 2013-02-23 Akim Demaille c++: rename b4_semantic_type_declare as b4_value_type_declare This is to match the names used in C and api.value.type, even if the parser actually defines semantic_type. * data/c++.m4 (b4_semantic_type_declare): Rename as... (b4_value_type_declare): this. * data/variant.hh: Likewise. 2013-02-23 Akim Demaille news: typo * NEWS: here. 2013-02-23 Akim Demaille style: space changes in the tests * tests/local.at: here. 2013-02-22 Akim Demaille style: formatting changes in the doc * doc/bison.texi: Use @file where appropriate. 2013-02-19 Akim Demaille tests: fix invalid C++11 code * tests/c++.at (Object): Somehow instances of Object were assigned YY_NULL, which is 0 most of the time (that case passes), but is nullptr in C++11, and there is nothing in Object to support such an assignment (failure). Use 0 as value, and provide the needed assignment operator. Also, use a more natural order within the class definition. 2013-02-19 Akim Demaille tests: fix failures with G++ 4.8 in Flex scanner * configure.ac (WARN_NO_NULL_CONVERSION_CXXFLAGS): Rename as... (FLEX_SCANNER_CXXFLAGS): this. Pass -Wno-zero-as-null-pointer-constant to G++ if it supports it. * examples/calc++/local.mk: Adjust. 2013-02-19 Akim Demaille regen 2013-02-19 Akim Demaille gnulib: update 2013-02-19 Akim Demaille style: rename variant private members * data/variant.hh (buffer, tname, as_, raw, align_me): Rename as... (yybuffer_, yytname_,yyas_, yyraw, yyalign_me): these. 2013-02-19 Akim Demaille style: space changes * data/variant.hh: Be sure to leave a space before arguments in function calls. 2013-02-19 Akim Demaille variant: fix G++ 4.4 warnings The changes by Théophile Ranquet about type punning issues need to be extend to in-place new to please G++ 4.4.7. * data/variant.hh (variant::as_): New, factors the casts that avoid compiler warnings. (as, build): Use them. 2013-02-18 Akim Demaille news: spell fixes * NEWS: here. 2013-02-18 Akim Demaille diagnostics: factor and enhance messages about duplicate rule directives When reporting a duplicate directive on a rule, point to its first occurrence: one.y:11.10-15: error: only one %empty allowed per rule %empty {} %empty ^^^^^^ one.y:11.3-8: previous declaration %empty {} %empty ^^^^^^ And consistently discard the second one. * src/complain.h, src/complain.c (duplicate_directive): New. * src/reader.c: Use it where appropriate. * src/symlist.h, src/symlist.c (symbol_list): Add a dprec_location member. * tests/actions.at: Adjust expected output. 2013-02-18 Akim Demaille style: no longer use backquotes * tests/actions.at, tests/atlocal.in, tests/c++.at, tests/calc.at, * tests/conflicts.at, tests/existing.at, tests/glr-regression.at, * tests/input.at, tests/java.at, tests/local.at, tests/sets.at, * tests/synclines.at, doc/bison.texi, lib/libiberty.h, lib/timevar.h: Use single quotes. 2013-02-18 Akim Demaille style: no longer use backquotes * README, REFERENCES, TODO, configure.ac, data/README, data/bison.m4, * data/c++.m4, data/c.m4, data/java.m4, data/lalr1.cc, * data/lalr1.java, data/yacc.c, doc/local.mk, etc/bench.pl.in, * src/conflicts.c, src/files.c, src/getargs.c, src/gram.h, src/lalr.c, * src/location.c, src/location.h, src/muscle-tab.c, src/muscle-tab.h, * src/output.c, src/parse-gram.c, src/parse-gram.y, src/print-xml.c, * src/print.c, src/reader.c, src/reduce.c, src/scan-skel.l, * src/symtab.h, src/system.h, src/tables.c: Use single quotes, as currently recommended by the GNU Coding Standards. 2013-02-18 Akim Demaille style: no longer use backquotes in messages * src/getargs.c (usage): Use single quotes. 2013-02-18 Akim Demaille doc: use %empty instead of /* empty */ * doc/bison.texi: Change the comments into explicit %empty. 2013-02-18 Akim Demaille doc: introduce %empty and -Wempty-rule * doc/bison.texi (Grammar Rules): Make it a @section which contains... (Rules Syntax): this new subsection (with the previous contents of "Grammar Rules". (Empty Rules): New subsection, extracted from the former "Grammar Rules". Document %empty. (Recursion): New a subsection of "Grammar Rules". Complete a few index entries. (Bison Options): Document -Wempty-rule. 2013-02-18 Akim Demaille report: use %empty to denote empty rules * src/gram.c (rule_rhs_print): Use %empty for empty rules. * tests/conflicts.at, tests/regression.at, tests/sets.at: Adjust. 2013-02-18 Akim Demaille diagnostics: %empty enables -Wempty-rule * src/complain.h, src/complain.c (warning_is_unset): New. * src/reader.c (grammar_current_rule_empty_set): If enabled -Wempty-rule, if not disabled. * tests/actions.at (Implicitly empty rule): Check this feature. Also check that -Wno-empty-rule does disable this warning. 2013-02-18 Akim Demaille -Wempty-rule: diagnose empty rules without %empty * src/complain.h, src/complain.c (warning_empty_rule, Wempty_rule): New warning category. (warnings_args, warnings_types): Adjust. * src/reader.c (grammar_rule_check): Check the empty rules are flagged by %empty. * tests/actions.at (Implicitly empty rule): New. * tests/existing.at: Add expected warnings. 2013-02-18 Akim Demaille tests: use %empty * tests/actions.at, tests/input.at, tests/reduce.at, * tests/regression.at: here. 2013-02-18 Akim Demaille regen 2013-02-18 Akim Demaille parser: use %empty Avoid that Bison's own use of "bison -Wall" trigger warnings. * src/parse-gram.y: Use %empty for every empty rule. 2013-02-18 Akim Demaille grammar: introduce %empty Provide a means to explicitly denote empty right-hand sides of rules: instead of exp: { ... } allow exp: %empty { ... } Make sure that %empty is properly used. With help from Joel E. Denny and Gabriel Rassoul. http://lists.gnu.org/archive/html/bison-patches/2013-01/msg00142.html * src/reader.h, src/reader.c (grammar_current_rule_empty_set): New. * src/parse-gram.y (%empty): New token. Use it. * src/scan-gram.l (%empty): Scan it. * src/reader.c (grammar_rule_check): Check that %empty is properly used. * tests/actions.at (Invalid uses of %empty, Valid uses of %empty): New. 2013-02-16 Akim Demaille getargs: minor simplification * src/getargs.c (flag_argmatch): Simplify the handling of "none". 2013-02-16 Akim Demaille style: move argument handling of -W into the diagnostics module This allows to reduce the number of public interfaces. * src/getargs.c (--yacc): Use warning_argmatch instead of tweaking directly warnings_flag (which will be private). (warning_argmatch, warnings_argmatch): Move to... * src/complain.h, src/complain.c: here. * src/getargs.h, src/getargs.c (warnings_args, warnings_types): Move to... * src/complain.c: here, now private. * src/complain.h (severity, warnings_flag): Move to... * src/complain.c: here, now private. 2013-02-16 Akim Demaille diagnostics: revamp the handling of -Werror Recent discussions with Joel E. Denny (http://lists.gnu.org/archive/html/bison-patches/2013-02/msg00026.html) show that it is desirable to tell the difference between an option that was explicitly disabled with -Wno-foo, as opposed to be left unset. The current framework does not allow this. Instead of having a first int to store which options are enabled, and another to store which are turned into errors, use an array that for each warning category tells its status: disabled, unset, warning, error. * src/complain.h, src/complain.c (warning_bit): New enum. (warnings): Use it. (severity): New enum. (warnings_flag): Now an array of severity. (errors_flag): Remove, now done by warnings_flag. (complain_init): New function, to initialie warnings_flag. (warnings_are_errors): New Boolean, for -Werror. * src/complain.c (warning_severity): New. (warnings_print_categories, complains): Use it. * src/getargs.c (warning_argmatch): Adjust to use warnings_flag. (warnings_argmatch): Ditto. Handle -Werror and -Wno-error here. (getargs): Adjust. * src/main.c (main): Call complain_init. * tests/input.at (Invalid options): Add more corner cases. 2013-02-14 Akim Demaille options: simplify the handling of -W * src/getargs.c (warnings_argmatch, warning_argmatch): Simplify by replacing function arguments with their actual values. (WARNING_ARGMATCH): Remove, useless. Adjust callers. 2013-02-14 Akim Demaille options: don't accept "error=" for -f and -r * src/getargs.c (warning_argmatch, warnings_argmatch, WARNINGS_ARGMATCH): New. Use them for -W/--warning. They are copied from... (flag_argmatch, flags_argmatch, FLAGS_ARGMATCH): these. Simplify by removing the support for "error". * tests/input.at (Invalid options): New. * TODO (Laxism in Bison invocation arguments): Remove. 2013-02-14 Akim Demaille diagnostics: factor the list of warning names * src/getargs.h, src/getargs.c (warnings_args, warnings_types): Make them public. * src/complain.h, src/complain.c (warnings_print_categories): Its only use outside complain.c was removed in a recent commit, so make it static. Simplify its implementation. Use warnings_args and warnings_types. * src/muscle-tab.c (muscle_percent_define_check_values): Make it silent. 2013-02-14 Akim Demaille diagnostics: no longer pretty-print rules in error messages, carets suffice * src/gram.c (grammar_rules_useless_report): Let -fcaret handle the pretty-printing of the guilty rules. (rule_print): Inline in its only use. * tests/conflicts.at, tests/existing.at, tests/reduce.at, * tests/regression.at: Adjust. * NEWS: Document. 2013-02-14 Akim Demaille options: no longer document warnings when diagnosing an invalid -W The argmatch functions accept prefixes of the alternatives (like getopt does for long options). Bison uses this to document the warning categories. This is troublesome: it duplicates the --help documentation, it is not gettextized, it is displayed with ugly quotes (because argmatch uses it to display the list of possible answers), and it prevents straighforward uses of the tables of valid warning categories (for instance so that warning diagnostics end with the name of the warning). The "hidden" option --trace uses the same trick, but it does not need to be translated, nor to be described in --help. * src/getargs.c (warnings_args): Remove pseudo documentation. Comment changes. 2013-02-11 Akim Demaille tests: enlarge the allowed duration for calc tests Hydra "often" fails on this test: 252. calc.at:658: 252. Calculator %glr-parser api.pure parse.error=verbose %debug %locations %defines api.prefix="calc" %verbose %yacc %parse-param {semantic_value *result} %parse-param {int *count} (calc.at:658): FAILED * tests/calc.at: Give 200s instead of 100s. Use AT_DEBUG_IF. 2013-02-11 Akim Demaille debug: improve the display of symbol lists * src/symtab.c (symbol_print): Remove useless quotes (the symbol already has quotes). Prefer fputs. * src/symlist.c (symbol_list_syms_print): Likewise. Fix separators. 2013-02-09 Akim Demaille style: minor changes * src/complain.c: Space changes. * src/reader.c: Comment changes. Avoid && in assertions. * src/location.c: Move comments to... * src/location.h: here. * src/symlist.h, src/symlist.c: Create a pseudo section for members that apply to the rule. 2013-02-08 Akim Demaille news: restructure, document variants for C++ * NEWS: here. 2013-02-08 Akim Demaille c++: api.token.constructor requires api.value.type=variant Eventually it should also support "union". * data/glr.cc: Move this check to... * data/c++.m4: here, as lalr1.cc is affected too. 2013-02-05 Akim Demaille build: restore C90 compatibility * src/parse-gram.y, src/parse-gram.c: Don't use // comments. 2013-02-05 Akim Demaille doc: use @group to improve page breaking * doc/bison.texi: here. 2013-02-04 Akim Demaille style: rename internal "stype" as "union_members" for clarity "stype" is quite unclear, and it also collides with the former %define variable that had the same name (replaced by api.value.type). * src/parse-gram.y (stype): Rename as... (union_members): this. * data/bison.m4: Adjust. (b4_user_stype): Rename as... (b4_user_union_members): this. * data/c++.m4, data/c.m4: Adjust. * src/parse-gram.c: regen. 2013-02-04 Akim Demaille tests: improve the language independance layer * tests/local.at (_AT_LANG_DISPATCH): New, shamelessly stolen from Autoconf's _AT_LANG_DISPATCH. (AT_LANG_DISPATCH): New. (AT_YYERROR_FORMALS, AT_YYERROR_PROTOTYPE, AT_YYERROR_DECLARE_EXTERN) (AT_YYERROR_DECLARE, AT_YYERROR_DEFINE, AT_MAIN_DEFINE, AT_COMPILE) (AT_FULL_COMPILE): Use AT_LANG_DISPATCH instead of an ad hoc m4_case. 2013-02-04 Akim Demaille regen 2013-02-04 Akim Demaille style: space changes in the parser * src/parse-gram.y: Fix spaces. 2013-02-04 Akim Demaille parser: use api.pure full * src/parse-gram.y: Use api.pure full instead of silly macro tricks. 2013-02-04 Akim Demaille style: use a for loop instead of a while loop, and scope reduction * src/reader.c (packgram): Improve readability. The parser calls grammar_current_rule_end at the end of every rhs, which adds a NULL to separate the rules. So there is no need to check whether "p" is non-null before proceeding. 2013-02-01 Theophile Ranquet variants: stylistic change * data/variant.hh (tname): Respect the GNU Coding Standards for this pointer's declaration. 2013-02-01 Theophile Ranquet grammar: free the association tracking graph The graph introduced by Valentin wasn't free'd after use. * src/symtab.c (assoc_free): New, clear the array of linked lists with... (linkedlist_free): This, new. (print_precedence_warnings): Call assoc_free when done. (print_assoc_warnings): Free used_assoc after use. 2013-02-01 Theophile Ranquet tests: use AT_FULL_COMPILE where possible * tests/c++.at (C++ Variant-based Symbol, Variants): Here. Rename the generated input files to use .y instead of .yy, as a requirement for using AT_FULL_COMPILE instead of a combination of AT_BISON_CHECK and AT_BISON_COMPILE_CXX. 2013-02-01 Theophile Ranquet variants: avoid type punning issue This is based on what is recommended by both Scott Meyers, in 'Effective C++', and Andrei Alexandrescu and Herb Sutter in 'C++ Coding Standards'. Use a static_cast on void* rather than directly use a reinterpret_cast, which can have nefarious effects on objects. However, even though following this guideline is good practice in general, I am not quite sure how relevant it is when applied to conversions from POD to objects. Actually, it might very well be the opposite: isn't this exactly what reinterpret_cast is for? What we really want *is* to transmit the memory map as a series of bytes, which, if I am correct, falls into the kind of "low level" hack for which this cast is meant. In any case, this silences the warning, which will be greatly appreciated by anyone using variants with a compiler supporting -fstrict-aliasing. * data/variant.hh (as): Here. * tests/c++.at (Exception safety, C++ Variant-based Symbols, Variants): Don't use NO_STRICT_ALIAS_CXXFLAGS (revert commit ddb9db15), as type punning is no longer an issue. * tests/atlocal.in, configure.ac (NO_STRICT_ALIAS_CXXFLAGS): Remove definition. * examples/local.mk (NO_STRICT_ALIAS_CXXFLAGS): Remove from AM_CXXFLAGS. * doc/bison.texi: Don't mention type punning issues. 2013-02-01 Theophile Ranquet todo: update Reformulate and give more details on my thoughts concerning the graphical visualization, and add an entry about a bug in the options processing for warnings as errors. * TODO: Here. 2013-02-01 Akim Demaille regen 2013-02-01 Akim Demaille location: pass the location first * src/location.h, src/location.c (location_print): For consistency with other data structures and other location_* routines, pass the location argument first. * src/complain.c: Adjust. (location_caret): Likewise. * src/parse-gram.y: Adjust. 2013-02-01 Akim Demaille symlist: use the right stream * src/symlist.c (symbol_list_syms_print): Use "f", not stderr. 2013-01-30 Akim Demaille tests: put two related tests together * tests/conflicts.at (Useless associativity warning): Move next to "Useless precedence warning". 2013-01-30 Akim Demaille news: name contributors * NEWS: here. 2013-01-30 Valentin Tolmer warnings: introduce -Wprecedence The new warning category "precedence" flags useless precedence and associativity. -Wprecedence can now be used, it is disabled by default. The warnings about precedence and associativity are grouped into one, and the testsuite was corrected accordingly. * src/complain.h (warnings): Introduce "precedence". * src/complain.c (warnings_print_categories): Adjust. * src/getargs.c (warnings_args, warning_types): Likewise. * src/symtab.h, src/symtab.c (print_associativity_warnings): Remove. * src/symtab.h (register_assoc): Correct arguments. * src/symtab.c (print_precedence_warnings): Print both warnings together. * doc/bison.texi (Bison options): Document the warnings and provide an example. * tests/conflicts.at, tests/existing.at, tests/local.at, * tests/regression.at: Adapt the testsuite for the new category (-Wprecedence instead of -Wother where appropriate). 2013-01-30 Akim Demaille build: avoid clang's colored diagnostics in the test suite The syncline tests, which try to recognize compiler diagnostics, are confused by escapes for colors. * configure.ac (warn_tests): New, to factor the warnings for both C and C++ tests. Add -fno-color-diagnostics to it. * tests/local.at (AT_TEST_TABLES_AND_PARSE): Do not remove glue together compiler flags. 2013-01-30 Akim Demaille build: please Clang++ 3.2+ on Flex scanners Clang++, with -Wall, rejects code generated by Flex (for C scanners): CXX examples/calc++/examples_calc___calc__-calc++-scanner.o In file included from examples/calc++/calc++-scanner.cc:1: error: implicit conversion of NULL constant to 'bool' [-Werror,-Wnull-conversion] if ( ! ( (yy_buffer_stack) ? (yy_buffer_stack)[(yy_buffer_stack_top)] : __null) ) { ~ ^~~~~~ false * configure.ac (WARN_NO_NULL_CONVERSION_CXXFLAGS): Compute it. * examples/calc++/local.mk (examples_calc___calc___CXXFLAGS): Use it. 2013-01-30 Valentin Tolmer grammar: record used associativity and print useless ones Record which symbol associativity is used, and display useless ones. * src/symtab.h, src/symtab.c (register_assoc, print_assoc_warnings): New * src/symtab.c (init_assoc, is_assoc_used): New * src/main.c: Use print_assoc_warnings * src/conflicts.c: Use register_assoc * tests/conflicts.at (Useless associativity warning): New. Due to the new warning, many tests had to be updated. * tests/conflicts.at tests/existing.at tests/regression.at: Add the associativity warning in the expected results. * tests/java.at: Fix the java calculator's grammar to remove a useless associativity. * doc/bison.texi (mfcalc example): Fix associativity to remove warning. 2013-01-29 Valentin Tolmer grammar: warn about unused precedence for symbols Symbols with precedence but no associativity, and whose precedence is never used, can be declared with %token instead. The used precedence relationships are recorded and a warning about useless ones is issued. * src/conflicts.c (resolve_sr_conflict): Record precedence relation. * src/symtab.c, src/symtab.h (prec_nodes, init_prec_nodes) (symgraphlink_new, register_precedence_second_symbol) (print_precedence_warnings): New. Record relationships in a graph and warn about useless ones. * src/main.c (main): Print precedence warnings. * tests/conflicts.at: New. 2013-01-29 Theophile Ranquet variants: remove the 'built' assertions When using %define parse.assert, the variants come with additional variables that are useful for development purposes. One is a Boolean indicating if the variant is built (to make sure we don't read a non-built variant), and the other is a string describing the stored type. There is no need to have both of these, the string is enough. * data/variant.hh (built): Remove. 2013-01-29 Akim Demaille style: indentation fixes * src/parse-gram.y: here. 2013-01-29 Akim Demaille maint: be sure to neutralize out-of-tree paths from our parser * tests/bison.in: Adjust to support fixed versions of ylwrap. 2013-01-29 Theophile Ranquet m4: generate a basic_symbol constructor for each symbol type Recently, there was a slightly vicious bug hidden in the make_ functions: parser::symbol_type parser::make_TEXT (const ::std::string& v) { return symbol_type (token::TOK_TEXT, v); } The constructor for symbol_type doesn't take an ::std::string& as argument, but a constant variant. However, because there is a variant constructor which takes an ::std::string&, this caused the implicit construction of a built variant. Considering that the variant argument for the symbol_type constructor was cv-qualified, this temporary variant was never destroyed. As a temporary solution, the symbol was built in two stages: symbol_type res (token::TOK_TEXT); res.value.build< ::std::string&> (v); return res; However, the solution introduced in this patch contributes to letting the symbols handle themselves, by supplying them with constructors that take a non-variant value and build the symbol's own variant with that value. * data/variant.hh (b4_symbol_constructor_define_): Use the new constructors rather than building in a temporary symbol. (b4_basic_symbol_constructor_declare, b4_basic_symbol_constructor_define): New macros generating the constructors. * data/c++.m4 (basic_symbol): Invoke the macros here. 2013-01-29 Theophile Ranquet c++: minor stylistic changes * data/c++m4: Remove useless comment lines. * data/variant.hh (self_type): Use this typedef instead of variant. (b4_symbol_constructor_define_): Remove commented-out line, and stylistic change (avoid blank line). 2013-01-29 Akim Demaille c++: please G++ 4.8 with -O3: type puning issue * tests/c++.at (Exception safety): Now that this test covers variants, pass -fno-strict-aliasing to g++. 2013-01-29 Akim Demaille c++: please G++ 4.8 with -O3: array bounds * data/c++.m4, data/lalr1.cc (by_state, by_type): Do not use -1 to denote the absence of value, as GCC then fears that this -1 might be used to dereference arrays (such as yytname). Use 0, which corresponds to $accept, which is valueless (the needed property: the symbol destructor must not try to reclaim the memory associated with the symbol). 2013-01-29 Akim Demaille c++: use more explicit types than int * data/c++.m4 (b4_public_types_declare): Declare token_number_type soon. Introduce symbol_number_type (wider than token_number_type). Clarify the requirement that kind_type from by_state and by_type denote the _input_ type (required by the constructor), not the stored type. Use symbol_number_type and token_number_type where appropriate, instead of int. * data/lalr1.cc: Adjust to these changes. Propagate "symbol_number_type". Invoke "type_get ()" instead of read "type" directly. 2013-01-29 Akim Demaille c++: value_type -> kind_type * data/c++.m4, data/lalr1.cc (by_type, by_state): Rename 'value_type' as 'kind_type', as it is clearer. 2013-01-29 Akim Demaille c++: improve the signature of yysyntax_error_ * data/lalr1.cc: This function is const. It takes a symbol_number_type. 2013-01-29 Akim Demaille c++: style changes * data/lalr1.cc: Formatting changes. And name changes. 2013-01-28 Akim Demaille doxygen: upgrade Doxyfile, and complete it * doc/Doxyfile.in: Let doxygen upgrade it. (INCLUDE_PATH): Point to lib too. (PROJECT_BRIEF): New. (EXCLUDE): Update to reflect the current file hierarchy. 2013-01-28 Akim Demaille maint: fix syntax-check issues * cfg.mk: Ignore strcmp in local.at. * tests/conflicts.at: Use AT_PARSER_CHECK. * tests/regression.at: Preserve the exit status of the generated parsers. * tests/local.mk ($(TESTSUITE)): Map @tb@ to a tabulation. * tests/c++.at, tests/input.at, tests/regression.at: Use @tb@. * cfg.mk: (space-tab): There are no longer exceptions. 2013-01-28 Akim Demaille tests: please C90 compilers * tests/actions.at, tests/conflicts.at: Use /* ... */ comments. Let "main" return a value. 2013-01-28 Akim Demaille maint: update todo * TODO: Remove fixed items. 2013-01-28 Akim Demaille news: minor improvements * NEWS: Name some more contributors. Restructure slightly. 2013-01-28 Akim Demaille tests: please clang and use ".cc", not ".c", for C++ input When fed with foo.c, clang++ 3.2 answers: clang: error: treating 'c' input as 'c++' when in C++ mode, this behavior is deprecated * tests/output.at (AT_CHECK_OUTPUT_FILE_NAME): Use *.cc and *.hh for C++. 2013-01-28 Akim Demaille tests: formatting changes * tests/local.at: Restore proper indentation. 2013-01-28 Theophile Ranquet c++: better inline expansion Many 'inline' keywords were in the declarations. They rather belong in definitions, so move them. * data/c++.m4 (basic_symbol, by_type): Many inlines here. * data/lalr1.cc (yytranslate_, yy_destroy_, by_state, yypush_, yypop_): Inline these as well. (move): Move the definition outside the struct, where it belongs. 2013-01-28 Akim Demaille tests: check that using variants is exception safe * tests/local.at: (Slightly) improve the regexp by escaping '.' when it denotes a point. (AT_VARIANT_IF): New. * tests/c++.at (Exception Safety): Run it for variants too. 2013-01-28 Akim Demaille tests: remove useless %defines Many tests were using %defines because C++ skeletons used to require it. * tests/actions.at, tests/c++.at, tests/input.at, tests/regression.at: Remove useless %defines. 2013-01-28 Akim Demaille c++: remove now-useless operators Now that symbols behaves properly, we can eliminate special routines that are no longer needed. * data/c++.m4, data/glr.cc, data/lalr1.cc, data/variant.hh: Remove useless assignment operators and copy constructors. As a consequence, remove useless includes for "abort". 2013-01-28 Akim Demaille tests: enable support for --debug * tests/c++.at (Variants): Here. And remove useless clutter when api.token.constructor is enabled. 2013-01-28 Akim Demaille c++: revamp the support for variants The current approach was too adhoc: the symbols were not sufficiently self-contained, in particular wrt memory management. The "new" guideline is the one that should have been followed from the start: let the symbols handle themslves, instead of leaving their users to it. It was justified by the will to avoid gratuitious moves and copies, but the current approach does not seem to be slower, yet it will probably be simpler to adjust to support move semantics from C++11. The documentation says that the %parse-param are available from the %destructor. In retrospect, that was a silly design decision, which we can break for variants, as its a new feature. It should be phased out for non-variants too. * data/variant.hh: A variant never knows if it stores something or not, it is up to its users to store this information. Yet, in parse.assert mode, make sure the empty/filled variants are properly used. (b4_symbol_constructor_define_): Don't call directly the symbol constructor, to save a useless temporary. * data/stack.hh (push): Steal the pushed value instead of duplicating it. This will simplify the callers of push, who handled this "move" approach themselves. * data/c++.m4 (basic_symbol): Let -1, as kind, denote the fact that a symbol is empty. This is needed for instance when shifting the lookahead: yyla is given as argument to "push", and its value is then moved on the stack. But then yyla must be declared "empty" so that its destructor won't be called. (basic_symbol::move): New. Move the responsibility of calling the destructor from yy_destroy to ~basic_symbol in the case of variants. * data/lalr1.cc (stack_symbol_type): Now a derived class from its previous value, so that we can add a constructor from a symbol_type. (by_state): State -1 means empty. (yypush_): Factor, by calling one overload from the other one, and using the new semantics of stack::push. No longer reclaim by hand the memory from rhs symbols, since now that we store objects with proper destructors, they will be reclaimed automatically. Conversely, be sure to delete yylhs. * tests/c++.at (C++ Variant-based Symbols): New "unit" test for symbols. 2013-01-28 Akim Demaille c++: formatting and comment changes * data/c++.m4, data/lalr1.cc, data/stack.hh, data/variant.hh: Fix indentation. Fix some comments. 2013-01-27 Valentin Tolmer tests: add token declaration order test * tests/conflicts.at: New test. 2013-01-27 Akim Demaille regen 2013-01-27 Valentin Tolmer grammar: preserve token declaration order In a declaration %token A B, the token A is declared before B, but in %left A B (or with %precedence or %nonassoc or %right), the token B was declared before A (tokens were declared in reverse order). * src/symlist.h, src/symlist.c (symbol_list_append): New. * src/parse-gram.y: Use it instead of symbol_list_prepend. * tests/input.at: Adjust expectations. 2013-01-25 Akim Demaille tests: improve test group titles * tests/local.at (AT_SETUP_STRIP): AT_SETUP does not behave properly with new-lines in its argument. Remove them. Fix the handling of %define with quotes. 2013-01-25 Akim Demaille c: no longer require stdio.h when locations are enabled Recent changes (in 2.7) introduced a dependency on both FILE and fprintf, which are "available" only in %debug mode. This was to define yy_location_print_, which is used only in %debug mode by the parser, but massively used by the test suite to output the locations in yyerror. Break this dependency: the test suite should define its own routines to display the locations. Eventually Bison will provide the user with a means to display locations, but not yet. * data/c.m4 (b4_yy_location_print_define): Use YYFPRINTF instead of fprintf directly. * data/yacc.c (b4_yy_location_print_define): Invoke it only in %debug mode, so that stdio.h is included (needed for FILE*), and YYFPRINTF is defined. * tests/local.at (AT_YYERROR_DECLARE, AT_YYERROR_DEFINE): Declare and define location_print and LOCATION_PRINT. * tests/actions.at, tests/existing.at, tests/glr-regression.at, * tests/input.at, tests/named-refs.at, tests/regression.at: Adjust to use them. Fix the expected line numbers (as the prologue's length has changed). 2013-01-25 Akim Demaille c: minor simplification in the debug code * data/c.m4 (yy_symbol_print): Minor factoring. 2013-01-25 Akim Demaille c++: display locations as C does See commit 3804aa260b956dd012adde3894767254422a5fcf. * data/location.cc (operator<<): Display location exactly as is done in C skeletons. * tests/local.at (AT_LOC_PUSHDEF, AT_LOC_POPDEF): Also define AT_FIRST_LINE, AT_LAST_LINE, AT_FIRST_COLUMN, AT_LAST_COLUMN. * tests/actions.at (Location Print): Also check C++ skeletons. 2013-01-25 Akim Demaille tests: highlight empty right-hand sides * tests/actions.at, tests/c++.at, tests/headers.at, * tests/input.at: here. 2013-01-25 Akim Demaille news: prepare for 2.8 * NEWS: Restructure. Name contributors. 2013-01-21 Akim Demaille tests: generalize default main for api.namespace * tests/local.at (AT_NAME_PREFIX): Also match api.namespace. (AT_MAIN_DEFINE): Take it into account. * tests/c++.at, tests/headers.at: Use AT_NAME_PREFIX. (AT_CHECK_NAMESPACE): Rename as... (AT_TEST): this. 2013-01-21 Akim Demaille tests: improve factoring of the main function * tests/local.at (AT_MAIN_DEFINE): If %debug is used, check if -d/--debug is passed to the generated parser, and enable the traces. Return exactly the result of yyparse, so that we can check exit code 2 too. * tests/actions.at, tests/glr-regression.at, tests/regression.at: Use AT_MAIN_DEFINE, helping AT_BISON_OPTION_PUSHDEFS where needed, preferably to option -t. 2013-01-21 Akim Demaille tests: factor the definition of main With Théophile Ranquet. * tests/local.at (AT_MAIN_DEFINE): New. (AT_YYERROR_DEFINE): Improve formatting. * tests/actions.at, tests/c++.at, tests/conflicts.at, * tests/glr-regression.at, tests/input.at, tests/regression.at, * tests/skeletons.at, tests/torture.at: Adjust. * tests/c++.at: Add missing %skeleton for a PUSHDEFS, and a missing PUSH/POPDEFS for another test. 2013-01-21 Akim Demaille tests: minor refactoring * tests/named-refs.at: Use AT_FULL_COMPILE where applicable. 2013-01-21 Akim Demaille diagnostics: avoid useless caret stuttering * src/scan-code.l: When reporting a missing ending ';', don't display the guilty action twice. 2013-01-21 Theophile Ranquet examples: please clang * doc/bison.texi (calc++-scanner.ll): Don't output useless yyinput function. 2013-01-21 Theophile Ranquet tests: better silencing of unused argument warnings input.yy:35:44: error: unused parameter 'msg' [-Werror,-Wunused-parameter] void yy::parser::error (std::string const& msg) ^ * tests/c++.at (C++ GLR parser identifier shadowing): Don't name unused argument, use YYUSE instead of a direct cast to void. 2013-01-21 Theophile Ranquet bench: compatibility for Bison <= 2.7 There used to be a bug in some skeletons, which caused the expansion of 'yylval' and 'yylloc', generating these errors: input.cc:547:16: error: expected ',' or '...' before '(' token #define yylval (yystackp->yyval) ^ input.yy:29:39: note: in expansion of macro 'yylval' int yylex (yy::parser::semantic_type *yylval) ^ This bug is fixed by 'skel: better aliasing of identifiers', but a workaround is useful when benchmarking against older versions of Bison, which are still affected by the bug. * etc/bench.pl.in: Rename yylval to yylvalp and yylloc to yyllocp in base grammar 'list'. 2013-01-15 Theophile Ranquet c++: remove useless inlines * data/c++.m4 (basic_symbol): Keep 'inline' in the prototypes, but don't duplicate it in the implementation. * data/variant.hh (variant): 'inline' is not needed when the implementation is provided in the class definition. 2013-01-15 Theophile Ranquet c++: m4 stylistic change * data/c++.m4 (syntax_error): Fix the indentation of 'inline'. 2013-01-14 Theophile Ranquet c++: silence warnings * data/c++.m4 (basic_symbol::operator=): Unused parameter. * tests/c++.at (C++ GLR parser identifier shadowing): Here too. - 2013-01-14 Theophile Ranquet news: typos * NEWS: Fix a typo, use YYSTYPE rather than semantic_type. 2013-01-12 Akim Demaille regen 2013-01-12 Akim Demaille maint: update copyright years Suggested by Stefano Lattarini. Run "make update-copyright". 2013-01-12 Akim Demaille build: fix VPATH issue * Makefile.am (update-b4-copyright, update-package-copyright-year): Fix path to build-aux. 2013-01-11 Theophile Ranquet carets: document default activation * NEWS: Announce it. * doc/bison.texi: Adjust. 2013-01-11 Theophile Ranquet carets: show them in more tests * tests/input.at, tests/named-refs.at: Here. 2013-01-11 Theophile Ranquet carets: activate by default * src/getargs.c (feature_flag): Here. * tests/local.at (AT_BISON_CHECK_, AT_BISON_CHECK_NO_XML): Deactivate carets for the testsuite, by default. * tests/input.at: Adjust the locations for command line definitions. 2013-01-11 Theophile Ranquet variants: document move and swap * data/variant.hh (swap): Doc. (build): Rename as... (move): This, more coherent naming with clearer meaning. * data/c++.m4 (move): Adjust. 2013-01-11 Theophile Ranquet c++: privatize variant blind copies * data/variant.hh (variant, operator=): Make private. * data/c++.m4 (operator=): New, to avoid needing a definition of that operator for each class member (such as a possible variant). * data/glr.cc, data/lalr.cc: Add the necessary include for the abort. 2013-01-11 Akim Demaille glr.c: fix an unused argument issue * data/glr.c (yyuserAction): "Use" yyrhslen, as in variant mode, we might not use it. 2013-01-11 Akim Demaille glr.c: style changes * data/glr.c (yyuserAction): Use a size_t for sizes. 2013-01-11 Akim Demaille c.m4: style fix * data/c.m4 (b4_parse_param_use): Add missing space before paren. 2013-01-11 Theophile Ranquet skel: better aliasing of identifiers * data/glr.c, data/yacc.c: Avoid emitting useless defines. * data/glr.cc: Restore prefixes for epilogue. 2013-01-11 Theophile Ranquet glr.cc: fatal if using api.token.ctor without variants * data/glr.cc: Here. 2013-01-11 Theophile Ranquet skel: correctly indent switch cases * data/bison.m4 (b4_type_action_): Here. 2013-01-11 Theophile Ranquet variants: assert changes * data/variant.hh (swap): More asserts can't hurt. Don't perform useless swaps. (build): Deactivate problematic asserts, pending further investigation. (variant): Prohibit copy construction. 2013-01-11 Theophile Ranquet lalr1.cc: use a vector for the symbol stack * data/lalr1.cc: Adjust includes. * data/stack.hh (push, pop): Use push_back and pop_back. (operator []): Access vector from the end. 2013-01-11 Theophile Ranquet lalr1.cc: change symbols implementation A "symbol" groups together the symbol type (INT, PLUS, etc.), its possible semantic value, and its optional location. The type is needed to access the value, as it is stored as a variant/union. There are two kinds of symbols. "symbol_type" are "external symbols": they have type, value and location, and are returned by yylex. "stack_symbol_type" are "internal symbols", they group state number, value and location, and are stored in the parser stack. The type of the symbol is computed from the state number. The class template symbol_base_type factors the code common to stack_symbol_type and symbol_type. It uses the Curiously Recurring Template pattern so that we can always (static_) downcast to the exact type. symbol_base_type features value and location, and delegates the handling of the type to its parameter. When trying to generalize the support for variant, a significant issue was revealed: because stack_symbol_type and symbol_type _derive_ from symbol_base_type, the type/state member is defined _after_ the value and location. In C++ the order of the definition of the members defines the order in which they are initialized, things go backward: the value is initialized _before_ the type. This is wrong, since the type is needed to access the value. Therefore, we need another means to factor the common code, one that ensures the order of the members. The idea is simple: define two (base) classes that code the symbol type ("by_type" codes it by its type, and "by_state" by the state number). Define basic_symbol as the class template that provides value and location support. Make it _derive_ from its parameter, by_type or by_state. Then define stack_symbol_type and symbol_type as basic_symbol, basic_symbol. The name basic_symbol was chosen by similarity with basic_string and basic_ostream. * data/c++.m4 (symbol_base_type): Remove, replace by... (basic_symbol): which derives from its parameter, one of... (by_state, by_type): which provide means to retrieve the actual type of symbol. (symbol_type): Is now basic_symbol. (stack_symbol_type): Is now basic_symbol. * data/lalr1.cc: Many adjustments. 2013-01-11 Theophile Ranquet bench: add %b directive to use a specific Bison For example, $ bench.pl -v '%s lalr1.cc & %d variant & ( %b ~/old-bison/bin/bison | %b ~/new-bison/bin/bison )' -g list -i 10000 * etc/bench.pl.in: Here. 2013-01-09 Akim Demaille regen 2013-01-04 Akim Demaille gnulib: update 2012-12-31 Akim Demaille doc: use deffn to declare the list of %define variables * doc/bison.texi (%define Summary): Use @deffn instead of @table, it spares a lot of width, especially in PDF, and looks nicer in the other formats too. It is also more consistent with the rest of the document. 2012-12-31 Akim Demaille doc: minor completion and fixes * doc/bison.texi (%define Summary): Provide more history to some variables. 2012-12-31 Akim Demaille java: stype is obsoleted by api.value.type This is consistent with the other %define variable names. * data/java.m4: Use api.value.type instead of stype. * doc/bison.texi, NEWS: Document that change. * src/muscle-tab.c (muscle_percent_variable_update): Provide backward compatibility. * tests/java.at: Adjust. 2012-12-31 Akim Demaille doc: fix html build * doc/local.mk (bison.html): Fix dependencies. 2012-12-31 Akim Demaille todo: remove erroneous task * tests/input.at: Check that there are no warnings about stray $ and @ in the epilogue. * TODO: Remove the correponding task. 2012-12-31 Akim Demaille gnulib: update 2012-12-28 Akim Demaille regen 2012-12-28 Akim Demaille syncline: one line is enough So far we were issuing two lines for each syncline change: /* Line 356 of yacc.c */ #line 1 "src/parse-gram.y" This is a lot of clutter, especially when reading diffs, as these lines change often. Fuse them into a single, shorter, line: #line 1 "src/parse-gram.y" /* yacc.c:356 */ * data/bison.m4 (b4_syncline): Issue a single line. Comment improvements. (b4_sync_start, b4_sync_end): Issue a shorter comment. * data/c++.m4 (b4_semantic_type_declare): b4_user_code must be on its own line as it might start with a "#line" directive. 2012-12-28 Akim Demaille regen 2012-12-28 Akim Demaille maint: restore ANSI 89 compliance * data/bison.m4, src/conflicts.c, src/files.c, src/output.c, * src/symtab.c: Use /* ... */ comments only. Declare variables before statements. 2012-12-28 Akim Demaille graph: minor simplification * src/gram.c (print_lhs): Use %*s to indent. * src/print_graph.c (print_lhs): Use obstack_printf. Became simple enough to be inlined in... (print_core): here. Use a "rule*" instead of an index in "rules[]". 2012-12-28 Akim Demaille closure, gram: add missing const * src/closure.h, src/closure.c, src/gram.h, src/gram.c: Add some missing const where appropriate. 2012-12-27 Theophile Ranquet carets: properly display when no line feed is present * src/location.c (location_caret): finish the line with one whether or not it is present in input. Rewrite code without getline. (cleanup_caret): Reset the caret_info global. * bootstrap.conf: No longer require getline. 2012-12-27 Theophile Ranquet scanner: reintroduce unput for missing end tokens Unput was no longer used since a POSIX-compatiblity issue with Flex 2.5.31, which has been adressed in newer versions of Flex. See this discussion: This partially reverts commit aa4180418fff518198e1b0f2c43fec6432210dc7. * src/scan-gram.l (unexpected_end): Here. * tests/input.at: Adjust for new order of error reports. 2012-12-27 Akim Demaille tables: scope reduction * src/tables.c (default_goto): Make it easier to understand. 2012-12-27 Akim Demaille regen 2012-12-27 Akim Demaille skeletons: fix comments The commit 38de4e570fdc7c8db9633c3b2054e565d8c1c6b9 underquoted the content of the comments, which resulted in losing square brackets in the comments. Besides, some other invocations were underquoting the effective arguments. * data/c.m4 (b4_comment_): Properly quote the comment. (b4_comment_, b4_comment): Move to... * data/c-like.m4: here, so that... * data/java.m4: can use it instead of its own copy. * data/bison.m4 (b4_integral_parser_tables_map): Fix some comments. * data/lalr1.cc, data/lalr1.java, data/yacc.c: Comment fixes. * data/lalr1.cc: Reorder a bit to factor some CPP directives. 2012-12-27 Akim Demaille maint: which -> whose Apparently, I was confusing both. * data/bison.m4, data/c.m4, data/glr.c, data/lalr1.cc, data/yacc.c: Use "whose" where appropriate. 2012-12-26 Akim Demaille tables: scope reduction * src/tables.c (matching_state): here. 2012-12-26 Akim Demaille tables: scope reduction * src/tables.c (token_actions): here. 2012-12-26 Akim Demaille tables: scope reduction * src/tables.c (save_row): here. 2012-12-26 Akim Demaille tables: scope reduction * src/tables.c (save_column, pack_vector): Reduce the scope to emphasize the structure of the code. Rename the returned value "res" to make understanding easier. 2012-12-26 Akim Demaille tables: use size_t where appropriate These changes aim at making the code easier to understand. * src/tables.c (tally): This is a size, always >= 0, so make it a size_t. 2012-12-26 Akim Demaille tables: style changes * src/tables.c: Prefer < to >. Fix/complete some comments. Remove useless parens. 2012-12-26 Akim Demaille skeletons: no longer call yylex via a CPP macro The YYLEX existed only to support YYLEX_PARAM, which is now removed. This macro was a nuisance, since incorrect yylex calls where pointed the macro _use_, instead of its definition. * data/c.m4 (b4_lex_formals, b4_lex): New. * data/glr.c, data/yacc.c: Use it. * data/lalr1.cc (b4_lex): New. Use it. squash! skeletons: no longer call yylex via a CPP macro 2012-12-26 Akim Demaille YYLEX_PARAM: drop support * data/yacc.c, doc/bison.texi: Remove YYLEX_PARAM support. * NEWS: Document it. 2012-12-26 Akim Demaille examples: minor improvements * examples/variant.yy: Don't use debug_stream(), obsoleted. Use <*>. 2012-12-26 Akim Demaille skeletons: factor comments about symbols * data/variant.hh (b4_char_sizeof_): Rename as... * data/bison.m4 (b4_symbol_tag_comment): this. Provide more documentation about b4_symbol_*. 2012-12-26 Akim Demaille c: improve the definition of public types * data/c.m4 (b4_token_enum): Improve comments. (b4_value_type_define, b4_location_type_define): New, extracted from... (b4_declare_yylstype): here. Separate the typedefs from the union/struct definitions. 2012-12-26 Akim Demaille doc: update variant usage * doc/bison.texi, examples/variant.yy: Use "%define api.value.type variant", instead of "%define variant". 2012-12-26 Akim Demaille tests: check the "%define variant" is deprecated. * tests/input.at: Rename some AT_SETUP to avoid that AT_SETUP_STRIP thinks they contain %define directives. ("%define" backward compatibility): Merge tests together to speed up the test suite, and to make maintenance easier (multiple AT_CHECK means multiple runs of the test suite to be sure to have updated all the error messages). Check the "%define variant" is properly obsoleted. 2012-12-23 Akim Demaille %define variables: support value changes in deprecation * src/muscle-tab.c (define_directive): Be robust to "assignment" containing '='. (muscle_percent_variable_update): Upgrade "variant" to "api.value.type". Support such upgrade patterns. Adjust callers. * data/bison.m4: Use api.value.type for variants. * tests/c++.at: Adjust tests. 2012-12-23 Akim Demaille diagnostics: treat obsolete %define variable names as obsolete directives Instead of warning: deprecated %define variable name: 'namespace', use 'api.namespace' [-Wdeprecated] display (in -fno-caret mode): warning: deprecated directive: '%define namespace foo', use '%define api.namespace foo' [-Wdeprecated] and (in -fcaret mode): warning: deprecated directive, use '%define api.namespace toto' [-Wdeprecated] %define namespace toto ^^^^^^^^^ This is in preparation of cases where not only the variable is renamed, but the values are too: warning: deprecated directive: '%define variant', use '%define api.value.type variant' [-Wdeprecated] * src/muscle-tab.c (define_directive): New. (muscle_percent_variable_update): Take the value as argument, and use it in the diagnostics. Loop with a pointer instead of an index. * tests/input.at (%define backward compatibility): Adjust. 2012-12-23 Akim Demaille diagnostics: factor the deprecated directive message * src/complain.h, src/complain.c (deprecated_directive): New. * src/muscle-tab.c: Use it. 2012-12-23 Akim Demaille variant: produce stable results Improve the output by ensuring a well defined order for type switches. * src/uniqstr.h: Style changes for macro arguments. (UNIQSTR_CMP): Replace by... (uniqstr_cmp): this. * src/uniqstr.c (uniqstr_cmp): New. Produce well defined results. * src/output.c: Use it. 2012-12-23 Akim Demaille uniqstr: formatting changes * src/uniqstr.h: Sort functions by object type. 2012-12-23 Akim Demaille skeletons: fix an error message * data/bison.m4 (b4_flag_if): Display the invalid value. 2012-12-23 Akim Demaille tests: improve titles * tests/local.at (AT_SETUP_STRIP): New. (AT_SETUP): Use it to shorten the test titles: remove %defines, %language and %skeleton whose arguments suffice. * tests/synclines.at: Use more precise AT_SETUP. 2012-12-23 Akim Demaille c++: comment changes * data/c++.m4, data/glr.cc, data/lalr1.cc: Convert some /* ... */ comments to //. 2012-12-23 Akim Demaille c++: use // comments in the output This is mostly used for the license header, the synclines, and the generated tables: - /* STOS_[STATE-NUM] -- The (internal number of the) accessing - symbol of state STATE-NUM. */ + // STOS_[STATE-NUM] -- The (internal number of the) accessing + // symbol of state STATE-NUM. static const unsigned char yystos_[]; * data/c.m4: Comment changes. (b4_comment_): Expand the text argument. Before this change, we were actually formatting M4 code as a C comment, and then expand it. (b4_comment): Fix the closing of comments: there is no reason to add the (line) prefix before the closing "*/". * data/c++.m4 (b4_comment): New. 2012-12-21 Akim Demaille maint: disable sc_prohibit_test_backticks * cfg.mk: here. And fix typos. Reported by Stefano Lattarini. 2012-12-21 Akim Demaille maint: more syntax-checks * cfg.mk (sc_prohibit_tab_based_indentation, sc_prohibit_test_backticks) (sc_preprocessor_indentation, sc_space_before_open_paren): New, stolen from Coreutils (2e9f5ca4ebbbdb6a9fa2dd3d5add3f7720a172d7). 2012-12-21 Akim Demaille debug: no longer generate tabs * src/closure.c, src/derives.c, src/nullable.c, tests/sets.at: Use spaces. 2012-12-21 Akim Demaille style changes: run cppi Run it in src/ for a start. * src/AnnotationList.h, src/InadequacyList.h, src/Sbitset.h, * src/closure.c, src/complain.h, src/flex-scanner.h, src/getargs.h, * src/gram.h, src/graphviz.h, src/ielr.h, src/location.h, * src/muscle-tab.h, src/named-ref.h, src/relation.h, src/scan-code.h, * src/state.h, src/symtab.h, src/system.h, src/uniqstr.h: Reindent preprocessor directives. 2012-12-21 Akim Demaille style changes: untabify * data/xslt/xml2dot.xsl, data/xslt/xml2text.xsl, m4/flex.m4, * tests/glr-regression.at, tests/torture.at: here. 2012-12-21 Akim Demaille tests: be robust to set -e. * examples/test (run): here. 2012-12-19 Akim Demaille variants: prohibit simple copies The "variant" structure provides a means to store, in a typeless way, C++ objects. Manipulating it without provide the type of the stored content is doomed to failure. So provide a means to copy in a type safe way, and prohibit typeless assignments. * data/c++.m4 (symbol_type::move): New. * data/lalr1.cc: Use it. * data/variant.hh (b4_variant_define): Provide variant::copy. Let variant::operator= abort. We cannot undefine it, yet, as it is still uses by the implicit assigment in symbols, which must also be disabled. 2012-12-19 Akim Demaille variant: more assertions Equip variants with more checking code. Provide a means to request includes. * data/variant.hh (b4_variant_includes): New. * data/lalr1.cc: Use it. * data/variant.hh (variant::built): Define at the end, as a private member. (variant::tname): New. Somewhat makes "built" useless, but let's keep both for a start, in case using "typeinfo" is considered unacceptable in some environments. Fix some formatting issues. 2012-12-19 Akim Demaille gnulib: update 2012-12-19 Akim Demaille skeletons: fix output directives * data/lalr1.cc, data/location.cc, data/glr.cc: Use b4_output_begin. Broken during a merge. 2012-12-19 Akim Demaille yacc.c: style changes * data/yacc.c (b4_lex_param): Provide arguments with a name. 2012-12-19 Akim Demaille glr.cc: simplifying the handling of parse/lex params The fact that glr.cc uses glr.c makes the handling of parse params more complex, as the parser object of glr.cc must be passed to the parse function of glr.c. Yet not all the functions need access to the parser object. * data/glr.cc (b4_parse_param_wrap): New. Use them. 2012-12-19 Akim Demaille glr: rename lex params * data/glr.c (b4_lex_param): Rename as... (b4_lex_formals): this, for consistency. Provide arguments a name. (LEX): Adjust. 2012-12-19 Akim Demaille glr.c: move function declaration earlier * data/glr.c (yypstack, yypdumpstack): Declare earlier, to make it easier to call them from other functions. 2012-12-19 Akim Demaille %define variables: backward compatibility * src/muscle-tab.c (muscle_percent_variable_update): Accept lex_symbol. Reported by Roland Levillain. 2012-12-16 Akim Demaille diagnostics: improve -fcaret for list of accepted values Instead of input.y:1.9-21: error: invalid value for %define variable 'api.push-pull': 'neither' %define api.push_pull "neither" ^^^^^^^^^^^^^ input.y:1.9-21: accepted value: 'pull' %define api.push_pull "neither" ^^^^^^^^^^^^^ input.y:1.9-21: accepted value: 'push' %define api.push_pull "neither" ^^^^^^^^^^^^^ input.y:1.9-21: accepted value: 'both' %define api.push_pull "neither" ^^^^^^^^^^^^^ report input.y:1.9-21: error: invalid value for %define variable 'api.push-pull': 'neither' %define api.push_pull "neither" ^^^^^^^^^^^^^ input.y:1.9-21: accepted value: 'pull' input.y:1.9-21: accepted value: 'push' input.y:1.9-21: accepted value: 'both' * src/complain.h (no_caret): New. * src/complain.c (error_message): Use it. * src/muscle-tab.c (muscle_percent_define_check_values): Use it. * src/scan-skel.l (flag): Ditto. * tests/input.at: Adjust and check. 2012-12-16 Akim Demaille skeletons: simplify the handling of default api.location.type * data/bison.m4 (b4_bison_locations_if): New. * data/glr.cc, data/lalr1.cc: Use it. 2012-12-16 Akim Demaille tests: address syntax-check failures * cfg.mk: Ignore failures in timevar (uses GCC style configuration, not gnulib's). * doc/local.mk: Space changes. * lib/main.c, tests/calc.at: Remove useless HAVE_ tests. 2012-12-15 Akim Demaille remove duplicate definitions * src/system.h: here, inherited from a merge. 2012-12-15 Akim Demaille tests: style changes * tests/glr-regression.at: Issue yyerror before yylex. 2012-12-15 Akim Demaille doc: fix dependencies * doc/local.mk: here. 2012-12-14 Akim Demaille doc: style fixes * doc/bison.texi: Add a couple of missing @var and @code. 2012-12-14 Theophile Ranquet doc: fix build dependencies Suggested by Nick Bowler * doc/local.mk: Avoid overwriting Automake's rules. 2012-12-14 Akim Demaille Merge branch 'origin/maint' * origin/maint: maint: credit Wojciech Polak maint: post-release administrivia version 2.7 yacc.c: scope reduction tests: C90 compliance fix C90 compliance glr.c: scope reduction gnulib: update 2012-12-14 Theophile Ranquet symtab: add missing initializations * src/symtab.c (semantic_type_new): Here. 2012-12-14 Theophile Ranquet symtab: fix some leaks * src/symlist.c (symbol_list_free): Deep free it. * src/symtab.c (symbols_free, semantic_types_sorted): Free it too. (symbols_do, sorted): Call by address. 2012-12-14 Theophile Ranquet tests: remove use of PARSE_PARAM * tests/header.at: Here. 2012-12-13 Akim Demaille maint: credit Wojciech Polak * NEWS, THANKS: He is the author of XML support (including XSLTs). 2012-12-12 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2012-12-12 Akim Demaille version 2.7 * NEWS: Record release date. 2012-12-12 Akim Demaille yacc.c: scope reduction * data/yacc.c (yysyntax_error): here. 2012-12-12 Akim Demaille tests: C90 compliance * tests/synclines.at: here. 2012-12-12 Akim Demaille fix C90 compliance * data/glr.c, src/graphviz.h, src/ielr.c, src/scan-gram.l, * src/system.h, tests/actions.at, tests/glr-regression.at: Do not use // comments. Do not introduce variables after statements. Provide "main" with a return value. 2012-12-12 Akim Demaille glr.c: scope reduction * data/glr.c (yyreportSyntaxError): Reduce the scope of yysize1 (now yysz). 2012-12-12 Akim Demaille gnulib: update 2012-12-10 Theophile Ranquet Merge remote-tracking branch 'origin/maint' * origin/maint: news: prepare for forthcoming release doc: explain how mid-rule actions are translated error: use better locations for unused midrule values doc: various minor improvements and fixes tests: ignore more useless compiler warnings tests: be robust to C being compiled with a C++11 compiler build: beware of Clang++ not supporting POSIXLY_CORRECT maint: post-release administrivia version 2.6.90 build: fix syntax-check error. cpp: simplify the Flex version checking macro news: improve the carets example and fix a typo cpp: improve the Flex version checking macro carets: improve the code maint: update news build: keep -Wmissing-declarations and -Wmissing-prototypes for modern GCCs build: drop -Wcast-qual gnulib: update 2012-12-09 Akim Demaille news: prepare for forthcoming release * NEWS: Fill paragraph. Reorder. Update examples. Remove line for 2.6.90. 2012-12-09 Akim Demaille doc: explain how mid-rule actions are translated * doc/bison.texi (Actions in Mid-Rule): Mention and use named references. Split into three subsections, among which... (Mid-Rule Action Translation): this new section. 2012-12-09 Akim Demaille error: use better locations for unused midrule values On %% exp: {;} {$$;} { $$ = $1; } instead of reporting (with -fcaret -Wmidrule-value) midrule.y:2.6-8: warning: unset value: $$ [-Wmidrule-values] exp: {;} {$$;} { $$ = $1; } ^^^ midrule.y:2.6-27: warning: unused value: $2 [-Wmidrule-values] exp: {;} {$$;} { $$ = $1; } ^^^^^^^^^^^^^^^^^^^^^^ report midrule.y:2.6-8: warning: unset value: $$ exp: {;} {$$;} { $$ = $1; } ^^^ midrule.y:2.10-14: warning: unused value: $2 exp: {;} {$$;} { $$ = $1; } ^^^^^ * src/reader.c (grammar_rule_check): When warning about the value of a midrule action, use the location of the midrule action instead of the location of the rule. the location of the part of the rule. * tests/actions.at (Default %printer and %destructor for mid-rule values): Adjust expectations * tests/input.at (Unused values with default %destructor): Ditto. (AT_CHECK_UNUSED_VALUES): Ditto. And use -fcaret. 2012-12-09 Akim Demaille doc: various minor improvements and fixes * doc/figs/example.dot, doc/figs/example.y: New. * doc/bison.texi: Prefer "token" to TOKEN. Use @group where appropriate. Adjust with style changes in the output (State 0, not state 0). Fix some @ref that were missing the third argument. Fix some incorrect line numbers. Use "nonterminal", not "non-terminal". Fix overfull and underfull TeX hboxes. Put the comments in the index. Remove duplicate index entries. Fuse glossary entries where appropriate. (Understanding): Improve the continuity between sections. Use example.dot to show the whole graph. * doc/Makefile.am: Adjust. 2012-12-09 Akim Demaille tests: ignore more useless compiler warnings * tests/synclines.at (AT_SYNCLINES_COMPILE): Ignore complains about using c++ to compile C. 2012-12-09 Akim Demaille tests: be robust to C being compiled with a C++11 compiler * tests/glr-regression.at: Use YY_NULL instead of NULL. Comment changes. 2012-12-09 Akim Demaille build: beware of Clang++ not supporting POSIXLY_CORRECT * m4/c-working.m4 (BISON_LANG_COMPILER_POSIXLY_CORRECT): New. (BISON_C_COMPILER_POSIXLY_CORRECT): Use it. For consistency with C++, also define BISON_C_WORKS. * m4/cxx.m4 (BISON_CXX_COMPILER_POSIXLY_CORRECT): New. * configure.ac: Use it. * tests/atlocal.in: Get its result. Propagate properly CXX values when used to compile C. When POSIXLY_CORRECT, adjust BISON_C_WORKS and BISON_CXX_WORKS. * tests/local.at (AT_COMPILE): Use BISON_C_WORKS. 2012-12-07 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2012-12-07 Akim Demaille version 2.6.90 * NEWS: Record release date. 2012-12-07 Akim Demaille build: fix syntax-check error. * cfg.mk: Exclude names-refs, it includes a "double" if (end of first line, first of second line below). test.y:43.12-44.59: symbol not found in production: if if-stmt-a: IF expr[cond] THEN stmt.list[then] ELSE stmt.list[else] FI ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 2012-12-07 Theophile Ranquet cpp: simplify the Flex version checking macro * src/flex-scanner,h (FLEX_VERSION): Consider YY_FLEX_SUBMINOR_VERSION defined. 2012-12-07 Theophile Ranquet news: improve the carets example and fix a typo * NEWS: Here. 2012-12-07 Theophile Ranquet cpp: improve the Flex version checking macro * src/flex-scanner.h (FLEX_VERSION): Here. 2012-12-07 Theophile Ranquet carets: improve the code * src/location.c: Remove duplicate documentations. (caret_info): Stylistic change. (location_caret): Many reworks. 2012-12-07 Akim Demaille maint: update news * NEWS: There is no 2.6.6, remove its stub. 2012-12-07 Akim Demaille build: keep -Wmissing-declarations and -Wmissing-prototypes for modern GCCs Fixes a -Werror failure of xalloc.h used in src. From Eric Blake. http://lists.gnu.org/archive/html/bug-gnulib/2012-12/msg00006.html * configure.ac: Check whether GCC pragma diagnostic push/pop works. Enable these warnings for bison if it does. Enable these warnings for the test suite anyway. 2012-12-07 Akim Demaille build: drop -Wcast-qual Suggested by Jim Meyering. http://lists.gnu.org/archive/html/bug-gnulib/2012-12/msg00017.html * configure.ac (warn_common): Remove -Wcast-qual. 2012-12-07 Akim Demaille gnulib: update 2012-12-06 Theophile Ranquet Merge remote-tracking branch 'origin/maint' * origin/maint: misc: pacify the Tiny C Compiler cpp: make the check of Flex version portable misc: require getline c++: support wide strings for file names doc: document carets tests: enhance existing tests with carets errors: show carets getargs: add support for --flags/-f 2012-12-06 Theophile Ranquet misc: pacify the Tiny C Compiler * src/graphviz.c (conclude_red): Remove a useless return. 2012-12-05 Theophile Ranquet cpp: make the check of Flex version portable This was problematic with tcc 0.9.25 * src/flex-scanner.h (FLEX_VERSION_GT): Rewrite and rename as... (FLEX_VERSION): This. 2012-12-05 Theophile Ranquet misc: require getline * bootstrap.conf: Here, used by src/location.c. * src/getargs.c (long_options): Rename --flags to --feature. 2012-12-05 Akim Demaille c++: support wide strings for file names Reported by Mark Boyall. http://lists.gnu.org/archive/html/help-bison/2011-08/msg00002.html * data/location.cc (operator<<): Be templated on the type of output stream. * tests/headers.at (Several parsers): Adjust. 2012-12-05 Theophile Ranquet doc: document carets * NEWS: Announce it. * doc/bison.texi (Bison Options): Here. 2012-12-05 Theophile Ranquet tests: enhance existing tests with carets * tests/actions.at: Unset value. * tests/conflicts.at: Rule useless due to conflicts. * tests/input.at: Missing terminator, unexpected end of file, command line redefinition of variable. * tests/named-refs.at: Many errors. * tests/reduce.at: Useless nonterminals and rules. * tests/regression.at: Large token. 2012-12-05 Theophile Ranquet errors: show carets * src/locations.c (caret_info): New, persistant information useful for... (location_caret): New, print a caret. (cleanup_caret): Release caret_info cleanly, call it... * src/main.c (main): Here. * src/complain.c (error_message): Call location_caret here. 2012-12-05 Theophile Ranquet getargs: add support for --feature/-f Introduce -fdiagnostics-show-caret * src/getargs.c (feature_flag): New global. * src/getargs.h (feature): New enum. 2012-12-05 Akim Demaille Merge remote-tracking branch 'origin/maint' * origin/maint: getargs: don't label --language/-l as experimental getargs: fix the locations of command-line input errors: indent missing action code semicolon warning 2012-12-04 Theophile Ranquet getargs: don't label --language/-l as experimental * NEWS: Announce it. * doc/bison.texi, src/getargs.c (usage): Here. 2012-12-03 Akim Demaille tests: minor improvements * tests/calc.at (AT_CHECK_SPACES): To speed up, accept several files at once, and factor some calls. 2012-12-03 Theophile Ranquet getargs: fix the locations of command-line input * src/getargs.c (command_line_location): Here. * tests/input.at: Adjust. 2012-12-03 Theophile Ranquet errors: indent missing action code semicolon warning Also, remove a duplicate #define. * src/scan-code.l (SC_RULE_ACTION): Here. * tests/actions.at: Adjust. 2012-12-03 Akim Demaille Merge remote-tracking branch 'origin/maint' * origin/maint: parser: accept #line NUM m4: use a safer pattern to enable/disable output tests: beware of gnulib's need for config.h gnulib: update yacc.c, glr.c: check and fix the display of locations formatting changes glr.c: remove stray macro 2012-12-03 Akim Demaille parser: accept #line NUM * src/scan-gram.l (scanner): Accept '#line NUM'. (handle_syncline): Adjust to the possible missing file name. 2012-12-03 Akim Demaille m4: use a safer pattern to enable/disable output Work on some other areas of Bison revealed that some macros expanded to be expanded only once were actually expanded several times. This was due to the fact that changecom was not properly restored each time, and macro names appearing in comments were then expanded. Introduce begin/end macros which are easier to match that changecom()/changecom(#). * data/bison.m4 (b4_output_begin, b4_output_end): New. * data/glr.c, data/glr.cc, data/lalr1.cc, data/lalr1.java, * data/location.cc, data/stack.hh, data/yacc.c: Use them. 2012-12-03 Akim Demaille tests: beware of gnulib's need for config.h * tests/skeletons.at, tests/torture.at: Be sure to include config.h where appropriate. 2012-11-30 Akim Demaille gnulib: update * lib/yyerror.c: Include config.h since the following stdio.h might be from gnulib. 2012-11-30 Akim Demaille yacc.c, glr.c: check and fix the display of locations In some case, negative column number could be displayed. Make YY_LOCATION_PRINT similar to bison's own implementation of locations. Since the macro is getting fat, make it a static function. Reported by Jonathan Fabrizio. * data/c.m4 (yy_location_print_define): Improve the implementation, and generate the yy_location_print_ function. Adjust YY_LOCATION_PRINT. * tests/actions.at (Location Print): New tests. 2012-11-30 Akim Demaille formatting changes * data/c.m4: Fix comments, put macros in a more natural order. Space changes (from M-x whitespace-cleanup). * src/location.c: Fix spaces. * tests/actions.at: Space changes. 2012-11-30 Akim Demaille glr.c: remove stray macro * data/glr.c (YYOPTIONAL_LOC): Remove, unused since commit 769a8ef9bcb5e14d0be9d0869f5dca20ab093930. 2012-11-29 Akim Demaille Merge remote-tracking branch 'origin/maint' * origin/maint: doc: minor fixes doc: improve the index doc: introduce api.pure full, rearrange some examples yacc.c: support "%define api.pure full" local.at: improvements 2012-11-29 Akim Demaille doc: minor fixes * doc/bison.texi: Use stderr for error messages. Meta-variables are usually spelled in lower case. Use @code for function names. 2012-11-29 Akim Demaille doc: improve the index * doc/bison.texi: Fix uses of "deffn" so that the arguments of the directives do not show in the index. Remove a duplicate entry for api.pure. 2012-11-29 Theophile Ranquet doc: introduce api.pure full, rearrange some examples * NEWS: Add entry. * doc/bison.texi (%define Summary): Show the old Yacc behaviour. (Parser Function): Move parse-param examples here. (Pure Calling): Remove parse-param examples. (Error Reporting): Don't show the old behavior, stick to 'full'. 2012-11-29 Theophile Ranquet yacc.c: support "%define api.pure full" This makes the interface for yyerror() pure without the need for a spurious parse_param. * data/yacc.c (b4_pure_if, b4_pure_flag): New definition, accept three states. (b4_yacc_pure_if): Rename as... (b4_yyerror_arg_loc_if): This, and use b4_pure_flag. * tests/actions.at (%define api.pure): Modernize. * test/calc.at (Simple LALR Calculator): Modernize. * tests/local.at (AT_YYERROR_ARG_LOC_IF): Adjust. 2012-11-28 Akim Demaille tests: check variants without locations * tests/c++.at (Variants): Support non-use of locations, and check its support. 2012-11-26 Theophile Ranquet local.at: improvements * tests/local.at (AT_YYERROR_FORMALS): Make llocp const. (AT_PURE_AND_LOC_IF, AT_GLR_OR_PARAM_IF): Remove, expand... (AT_YYERROR_ARG_LOC_IF): Here, and use m4_join for readability. 2012-11-26 Akim Demaille tests: use -fno-strict-aliasing with variants Reported by Théophile Ranquet. * configure.ac (NO_STRICT_ALIAS_CXXFLAGS): New. * tests/c++.at, tests/atlocal.in, examples/local.mk: Use it. 2012-11-26 Akim Demaille tests: remove leftover * tests/atlocal.in: Remove duplicate handling of --compile-c-with-cxx. 2012-11-26 Akim Demaille doc: use %precedence instead of nonassoc when associativity is not wanted * doc/bison.texi: here. Formatting changes in some grammars. Fix a %prec into %precedence. 2012-11-26 Akim Demaille Merge remote-tracking branch 'origin/maint' * origin/maint: yacc.c: always initialize yylloc scanner: issue a single error for groups of invalid characters tests: formatting changes doc: one of the fixes for an ambiguous grammar was ambiguous too doc: fix the dangling else with precedence directives doc: prefer "token" to TOKEN doc: formatting changes scanner: use explicit "ignore" statements 2012-11-26 Akim Demaille Merge remote-tracking branch 'origin/branch-2.6' into maint * origin/branch-2.6: yacc.c: always initialize yylloc doc: one of the fixes for an ambiguous grammar was ambiguous too doc: fix the dangling else with precedence directives doc: prefer "token" to TOKEN doc: formatting changes 2012-11-23 Theophile Ranquet yacc.c: always initialize yylloc The initial location might be used if the parser starts by an empty reduction, so really ensure proper initialization of the initial location. The previous approach fails for PostgreSQL, which uses Reported by Peter Eisentraut. http://lists.gnu.org/archive/html/bug-bison/2012-11/msg00023.html With help from Théophile Ranquet. * data/yacc.c (b4_declare_scanner_communication_variables): Be sure to initialize yylloc, even when its structure is unknown. (yyparse): Simplify the call to b4_dollar_pushdef. * tests/actions.at (Initial location): Check of similar pattern as in the case of PostgreSQL. 2012-11-23 Akim Demaille scanner: issue a single error for groups of invalid characters * src/scan-gram.l: Scan groups of invalid characters together. * tests/input.at, tests/named-refs.at: Adjust. 2012-11-23 Akim Demaille tests: formatting changes * tests/named-refs.at: Here. 2012-11-23 Akim Demaille doc: one of the fixes for an ambiguous grammar was ambiguous too Reported by Аскар Сафин. http://lists.gnu.org/archive/html/bug-bison/2012-11/msg00024.html * doc/bison.texi (Reduce/Reduce): Fix the resulting ambiguity using precedence/associativity directives. 2012-11-22 Akim Demaille doc: fix the dangling else with precedence directives * doc/bison.texi (Non Operators): New node. (Shift/Reduce): Point to it. Don't promote "%expect n" too much. 2012-11-22 Akim Demaille doc: prefer "token" to TOKEN This is more readable in short examples. * doc/bison.texi (Shift/Reduce): here. Make "win" and "lose" action more alike. 2012-11-22 Akim Demaille doc: formatting changes * doc/bison.texi: Use @group. 2012-11-14 Akim Demaille scanner: use explicit "ignore" statements * src/scan-gram.l: here. 2012-11-13 Akim Demaille Merge remote-tracking branch 'origin/maint' * origin/maint: tests: close files in glr-regression xml: match DOT output and xml2dot.xsl processing xml: factor xslt space template graph: fix a memory leak xml: documentation output: capitalize State 2012-11-12 Theophile Ranquet tests: close files in glr-regression * tests/glr-regression.at: Here. 2012-11-12 Theophile Ranquet xml: match DOT output and xml2dot.xsl processing Make the DOT produced by XSLT processing equivalent to the one made with the --graph option. * data/xslt/xml2dot.xsl: Stylistic changes, and add support for reductions. * doc/bison.texi (Xml): Update. * src/graphviz.c (conclude_red): Minor stylistic changes to DOT internals. (output_red): Swap enabled and disabled reductions output, for coherence with XSLT output. * src/print_graph.c (print_core): Minor stylistic change to States' output. (print_actions): Swap order of output for reductions and transitions. * tests/local.at (AT_BISON_CHECK_XML): Ignore differences in order. * tests/output.at: Adjust to changes in DOT internals. 2012-11-12 Theophile Ranquet xml: factor xslt space template * data/xslt/bison.xsl (space): New, import from... * data/xslt/xml2text.xsl: Here. 2012-11-12 Theophile Ranquet graph: fix a memory leak * src/graphviz.c (output_red): Here. 2012-11-12 Theophile Ranquet xml: documentation The XML output combined with the XSL Transformations provided in data/ are incredibly useful, they should be documented. * doc/bison.texi (Xml): New node. 2012-11-12 Theophile Ranquet output: capitalize State * src/print.c (print_state): Here. * tests/conflicts.at, tests/existing.at, tests/local.at, tests/reduce.at, tests/regression.at, tests/sets.at: Adjust. 2012-11-12 Akim Demaille tests: fix syntax-check errors Reported by Théophile Ranquet. * tests/c++.at: Use AT_PARSER_CHECK. Avoid using "strcmp", which triggers an error from syntax-check. 2012-11-12 Akim Demaille Merge remote-tracking branch 'origin/maint' * origin/maint: maint: address syntax-check errors. tests: use valgrind where appropriate tests: use valgrind where appropriate tests: don't expect $EGREP to support -w tests: more possible error compiler messages for "#error" 2012-11-12 Akim Demaille maint: address syntax-check errors. * cfg.mk: Ignore the "error" call in tests/c++.at, it is not to be translated. * doc/bison.texi: Fix incorrect @pxref use. * po/POTFILES.in: Add missing file. * src/print_graph.c: Remove useless include. 2012-11-12 Akim Demaille tests: use valgrind where appropriate Reported by Théophile Ranquet. * cfg.mk (sc_at_parser_check): New. * tests/c++.at: Fix use of AT_CHECK vs. AT_PARSER_CHECK. 2012-11-12 Akim Demaille Merge remote-tracking branch 'origin/branch-2.6' into maint * origin/branch-2.6: tests: use valgrind where appropriate tests: don't expect $EGREP to support -w 2012-11-10 Akim Demaille tests: use valgrind where appropriate Reported by Théophile Ranquet. * tests/glr-regression.at: Rewrite some test cases so that AT_PARSER_CHECK, which runs valgrind, is exposed with the parser, not with "echo". * tests/local.at, tests/regression.at, tests/headers.at: Use AT_PARSER_CHECK for generated parsers. 2012-11-08 Akim Demaille tests: don't expect $EGREP to support -w Does not work on Solaris 10. Reported by Dennis Clarke. http://lists.gnu.org/archive/html/bug-bison/2012-11/msg00009.html * tests/headers.at (Several parsers): Use Perl instead. While at it, run it only once, on all the generated headers. Adjust to YY_NULL be defined in position.hh. 2012-11-08 Akim Demaille tests: more possible error compiler messages for "#error" * tests/synclines.at (AT_SYNCLINES_COMPILE): Adjust for Clang. Verified with GCC 4.0, 4.2 to 4.8, and Clang 2.9, 3.2: none skip. 2012-11-08 Akim Demaille regen 2012-11-08 Akim Demaille Merge branch 'maint' * origin/maint: regen maint: post-release administrivia version 2.6.5 regen tests: syntax-check tests: beware of compilers that do not support POSIXLY_CORRECT gnulib: update 2012-11-08 Akim Demaille regen 2012-11-08 Akim Demaille Merge branch 'branch-2.6' into maint * origin/branch-2.6: maint: post-release administrivia version 2.6.5 regen tests: syntax-check tests: beware of compilers that do not support POSIXLY_CORRECT gnulib: update 2012-11-07 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2012-11-07 Akim Demaille version 2.6.5 * NEWS: Record release date. 2012-11-07 Akim Demaille regen 2012-11-07 Akim Demaille tests: syntax-check * tests/actions.at: Fix typo. 2012-11-07 Akim Demaille tests: beware of compilers that do not support POSIXLY_CORRECT Running "maintainer-release-check" on OS X with Clang 2.9 fails, because "clang-mp-2.9 -o test -g test.c" launches "/usr/bin/dsymutil test -o test.dSYM" which fails with "error: unable to open executable '-o'". * m4/c-working.m4 (BISON_CHECK_WITH_POSIXLY_CORRECT) (BISON_C_COMPILER_POSIXLY_CORRECT): New. * configure.ac: Use the latter. * tests/atlocal.in (POSIXLY_CORRECT_IS_EXPORTED): New. * tests/local.at (AT_BISON_CHECK_WARNINGS_): Use it instead of computing its value each time. (AT_QUELL_VALGRIND): Skip tests that cannot work because of compilers that do not support POSIXLY_CORRECT. 2012-11-07 Akim Demaille gnulib: update 2012-11-06 Akim Demaille Merge remote-tracking branch 'origin/maint' * origin/maint: (24 commits) tests: calc: modernize the use of locations tests: remove useless location initializations lalr1.cc: always initialize yylval. tests: check that C and C++ objects can be linked together. yacc.c: also disable -Wuninitialized. glr.cc, yacc.c: initialize yylloc properly yacc.c, glr.c: a better YY_LOCATION_PRINT yacc.c: simplify initialization doc: formatting changes c++: fix position operator signatures tests: remove useless location initialization. tests: fix locations in C tests: handle %parse-param in the generated yyerror tests: simplifications grammars: fix display of nul character in error message tests: sort tests: cosmetic changes comment changes autoconf: update gnulib: update ... 2012-11-06 Akim Demaille Merge branch 'branch-2.6' into maint * origin/branch-2.6: (24 commits) tests: calc: modernize the use of locations tests: remove useless location initializations lalr1.cc: always initialize yylval. tests: check that C and C++ objects can be linked together. yacc.c: also disable -Wuninitialized. glr.cc, yacc.c: initialize yylloc properly yacc.c, glr.c: a better YY_LOCATION_PRINT yacc.c: simplify initialization doc: formatting changes c++: fix position operator signatures tests: remove useless location initialization. tests: fix locations in C tests: handle %parse-param in the generated yyerror tests: simplifications grammars: fix display of nul character in error message tests: sort tests: cosmetic changes comment changes autoconf: update gnulib: update ... 2012-11-06 Akim Demaille tests: calc: modernize the use of locations * tests/calc.at: Don't initialize the location, let the parser do it. Use a $printer. Change some testing input to be easier to distinguish (instead of always "0 0" for instance). 2012-11-06 Akim Demaille tests: remove useless location initializations * tests/actions.at, tests/calc.at: here. 2012-11-06 Akim Demaille lalr1.cc: always initialize yylval. * data/lalr1.cc: here. 2012-11-06 Akim Demaille tests: check that C and C++ objects can be linked together. * tests/local.at (AT_SKIP_IF_CANNOT_LINK_C_AND_CXX): New. * tests/headers.at (Several parsers): Use it. 2012-11-06 Akim Demaille yacc.c: also disable -Wuninitialized. * data/yacc.c (YY_IGNORE_MAYBE_UNINITIALIZED_BEGIN): For some versions of GCC, -Wmaybe-uninitialized alone does not suffice. 2012-11-06 Akim Demaille glr.cc, yacc.c: initialize yylloc properly There are several issues to address here. One is that yylloc should be initialized when possible. Another is that the push parser needs to update yypushed_loc when the user modified it. And if the parser starts by a reduction of an empty, it uses the first location on the stack, which, therefore, must also be initialized to this initial location. This is getting complex, especially since because initializing a global (impure interface) is different from initializing a local variable. To simplify, the local yylloc is not initialized during its definition. * data/c.m4 (b4_yyloc_default_define): Replace by... (b4_yyloc_default): this. Adjust dependencies. * data/glr.cc: Initialize yylloc. * data/yacc.c (b4_declare_scanner_communication_variables): Initialize yylloc during its definition. Don't define yyloc_default. (yypush_parse): The location formal is not const, as we might initialize it. (yyparse): Define yyloc_default. Use it before running the user initial action. Possibly update the first location on the stack, and the pushed location after the user initial action. * tests/actions.at (Initial locations): Check that the initial location is correct. 2012-11-06 Akim Demaille yacc.c, glr.c: a better YY_LOCATION_PRINT * data/c.m4 (b4_yy_location_print_define): New. Now issues "short" locations, e.g., "1.1" instead of "1.1-1.1". Was initially a function, but then we face "static but unused" warnings. Simpler as a macro. * tests/local.at, data/glr.c, data/yacc.c: Use it instead of duplicating. * tests/actions.at: Adjust expectations. 2012-11-06 Akim Demaille yacc.c: simplify initialization * data/yacc.c: Fuse the initializations of yyssp, yyss and the like. Remove an obsolete comment: we do initialize these initial stack members (in some cases). 2012-11-05 Akim Demaille doc: formatting changes * doc/bison.texi: In a pointer type. 2012-11-05 Akim Demaille c++: fix position operator signatures * data/location.cc (operator+=, operator-=): Remove const from return type. 2012-11-05 Akim Demaille tests: remove useless location initialization. * tests/glr-regression.at: here. glr.c does initialize yylloc. 2012-11-05 Akim Demaille tests: fix locations in C * tests/local.at (AT_YYERROR_DEFINE): Don't display the end of the location if it is not after its beginning. * tests/actions.at, tests/cxx-type.at: Adjust the expected output. 2012-11-05 Akim Demaille tests: handle %parse-param in the generated yyerror * tests/local.at (AT_PARSE_PARAMS): New. (AT_YYERROR_FORMALS, AT_YYERROR_DEFINE): Use it to add the parse-param to yyerror. * tests/calc.at, tests/regression.at: Use AT_YYERROR_DEFINE and AT_YYERROR_DECLARE, now that they handle properly the parse-params. Be sure to let AT_BISON_OPTION_PUSHDEFS now what parse-params are used. 2012-11-05 Akim Demaille tests: simplifications * tests/actions.at (Exotic Dollars): Formatting changes. Use AT_FULL_COMPILE. (AT_CHECK_PRINTER_AND_DESTRUCTOR): Remove useless initialization of @$. 2012-11-01 Akim Demaille lalr1.cc: rename lex_symbol as api.token.constructor * data/bison.m4 (b4_lex_symbol_if): Rename as... (b4_token_ctor_if): this. Depend upon api.token.constructor. * data/c++.m4, data/lalr1.cc: Adjust. * doc/bison.texi: Fix all the occurrences of lex_symbol. * etc/bench.pl.in: Adjust. * examples/variant.yy: Likewise. * tests/local.at (AT_BISON_OPTION_PUSHDEFS, AT_BISON_OPTION_POPDEFS): Handle AT_TOKEN_CTOR_IF. * tests/c++.at: Adjust to using api.token.constructor and AT_TOKEN_CTOR_IF. Simplify the test of both build call styles. (AT_CHECK_VARIANTS): Rename as... (AT_TEST): this. And undef when done. 2012-11-01 Akim Demaille examples: simplify/improve * examples/variant.yy: Put yylex in yy::, and simplify accordingly. Minor formatting changes. 2012-11-01 Akim Demaille bison.m4: support b4_*_if macros whose name differ from their variable * data/bison.m4 (b4_percent_define_if_define_, b4_percent_define_if_define): Accept a second argument. 2012-11-01 Akim Demaille grammars: fix display of nul character in error message Reported by Marc Mendiola. http://lists.gnu.org/archive/html/help-bison/2012-10/msg00017.html * gnulib: Update to get quote_mem. * src/scan-gram.l: Use it. * tests/input.at (Invalid inputs): Additional checks. * tests/named-refs.at: Likewise. 2012-11-01 Akim Demaille tests: sort * tests/regression.at (Invalid inputs, Invalid inputs with {}): Move to... * tests/input.at: here, for consistency. 2012-11-01 Akim Demaille tests: cosmetic changes * tests/actions.at (AT_CHECK_PRINTER_AND_DESTRUCTOR): Improve the displayed title. 2012-11-01 Akim Demaille comment changes * data/lalr1.cc: here. 2012-11-01 Akim Demaille autoconf: update There are comment changes only in the files we use. 2012-11-01 Akim Demaille gnulib: update 2012-10-28 Akim Demaille regen 2012-10-28 Akim Demaille yacc.c: initialize yylval and yylloc. When generating a pure push parser, the initialization of yylval and yylloc may not be visible to the compiler. With warnings enabled, GCC 4.3.6, 4.4.7, 4.5.4, and 4.6.3 report uninitialized uses of yylval/yylloc. Using local pragmas to disable these warnings is not supported before 4.6, and 4.6 does not support it properly. So initialize yylval and yylloc at their definition. Reported by Peter Simons. See http://lists.gnu.org/archive/html/bison-patches/2012-10/msg00133.html * data/c.m4 (b4_yyloc_default_define): New. * data/yacc.c: Use it when locations are requested. (YYLVAL_INITIALIZE): Replace by... (YY_INITIAL_VALUE): this. (yyparse): Initialize yylloc and yylval. Therefore, remove the initialization of yylloc's field. * data/glr.c: Likewise. 2012-10-26 Theophile Ranquet graphs: fix spacing refactoring * src/print_graph.c (print_lhs, print_core): Here. 2012-10-26 Theophile Ranquet tests: make deprecation tests more specific * tests/input.at (Deprecated directives): Here, don't generate unrelated errors or warnings. 2012-10-26 Theophile Ranquet tests: fix AT_BISON_CHECK_WARNINGS_ stderr rewriting * tests/input.at (Deprecated directives): Avoid spurious error. * tests/locat.at (AT_BISON_CHECK_WARNINGS): Adjust for recent changes. 2012-10-26 Theophile Ranquet scan-skel.l: consider m4 notes as related to "complaint" errors * src/scan-skel.l (flag): Here. 2012-10-26 Theophile Ranquet warnings: distinguish context information based on warning type * src/scan-code.l (show_sub_message, show_sub_messages): Take a new warnings argument. 2012-10-26 Theophile Ranquet warnings: fix early exit of warnings treated as errors Treating warnings as errors caused Bison to exit earlier than needed, making it hide warnings that would have been printed had -Werror not been set. Also, fix a bug that caused some context information of errors to not be shown. * src/complain.c (complaint_issued): Rename as... (complaint_status): This, and change its type from boolean to * src/complain.h (err_status): This, new enumeration. * src/main.c (main): Adjust (only finish early if an actual complaint was risen, not a mere warning treated an error). * src/reader.c: Adjust. 2012-10-26 Theophile Ranquet tests: reindent for legibility * tests/local.at (AT_BISON_CHECK_WARNINGS_): Here. 2012-10-26 Akim Demaille build: fix Texinfo compilation * doc/local.mk: fix dependencies. 2012-10-26 Akim Demaille regen 2012-10-26 Akim Demaille Merge remote-tracking branch 'origin/maint' * origin/maint: (46 commits) doc: minor style change maint: use gendocs's new -I option regen yacc.c: do not define location support when not using locations maint: be compilable with GCC 4.0 tests: address a warning from GCC 4.4 tests: don't use options that Clang does not support tests: restore the tests on -Werror regen parse-gram: update the Bison interface fix comment maint: post-release administrivia version 2.6.4 regen 2.6.4: botched 2.6.3 maint: post-release administrivia version 2.6.3 gnulib: update tests: check %no-lines NEWS: warnings with clang ... 2012-10-26 Akim Demaille Merge branch 'branch-2.6' into maint * origin/branch-2.6: regen yacc.c: do not define location support when not using locations maint: be compilable with GCC 4.0 tests: address a warning from GCC 4.4 tests: don't use options that Clang does not support tests: restore the tests on -Werror regen parse-gram: update the Bison interface fix comment 2012-10-26 Akim Demaille doc: minor style change * doc/figs/example-reduce.txt: here. 2012-10-26 Akim Demaille maint: use gendocs's new -I option * gnulib: Update gendocs. * cfg.mk (gendocs_options_): New. 2012-10-26 Akim Demaille regen 2012-10-26 Akim Demaille yacc.c: don't use _Pragma GCC diagnostic with 4.6 Reported by Peter Simons. http://lists.gnu.org/archive/html/bug-bison/2012-10/msg00033.html * data/yacc.c (b4_declare_scanner_communication_variables): 4.7 seems fine though. 2012-10-26 Akim Demaille regen 2012-10-26 Akim Demaille yacc.c: do not define location support when not using locations * data/yacc.c (YYLLOC_DEFAULT, YYRHSLOC): Don't define when not using locations. 2012-10-26 Akim Demaille maint: be compilable with GCC 4.0 The "shadows a global declaration" warning in GCC 4.0 was a bit annoying. It does not like that a type name be used in a prototype of a function (not the implementation, just the declaration): In file included from src/LR0.c:38: src/reader.h:56: warning: declaration of 'named_ref' shadows a global declaration src/named-ref.h:35: warning: shadowed declaration is here It does not like either when a global variable name is used in a prototype. Flex 2.5.37 generates this prototype: void gram_set_debug (int debug_flag ); * src/getargs.h, src/getargs.c (debug_flag): Rename as... (debug): this. Adjust dependencies. * src/reader.h: Don't use "named_ref" as a formal argument name. 2012-10-26 Theophile Ranquet misc: document TESTSUITEFLAGS in README-hacking * README-hacking: Document -j and -k flags. 2012-10-26 Theophile Ranquet deprecation: add tests * tests/input.at (Deprecated directives warn, Non-deprecated directives don't, Unput doesn't mess up locations): New tests. 2012-10-25 Akim Demaille tests: address a warning from GCC 4.4 236. torture.at:465: testing Exploding the Stack Size with Alloca ... ../../../tests/torture.at:474: bison -o input.c input.y ../../../tests/torture.at:474: $CC $CFLAGS $CPPFLAGS $LDFLAGS -o input input.c $LIBS stderr: cc1: warnings being treated as errors input.y: In function 'main': input.y:60: error: 'status' may be used uninitialized in this function * tests/torture.at (AT_DATA_STACK_TORTURE): Initial status to avoid the previous error. 2012-10-25 Akim Demaille tests: don't use options that Clang does not support * configure.ac (WARN_CFLAGS, WARN_CXXFLAGS): Do not include options that Clang does not support. 2012-10-25 Akim Demaille tests: restore the tests on -Werror When run as /bin/sh, Bash sets the shell variable POSIXLY_CORRECT to y. The test suite checks for the envvar POSIXLY_CORRECT to turn of some tests not supported in POSIX mode. Restore these tests. Reported by the Hydra build farm, from Rob Vermaas. * tests/local.at (AT_BISON_CHECK_WARNINGS_): Check the envvar POSIXLY_CORRECT, not the shell variable. 2012-10-25 Akim Demaille regen 2012-10-25 Akim Demaille parse-gram: update the Bison interface * src/parse-gram.y (%pure-parser, %name-prefix): Replace with... (%define api.pure, %define api.prefix) * src/location.h, src/scan-gram.h: Adjust to api.prefix. 2012-10-25 Akim Demaille fix comment * data/c.m4 (b4_YYDEBUG_define): here. 2012-10-24 Theophile Ranquet regen 2012-10-24 Theophile Ranquet deprecation: issue warnings in scanner * src/parse-gram.y: Move the handling of (three) deprecated constructs ... * src/scan-gram.l: ...Here, and issue warnings. (DEPRECATED): New. 2012-10-23 Akim Demaille Merge branch 'branch-2.6' into maint * origin/branch-2.6: maint: post-release administrivia version 2.6.4 regen 2.6.4: botched 2.6.3 2012-10-23 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2012-10-23 Akim Demaille version 2.6.4 * NEWS: Record release date. 2012-10-22 Akim Demaille regen 2012-10-22 Akim Demaille 2.6.4: botched 2.6.3 * NEWS: here. 2012-10-22 Akim Demaille Merge branch '2.6.3' into maint * 2.6.3: (22 commits) maint: post-release administrivia version 2.6.3 gnulib: update tests: check %no-lines NEWS: warnings with clang warnings: avoid warnings from clang tests: no longer disable -O compiler options yacc.c: initialize yylval in pure-parser mode skeletons: style changes tests: minor improvements tests: use $PERL instead of perl build: look for Perl in configure. tests: fix sed portability issues tests: diff -u is not portable maint: word changes lalr1.cc: fix test suite portability maint: fix an erroneous include tests: check that headers are self contained doc: add missing documentation for --report headers: move CPP guards into YY_*_INCLUDED to avoid collisions ... 2012-10-22 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2012-10-22 Akim Demaille version 2.6.3 * NEWS: Record release date. 2012-10-22 Akim Demaille gnulib: update 2012-10-22 Akim Demaille tests: check %no-lines * tests/synclines.at: here. 2012-10-22 Akim Demaille NEWS: warnings with clang * NEWS: here. 2012-10-22 Akim Demaille warnings: avoid warnings from clang Fix the following warning parse-gram.c:2078:14: error: equality comparison with extraneous parentheses [-Werror,-Wparentheses-equality] if (((yyn) == (-91))) ~~~~~~^~~~~~~~ parse-gram.c:2078:14: note: remove extraneous parentheses around the comparison to silence this warning if (((yyn) == (-91))) ~ ^ ~ parse-gram.c:2078:14: note: use '=' to turn this equality comparison into an assignment if (((yyn) == (-91))) ^~ = 1 error generated. and the following one: input.cc:740:1: error: function declared 'noreturn' should not return [-Werror,-Winvalid-noreturn] static void yyMemoryExhausted (yyGLRStack* yystackp) __attribute__ ((__noreturn__)); static void yyMemoryExhausted (yyGLRStack* yystackp) { YYLONGJMP (yystackp->yyexception_buffer, 2); } ^ 1 warning and 1 error generated. This is Apple clang version 3.1 (tags/Apple/clang-318.0.61). * data/c.m4 (b4_table_value_equals): Use (!!(A == B)) instead of (A == B) to avoid this warning. Any reasonable compiler should generate the same code. * src/uniqstr.h (UNIQSTR_EQ): Likewise. * data/glr.c (LONGJMP): abort after longjmp to pacify clang. 2012-10-22 Akim Demaille tests: no longer disable -O compiler options Tests are running without -O since f377f69fec28013c79db4efe12bbb9d48987fb2c because some warnings (about yylval not being initialized) show only when GCC is given -O2. The previous patch fixes the warnings. Run the test suite with compiler options unmodified. * tests/atlocal.in (O0CFLAGS, O0CXXFLAGS): Remove, use CFLAGS and CXXFLAGS. 2012-10-22 Paul Eggert yacc.c: initialize yylval in pure-parser mode See http://lists.gnu.org/archive/html/bison-patches/2012-08/msg00024.html (spreading over September and October). * data/yacc.c (YY_IGNORE_MAYBE_UNINITIALIZED_BEGIN) (YY_IGNORE_MAYBE_UNINITIALIZED_END, YYLVAL_INITIALIZE): New macros. Use them to suppress an unwanted GCC diagnostic. 2012-10-22 Akim Demaille skeletons: style changes * data/yacc.c, data/glr.c: Prefer Title case for (CPP) macro arguments. 2012-10-22 Akim Demaille tests: minor improvements * tests/c++.at: Space changes. Use AT_YYERROR_DEFINE. * tests/local.at (AT_YYERROR_DEFINE): Issue errors on unknown languages. 2012-10-22 Akim Demaille tests: use $PERL instead of perl * tests/atlocal.in (PERL): New. Sort. * tests/calc.at, tests/input.at, tests/local.at, tests/regression.at, * tests/skeletons.at, tests/synclines.at, tests/torture.at: here. 2012-10-22 Akim Demaille build: look for Perl in configure. Bison uses "/usr/bin/perl" or "perl" in several places, and it does not appear to be a problem. But, at least to make it simpler to change PERL on the make command line, check for perl in configure. * configure.ac (PERL): New. * doc/Doxyfile.in, doc/Makefile.am, tests/bison.in: Use it. 2012-10-22 Akim Demaille tests: fix sed portability issues Reported by Didier Godefroy, . * tests/calc.at (AT_CHECK_SPACES): Use Perl. 2012-10-22 Akim Demaille tests: diff -u is not portable Reported by Didier Godefroy . * tests/existing.at (AT_LALR1_DIFF_CHECK): Skip if diff -u does not work. 2012-10-22 Akim Demaille maint: word changes * README-hacking (Typical errors): Improve wording. 2012-10-22 Akim Demaille lalr1.cc: fix test suite portability Reported by Rob Vermaas' Hydra build farm on x86_64-darwin 10.2.0 with G++ 4.6.3. * tests/headers.at (Several parsers): Include AT_DATA_SOURCE_PROLOGUE in the files to compile. * data/location.cc: Do not include twice string and iostream (once by position.hh, and then by location.hh). * README-hacking (Typical errors): Some hints for other maintainers. 2012-10-22 Theophile Ranquet maint: fix an erroneous include This fixes test 130 (Several parsers). * data/location.cc: Include rather than since we really need << on strings for instance. * NEWS: Document this. 2012-10-22 Akim Demaille tests: check that headers are self contained Reported by Alexandre Duret-Lutz. * tests/headers.at (Several parsers): here. 2012-10-22 Akim Demaille doc: add missing documentation for --report * doc/bison.texi (Bison Options): Document --report's "solved", "all", and "none". 2012-10-22 Akim Demaille headers: move CPP guards into YY_*_INCLUDED to avoid collisions See . * data/c.m4 (b4_cpp_guard): Prepend YY_ and append _INCLUDED. * tests/headers.at: Adjust. * NEWS, doc/bison.texi: Document. 2012-10-22 Akim Demaille minor changes. * NEWS: Word changes. * doc/bison.texi: Spell check. Fix minor issues. * tests/headers.at: Comment and formatting changes. 2012-10-22 Akim Demaille gnulib: update 2012-10-19 Akim Demaille gnulib: update 2012-10-19 Akim Demaille xml: slight improvement of the DOT output This was completely forgotten... Nothing about XML is actually documented... * data/xslt/xml2dot.xsl: Use boxes, and Courier font. 2012-10-19 Akim Demaille maint: check for dot before using it * configure.ac: here. * doc/Makefile.am: Use $(DOT). Ship the generated files, to spare the user the need for Graphviz. 2012-10-18 Theophile Ranquet graphs: documentation Note that 'make web-manual' fails. * NEWS: Document these changes. * doc/Makefile.am: Adjust to generate example files. * doc/bison.texi: Add a Graphviz section after "Understanding::", the section describing the .output file, because these are similar. * doc/figs/example-reduce.dot, doc/figs/example-reduce.txt, doc/figs/example-shift.dot, doc/figs/example-shift.txt: New, minimal examples to illustrate the documentation. 2012-10-18 Theophile Ranquet graphs: add tests, introducing -k graph * tests/output.at (AT_TEST): New. Use it to add 6 --graph tests. 2012-10-18 Theophile Ranquet graphs: change the output format of the rules Use something similar to the report file. * src/print_graph.c (print_lhs): New, obstack equivalent of rule_lhs_print. (print_core): Use here. 2012-10-18 Theophile Ranquet graphs: style changes * src/graphviz.c (start_graph): Use courier font. (conclude_red): Use commas to separate attributes. Show the acceptation as a special reduction, with a blue color and an "Acc" label. Show the lookahead tokens between square brackets. (output_red): No longer label default reductions. * src/print_graph.c (print_core): Refactor spacing, and print an additional space between a rule's rhs and its lookahead tokens. Also, capitalize "State". (print_actions): Style, move a declaration. 2012-10-18 Theophile Ranquet graphs: address an issue with R/R conflicts All disabled reductions should now be shown as such. * src/graphviz.c (output_red): Here. (conclude_red): New. 2012-10-16 Akim Demaille news: spell check * NEWS: here. 2012-10-16 Akim Demaille Merge branch 'maint' * origin/maint: java: use api.location.type and api.position.type 2012-10-16 Akim Demaille variables: use singular in %define variable names See http://lists.gnu.org/archive/html/bison-patches/2012-02/msg00045.html * doc/bison.texi, src/lalr.c, src/main.c, src/muscle-tab.c, * src/print.c, src/reader.c, src/tables.c, tests/conflicts.at, * tests/input.at, tests/reduce.at: s/lr.default-reductions/lr.default-reduction/ s/lr.keep-unreachable-states/lr.keep-unreachable-state/. * NEWS: Document. 2012-10-16 Akim Demaille java: fixes * data/java.m4: Remove stray M4 characters. 2012-10-16 Akim Demaille api.tokens.prefix -> api.token.prefix See http://lists.gnu.org/archive/html/bison-patches/2012-02/msg00045.html Note that api.tokens.prefix has not been released, yet. * NEWS, data/bison.m4, doc/bison.texi, tests/c++.at, * tests/calc.at, tests/java.at, tests/local.at: Do it. * src/muscle-tab.c (muscle_percent_variable_update): Ensure backward compatibility. 2012-10-15 Theophile Ranquet scan-skel.l: shift complain_args arguments Because argv[0] is never used, shift it out from the argument list. * src/complain.c (complain_args): Here. * src/scan-skel.l (at_complain): Adjust argv and argc. 2012-10-15 Theophile Ranquet scan-skel.l: formatting changes * src/scan-skel.l (fail_for_at_directive_too_few_args): Here. 2012-10-12 Akim Demaille java: use api.location.type and api.position.type * data/java.m4: here. * NEWS, doc/bison.texi, tests/java.at: Adjust. 2012-10-12 Akim Demaille Merge branch 'maint' * origin/maint: tests: check %no-lines tests: minor simplification graphs: stylistic changes. graphs: minor style changes graphs: show reductions graphs: style: prefix state number with "state" graphs: style: use left justification for states graphs: style: prefix rules and change shapes obstack: import obstack_finish0 from master c++: api.location.type muscles: a function for backward compatibility maint: more macros 2012-10-12 Akim Demaille tests: check %no-lines * tests/synclines.at: here. 2012-10-12 Akim Demaille tests: minor simplification * tests/headers.at (Several parsers): Use *.y even for C++. 2012-10-11 Akim Demaille graphs: stylistic changes. * src/graphviz.c (output_red): Comment and formatting changes. 2012-10-11 Theophile Ranquet graphs: minor style changes * src/graphviz.c (output_red): Fix C90 issues. Reduce variable scopes. 2012-10-11 Theophile Ranquet graphs: show reductions * src/graphviz.c (output_red): New, show reductions on the graph. (no_reduce_bitset_init): New, initialize a bitset. (print_token): New, print a lookahead token. (escape): New, print "foo" as \"foo\" because Dot doesn't like quotes within a label. * src/graphviz.h : Adjust. * src/print_graph.c (print_actions): Call output_red here. 2012-10-11 Theophile Ranquet graphs: style: prefix state number with "state" * src/print_graph.c (print_core): Here. 2012-10-11 Theophile Ranquet graphs: style: use left justification for states The label text of nodes is centered "by default" (by the use of '\n' as a line feed). This gives bad readability to the grammar rules shown in state nodes, a left justification is much nicer. This is done by using '\l' as the line feed. In order to allow \l in the DOT file, changes to the quoting system seem necessary. * src/print_graph.c (print_core): Escape tokens here, instead of... * src/graphviz.c (output_node): Here... (escape): Using this, new. 2012-10-11 Theophile Ranquet graphs: style: prefix rules and change shapes * src/graphviz.c (start_graph): Use box rather than ellipsis. * src/print_graph.c (print_core): Prefix rules with their number. 2012-10-11 Theophile Ranquet obstack: import obstack_finish0 from master * src/system.h (obstack_finish0): New. 2012-10-11 Akim Demaille Merge branch 'maint' * origin/maint: NEWS: warnings with clang warnings: avoid warnings from clang tests: no longer disable -O compiler options yacc.c: initialize yylval in pure-parser mode skeletons: style changes lalr1.cc: document exception safety lalr1.cc: check exception safety of error handling lalr1.cc: check (and fix) %printer exception safety lalr1.cc: check (and fix) %initial-action exception safety lalr1.cc: fix exception safety lalr1.cc: check exception safety. lalr1.cc: indentation fixes. lalr1.cc: don't leave macros define to nothing tests: minor improvements tests: use $PERL instead of perl build: look for Perl in configure. tests: fix sed portability issues tests: diff -u is not portable 2012-10-09 Akim Demaille c++: api.location.type This feature was introduced in 95a2de5695670ae0df98cb3c42141cad549f0204 (which is part of 2.5), but not documented. Give it a proper name, and make it public. * data/c++.m4, data/lalr1.cc, data/glr.cc, data/java.m4: Use api.location.type instead of location_type. * src/muscle-tab.c (muscle_percent_variable_update): Map the latter to the former. * tests/local.at: Adjust. * tests/calc.at: Use api.location.type. Leave tests/java.at with location_type, at least for the time being, to cover both names. * doc/bison.texi: Document api.location.type. (User Defined Location Type): New. * NEWS: Update. 2012-10-09 Akim Demaille muscles: a function for backward compatibility Based on commit 171ad99d6421935a278656be6dc7161591835d00 from master. * src/muscle-tab.c (muscle_percent_variable_update): New. (muscle_percent_define_insert): Use it. Define the variables with their initial value. 2012-10-09 Akim Demaille maint: more macros * src/output.c (ARRAY_CARDINALITY): Move to... * src/system.h: here. (STREQ, STRNEQ): new. 2012-10-08 Akim Demaille NEWS: warnings with clang * NEWS: here. 2012-10-08 Akim Demaille warnings: avoid warnings from clang Fix the following warning parse-gram.c:2078:14: error: equality comparison with extraneous parentheses [-Werror,-Wparentheses-equality] if (((yyn) == (-91))) ~~~~~~^~~~~~~~ parse-gram.c:2078:14: note: remove extraneous parentheses around the comparison to silence this warning if (((yyn) == (-91))) ~ ^ ~ parse-gram.c:2078:14: note: use '=' to turn this equality comparison into an assignment if (((yyn) == (-91))) ^~ = 1 error generated. and the following one: input.cc:740:1: error: function declared 'noreturn' should not return [-Werror,-Winvalid-noreturn] static void yyMemoryExhausted (yyGLRStack* yystackp) __attribute__ ((__noreturn__)); static void yyMemoryExhausted (yyGLRStack* yystackp) { YYLONGJMP (yystackp->yyexception_buffer, 2); } ^ 1 warning and 1 error generated. This is Apple clang version 3.1 (tags/Apple/clang-318.0.61). * data/c.m4 (b4_table_value_equals): Use (!!(A == B)) instead of (A == B) to avoid this warning. Any reasonable compiler should generate the same code. * src/uniqstr.h (UNIQSTR_EQ): Likewise. * data/glr.c (LONGJMP): abort after longjmp to pacify clang. 2012-10-08 Akim Demaille tests: no longer disable -O compiler options Tests are running without -O since f377f69fec28013c79db4efe12bbb9d48987fb2c because some warnings (about yylval not being initialized) show only when GCC is given -O2. The previous patch fixes the warnings. Run the test suite with compiler options unmodified. * tests/atlocal.in (O0CFLAGS, O0CXXFLAGS): Remove, use CFLAGS and CXXFLAGS. 2012-10-08 Paul Eggert yacc.c: initialize yylval in pure-parser mode See http://lists.gnu.org/archive/html/bison-patches/2012-08/msg00024.html (spreading over September and October). * data/yacc.c (YY_IGNORE_MAYBE_UNINITIALIZED_BEGIN) (YY_IGNORE_MAYBE_UNINITIALIZED_END, YYLVAL_INITIALIZE): New macros. Use them to suppress an unwanted GCC diagnostic. 2012-10-08 Akim Demaille skeletons: style changes * data/yacc.c, data/glr.c: Prefer Title case for (CPP) macro arguments. 2012-10-06 Akim Demaille lalr1.cc: document exception safety * NEWS: here. * doc/bison.texi (Destructor Decl, C++ Parser Interface): and there. 2012-10-06 Akim Demaille lalr1.cc: check exception safety of error handling * tests/c++.at (Exception safety): Don't use swap here, it is useless. Cover more test cases: yyerror, YYERROR, YYABORT, and error recovery. (Object): Instead of just keeping a counter of instances, keep a list of them. 2012-10-06 Akim Demaille lalr1.cc: check (and fix) %printer exception safety * tests/c++.at (Exception safety): Let the parser support the --debug option. On 'p', throw an exception from the %printer. * data/lalr1.cc (yyparse): Do not display the values we discard, as it uses %printer, which might have thrown the exception. 2012-10-06 Akim Demaille lalr1.cc: check (and fix) %initial-action exception safety * data/lalr1.cc: Check size > 1, rather than size != 1, when cleaning the stack, as at the beginning, size is 0. * tests/c++.at (Exception safety): Check exception safety in %initial-action. 2012-10-06 Akim Demaille lalr1.cc: fix exception safety lalr1.cc does not reclaim its memory when ended by an exception. Reported by Oleksii Taran: http://lists.gnu.org/archive/html/help-bison/2012-09/msg00000.html * data/lalr1.cc (yyparse): Protect the whole yyparse by a try-catch block that cleans the stack and the lookahead. 2012-10-06 Akim Demaille lalr1.cc: check exception safety. * tests/c++.at (Exception safety): New. 2012-10-06 Akim Demaille lalr1.cc: indentation fixes. * data/lalr1.cc (yyparse): here. Untabify a block of code. 2012-10-06 Akim Demaille lalr1.cc: don't leave macros define to nothing * data/lalr1.cc (YY_SYMBOL_PRINT, YY_REDUCE_PRINT, YY_STACK_PRINT): Define to something so that, for instance, "if (foo) YY_SYMBOL_PRINT" is valid even when !YYDEBUG. 2012-10-06 Akim Demaille tests: minor improvements * tests/c++.at: Space changes. Use AT_YYERROR_DEFINE. * tests/local.at (AT_YYERROR_DEFINE): Issue errors on unknown languages. 2012-10-05 Akim Demaille tests: use $PERL instead of perl * tests/atlocal.in (PERL): New. Sort. * tests/calc.at, tests/input.at, tests/local.at, tests/regression.at, * tests/skeletons.at, tests/synclines.at, tests/torture.at: here. 2012-10-05 Akim Demaille build: look for Perl in configure. Bison uses "/usr/bin/perl" or "perl" in several places, and it does not appear to be a problem. But, at least to make it simpler to change PERL on the make command line, check for perl in configure. * configure.ac (PERL): New. * doc/Doxyfile.in, doc/Makefile.am, tests/bison.in: Use it. 2012-10-05 Akim Demaille tests: fix sed portability issues Reported by Didier Godefroy, . * tests/calc.at (AT_CHECK_SPACES): Use Perl. 2012-10-05 Akim Demaille tests: diff -u is not portable Reported by Didier Godefroy . * tests/existing.at (AT_LALR1_DIFF_CHECK): Skip if diff -u does not work. 2012-10-04 Akim Demaille Merge branch 'maint' * origin/maint: maint: word changes lalr1.cc: fix test suite portability maint: fix an erroneous include tests: check that headers are self contained doc: add missing documentation for --report 2012-10-04 Akim Demaille scan-skel: use the scanner to reject all invalid directives * src/scan-skel.l: Use a simpler and more consistent pattern escaping scheme. Catch all the invalid directives here by just removing the previous catch-all-but-alphabetical rule. 2012-10-04 Theophile Ranquet scan-skel: recognize the @directives directly in scanner * src/scan-skel.l (at_directive, at_init): New. (at_ptr): New, function pointer used to call the right at_directive function (at_basename, etc.). (outname): Rename as... (out_name): this, for consistency with out_lineno. 2012-10-04 Theophile Ranquet scan-skel: split @directive functions * src/scan-skel.l (at_directive_perform): Split as... (at_basename, at_complain, at_output): these. 2012-10-04 Theophile Ranquet errors: support indented context info in m4 macros * TODO: Address the issue, so remove it. * data/bison.m4: Use b4_error with [[note]] rather than a complain_at for context information. * src/complain.c (complain_args): Take an additional argument, an indentation pointer, to allow the dispatching of context information. * src/complain.h (complain_args): Adjust prototype. * src/scan-skel.l (at_directive_perform): Recognize the new @note mark. * tests/input.at: Adjust. 2012-10-04 Theophile Ranquet errors: factor b4_error @directives Instead of @complain, @warn, and @fatal, use a unique @complain directive. This directive's first argument is "complain", "warn", etc. * data/bison.m4 (m4_error): Here. * src/scan-skel.l (at_directive_perform): Adjust. (flag): Replace the switch by safer and more explicit if branches. 2012-10-04 Theophile Ranquet errors: pointerize complain_at_indent * src/complain.c (complain_at_indent): Rename as... (complaint_indent): This, and take the location as a pointer. * src/complain.h, src/muscle-tab.c, src/reader.c, src/scan-code.l, src/symtab.c: Adjust. 2012-10-04 Akim Demaille maint: word changes * README-hacking (Typical errors): Improve wording. 2012-10-04 Akim Demaille lalr1.cc: fix test suite portability Reported by Rob Vermaas' Hydra build farm on x86_64-darwin 10.2.0 with G++ 4.6.3. * tests/headers.at (Several parsers): Include AT_DATA_SOURCE_PROLOGUE in the files to compile. * data/location.cc: Do not include twice string and iostream (once by position.hh, and then by location.hh). * README-hacking (Typical errors): Some hints for other maintainers. 2012-10-03 Theophile Ranquet maint: fix an erroneous include This fixes test 130 (Several parsers). * data/location.cc: Include rather than since we really need << on strings for instance. * NEWS: Document this. 2012-10-03 Akim Demaille tests: check that headers are self contained Reported by Alexandre Duret-Lutz. * tests/headers.at (Several parsers): here. 2012-10-03 Akim Demaille doc: add missing documentation for --report * doc/bison.texi (Bison Options): Document --report's "solved", "all", and "none". 2012-10-01 Akim Demaille build: use gnulib's non-recursive-gnulib-prefix-hack Suggested by Jim Meyering. * etc/prefix-gnulib-mk: Remove, as it is now provided by... * bootstrap.conf (modules): the non-recursive-gnulib-prefix-hack module. * build-aux/.gitignore, configure.ac, m4/.gitignore: Adjust. 2012-10-01 Akim Demaille warnings: simplify the m4 interface * data/bison.m4 (error_at): Replace... (error): this. (b4_warn, b4_complain, b4_fatal): Bounce to their _at equivalent, with empty location. * src/scan-skel.l (at_directive_perform): Simplify accordingly. 2012-10-01 Theophile Ranquet warnings: separate flags_argmatch This function is now a mere iterator that calls flag_argmatch, a new function, that matches a single option parameter. * src/getargs.c (flag_argmatch): New, taken from... (flags_argmatch): Here. 2012-10-01 Theophile Ranquet warnings: refactoring The code here was too confusing, this seems more natural. * src/complain.c (error_message): Move the indentation check and the category output to complains. Also, no longer take a 'warnings' argument. (complains): Factor calls to error_message. 2012-10-01 Theophile Ranquet formatting changes * src/complain.c: Here. 2012-10-01 Theophile Ranquet warnings: organize variadic complaints call Move the dispatch of variadic complains to complain.c, rather than do it in a scanner. * src/complain.h, src/complain.c (complain_args): New. * src/scan-skel.l (at_directive_perform): Use it. 2012-10-01 Theophile Ranquet warnings: fusion of complain and complain_at These functions are very similar, and keeping them seperate makes future improvements difficult, so merge them. This impacts 89 calls. * src/bootstrap.conf: Adjust. * src/complain.c (complain, complain_at): Merge into... (complain): this. (complain_args): Adjust. * src/complain.h, src/conflicts.c, src/files.c, src/getargs.c, * src/gram.c, src/location.c, src/muscle-tab.c, src/parse-gram.y, * src/reader.c, src/reduce.c, src/scan-code.l, src/scan-gram.l, * src/scan-skel.l, src/symlist.c, src/symtab.c: Adjust. 2012-10-01 Theophile Ranquet warnings: remove spurious suffixes on context Rectify a bug that introduced suffixes out of place. * src/complainc.c (complains): Handle all three special warning bits. * src/scan-code.l (show_sub_message): Remove useless argument. * tests/named-refs.at: Adjust. 2012-10-01 Akim Demaille Merge remote-tracking branch 'origin/maint' * origin/maint: headers: move CPP guards into YY_*_INCLUDED to avoid collisions minor changes. 2012-10-01 Akim Demaille Merge remote-tracking branch 'origin/maint' * origin/maint: gnulib: update errors: indent "user token number redeclaration" context 2012-10-01 Akim Demaille headers: move CPP guards into YY_*_INCLUDED to avoid collisions See . * data/c.m4 (b4_cpp_guard): Prepend YY_ and append _INCLUDED. * tests/headers.at: Adjust. * NEWS, doc/bison.texi: Document. 2012-10-01 Akim Demaille minor changes. * NEWS: Word changes. * doc/bison.texi: Spell check. Fix minor issues. * tests/headers.at: Comment and formatting changes. 2012-09-28 Akim Demaille gnulib: update 2012-09-28 Theophile Ranquet errors: indent "user token number redeclaration" context This is the continuation of the work on the readability of errors context. * src/symtab.c (user_token_number_redeclaration): Use complain_at_indent to output with increased indentation level. * tests/input:at: Apply this change. 2012-09-27 Theophile Ranquet errors: don't display "warnings treated as errors" This line doesn't add any meaningful information anymore, the appended [-Werror=CATEGORY] is enough. It is actually more insightful, as it allows to distinguish warnings treated as errors from those that aren't. This line is also removed by gcc 4.8. * src/complain.c (set_warnings_issued): The only action left was checking if the error bit corresponding to the warning issued was set, and that function was only called once. Therefore, remove it, and do its job directly in the caller... (complains): here. * src/complains.h: Adjust. * tests/input.at: Adjust. * NEWS: Document this change. 2012-09-27 Akim Demaille errors: change output, and improve -y coherence The prefix of warnings treated as errors is now "error: ". Also, their suffix now reflects the changes in the Werror option format. An output for -Werror=other used to be: bison: warnings being treated as errors input.y:1.1: warning: stray ',' treated as white space [-Wother] It is now: bison: warnings being treated as errors input.y:1.1: error: stray ',' treated as white space [-Werror=other] The line "warnings being treated as errors" no longer adds any info, it will be removed in a forthcoming change. * NEWS: Add entry "Enhancement of the -Werror" * doc/bison.texi: Move the warnings-as-error to a new bullet. * src/complain.c (complains): Refactor, change the prefix of warnings that are treated as errors. (warnings_print_categories): Support for [-Werror=CATEGORY] display * src/getargc.c (getargs): -y implies -Werror=yacc * tests/input.at: Update expected --yacc output for coherence. 2012-09-27 Theophile Ranquet errors: introduce the -Werror=CATEGORY option This new option is a lot more flexible than the previous one. Its details will be discussed in the NEWS and info file, in a forthcoming change. If no category is specified (ie: used as simply "-Werror"), the functionality is the same as before. * src/complain.c (errors_flag): New variable. (set_warning_issued): Accept warning categories as an argument. * src/complain.h (Wall): Better definition. * src/getargs.c (flags_argmatch): Support for the new format. (usage): Update -Werror to -Werror[=CATEGORY] format. * src/complain.c (errors_flag): New variable. (set_warning_issued): Accept warning categories as an argument. * src/complain.h (Wall): Better definition. * src/getargs.c (flags_argmatch): Support for the new format. (usage): Update -Werror to -Werror=[CATEGORY] format. 2012-09-26 Akim Demaille Merge branch 'maint' * maint: warnings: introduce -Wdeprecated in the usage info errors: prefix the output with "error: " errors: indent "invalid value for %define" context errors: indent "%define var" redefinition context errors: indent "symbol redeclaration" context errors: indent "result type clash" error context 2012-09-26 Theophile Ranquet warnings: introduce -Wdeprecated in the usage info The deprecated warning, introduced some time ago, was not displayed in the usage message. This patch addresses the issue. * src/getargs.c (usage): Insert here. 2012-09-26 Akim Demaille regen 2012-09-26 Akim Demaille Merge remote-tracking branch 'origin/maint' * origin/maint: regen yacc: fix handling of CPP guards when no header is generated gnulib: update 2012-09-26 Theophile Ranquet errors: prefix the output with "error: " This improves readability. This is also what gcc does. * NEWS: Document this change. * src/complain.c (complain_at): Prefix all errors with "error: ". (complain_at_indent, warn_at_indent): Do not prefix the context information of errors, which are basically just indented errors. * tests/conflicts.at, tests/glr-regression.at, tests/input.at, tests/named-refs.at, tests/output.at, tests/push.at, tests/regression.at, tests/skeletons.at: Apply this change. 2012-09-26 Theophile Ranquet errors: indent "invalid value for %define" context This is the continuation of the work on the readability of errors context. For example, what used to be: input.y:1.9-29: invalid value for %define variable 'foo' : 'bar' input.y:1.9-29: accepted value: 'most' is now: input.y:1.9-29: invalid value for %define variable 'foo' : 'bar' input.y:1.9-29: accepted value: 'most' * src/muscle-tab.c (muscle_percent_define_check_values): Use complain_at_indent to output with increased indentation level. * tests/input:at: Apply this change. 2012-09-26 Theophile Ranquet errors: indent "%define var" redefinition context This is the continuation of the work on the readability of errors context. For example, what used to be: input.y:2.9-11: %define variable 'var' redefined input.y:1.9-11: previous definition is now: input.y:2.9-11: %define variable 'var' redefined input.y:1.9-11: previous definition * src/muscle-tab.c (muscle_percent_define_insert): Use complain_at_indent to output with increased indentation level. * tests/input.at: Apply this change. 2012-09-26 Theophile Ranquet errors: indent "symbol redeclaration" context This is the continuation of the work on the readability of errors context. For example, what used to be: input.y:5.10-24: %printer redeclaration for input.y:3.11-25: previous declaration is now: input.y:5.10-24: %printer redeclaration for input.y:3.11-25: previous declaration * NEWS: Document this change. * src/symtab.c (symbol_redeclaration, semantic_type_redeclaration, user_token_number_redeclaration, default_tagged_destructor_set, default_tagless_destructor_set, default_tagged_printer_set, default_tagless_printer_set): Use complain_at_indent to output with increased indentation level. * tests/input.at: Apply this change. 2012-09-26 Theophile Ranquet errors: indent "result type clash" error context This used to be the format of the error report: input.y:6.5-10: result type clash on merge function 'merge': [...] input.y:2.4-9: previous declaration In order to distinguish the actual error from the context provided, we rather this new output: input.y:6.5-10: result type clash on merge function 'merge': [...] input.y:2.4-9: previous declaration Another patch will introduce an "error: " prefix to all non-indented lines, giving yet better readability to the reports. * src/complain.h (SUB_INDENT): Move to here. * src/reader.c (record_merge_function_type): Use complain_at_indent to output with increased indentation level. * src/scan-code.l (SUB_INDENT): Remove from here. * tests/glr-regression.at: Apply this change. 2012-09-25 Akim Demaille warnings: use the regular interface for s/r and r/r conflicts The current routines used to display s/r and r/r conflicts are both inconvenient from the programmer point of view (they do not use the warning infrastructure) and for the user (the messages are rather terse, not necessarily pleasant to read, and because they don't use the same routines, they look different). It was due to the belief (dating back to the initial checked-in version of Bison) that, at some point, POSIX Yacc mandated the format for these messages. Today, the Open Group's manual page for Yacc, , explicitly states that the format of these messages is unspecified. See commit be7280480c175bed203883f524c7dcd6cf37c13d and . For a discussion on the chosen warning format, see http://lists.gnu.org/archive/html/bison-patches/2012-09/msg00039.html In an effort to factor the handling of errors and warnings, use the Bison warning routines to report these messages. * src/conflicts.c (conflicts_print): Rewrite with clearer sections about S/R and then R/R conflicts. (conflict_report): Remove, inlined in its sole caller... (conflicts_output): here. * tests/conflicts.at, tests/existing.at, tests/glr-regression.at, * tests/reduce.at, tests/regression.at: Adjust the expected results. * NEWS: Update. 2012-09-25 Akim Demaille regen 2012-09-25 Akim Demaille yacc: fix handling of CPP guards when no header is generated When no header was to be generated, Bison would issue: /* In a future release of Bison, this section will be replaced by #include "". */ #ifndef YY_ # define YY_ It now properly generates nothing. * data/c.m4 (b4_cpp_guard_open, b4_cpp_guard_close): Issue nothing when the file name is empty. * data/yacc.c: Do not generate the above comment when there is no header to generate. * NEWS: Update. 2012-09-25 Akim Demaille gnulib: update 2012-09-21 Akim Demaille conflicts: refactor the counting routines * src/conflicts.c (count_sr_conflicts, count_rr_conflicts): Rename as... (count_sr_conflicts, count_rr_conflicts): these. Use size_t for counts. (count_sr_conflicts, count_rr_conflicts): New. Use them. 2012-09-21 Akim Demaille %expect-rr is for GLR only * src/conflicts.c (conflicts_print): Complain about %expect-rr if not in GLR mode, regardless of the number of reduce/reduce conflicts. * tests/conflicts.at (%expect-rr non GLR): New test. * NEWS: Update. 2012-09-21 Akim Demaille TODO: lalr1.cc master vs maint * TODO: here. 2012-09-21 Akim Demaille c++: coding style fixes * data/lalr1.cc, tests/c++.at: Formatting changes. 2012-09-21 Akim Demaille Revert "introduced a GCC-like -Werror=type" This reverts commit 981c53e257f1974854edc4f6ad0e88c7f18e2bea. 2012-09-21 Akim Demaille Revert "made previous commit less hairy" This reverts commit fd01e1d05ea3e627033d148b5400b99a18ac7ba3. 2012-09-21 Akim Demaille Revert "new Werror report format fixed in a test" This reverts commit 41511178a71ebaf81bde7ffc682c575537af45cb. 2012-09-21 Theophile Ranquet new Werror report format fixed in a test * tests/input.at : replaced [-Wyacc] with [-Werror=yacc] todo: fix the other failed test of the suite, tests/conflicts.at:1554 2012-09-21 Theophile Ranquet made previous commit less hairy * src/getargs.c : here 2012-09-20 Theophile Ranquet introduced a GCC-like -Werror=type * src/complain.h : errors_flag variable * src/complain.c : actual stuff happens here * src/conflits.c : differentiated SR and RR conflicts * src/getargs.c : flags_argmatch recognizes the new -Werror format 2012-09-04 Akim Demaille regen 2012-09-04 Akim Demaille Merge remote-tracking branch 'origin/maint' * origin/maint: maint: remove useless file update files to ignore remove useless include use locale-indep. c_is* functions for parsing, not isspace, isprint etc gnulib: update --help: include a place to report translation issues tests: style changes tests: fix push-pull test yacc.c: style changes 2012-09-04 Akim Demaille maint: remove useless file * externals/bootstrap.cfg: Remove. This file was used by a specific build system. It was added to the master repository by accident. 2012-09-04 Akim Demaille update files to ignore * doc/.gitignore: Don't ignore split info files as we don't split our info file. See . 2012-09-04 Akim Demaille remove useless include * src/system.h: Don't include sys/types.h. Reported by Eric Blake, . (FUNCTION_PRINT): Remove, unused. 2012-09-03 Jim Meyering use locale-indep. c_is* functions for parsing, not isspace, isprint etc * src/parse-gram.y: Include "c-ctype.h". (add_param): Parse with c_isspace, not isspace. * src/parse-gram.c: Likewise. * src/scan-gram.l: Include c-ctype.h, not ctype.h. (SC_ESCAPED_STRING,SC_ESCAPED_CHARACTER): Use c_isspace and c_isprint, not ctype.h's locale-dependent functions. 2012-09-03 Akim Demaille gnulib: update 2012-09-03 Akim Demaille --help: include a place to report translation issues http://lists.gnu.org/archive/html/bug-bison/2012-08/msg00007.html shows that it is useful to help users report translation issues. While at it, include other informative bits that the coreutils shows. * src/getargs.c (usage): Report more URLs where the user can refer to. Mostly copied/pasted from coreutils' emit_ancillary_info function. 2012-09-03 Akim Demaille news: style changes * NEWS: Minor improvements. 2012-09-03 Akim Demaille use -Wdeprecated for obsolete %define variable names * src/muscle-tab.c (muscle_percent_variable_update): Here. * tests/input.at (%define backward compatibility): Update expectations. 2012-09-03 Akim Demaille introduce -Wdeprecated GCC seems to be using "deprecated" consistently over "obsoleted", so use -Wdeprecated rather than -Wobsolete. * src/complain.h (warnings): Add Wdeprecated. * src/complain.c (warnings_print_categories): Adjust. * src/getargs.c: Likewise. * doc/bison.texi: Document it. * src/scan-code.l: Use this category for the trailing ';' support. * tests/actions.at: Adjust expected output. 2012-09-03 Akim Demaille undefined but unused is a warning * src/symtab.c (symbol_check_defined): Undeclared symbols are only a warning. * tests/input.at (Undeclared symbols used for a printer or destructor): Rename as... (Undefined symbols): this, and check this case. * NEWS: Doc it. 2012-09-03 Akim Demaille glr.cc: %defines is no longer mandatory * data/glr.cc: No longer require %defines. When it is not given, define the position and location classes instead of including their headers. (b4_shared_declarations): Use the original parse-params. * data/glr.c (b4_shared_declarations): Define only if undefined. * tests/actions.at, tests/calc.at: No longer force the use of %defines for glr.cc. * NEWS: Doc it. 2012-09-03 Akim Demaille todo: check push parsers 2012-09-03 Akim Demaille style: remove useless C++ provisio * src/complain.h: here. 2012-09-03 Akim Demaille parser: style changes * src/parse-gram.y: Avoid deprecated directives. 2012-09-03 Akim Demaille doc: address a fixme * doc/bison.texi (Calc++ Parser): Add a cross-reference. 2012-09-03 Akim Demaille style changes * data/glr.cc, tests/actions.at: Fix comments. * src/symtab.h, src/symtab.c: Fix indentation/comments. * src/symlist.c: Fix indentation. 2012-08-31 Akim Demaille tests: style changes * tests/torture.at (AT_DATA_STACK_TORTURE): M4 style changes to improve readability. Fix an assertion which, because of a <= instead of ==, did not check new_status as visibly meant. (get_args): New. 2012-08-31 Akim Demaille tests: fix push-pull test * tests/torture.at: %push-pull-parser is no longer supported. 2012-08-31 Akim Demaille yacc.c: style changes * data/yacc.c: (yytoken): Define with initial value. 2012-08-12 Akim Demaille refactoring: define variables with a value * src/muscle-tab.c: Where possible, fuse definition and initial assignment. 2012-08-12 Akim Demaille minor refactoring: shorten variable names * src/scan-skel.l (at_directive_argc, at_directive_argv) (AT_DIRECTIVE_ARGC_MAX): Rename as... (argc, argv, ARGC_MAX): these, as there is no possible confusion. (flags): New. (QPUTS): Remove, inline its only use. 2012-08-12 Akim Demaille obstacks: simplifications * src/system.h (obstack_finish0): New. Use it to simplify several uses. * src/muscle-tab.h (MUSCLE_INSERTF): New. * src/muscle-tab.c: Use obstack_printf where simpler. 2012-08-12 Akim Demaille minor refactoring in user code scanning * src/scan-code.l (show_sub_message, show_sub_messages): Instead of a Boolean, take a "warnings" argument. Avoid storing printf-like format strings in a variable, so that GCC can check them. 2012-08-12 Akim Demaille minor refactoring in user code scanning * src/scan-code.l (show_sub_message): New, extracted from... (show_sub_messages): here. 2012-08-03 Akim Demaille tests: strengthen the trailing spaces check * tests/calc.at: here. * data/glr.c: Fix accordingly. 2012-08-03 Akim Demaille regen 2012-08-03 Akim Demaille Merge branch 'maint' * origin/maint: maint: post-release administrivia version 2.6.2 NEWS: update. yacc: remove trailing end of line at end of file thanks: fix a contributor name gnulib: update tests: synch line -> syncline, for consistency tests: synclines: style changes tests: synclines: fix perl invocation regen c++: trailing end-of-lines in %parse-param tests: simplify 2012-08-03 Akim Demaille regen 2012-08-03 Akim Demaille remove support for lint Basically revert commit 12ce2df60d16961eaa03a5aa009eeaa645e4e1cb. http://lists.gnu.org/archive/html/bison-patches/2012-08/msg00004.html * data/c.m4, data/glr.c, data/yacc.c (YYID): Remove. No longer use ARGSUSED. * src/getargs.c: Restore simpler inclusion of getopt.h (anyway, since then we now use gnulib which certainly protects us from such issues). 2012-08-03 Akim Demaille skeletons: renamings after knr removal * data/c.m4 (b4_yydestruct_generate, b4_yy_symbol_print_generate): Rename as... (b4_yydestruct_define, b4_yy_symbol_print_define): these, for consistency. * data/glr.c, data/glr.cc, data/yacc.c: Adjust. 2012-08-03 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2012-08-03 Akim Demaille version 2.6.2 * NEWS: Record release date. 2012-08-02 Akim Demaille c++: fix a comment * data/c++.m4: Be sure to attach a ';' to its declaration, otherwise it appears in the preceding comment. 2012-08-02 Akim Demaille skeletons: simplify after knr removal * data/c.m4 (b4_yydestruct_generate, b4_yy_symbol_print_generate): They no longer need an argument, it has a single possible value. * data/glr.c, data/yacc.c: Adjust. 2012-08-02 Akim Demaille skeletons: renamings after knr removal * data/c.m4 (b4_c_comment_, b4_c_args, b4_c_function_def) (b4_c_function_decl, b4_c_formals, b4_c_call, b4_c_arg): Rename as... (b4_comment, b4_args, b4_function_define, b4_function_declare, b4_formals, b4_function_call, b4_arg): these. * data/glr.c, data/glr.cc, data/lalr1.cc, data/yacc.c: Adjust. 2012-08-02 Akim Demaille skeletons: b4_args -> b4_join to prepare forthcoming changes * data/bison.m4 (b4_args, _b4_args): Rename as... (b4_join, _b4_join): these. * data/c++.m4, data/lalr1.cc, data/variant.hh: Adjust. 2012-08-02 Akim Demaille regen 2012-08-02 Akim Demaille YYPARSE_PARAM: drop support * data/yacc.c: No longer support it. * doc/bison.texi, tests/headers.at: Adjust. * NEWS: Document. 2012-08-02 Akim Demaille skeletons: remove K&R C support * data/c.m4 (b4_c_modern, b4_c_knr_formal_names, b4_c_knr_formal_decls) (b4_c_knr_formal_decl, b4_c_formal_names, b4_c_formal_decls) (b4_c_formal_decl): Remove. (b4_c_ansi_formal_names, b4_c_ansi_formal_decls, b4_c_ansi_formal_decl): Rename as... (b4_c_formal_names, b4_c_formal_decls, b4_c_formal_decl): these. * data/glr.c, data/glr.cc, data/yacc.c: Adjust. 2012-08-02 Akim Demaille NEWS: update. * NEWS: Catch up with the other changes from 2.6.1. 2012-08-02 Akim Demaille yacc: remove trailing end of line at end of file There are still spurious spaces at the end of some lines. But this is addressed in the master branch, and I am reluctant to try to backport this. * data/yacc.c, data/glr.c, data/lalr1.cc, data/glr.cc: here. * tests/calc.at (AT_CHECK_SPACES): New. Use it. Be sure not to introduce trailing empty lines in the *.y files. * NEWS: Doc it. * cfg.mk (syntax-check): Remove the exception. 2012-08-02 Akim Demaille thanks: fix a contributor name * THANKS: On his request. 2012-08-01 Akim Demaille gnulib: update 2012-08-01 Akim Demaille tests: synch line -> syncline, for consistency * tests/synclines.at: Do it, as "syncline" is used consistently everywhere else in Bison. 2012-08-01 Akim Demaille tests: synclines: style changes * tests/synclines.at (AT_TEST_SYNCLINE): Rename as... (AT_TEST): this. Use pushdef/popdef. Formatting changes. Use '+' instead of '*' where appropriate. 2012-08-01 Akim Demaille tests: synclines: fix perl invocation Reported by Summum Bonum. * tests/synclines.at: Fix Perl invocation: its -f is not like sed's. 2012-08-01 Akim Demaille regen 2012-08-01 Akim Demaille c++: trailing end-of-lines in %parse-param * src/parse-gram.y (add_param): No only skip ' ' and '\t', skip all leading and trailing spaces. * tests/regression.at (Lex and parse params): Check it. * NEWS: Document it. 2012-08-01 Akim Demaille tests: simplify * tests/regression.at: Remove useless compilations: AT_FULL_COMPILE includes the compilation by bison. 2012-07-31 Akim Demaille todo: more items * TODO: $ in the epilogue, and obstack_copy. 2012-07-31 Akim Demaille Merge branch 'maint' * maint: use obstack_printf scanner: restore a missing start condition gnulib: update maint: post-release administrivia version 2.6.1 gnulib: update maint: fix some syntax-check issues tests: do not depend on __cplusplus to decide for C++ or C output 2012-07-31 Akim Demaille tests: comment changes * tests/actions.at, tests/input.at: here. 2012-07-31 Akim Demaille tests: really check the set of generated files * tests/output.at (AT_CHECK_OUTPUT): It used to check that the expected files are indeed generated, but it did not check that there are no additional ones. Do that, and adjust expectations (in particular alphabetical order). 2012-07-31 Akim Demaille lalr1.cc: do not create stack.hh without %defines * data/stack.hh (b4_stack_define): New. * data/lalr1.cc: Use it when %defines is not passed. * tests/output.at: Adjust expected output. 2012-07-31 Akim Demaille lalr1.cc: location.hh and position.hh are not generated without %defines * data/location.cc (b4_position_define, b4_location_define): New. (location.hh, position.hh): Generate only if %defines. * data/lalr1.cc: therefore, define these classes when locations are needed, but headers are not generated. * tests/output.at: Check that these files are not generated. * NEWS: Document. 2012-07-31 Akim Demaille lalr1.cc: no longer require %defines. * data/lalr1.cc: Generate the parser header only when %defines is passed. * tests/calc.at: Check it. 2012-07-31 Akim Demaille skeletons: style changes * data/glr.c, data/lalr1.cc: Use more consistent comments, and YY_NULL declaration. 2012-07-31 Akim Demaille glr.cc, lalr1.cc: define b4_shared_declarations * data/glr.cc, data/lalr1.cc: here. The name is no longer right, but at least it is consistent with the other skeletons. 2012-07-31 Akim Demaille glr.cc: no longer require location support * data/glr.cc: Use b4_locations_if where appropriate. * data/lalr1.cc: M4 quotation changes to highlight code duplication with glr.cc. * tests/calc.at: Check glr.cc with and without %location. While at it, fuse multiple %parse-params into one. * tests/actions.at: Simplify. * NEWS: Doc this. Some other wording changes. 2012-07-31 Akim Demaille use obstack_printf This is not just nicer, it is also much safer, since we were using sprintf... * bootstrap.conf: Require it. * src/system.h (obstack_fgrow1, obstack_fgrow2, obstack_fgrow3) (obstack_fgrow4): Remove. Adjust dependencies. 2012-07-30 Akim Demaille scanner: restore a missing start condition $ flex src/scan-skel.l src/scan-skel.l:145: multiple <> rules for start condition SC_AT_DIRECTIVE_ARGS src/scan-skel.l:145: multiple <> rules for start condition SC_AT_DIRECTIVE_SKIP_WS This is warning, and it seems there are no means to make it an error. * src/scan-skel.l: Restore the start-condition INITIAL for an <> clause. 2012-07-30 Akim Demaille gnulib: update 2012-07-30 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2012-07-30 Akim Demaille version 2.6.1 * NEWS: Record release date. 2012-07-30 Akim Demaille gnulib: update 2012-07-27 Akim Demaille maint: fix some syntax-check issues * cfg.mk: Nuke the following warnings which are confused by our text reports (that state that the error token is number 256). prohibit_magic_number_exit ../../doc/bison.texi:8170:error (256) ../../tests/conflicts.at:570:error (256) ../../tests/conflicts.at:673:error (256) ../../tests/conflicts.at:811:error (256) ../../tests/conflicts.at:1154:error (256) ../../tests/regression.at:281:error (256) ../../tests/regression.at:582:error (256) maint.mk: use EXIT_* values rather than magic number 2012-07-27 Akim Demaille regen 2012-07-27 Akim Demaille tests: do not depend on __cplusplus to decide for C++ or C output Since we do support compiling C code with a C++ compiler. * tests/actions.at (Qualified $$ in actions): Use AT_SKEL_CC_IF. 2012-07-27 Akim Demaille Merge remote-tracking branch 'origin/maint' * origin/maint: (29 commits) regen synclines: remove spurious empty line also support $$ in the %initial-action skeletons: b4_dollar_pushdef and popdef to simpify complex definitions regen printer/destructor: translate only once factor the handling of m4 escaping news: schedule the removal of the ";" hack style changes in the scanners regen support $$ in printers and destructors scan-code: factor the handling of the type in $$ muscles: fix another occurrence of unescaped type name glr.cc: fix the handling of yydebug gnulib: update formatting changes tests: fix an assertion tests: adjust to GCC 4.8, which displays caret errors be sure to properly escape type names obstack_quote: escape and quote for M4 muscles: shuffle responsabilities muscles: make private functions static muscles: rename private functions/macros obstack_escape: escape M4 characters remove dead macro maint: style changes doc: avoid problems with case insensitive file systems configure: fix botched quoting news: fix typo. 2012-07-27 Akim Demaille regen 2012-07-27 Akim Demaille synclines: remove spurious empty line * data/bison.m4 (b4_syncline): Do not start with an empty line. 2012-07-27 Akim Demaille also support $$ in the %initial-action scan-code.l is already passing argument to b4_dollar_dollar for the initial acton, but its definition (of b4_dollar_dollar) does not use this argument. Generalize this definition, and use it for the %initial-action too. * data/c.m4 (b4_dollar_dollar_, b4_dollar_pushdef, b4_dollar_popdef): Instead of expecting a pointer, require a value, and use ".". Since they are now generic enough, move to... * data/c-like.m4: this new file. * data/c.m4, data/java.m4: Load it. * data/glr.c, data/lalr1.cc, data/lalr1.java, data/yacc.c: Use b4_dollar_pushdef for the %initial-action. * tests/actions.at: Check that. * data/Makefile.am: Adjust. * NEWS, doc/bison.texi: Document. 2012-07-27 Akim Demaille skeletons: b4_dollar_pushdef and popdef to simpify complex definitions M4 is really making it uselessly hard to define macros that define macros. * data/c.m4 (b4_dollar_pushdef, b4_dollar_popdef): New. Use it. 2012-07-27 Akim Demaille regen 2012-07-27 Akim Demaille printer/destructor: translate only once Currently "%printer {...} a b c d e f" translates the {...} six times. Not only is this bad for time and space, it also issues six times the same warnings. * src/symlist.h, src/symlist.c (symbol_list_destructor_set) (symbol_list_printer_set): Take the action as code_props instead of const char *. * src/parse-gram.y: Translate these actions here. * src/scan-code.h: Comment change. * tests/input.at: Check that warnings are issued only once. 2012-07-27 Akim Demaille factor the handling of m4 escaping The conversion from @ to @@ and so forth is coded is too many different places. Factor, a bit. * src/scan-code.l: Instead of duplicating the logic of obstack_escape, use it. It sure is less efficient, but the cost is negligible. This allows to factor rules that are alike. And to factor some start-condition clauses. * tests/input.at (Stray $ or @): New. * NEWS: Document it. 2012-07-27 Akim Demaille news: schedule the removal of the ";" hack scan-code.l is significantly more complex because of this. * NEWS: Doc it. 2012-07-27 Akim Demaille style changes in the scanners * src/scan-code.l, src/scan-skel.l: Use a more traditional indentation style for start-conditions. Prefer "continue" to a comment, for empty actions. Strip useless {}. Remove useless start-condition clauses. 2012-07-26 Akim Demaille regen 2012-07-26 Akim Demaille support $$ in printers and destructors * src/scan-code.l (SC_SYMBOL_ACTION): Accept $$, not just $$. * data/c.m4 (b4_dollar_dollar_): New. (b4_symbol_actions): Let b4_dollar_dollar use b4_dollar_dollar_. * NEWS, doc/bison.texi: Document it. * tests/actions.at: Check this for C and C++. 2012-07-26 Akim Demaille scan-code: factor the handling of the type in $$ * src/scan-code.l (fetch_type_name): New. (handle_action_dollar): Use it. (gt_ptr): Remove, useless. 2012-07-26 Akim Demaille muscles: fix another occurrence of unescaped type name * src/output.c (quoted_output): Split into... (quoted_output, string_output): these. Use the former when outputting a type_name. * tests/input.at: Check this case. * src/symtab.h: Comment changes. 2012-07-26 Akim Demaille glr.cc: fix the handling of yydebug * data/glr.cc (yydebug_): Remove, unused. (set_debug_level, debug_level): Work on yydebug instead. * doc/bison.texi, NEWS: Document this. 2012-07-26 Akim Demaille gnulib: update 2012-07-26 Akim Demaille formatting changes * src/symtab.h: here. 2012-07-26 Akim Demaille tests: fix an assertion * tests/local.at (AT_YYLEX_DEFINE): Be sure to check the array against its length, not its size in bytes. 2012-07-26 Akim Demaille tests: adjust to GCC 4.8, which displays caret errors With GCC 4.8, the tests on synclines are skipped. Transform input.y:1:2: error: #error "1" #error "1" ^ into input.y:1: #error "1" * tests/synclines.at (AT_SYNCLINES_COMPILE): Do it, using Perl instead of sed. 2012-07-26 Akim Demaille be sure to properly escape type names * src/scan-code.l: Use obstack_quote when passing type_name to m4. * tests/input.at (Code injection): New. * NEWS: Document it. Thanks to Paul Eggert for the wording. 2012-07-26 Akim Demaille obstack_quote: escape and quote for M4 * src/system.h (obstack_quote): New. * src/muscle-tab.c: Use it instead of obstack_escape where applicable. * src/scan-code.l: Since obstack_quote supports NULL, leave type_name as NULL instead of defaulting to "". 2012-07-26 Akim Demaille muscles: shuffle responsabilities * src/muscle-tab.c (muscle_boundary_grow): Be in charge of quotation, instead of leaving this to the caller. 2012-07-26 Akim Demaille muscles: make private functions static * src/muscle-tab.h, src/muscle-tab.c (muscle_boundary_grow) (muscle_location_grow): Now static. 2012-07-26 Akim Demaille muscles: rename private functions/macros * src/muscle-tab.c (MUSCLE_COMMON_DECODE, muscle_string_decode) (muscle_location_decode): Not related to muscles, rename as... (COMMON_DECODE, string_decode, location_decode): these. 2012-07-26 Akim Demaille obstack_escape: escape M4 characters * src/muscle-tab.h (MUSCLE_OBSTACK_SGROW): This is not related to muscles, so move to, and rename as... * src/system.h (obstack_escape): this. Adjust dependencies. 2012-07-26 Akim Demaille remove dead macro * src/system.h (DEFAULT_TMPDIR): Remove, unused. 2012-07-26 Akim Demaille maint: style changes * src/scan-code.l: Remove useless braces. Formatting changes. Prefer NULL to 0. * src/muscle-tab.c, src/system.h: Formatting changes. 2012-07-24 Akim Demaille doc: avoid problems with case insensitive file systems makeinfo --html generates index.html, and the node "Index" will result in Index.html. On case insensitive file systems, such as on Mac OS X by default, this results in a single, invalid, file (Texinfo 4.13). See http://lists.gnu.org/archive/html/bug-texinfo/2012-07/msg00032.html * doc/bison.texi (Index): Rename as... (Index of Terms): this. 2012-07-24 Stefano Lattarini (tiny change) configure: fix botched quoting * configure.ac: In the AC_SUBST call on 'VALGRIND_PREBISON'. Without this change, when running ./configure, I see: ... checking for valgrind... valgrind ./configure: line 35221: -q: command not found checking for Java compiler... gcj -C -fsource=1.3 -ftarget=1.4 ... 2012-07-24 Akim Demaille regen 2012-07-24 Akim Demaille yystype, yyltype: remove. * data/c.m4: here. * NEWS: Doc it. 2012-07-22 Akim Demaille regen 2012-07-22 Akim Demaille YYFAIL: remove. * data/lalr1.java, data/yacc.c, src/scan-code.l: Remove YYFAIL support. * NEWS, TODO: Update. 2012-07-22 Akim Demaille todo: update. * TODO: obsolete items. 2012-07-22 Akim Demaille regen. 2012-07-22 Akim Demaille space changes. * data/bison.m4 (b4_symbol_action): Remove spurious eol in the output. 2012-07-22 Akim Demaille parser: fix %printer usage. * src/parse-gram.y: Instead of stderr, using yyo. 2012-07-22 Akim Demaille parser: factor the handling of code_props * src/parse-gram.y: Now that %printer and %destructor are treated equally, let... (code_props_type): handle them. 2012-07-22 Akim Demaille parser: factor handling of type tags * src/parse-gram.y: Now that <*> and <> are processed like regular tags, let... (tag): handle them. 2012-07-22 Akim Demaille regen. 2012-07-22 Akim Demaille simplify the handling of <> and <*>'s code_props. Currently they are treated in separated variables, contrary to other code_props. This duplicates code (and messages for translators) uselessly, as demonstrated by the fact that thanks to this patch, now useless <*> and <> code_props are reported like the others. * src/parse-gram.y (generic_symlist_item): Treat "<*>" and "<>" as regular type tags. * src/symlist.h, src/symlist.c (symbol_list_default_tagged_new) (symbol_list_default_tagless_new,SYMLIST_DEFAULT_TAGGED) (SYMLIST_DEFAULT_TAGLESS): Remove. * src/symtab.h, src/symtab.c (default_tagged_code_props) (default_tagless_code_props, default_tagged_code_props_set) (default_tagless_code_props_set): Remove. (symbol_code_props_get): Default to <*> or <>'s code_props. * tests/actions.at: Complete expected errors: there are new warnings. * tests/input.at: Likewise. (Useless printers or destructors): Extend. 2012-07-22 Akim Demaille allow modification on retrieved code_props. The logic to compute the %printer or %destructor to used (i.e., a code_props) is implemented twice: one, of course, in symbol_code_props_get, and another time in symbol_check_defined to record the fact that a code_props is used (so that we can reported unused ones). Let the former use the latter. I would probably use "mutable" in C++ and keep these guys const, but this is C. And casting away constness triggers warnings. * src/scan-code.h, src/scan-code.l (code_props_none): Is not const. * src/symtab.h, src/symtab.c (symbol_code_props_get): The symbol is not const. (symbol_check_defined): Use it. 2012-07-22 Akim Demaille maint: regen. 2012-07-22 Akim Demaille maint: fix bison's own header guards. Because I'm using a VPATH build with an absolute srcdir, I have GRAM__USERS_AKIM_SRC_GNU_BISON_SRC_PARSE_GRAM_H. Before, I was using a relative srcdir, and had GRAM_______SRC_PARSE_GRAM_H (coming from ../../). Let it be GRAM_SRC_PARSE_GRAM_H. * tests/bison.in: Do not depend on the value of $top_srcdir for Bison itself. If we were to use relative paths from .c to .y, we would not have this problem. 2012-07-22 Akim Demaille maint: add missing const. * src/symtab.h, src/symtab.c (symbol_print): here. 2012-07-22 Akim Demaille style changes. * src/parse-gram.y, src/symtab.c: Space changes. * src/symtab.h: Comment changes. 2012-07-22 Akim Demaille autoconf: update. * submodules/autoconf: here. No significant changes for our use of m4sugar.m4. 2012-07-22 Akim Demaille output: no longer use b4_tokens. * data/glr.c, data/glr.cc, data/lalr1.cc, data/lalr1.java, data/yacc.c: Since the previous commit, b4_tokens_define and the like no longer need b4_tokens. * src/output.c (token_definitions_output): Remove. 2012-07-22 Akim Demaille output: use the token list to define the yytokentype There are currently two systems used to pass information about tokens to m4: the original one, and another, which is used for instance for printers and destructors, variants etc. Move to using only the latter. * data/bison.m4 (b4_symbol_map, b4_token_visible_if) (b4_token_has_definition, b4_any_token_visible_if, b4_token_format): New. * data/c++.m4, data/c.m4, data/glr.c, data/java.m4: Adjust to use them. 2012-07-20 Akim Demaille tests: fix VPATH issue * examples/test: With an absolute VPATH build, "../" is incorrect. 2012-07-20 Akim Demaille news: fix typo. * NEWS: here. Reported by Ben Pfaff. 2012-07-19 Akim Demaille Merge remote-tracking branch 'origin/maint' * origin/maint: maint: update gnu-web-doc-update. maint: post-release administrivia version 2.6 maint: prepare for release 2.6 maint: post-release administrivia version 2.5.91 maint: prepare NEWS. maint: fix spaces. tests: adjust to case where the C compiler is actually a C++ compiler tests: fix dependencies doc: fix Texinfo command maint: Valgrind on OS X. tests: be sure that backups are safe. maint: dead comment. tests: refactor for legibility. tests: refactor the bison invocations. maint: fix syntax-check ignore patterns. gnulib: update gnulib: update. gnulib: update 2012-07-19 Akim Demaille maint: update gnu-web-doc-update. * gnulib: here. 2012-07-19 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2012-07-19 Akim Demaille version 2.6 * NEWS: Record release date. 2012-07-19 Akim Demaille maint: prepare for release 2.6 * NEWS: here. 2012-07-18 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2012-07-18 Akim Demaille version 2.5.91 * NEWS: Record release date. 2012-07-18 Akim Demaille maint: prepare NEWS. 2012-07-18 Akim Demaille maint: fix spaces. * build-aux/Makefile.am: here. 2012-07-18 Akim Demaille tests: adjust to case where the C compiler is actually a C++ compiler * tests/atlocal.in (CC_IS_CXX): New. * tests/headers.at (Several parsers): Use it. 2012-07-18 Akim Demaille tests: fix dependencies * tests/Makefile.am: we need atconfig and atlocal to be up to date when calling testsuite. 2012-07-18 Akim Demaille doc: fix Texinfo command * doc/bison.texi: In parens, use @pxref. 2012-07-18 Akim Demaille maint: Valgrind on OS X. * configure.ac (VALGRIND_PREBISON): New. * tests/Makefile.am (maintainer-check-valgrind): Use it. * etc/darwin11.4.0.supp: New. * configure.ac, etc/Makefile.am: Use it. * configure.ac: Disable Valgrind on Mac OS X. * README-hacking: Explain why. 2012-07-17 Akim Demaille tests: be sure that backups are safe. * tests/local.at (at_save_special_files): here. 2012-07-17 Akim Demaille maint: dead comment. * etc/README: here. 2012-07-17 Akim Demaille tests: refactor for legibility. * tests/local.at (AT_BISON_CHECK_WARNINGS, AT_BISON_CHECK_WARNINGS_): New. 2012-07-17 Akim Demaille tests: refactor the bison invocations. * tests/local.at (m4_null_if, AT_BISON_CHECK_): New. 2012-07-17 Akim Demaille maint: fix syntax-check ignore patterns. * cfg.mk: here. 2012-07-17 Akim Demaille gnulib: update 2012-07-16 Akim Demaille gnulib: update. * gnulib: Update so that gitlog-to-changelog support --srcdir. * Makefile.am: Use it. 2012-07-10 Akim Demaille gnulib: update * bootstrap, build-aux/.gitignore, gnulib, m4/.gitignore: update. 2012-07-06 Akim Demaille maint: minor fixes * NEWS: restore missing entry. * cfg.mk: Adjust to *.texinfo -> *.texi. * src/symtab.c: Spaces fixes. 2012-07-06 Akim Demaille tests: address g++-4.8 warnings. list.yy: In function 'yy::parser::symbol_type yylex()': list.yy:107:29: error: typedef 'token' locally defined but not used [-Werror=unused-local-typedefs] typedef yy::parser::token token; ^ * tests/c++.at (AT_CHECK_VARIANTS): here. 2012-07-06 Akim Demaille Merge remote-tracking branch 'origin/maint' * origin/maint: maint: update release instructions maint: post-release administrivia version 2.5.90 build: fix gen-ChangeLog call. gnulib: update. tests: fix SKIP_IF for Java. api.prefix: incompatible with %name-prefix. api.prefix: strengthen the tests and fix push-parsers. skeletons: style changes. NEWS: minor changes. api.prefix: improve the documentation for YYDEBUG. gnulib: update. 2012-07-06 Akim Demaille maint: update release instructions * README-hacking: here. 2012-07-05 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2012-07-05 Akim Demaille version 2.5.90 * NEWS: Record release date. 2012-07-05 Akim Demaille build: fix gen-ChangeLog call. * Makefile.am: Be sure to catch errors, and fix option name 2012-07-05 Akim Demaille gnulib: update. * gnulib/build-aux/do-release-commit-and-tag: Fix. 2012-07-05 Akim Demaille tests: fix SKIP_IF for Java. * tests/local.at (AT_JAVA_COMPILE): here. 2012-07-05 Akim Demaille api.prefix: incompatible with %name-prefix. * data/bison.m4: Make it incompatible. * tests/input.at: Check that it is. 2012-07-05 Akim Demaille api.prefix: strengthen the tests and fix push-parsers. * tests/calc.at: Check api.prefix in addition to %name-prefix. * tests/headers.at: Check push parsers and pure interface. * tests/local.at: Use YYLTYPE renamed. * data/yacc.c (b4_declare_yyparse_push_): Handle api.prefix. * doc/bison.texi: Style changes. 2012-07-05 Akim Demaille skeletons: style changes. * data/bison.m4: Define default values after having defined the support macros. Kill a dead comment. 2012-07-05 Akim Demaille NEWS: minor changes. * NEWS: style changes. 2012-07-05 Akim Demaille api.prefix: improve the documentation for YYDEBUG. * doc/bison.texi: Explain how api.prefix is applied to YYDEBUG. 2012-07-05 Akim Demaille gnulib: update. * bootstrap, gnulib: Update. * cfg.mk (syntax-check): Don't check "error" usage in bison.texi. 2012-07-04 Akim Demaille Merge remote-tracking branch 'origin/maint' * origin/maint: tests: headers.at: strengthen. glr.cc: do not override C++ definitions by C macros. YYLLOC_DEFAULT: factor, and don't export it in headers. api.prefix: do not use #define to handle YYSTYPE_IS_TRIVIAL etc. tests: portability fixes. c++: fewer #includes in the headers. glr.cc: formatting changes. tests: more logs. api.prefix: also rename YYDEBUG. 2012-07-04 Akim Demaille tests: headers.at: strengthen. * tests/headers.at (Several headers): Be stricter when checking the exported macros. 2012-07-04 Akim Demaille glr.cc: do not override C++ definitions by C macros. * data/glr.c: here. * data/glr.cc: Fix overquotation. * tests/headers.at: Comment changes. 2012-07-04 Akim Demaille YYLLOC_DEFAULT: factor, and don't export it in headers. * data/c++.m4, data/c.m4 (b4_yylloc_default_define): New. * data/glr.c, data/glr.cc, data/lalr1.cc, data/yacc.c: Use it. * data/glr.cc: Do not define YYLLOC_DEFAULT in the header file, but in the implementation one. 2012-07-04 Akim Demaille api.prefix: do not use #define to handle YYSTYPE_IS_TRIVIAL etc. The following mixture is insane: #define YYSTYPE_IS_TRIVIAL PREFIX_STYPE_IS_TRIVIAL #if (defined YYSTYPE_IS_TRIVIAL && YYSTYPE_IS_TRIVIAL) since, of course YYSTYPE_IS_TRIVIAL is defined. Instead we could define YYSTYPE_IS_TRIVIAL as PREFIX_STYPE_IS_TRIVIAL only when the later is defined, but let's avoid stacking CPP on top of M4: rather, use #if (defined PREFIX_STYPE_IS_TRIVIAL && PREFIX_STYPE_IS_TRIVIAL) * data/glr.c, data/yacc.c: Use YYSTYPE_IS_TRIVIAL, YYSTYPE_IS_DECLARED, YYLTYPE_IS_TRIVIAL and YYLTYPE_IS_DECLARED under their api.prefix-renamed name. 2012-07-04 Akim Demaille tests: portability fixes. Reported by Hydra. * tests/headers.at (Several headers): Be sure to include config.h in the files to compile. 2012-07-04 Akim Demaille c++: fewer #includes in the headers. * data/lalr1.cc: Define YY_NULL in the *.cc file, it is not needed in the header. * data/location.cc: iosfwd suffices. 2012-07-04 Akim Demaille glr.cc: formatting changes. * data/glr.cc: here. 2012-07-04 Akim Demaille tests: more logs. * tests/headers.at (Several parsers): Here. 2012-07-04 Akim Demaille api.prefix: also rename YYDEBUG. The testsuite in master has shown weird errors for the "Mulitple Parsers" tests: the caller of p5.parse() received some apparently random value, while tracing p5.parse() showed that the function was consistently returning 0. It happens when mixing several parser headers, some generated without %debug, others with. In particular the C++ parser was generated with %debug, i.e., with: #ifndef YYDEBUG # define YYDEBUG 1 #endif and compiled separatedly. Yet, its header was included after the one of another parser, this time without %debug, i.e., with #ifndef YYDEBUG # define YYDEBUG 0 #endif in its header. As a result, the parser was compiled with YYDEBUG set, but its header was used without. Since the layout of the objects are then completely different, boom. Therefore, do not change the value of YYDEBUG. Rather, use it as a default value for DEBUG. * data/c.m4 (b4_YYDEBUG_define): New. (b4_declare_yydebug): Rename as... (b4_yydebug_declare): this, for consistency. * data/glr.c, data/glr.cc, data/lalr1.cc, data/yacc.c: Use it. * NEWS: Document it. 2012-07-02 Akim Demaille formatting changes. * data/lalr1.cc: here. 2012-07-02 Akim Demaille NEWS: spell fixes. * NEWS: here. Reported by Stefano Lattarini. 2012-07-02 Akim Demaille Merge remote-tracking branch 'origin/maint' * origin/maint: NEWS: spell check. api.prefix. 2012-07-02 Akim Demaille NEWS: spell check. * NEWS: here. 2012-06-29 Akim Demaille api.prefix. * data/c.m4 (b4_api_prefix, b4_api_PREFIX): New. (b4_prefix, b4_union_name, b4_token_enums, b4_declare_yylstype): Use them. * data/glr.c, data/yacc.c, data/glr.cc, data/lalr1.cc: Use them to change the prefix of exported preprocessor symbols. * src/getargs.c (usage): Ditto. * tests/headers.at (Several parsers): New. * tests/local.at (AT_API_PREFIX): New. AT_YYSTYPE, AT_YYLTYPE): Adjust. * doc/bison.texi (Multiple Parsers): Move documentation of %name-prefix to... (Table of Symbols): here. (Multiple Parsers): Document api.prefix. (%define Summary): Point to it. Use @code for variable names. (Bison Options): -p/--name-prefix are obsoleted. * NEWS: Announce api.prefix. 2012-06-29 Victor Santet warnings: display warnings categories * src/complain.c (error_message): Call 'warnings_print_categories'. * src/gram.c (grammar_rules_useless_report): Display itself warning category. * tests/actions.at, tests/conflicts.at, tests/existing.at, tests/input.at, tests/named-refs.at, tests/output.at, tests/reduce.at, tests/regression.at, tests/skeletons.at: Adjust. * NEWS: Document this. 2012-06-29 Victor Santet warnings: be ready to print warnings categories A function to print warnings categories, like -Wyacc, -Wother, etc. * src/complain.h, src/complain.c (print_warning_categories): New function. * src/output.c (ARRAY_CARDINALITY): Move it to file 'src/system.h'. * src/complain.h (enum warnings): New value, 'silent', "complain" must not display the warning type. 2012-06-29 Akim Demaille maint: prepare forthcoming changes * src/gram.c (rule_rhs_print): Do not print new line anymore. (rule_print): Make it static. * src/closure.c, src/derives.c, src/gram.c: Adjust. 2012-06-29 Victor Santet style changes * src/complain.c, src/reader.c, src/reduce.c, src/main.c: Fix indentation. Simplify a bit. 2012-06-28 Akim Demaille regen. 2012-06-28 Victor Santet warnings: factoring: complaints * src/complain.c (error_message): Accept warning categories (an integer) as argument. Location is a 'const location *' instead of 'location *'. (ERROR_MESSAGE): Delete it. * src/complain.c, src/complain.h (complains): New function. (complain, complain_at, complain_at_indent): Generic functions for complaints. Call 'complains'. (warn_at, warn_at_indent, warn, yacc_at, midrule_value_at) (fatal_at, fatal): Delete them. Adjust dependencies. * src/complain.h (enum warnings): New fields 'complaint' and 'fatal'. * bootstrap.conf (XGETTEXT_OPTIONS): Adjust. 2012-06-28 Victor Santet warnings: move them to complain.c. * src/getargs.h, src/getargs.c (warnings, warnings_flags): Move to... * src/complain.h, src/complain.c: Here. 2012-06-28 Victor Santet warnings: rename the categories Forthcoming changes will use the warning categories much more often, so shortening them will improve readability. * src/complain.c, src/complain.h, src/conflicts.c, * src/getargs.c, src/getargs.h, src/gram.c (enum warnings): s/warnings_/W/g. 2012-06-28 Akim Demaille fix merge. * data/bison.m4: Use b4_error_verbose_if after it was defined. 2012-06-28 Akim Demaille Merge remote-tracking branch 'origin/maint' * origin/maint: tests: use the generalized default yylex. tests: AT_YYERROR_DEFINE: prepare for list of ints. skeletons: no longer define YYLSP_NEEDED. c++: do not export YYTOKEN_TABLE and YYERROR_VERBOSE. 2012-06-28 Akim Demaille tests: use the generalized default yylex. * tests/actions.at, tests/glr-regression.at, tests/regression.at: here. 2012-06-28 Akim Demaille tests: AT_YYERROR_DEFINE: prepare for list of ints. * tests/local.at (AT_YYERROR_DEFINE): Don't add quotes, check their presence to detect char/int types. * tests/actions.at, tests/conflicts.at, tests/glr-regression.at, * tests/push.at, tests/regression.at: Adjust. 2012-06-27 Akim Demaille skeletons: no longer define YYLSP_NEEDED. * data/c.m4, data/glr.cc: here. * NEWS, TODO: Adjust. 2012-06-27 Akim Demaille c++: do not export YYTOKEN_TABLE and YYERROR_VERBOSE. * src/output.c (prepare_symbols): Do not define b4_token_table. (prepare): Define b4_token_table_flag. * data/bison.m4 (b4_token_table_if): New. Arm it when error-verbose. * data/glr.c, data/yacc.c (YYTOKEN_TABLE): Remove. Use m4. * data/lalr1.cc: Likewise. (YYERROR_VERBOSE): Remove. * NEWS, doc/bison.texi: Document this. 2012-06-26 Akim Demaille Merge remote-tracking branch 'origin/maint' * origin/maint: maint: use *.texi. 2012-06-26 Akim Demaille maint: use *.texi. This is more consistent with the other packages, and Automake-NG supports only *.texi. * doc/bison.texinfo: Rename as... * doc/bison.texi: this. * doc/Makefile.am, examples/calc++/Makefile.am: Adjust. 2012-06-26 Akim Demaille Merge remote-tracking branch 'origin/maint' * origin/maint: tests: do not output m4 set up. tests: use the generic yyerror function. tests: use assert instead of plain abort. tests: improve the generic yylex implementation. tests: generalize the compilation macros. tests: fix confusion between api.prefix and name-prefix. maint: gitignores. yacc: work around the ylwrap limitation. 2012-06-26 Victor Santet warnings: raise warning for useless printers or destructors * src/scan-code.h (code_props): Add field 'is_used'. (CODE_PROPS_NONE_INIT): Adjust. * src/scan-code.l (code_props_plain_init, code_props_symbol_action_init) (code_props_rule_action_init): Instead of implementing several times the initialization of the code_props structures, use code_props_none_init. * src/symtab.c (symbol_check_defined): If a symbol does not have a destructor (resp. printer) but has a type which has a destructor (resp. printer), then set field 'is_used' to true. (semantic_type_check_defined): If a type has a destructor (resp. printer) but all symbols of this type have already a destructor (resp. printer), then raise a warning. * tests/input.at (Useless printers or destructors): New. 2012-06-26 Akim Demaille tests: do not output m4 set up. * tests/local.at (AT_BISON_OPTION_PUSHDEFS, AT_BISON_OPTION_POPDEFS): Use a diversion to avoid outputting comments etc. Removes 17k lines from testsuite (10% of the number of lines). 2012-06-26 Akim Demaille tests: use the generic yyerror function. * tests/actions.at (_AT_CHECK_PRINTER_AND_DESTRUCTOR): Factor. Use AT_YYERROR_DEFINE. Therefore, instead of using stdout, use and check stderr. * tests/glr-regression.at (Uninitialized location when reporting ambiguity): Use AT_YYERROR_DEFINE. 2012-06-26 Akim Demaille tests: use assert instead of plain abort. * tests/actions.at, tests/calc.at, tests/conflicts.at, * tests/cxx-type.at, tests/glr-regression.at, tests/input.at, * tests/named-refs.at, tests/regression.at, tests/torture.at, * tests/local.at: Prefer assert to abort. 2012-06-26 Akim Demaille tests: improve the generic yylex implementation. * tests/local.at (AT_YYSTYPE, AT_YYLTYPE): New. (AT_YYLEX_FORMALS): Use them. (AT_YYLEX_DEFINE): Be independent of the location implementation. 2012-06-26 Akim Demaille tests: generalize the compilation macros. * tests/local.at (AT_COMPILE, AT_COMPILE_CXX): If OUTPUT ends with ".o", then append the "natural" extension for the input file (.c or .cc). If there is no source, pass -c. * tests/headers.at, tests/input.at, tests/regression.at: Adjust. 2012-06-26 Akim Demaille tests: fix confusion between api.prefix and name-prefix. * tests/local.at (AT_NAME_PREFIX): Take api.prefix into account. (AT_API_PREFIX): Rename as... (AT_API_prefix): this. Do not take %name-prefix into account. Fix misuses. 2012-06-26 Akim Demaille maint: gitignores. 2012-06-25 Victor Santet warnings: useless semantic types * src/symtab.h (symbol_list): Represent semantic types as structure 'semantic_type'. * src/symlist.c (symbol_list_type_new): Allocate this structure. (symbol_list_code_props_set): Set this semantic type's status to used if it was not declared. * src/symtab.c (semantic_types_sorted): New. (semantic_type_new): Set the new semantic type's location appropriately. (symbol_check_defined): If a symbol has a type, then set this type's status to "declared". (semantic_type_check_defined, semantic_type_check_defined_processor): Same as symbol_check_defined and symbol_check_defined_processor, but for semantic types. (symbol_check_defined): Check semantic types usefulness. * src/symtab.h (semantic_type): New fields 'location' and 'status'. * src/symtab.h, src/symtab.c (semantic_type_new) (semantic_type_from_uniqstr, semantic_type_get): Accept a location as a supplementary argument. * tests/input.at (Unassociated types used for printer of destructor): New. * tests/c++.at (AT_CHECK_VARIANTS): Fix an error caught by this commit. 2012-06-25 Akim Demaille yacc: work around the ylwrap limitation. * data/yacc.c (b4_shared_declarations): Include the header guards. Do not include the header in the *.c file, duplicate it. * NEWS (Future Changes): Extend, and announce the forthcoming change about the use of the parser header. 2012-06-25 Akim Demaille fix for printers and destructors. The previous "code_props: factor more" patch sends has_%printer etc. to m4, instead of has_printer. * src/output.c (prepare_symbol_definitions): Fix value of pname. 2012-06-25 Akim Demaille Merge remote-tracking branch 'origin/maint' * origin/maint: tests: more uniformity. tests: handle locations in a more generic way. tests: handle locations in the generic yyerror functions. tests: fix AT_CHECK_CALC. tests: improve infrastructure tests: factor. skeletons: minor style changes tests: AT_LANG. c skeletons: factor the declaration of yylloc and yylval. news: condemn YYPARSE_PARAM and YYLEX_PARAM. maint: regen. 2012-06-22 Akim Demaille code_props: factor more. * src/symtab.h, src/symtab.c (code_props_type_string): No longer static. * src/output.c (CODE_PROPS): Remove, we can now iterate on both the destructor and the printer. (SET_KEY2): New. 2012-06-22 Victor Santet maint: factor the handling of %printer and %destructor There is too much code duplication between %printer and %destructor. We used to have two functions for each action: the first one for destructors, the second one for printers. Factor using a 'code_props_type', and an array of code_props instead of two members. * src/symlist.h, src/symlist.c (symbol_list_destructor_set) (symbol_list_printer_set): Fuse into... (symbol_list_code_props_set): this. * src/symtab.h, src/symtab.c (default_tagged_destructor) (default_tagged_printer): Fuse into... (default_tagged_code_props): this. (default_tagless_destructor, default_tagless_printer) (default_tagless_code_props): Likewise. (code_props_type_string): new. (symbol_destructor_set, symbol_destructor_get, semantic_type_destructor_set) (default_tagged_destructor_set, default_tagless_destructor_set) (symbol_printer_set, symbol_printer_get, semantic_type_printer_set) (default_tagged_printer_set, default_tagless_printer_set): Replace by... (symbol_code_props_set, symbol_code_props_get, semantic_type_code_props_set) (default_tagged_code_props_set, default_tagless_code_props_set): these. * src/parse-gram.y (grammar_declaration): Adjust. * src/output.c (CODE_PROP, grammar_declaration): Ditto. * src/reader.c (symbol_should_be_used): Ditto. 2012-06-22 Akim Demaille tests: more uniformity. * tests/local.at (AT_LEX_FORMALS, AT_LEX_ARGS, AT_LEX_PRE_FORMALS) (AT_LEX_PRE_ARGS): Rename as... (AT_YYLEX_FORMALS, AT_YYLEX_ARGS, AT_YYLEX_PRE_FORMALS) (AT_YYLEX_PRE_ARGS): these, for consistency. (AT_API_PREFIX): Take %name-prefix into account. (AT_YYLEX_PROTOTYPE): New. Use it. * tests/actions.at, tests/calc.at, tests/cxx-type.at: Adjust to use them. 2012-06-22 Akim Demaille tests: handle locations in a more generic way. * tests/local.at (AT_YYERROR_PROTOTYPE): New. Use it. * tests/cxx-type.at: Extensive revamp to use a more traditional quotation scheme, and to use the generic yyerror implementation. Prefer Autotest macros to CPP macros. * tests/java.at: . 2012-06-22 Akim Demaille tests: handle locations in the generic yyerror functions. * tests/local.at (AT_YYERROR_DECLARE_EXTERN, AT_YYERROR_DECLARE) (AT_YYERROR_DEFINE): Handle locations for C and C++. * tests/calc.at: Use it for C++ (as C has extra arguments which are not yet handled by AT_BISON_OPTION_PUSHDEFS). * tests/actions.at: Adjust. 2012-06-22 Akim Demaille tests: fix AT_CHECK_CALC. * tests/calc.at (AT_CHECK_CALC): Contrary to its documentation, the test was skipped if given a second argument. Unused feature, remove it. 2012-06-22 Akim Demaille tests: improve infrastructure * tests/local.at (AT_LANG): Use c++ instead of cxx for C++. Adjust dependencies. (AT_YYERROR_DECLARE_EXTERN, AT_YYERROR_DECLARE): Issue nothing for C++/Java. (AT_YYERROR_DEFINE): Use m4_case. (AT_JAVA_COMPILE): Use AT_SKIP_IF. 2012-06-21 Akim Demaille tests: factor. * tests/glr-regression.at, tests/output.at, tests/push.at, * tests/regression.at, tests/torture.at, tests/actions.at: Use AT_YYLEX_* and AT_YYERROR_*. 2012-06-21 Akim Demaille skeletons: minor style changes * data/glr.c, data/yacc.c: here. 2012-06-21 Akim Demaille tests: AT_LANG. * tests/local.at (AT_BISON_OPTION_PUSHDEFS, AT_BISON_OPTION_POPDEFS): Define/undefine AT_LANGE (AT_LANG_COMPILE): New. (AT_FULL_COMPILE): Use AT_LANG. 2012-06-21 Victor Santet symtab: refactoring Prepares forthcoming changes. * src/symtab.c (symbols_do): Accept the hash table and the sorted list as arguments. Adjust dependencies. 2012-06-21 Akim Demaille maint: address syntax-check issues. * examples/calc++/local.mk: Space changes. * src/files.c: Avoid unmarked_diagnostics. * src/output.c: Remove useless include. 2012-06-21 Akim Demaille c skeletons: factor the declaration of yylloc and yylval. There is one difference: now, even without --defines, we generate extern declarations for these variables. The factoring is worth it. * data/c.m4 (b4_declare_yylstype): Declare them. * data/glr.c, data/yacc.c: Adjust. 2012-06-21 Akim Demaille news: condemn YYPARSE_PARAM and YYLEX_PARAM. * NEWS: here. (Bison 1.875): Add %parse-param and %lex-param. * doc/bison.texinfo: Spello. 2012-06-20 Akim Demaille fix warnings for useless %printer/%destructor The previous commit, which turns into a warning what used to be an error: %printer {} foo; %% exp: '0'; has two shortcomings: the warning is way too long (foo is reported to be useless later), and besides, it also turns into a warning much more serious errors: %printer {} foo; %% exp: foo; Reduce the amount to warnings in the first case, restore the error in the second. * src/symtab.h (status): Add a new inital state: undeclared. * src/symtab.c (symbol_new): Initialize to undeclared. (symbol_class_set): Simplify the logic of the code that neutralize the "redeclared" warning after the "redefined" one. (symbol_check_defined): "undeclared" is also an error. * src/reader.c (grammar_current_rule_symbol_append): Symbols appearing in a rule are "needed". * src/symlist.c (symbol_list_destructor_set, symbol_list_printer_set): An unknown symbol appearing in a %printer/%destructor is "used". * src/reduce.c (nonterminals_reduce): Do not report as "useless" symbols that are not used (e.g., those that for instance appeared only in a %printer). * tests/input.at (Undeclared symbols used for a printer or destructor): Improve the cover the cases described above. 2012-06-20 Akim Demaille gitignore: test-driver. 2012-06-20 Akim Demaille maint: style changes. * src/reduce.c (reduce_grammar_tables): Define variables with their initial value. 2012-06-20 Akim Demaille news: fixes. * NEWS: Fix spelling. 2012-06-20 Victor Santet warnings: used but undeclared symbols are warnings We used to raise an error if a symbol appears only in a %printer or %destructor. Make it a warning. * src/symtab.h (status): New enum. (symbol): Replace the binary "declared" with the three-state "status". Adjust dependencies. * src/symtab.c (symbol_check_defined): Needed symbols are an error, whereas "used" are simply warnings. * src/symlist.c (symbol_list_destructor_set, symbol_list_printer): Set symbol status to 'used' when associated to destructors or printers. * input.at (Undeclared symbols used for a printer or destructor): New. 2012-06-20 Akim Demaille maint: regen. * Makefile.am (regen): New target. 2012-06-20 Akim Demaille regen. 2012-06-20 Akim Demaille maint: regen. * Makefile.am (regen): New target. 2012-06-19 Akim Demaille tests: enhance AT_YYERROR_DEFINE. * tests/local.at: Handle the fact that locations are no longer needed with lalr1.cc. 2012-06-19 Akim Demaille Merge remote-tracking branch 'origin/maint' * origin/maint: maint: formatting changes. tests: support api.prefix. tests: pacify font-lock-mode. tests: remove test covered elsewhere. tests: factor the declaration/definition of yyerror and yylex. regen. tests: portability issues. tests: call the parser from another compilation unit. glr.c, yacc.c: declare yydebug in the header. skeletons: use header guards. tests: improve AT_FULL_COMPILE. tests: reorder. tests: strengthen the test on generated headers inclusion yacc.c: instead of duplicating y.tab.h inside y.tac.c, include it. yacc.c: factor. 2012-06-19 Akim Demaille maint: formatting changes. * NEWS: Fix indentation of code snippets. Untabify. 2012-06-17 Akim Demaille tests: support api.prefix. * tests/local.at (AT_BISON_OPTION_PUSHDEFS, AT_BISON_OPTION_POPDEFS): Define AT_API_PREFIX. (AT_YYERROR_DEFINE, AT_YYERROR_DECLARE_EXTERN, AT_YYLEX_DECLARE_EXTERN) (AT_YYLEX_DEFINE): Use it. * tests/input.at, tests/regression.at, tests/torture.at: Add AT_BISON_OPTION_PUSHDEFS/POPDEFS. 2012-06-17 Akim Demaille tests: pacify font-lock-mode. * tests/local.at: here. 2012-06-17 Akim Demaille tests: remove test covered elsewhere. * tests/headers.at (%union and --defines): Remove, pretty useless and insignificant. 2012-06-17 Akim Demaille tests: factor the declaration/definition of yyerror and yylex. * tests/local.at (AT_YYERROR_DECLARE, AT_YYERROR_DECLARE_EXTERN) (AT_YYERROR_DEFINE, AT_YYLEX_DECLARE, AT_YYLEX_DECLARE_EXTERN) (AT_YYLEX_DEFINE): New. Must be used inside AT_BISON_OPTION_PUSHDEFS/POPDEFS pair. * tests/actions.at, tests/conflicts.at, tests/glr-regression.at, * tests/headers.at, tests/input.at, tests/named-refs.at, * tests/regression.at, tests/skeletons.at, tests/synclines.at, * tests/torture.at: Use them. 2012-06-17 Akim Demaille regen. 2012-06-17 Akim Demaille tests: portability issues. * tests/calc.at (AT_CALC_MAIN): Missing include reported by Hydra. 2012-06-15 Akim Demaille tests: call the parser from another compilation unit. In order to improve the testing of %defines, which exports the interface of the generated parser, change the calc.at tests so that when %defines is passed, main will be in another compilation unit. It loads the generated header. * tests/calc.at (AT_CALC_MAIN): New. Includes the definition of the global variables. Therefore, now declare them from the %requires section of the parser. Adjust to yydebug and yyparse being renamed by %name-prefix. 2012-06-15 Akim Demaille glr.c, yacc.c: declare yydebug in the header. * data/c.m4 (b4_declare_yydebug): New. * data/glr.c, data/yacc.c (b4_shared_declarations): Use it. Remove the corresponding code from the parser body. * NEWS: Doc this. 2012-06-15 Akim Demaille skeletons: use header guards. * data/glr.c, data/glr.cc, data/yacc.c: here. * NEWS: Document it. 2012-06-15 Akim Demaille tests: improve AT_FULL_COMPILE. * tests/local.at: Accept a third argument. Simplify quotation pattern. Calls for better refactoring, but will suffice for a while. 2012-06-15 Akim Demaille tests: reorder. * tests/calc.at (power): Move its definition, as a preparation for forthcoming changes. And space changes. 2012-06-15 Akim Demaille tests: strengthen the test on generated headers inclusion * tests/headers.at (AT_TEST_CPP_GUARD_H): Accept Bison directives. (Invalid CPP headers): Check glr. 2012-06-15 Akim Demaille yacc.c: instead of duplicating y.tab.h inside y.tac.c, include it. This is already what glr.c and lalr1.cc do. * data/yacc.c: here. 2012-06-15 Akim Demaille maint: xfdopen, and scope reduction. * src/files.h, src/files.c (xfdopen): New. * src/output.c (output_skeleton): Use it. Reduce the scope of argv. 2012-06-15 Akim Demaille maint: space changes * configure.ac, src/complain.c: space changes. 2012-06-13 Akim Demaille yacc.c: factor. yacc.c used to include two almost identical sections: one for the *.h file, and another for the *.c file. The main difference is that in the *.c file we used the yy* names (as %name-prefix is handled by "#define yy* *" before), while the *.hh used * names. Keep only the later. If this is troublesome, b4_shared_declarations can easily take the desired prefix as argument. * data/yacc.c (b4_shared_declarations): New. Use it to factor duplicated declarations. 2012-06-13 Stefano Lattarini (tiny change) cosmetics: prettify names for compiled object for bison * src/local.mk (src_bison_SHORTNAME): Define to "bison". 2012-06-13 Akim Demaille lalr1.cc: spello. * data/lalr1.cc: Reported by Gilles Espinasse. 2012-06-13 Akim Demaille Merge remote-tracking branch 'origin/maint' * origin/maint: skeletons: factor yacc.c and glr.c. glr.c: minor refactoring. tests: remove all the -On flags. maint: fix spello. maint: improve release procedure instructions. gnulib: update readme-release. maint: cfg.mk: manual title. maint: cfg.mk: simplify maint: post-release administrivia 2012-06-12 Akim Demaille skeletons: factor yacc.c and glr.c. yacc.c and glr.c share common declarations. Their YYLTYPE are exactly equal, and their YYSTYPE are sufficiently alike to be fused (its declaration was protected by YYSTYPE_IS_DECLARED in yacc.c, but not in glr.c). Besides, yacc.c duplicated the definitions of YYLTYPE and YYSTYPE (*.h/*.c). * data/c.m4 (b4_declare_yylstype): New. * data/yacc.c, data/glr.c: Use it. 2012-06-12 Akim Demaille glr.c: minor refactoring. * data/glr.c (b4_shared_declarations): Move from the generated file section, to the M4 prologue. 2012-06-12 Akim Demaille tests: remove all the -On flags. * tests/atlocal.in: Here. Reported by Gilles Espinasse. 2012-06-12 Akim Demaille maint: fix spello. * README-hacking: Here. * THANKS: Reported by Gilles Espinasse. 2012-06-12 Akim Demaille maint: improve release procedure instructions. * gnulib: Update, in particular (README-release). * bootstrap.conf: don't require gendocs, provided by gnu-web-doc-update, provided by readme-release. * README-hacking: Update accordingly. 2012-06-07 Akim Demaille gnulib: update readme-release. * gnulib (readme-release): Now includes the modules it promotes. * bootstrap.conf: Simplify accordingly. 2012-06-07 Akim Demaille maint: cfg.mk: manual title. * cfg.mk (manuel_title): New. 2012-06-07 Akim Demaille maint: cfg.mk: simplify * cfg.mk: Remove bits provided by maint.mk. 2012-06-07 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2012-06-05 Akim Demaille maint: an envvar equal to "00" is 0. * src/output.c (prepare): here. Reported by Paul Eggert. 2012-06-05 Akim Demaille maint: don't use mbsr?chr. Basically, revert ba60c39547a445dee3e07920931b4d7a81843868's move to mbs* functions, which was prompted by -DGNULIB_POSIXCHECK. See and following. * bootstrap.conf: No longer ask for them. * src/files.c, src/getargs.c, src/location.c, * src/parse-gram.c, src/parse-gram.y, src/scan-gram.l, * src/symtab.c: s/mbs(r?chr)/str$1/g. 2012-06-05 Akim Demaille maint: use xconcat-filename. * bootstrap.conf (gnulib_modules): Request it. * src/output.h, src/output.c (compute_pkgdatadir): Rename as... (pkgdatadir): this. Adjust dependencies. * src/output.c (output_skeleton): Reduce the scope of "in". Use xconcatenated_filename to simplify the construction of the qualified paths to m4sugar.m4, bison.m4, and the selected skeleton. There are a few minor differences: the new code uses strchr instead of mbschr (but this was not really justified), and the new code does not garantee a single slash even if $BISON_PKGDATADIR ends with several (which was considered more accurate). See the discussion at . 2012-06-05 Akim Demaille maint: minor simplification * src/output.c (prepare): Assign use_push_for_pull_flag's value at its declaration. 2012-06-05 Akim Demaille Merge remote-tracking branch 'origin/maint' * origin/maint: version 2.5.1 NEWS: prepare for 2.5.1. maint: update release procedure maint: fix comment typos maint: post-release administrivia 2012-06-05 Akim Demaille version 2.5.1 * NEWS: Record release date. 2012-06-05 Akim Demaille NEWS: prepare for 2.5.1. * NEWS: Be compliant with do-release-commit-and-tag. 2012-06-05 Akim Demaille maint: update release procedure * bootstrap.conf: Request do-release-commit-and-tag and readme-release. * README-hacking: Adjust. 2012-06-05 Jim Meyering maint: fix comment typos Using http://github.com/lyda/misspell-check, massage its output into sed commands to perform the suggested changes. Initially, I filtered out the THRU->Through changes, because that failed to retain capitalization in the grammar token. Instead, do this manually, beforehand: sed -i s/THRU/THROUGH/ tests/existing.at git ls-files|misspellings -f -|perl -nl \ -e '/^(.*?)\[(\d+)\]: (\w+) -> "(.*?)"$/ or next;' \ -e '($file,$n,$l,$r)=($1,$2,$3,$4); $q="'\''"; $r=~s/$q/$q\\$q$q/g;'\ -e 'print "sed -i $q${n}s!$l!$r!$q $file"'|bash 2012-05-24 Akim Demaille build: regen. 2012-05-24 Akim Demaille Merge tag 'v2.5.1_rc2' Bison 2.5.1_rc2. * tag 'v2.5.1_rc2': (34 commits) Bison 2.5.1_rc2. doc: fixes. build: fix ChangeLog generation. c++: compute the header guards. skeletons: remove support for unused directive. lalr1.cc: improve Doxygen documentation. lalr1.cc: extract stack.hh. news: convert to double quotes. space changes. build: do not prototype flex-generated functions. build: fix ChangeLog generation. Bison 2.5.1_rc1. tests: save/restore Autotest special files when checking XML support. tests: AT_SAVE_SPECIAL_FILES / AT_RESTORE_SPECIAL_FILES. tests: honor TESTSUITEFLAGS in all the check targets. build: do not enable c++ warnings on 0 when nullptr is not supported. maint: update gnulib. build: config.in.h. build: move silent rules. glr.c: reduce variable scopes. maint: maintainer-release-check. maint: shush a syntax-check. maint: prefer "commit message" to "log entry". command line: fix minor leaks. maint: we no longer maintain the ChangeLog. maint: fix the generation of the synclines for bison's parser. maint: regen. maint: import the xmemdup0 gnulib module. maint: remove left-over gnulib modules. maint: ignore files imported by autopoint. build: AC_PROG_LEX: use more readable variable names. maint: regen src/parse-gram.[ch] maint: simplify parse-gram.y maint: s/strncpy/memcpy/, when equivalent 2012-05-23 Akim Demaille maint: post-release administrivia * NEWS: Add header line for next release. * .prev-version: Record previous version. * cfg.mk (old_NEWS_hash): Auto-update. 2012-05-23 Akim Demaille Bison 2.5.1_rc2. * NEWS: Update. 2012-05-23 Akim Demaille doc: fixes. * doc/bison.texinfo: Fix errors spotted by syntax-check. 2012-05-23 Akim Demaille build: fix ChangeLog generation. * gnulib: Update to get newest gitlog-to-changelog. * bootstrap: Update. * Makefile.am (gen-ChangeLog): Fix for Bison's git log style. 2012-05-21 Akim Demaille c++: compute the header guards. This is a frequent request. Recently pointed out by Wei Song, . * data/c.m4 (b4_tocpp, b4_cpp_guard, b4_cpp_guard_open) (b4_cpp_guard_close): New. * data/lalr1.cc, data/location.cc, data/stack.hh: Use them. * TODO (Header Guards): Move to... * NEWS: here. Formatting changes. 2012-05-21 Akim Demaille skeletons: remove support for unused directive. * src/scan-skel.l (@dir_prefix@): Remove support, has never been used, not even in the commit that introduced it, 2b81e969ea04c1a6502928ba7e847ec8ff7dcb2f. 2012-05-21 Akim Demaille lalr1.cc: improve Doxygen documentation. * data/location.cc: Qualify file names with directory name. 2012-05-21 Akim Demaille lalr1.cc: extract stack.hh. See commit 51bacae6b58fd5c6cce861f00440dc917384625e. * data/stack.hh: New, extracted from... * data/lalr1.cc: here. * data/Makefile.am: Adjust. 2012-05-21 Akim Demaille news: convert to double quotes. * NEWS: Convert from `quoted' to "quoted". Reported by Stefano Lattarini. http://lists.gnu.org/archive/html/bison-patches/2012-05/msg00039.html 2012-05-21 Akim Demaille space changes. * src/flex-scanner.h: Indent nested cpp directives. 2012-05-21 Akim Demaille build: do not prototype flex-generated functions. Some versions of Flex, possibly modified by the distribution package maintainers, have incompatible signatures. Since newer versions of Flex prototype their functions, avoid the conflicts in that case. Reported by Stefano Lattarini. . * src/flex-scanner.h (FLEX_VERSION_GT): New. Use it to issue prototypes for flex-generated functions only for versions up to 2.5.31, in accordance with the comment. See commit dc9701e848f27ae64b6ddcf809580998667d60f2. Use it to define yylex_destroy when needed. 2012-05-16 Akim Demaille build: fix ChangeLog generation. * Makefile.am (gen-ChangeLog): Fix for VPATH builds. 2012-05-14 Akim Demaille Bison 2.5.1_rc1. * NEWS: Update. * src/parse-gram.c, src/parse-gram.h: Regen. 2012-05-11 Akim Demaille tests: save/restore Autotest special files when checking XML support. Currently the test 248, "parse-gram.y: LALR = IELR", fails BISON_TEST_XML is set. * tests/local.at (AT_BISON_CHECK_XML): Belt: Save/restore files. * tests/regression.at (parse-gram.y: LALR = IELR): Suspenders: Don't rely on expout. Each one of these changes suffices. 2012-05-11 Akim Demaille tests: AT_SAVE_SPECIAL_FILES / AT_RESTORE_SPECIAL_FILES. Some of our macros play with expout and other Autotest special files, which may break their callers (e.g., currently TESTSUITEFLAGS='248 BISON_TEST_XML=1' fails). There is already some support for this. Expand it to be ready to use it elsewhere. * tests/local.at (AT_RESTORE_SPECIAL_FILES, AT_SAVE_SPECIAL_FILES) (at_save_special_files, at_restore_special_files): New. (AT_BISON_CHECK_NO_XML): Use them. 2012-05-11 Akim Demaille tests: honor TESTSUITEFLAGS in all the check targets. * tests/Makefile.am (installcheck-local): Simplify. (maintainer-check-posix, maintainer-check-valgrind): Honor $(TESTSUITEFLAGS). 2012-05-11 Akim Demaille build: do not enable c++ warnings on 0 when nullptr is not supported. * configure.ac (WARN_CXXFLAGS): Enable -Wzero-as-null-pointer-constant only when nullptr is supported.. 2012-05-11 Akim Demaille maint: update gnulib. * bootstrap, gnulib: Update. 2012-05-09 Akim Demaille build: config.in.h. Historically we used config.hin (where everybody else used config.h.in) to please DOS. Now that we use gnulib, there are already tons of files with several dots, especially *.in.h. * configure.ac: Rename config.hin as config.in.h. 2012-05-09 Akim Demaille build: move silent rules. * tests/Makefile.am: In the generation of the test suite. 2012-05-09 Akim Demaille glr.c: reduce variable scopes. * data/glr.c: Where appropriate, fuse variable declarations followed by assignments by variable declarations with a value. Where appropriate, introduce new scopes to limit variable spans. 2012-05-08 Akim Demaille maint: maintainer-release-check. * tests/Makefile.am (maintainer-release-check): New. * Makefile.am (MAINTAINER_CHECKS): New. Support maintainer-release-check. * README-hacking: Document it, and syntax-check too. 2012-05-08 Akim Demaille maint: shush a syntax-check. * cfg.mk: lib/timevar is not planned to be gnulib'ed, as it comes from GCC. 2012-05-08 Akim Demaille maint: prefer "commit message" to "log entry". * README-hacking: here. Suggested by Stefano Lattarini. 2012-05-08 Akim Demaille command line: fix minor leaks. * src/getargs.c (getargs): Free pointers before allocating them new content. 2012-05-08 Akim Demaille maint: we no longer maintain the ChangeLog. * .gitattributes: No need to merge it. * README-hacking: Update release instructions. 2012-05-06 Akim Demaille maint: fix the generation of the synclines for bison's parser. * tests/bison.in: Import from master the changes that make this script generate synclines that are independant of the builddir/srcdir user's set up. 2012-05-06 Akim Demaille maint: regen. * src/parse-gram.c, src/parse-gram.h: Regen. 2012-05-06 Akim Demaille maint: import the xmemdup0 gnulib module. * bootstrap.conf: Require this module. * src/parse-gram.y: Include xmemdup0.h. 2012-05-06 Akim Demaille maint: remove left-over gnulib modules. * bootstrap.conf (gnulib_modules): Remove pipe-posix. * lib/.gitignore, m4/.gitignore: Remove files that we no longer use. 2012-05-06 Akim Demaille maint: ignore files imported by autopoint. * m4/.gitignore: here. 2012-05-06 Akim Demaille build: AC_PROG_LEX: use more readable variable names. * m4/flex.m4 (AC_PROG_LEX): Prefer LEX_IS_FLEX to FLEX. Prefer true/false to yes/no for such variables. * configure.ac: Adjust. 2012-05-06 Jim Meyering maint: regen src/parse-gram.[ch] 2012-05-06 Jim Meyering Akim Demaille maint: simplify parse-gram.y * src/parse-gram.y (add_param): Use xmemdup0 in place of xmalloc+memcpy, and strspn in place of an open-coded loop. 2012-05-06 Jim Meyering maint: s/strncpy/memcpy/, when equivalent * src/output.c (output_skeleton): Use memcpy, not strncpy, since the source is known to fit in the destination buffer. * src/parse-gram.y (%skeleton): Likewise. 2012-05-04 Akim Demaille glr.c: formatting changes. * data/glr.c: Fix indentation. 2012-05-04 Akim Demaille Merge remote-tracking branch 'origin/master' * origin/master: glr.c: untabify. glr.cc: untabify. glr.cc: formatting changes. glr.cc: remove unused signature. glr.cc: properly declare locations are const where appropriate. doc: fix @xref. 2012-05-04 Akim Demaille glr.c: untabify. * data/glr.c: here. 2012-05-04 Akim Demaille glr.cc: untabify. * data/glr.cc: here. 2012-05-04 Akim Demaille glr.cc: formatting changes. * data/glr.cc: Fit in 80 columns. 2012-05-04 Akim Demaille glr.cc: remove unused signature. * data/glr.cc (yydestruct_): Not used, remove. It is yydestruct which is used. 2012-05-04 Akim Demaille glr.cc: properly declare locations are const where appropriate. * data/glr.cc (yyerror): The location is const. 2012-05-04 Akim Demaille doc: fix @xref. * doc/bison.texinfo: here. 2012-05-04 Akim Demaille Merge remote-tracking branch 'origin/maint' * origin/maint: (22 commits) tests: ignore code coverage/profiling failure messages doc: fix some invalid @ref. build: fix previous commit. install-pdf: fix. NEWS: Update. %printer: support both yyo and yyoutput. doc: mfcalc: demonstrate %printer. tests: style changes. build: require Flex. build: flex.m4: check for Flex. build: flex.m4: quote properly. build: flex.m4. build: autoconf: update. glr: eliminate last bits of unwanted locations. NEWS: 2.6 will drop K&R. TODO: remove dead items. TODO: import from master. gnulib: update. maint: update NEWS. doc: fix index. doc: fix documentation of YYERROR. c++: more YY_NULL 2012-05-02 Akim Demaille tests: ignore code coverage/profiling failure messages The Hydra buildfarm provides code coverage analysis. For some reason, in some test cases, code coverage data seem to be incompatible, and generate error messages at parser run-time. Ignore these messages so that (i) these tests do pass, (ii) coverage results be provided by Hydra. * tests/local.at (AT_PARSER_CHECK): Ignore messages for failed merges of code coverage/profiling results. 2012-04-16 Akim Demaille doc: fix some invalid @ref. * doc/bison.texinfo: Fix incorrect @ref uses. 2012-04-16 Akim Demaille build: fix previous commit. * bootstrap: Update from gnulib. 2012-04-16 Akim Demaille install-pdf: fix. * gnulib: Fix install-pdf in po/ and runtime-po/. Reported by Hans Aberg. Fixed by Joel E. Denny. http://lists.gnu.org/archive/html/bug-bison/2011-05/msg00008.html 2012-04-16 Akim Demaille NEWS: Update. * NEWS: Spell check. (%printer): is now documented. 2012-04-16 Akim Demaille %printer: support both yyo and yyoutput. lalr1.cc used to support yyo, but not yyoutput. Support both, but document only yyoutput (at least until there is some consensus on this). * data/c.m4 (yy_symbol_value_print): Also support yyo. * data/glr.cc (yy_symbol_value_print_): Support both yyo and yyoutput. * data/lalr1.cc: Also support yyoutput. * doc/bison.texinfo: Explicitly use yyoutput in the examples. * examples/mfcalc/mfcalc.test: Test the -p option. 2012-04-16 Akim Demaille doc: mfcalc: demonstrate %printer. * doc/bison.texinfo (Printer Decl): New. Number mfcalc.y snippets so that they are output in the proper order. (The mfcalc Main): Use yydebug. (Debugging): Simplify the text. (Enabling Traces, Mfcalc Traces, The YYPRINT Macro): New. (Table of Symbols): Document YYPRINT and YYFPRINTF. 2012-04-16 Akim Demaille tests: style changes. * tests/input.at: Use "print" in %printer instead of "destroy". It is unused, so we don't care, yet it is less surprising. * tests/actions.at: Comment changes. 2012-04-16 Akim Demaille %printer: support both yyo and yyoutput. lalr1.cc used to support yyo, but not yyoutput. Support both, but document only yyoutput (at least until there is some consensus on this). * data/c.m4 (yy_symbol_value_print): Also support yyo. * data/glr.cc (yy_symbol_value_print_): Support both yyo and yyoutput. * data/lalr1.cc: Also support yyoutput. * doc/bison.texinfo: Explicitly use yyoutput in the examples. * examples/mfcalc/mfcalc.test: Test the -p option. 2012-04-16 Akim Demaille doc: mfcalc: demonstrate %printer. * doc/bison.texinfo (Printer Decl): New. Number mfcalc.y snippets so that they are output in the proper order. (The mfcalc Main): Use yydebug. (Debugging): Simplify the text. (Enabling Traces, Mfcalc Traces, The YYPRINT Macro): New. (Table of Symbols): Document YYPRINT and YYFPRINTF. 2012-04-16 Akim Demaille tests: style changes. * tests/input.at: Use "print" in %printer instead of "destroy". It is unused, so we don't care, yet it is less surprising. * tests/actions.at: Comment changes. 2012-04-10 Akim Demaille build: require Flex. * configure.ac: Require Flex. 2012-04-10 Akim Demaille build: flex.m4: check for Flex. * m4/flex.m4 (_AC_PROG_LEX_YYTEXT_DECL): Check that $LEX supports some of the Flex options, and exclusive start conditions. Define FLEX to 'yes'/'', as AC_PROG_CC does for GCC. 2012-04-10 Akim Demaille build: flex.m4: quote properly. * m4/flex.m4: Use quotes more systematically. 2012-04-10 Akim Demaille build: flex.m4. * m4/flex.m4: New. An exact copy of what is in Autoconf currently. 2012-04-10 Akim Demaille build: autoconf: update. * submodules/autoconf: Update. There are no changes in data/m4sugar/foreach.m4, and the changes in data/m4sugar/m4sugar.m4 are minor. 2012-04-10 Akim Demaille glr: eliminate last bits of unwanted locations. * data/glr.c (YYLTYPE): Do not define when locations are not demanded. Adjust all dependencies. 2012-04-10 Akim Demaille NEWS: 2.6 will drop K&R. * NEWS: here. (glr.c): Fix a spello. 2012-04-09 Akim Demaille TODO: remove dead items. * TODO (Documentation, %printer, Java): Remove, already done (or just waiting for approval). (Fortran, BTYacc): Remove, there does not seem to be demand. 2012-04-09 Akim Demaille TODO: import from master. * TODO: Copy the current version. 2012-04-08 Akim Demaille tests: fix bison wrapper. * tests/bison.in (PERL): Fix. 2012-04-08 Akim Demaille doc: fix some invalid @ref. * doc/bison.texinfo: Fix incorrect @ref uses. 2012-04-08 Akim Demaille build: extexi: support out-of-order blocks. * examples/extexi (%file_output): Remove. (&process): Accept "FILE: BLOCK-NUM" requests. 2012-04-08 Akim Demaille build: look for Perl in configure. Bison uses "/usr/bin/perl" or "perl" in several places, and it does not appear to be a problem. But, at least to make it simpler to change PERL on the make command line, check for perl in configure. * configure.ac (PERL): New. * doc/Doxyfile.in, doc/local.mk, examples/local.mk, * tests/bison.in: Use it. 2012-04-08 Akim Demaille maint: rewrite extexi in Perl. * examples/extexi: Rewrite in Perl. * examples/local.mk (extract): Adjust. 2012-04-07 Akim Demaille build: simplify clean. * doc/local.mk (CLEANFILES): Since the previous commit, there a no longer such files. * Makefile.in (CLEANFILES): Initialize here. 2012-04-07 Akim Demaille gnulib: update. * bootstrap.conf (bootstrap_sync): True again. It was disabled while waiting for changes to be integrated in gnulib's bootstrap, which was done long ago. * bootstrap, gnulib: Update. 2012-04-04 Akim Demaille maint: update NEWS. * NEWS: Fix entry about __attribute__. Reorder by "decreasing" order of importance. 2012-04-04 Akim Demaille doc: fix index. http://lists.gnu.org/archive/html/bison-patches/2012-04/msg00006.html * doc/bison.texinfo: Avoid using @def* variant with more than the defined entity as main entity, as it results in an incorrect index. For instance, don't document {return YYERROR;}, which results in a single index entry "return YYERROR;", but rather as typed function whose return type is "type", and whose argument list is ";". 2012-04-04 Akim Demaille doc: fix documentation of YYERROR. * doc/bison.texinfo (Table of Symbols): Fix the documentation of YYERROR by copying that from "Action Features". 2012-04-01 Akim Demaille build: fix distcheck issues. For some reason it seems that texi2dvi -o no longer forces --clean mode, so we have stray TeX compilation files staying in top_builddir since TeX is run from there. While at it, upgrade the generation of the (completely obsolete) reference card. Target PDF. * doc/local.mk (TEXI2DVI): Pass --build-dir. (CLEANDIRS): More accurate. (doc/refcard.dvi): Replace with... (doc/refcard.pdf): this. Adjust dependencies. 2012-04-01 Akim Demaille build: don't rely on $< in non-pattern rules. * doc/local.mk, tests/local.mk: here. Reported by Stefano Lattarini. 2012-04-01 Akim Demaille doc: upgrade Doxyfile. * doc/Doxyfile.in: Run doxygen -u. Prompted by Tim Landscheidt. 2012-04-01 Akim Demaille doc: help Doxygen find our files. * doc/Doxyfile.in (INCLUDE_PATH): here. 2012-04-01 Tim Landscheidt Fix Doxygen generation and clean-up. * doc/Doxyfile.in: Amend OUTPUT_DIRECTORY. * doc/local.mk (html-local): Amend working directory. (CLEANDIRS): Fix "html", remove "latex". 2012-04-01 Tim Landscheidt Fix Doxygen comment. * src/InadequacyList.h: s###. 2012-04-01 Akim Demaille c++: more YY_NULL Caught by maintainer-check-g++. * data/glr.c, data/lalr1.cc, data/yacc.c, tests/cxx-type.at, * tests/glr-regression.at, tests/push.at: When simple to do, avoid expliciting the null ptr. Otherwise use YY_NULL. 2012-04-01 Akim Demaille c++: more YY_NULL Caught by maintainer-check-g++. * data/glr.c, data/lalr1.cc, data/yacc.c, tests/cxx-type.at, * tests/glr-regression.at, tests/push.at: When simple to do, avoid expliciting the null ptr. Otherwise use YY_NULL. 2012-04-01 Akim Demaille Merge remote-tracking branch 'origin/maint' * origin/maint: bump to 2012 in skeletons. build: remove ancient Autoconf tests. doc: c++: complete the location documentation. c++: locations: provide convenience constructors. c++: locations: remove useless "inline". glr: do not use locations when they are not requested c++: use nullptr for C++11. build: simplify and improve the compiler warnings for tests. gnulib: update. maint: formatting changes. NEWS: update. Java: Fix syntax error handling without error token. tests: beware of -pedantic on large #line numbers. tests: when using the C++ compiler, use its flags too. 2012-04-01 Akim Demaille bump to 2012 in skeletons. * data/glr.c, data/glr.cc, data/lalr1.cc, data/lalr1.java, * data/location.cc, data/yacc.c: Bump copyright year ranges. 2012-04-01 Akim Demaille build: remove ancient Autoconf tests. lib/subpipe.c was removed in 47fa574761319b0a422691223c9b8a9a72f36aa2. * m4/subpipe.m4: Remove. * configure.ac (BISON_PREREQ_SUBPIPE): Remove. 2012-03-31 Akim Demaille doc: c++: complete the location documentation. * data/location.cc (position::initialize, location::initialize): Also accept line and column, with default values. * doc/bison.texinfo (C++ position, C++ location): New nodes. Describe more thoroughly these classes. Fix several Texinfo misuses. 2012-03-31 Akim Demaille c++: locations: provide convenience constructors. * data/location.cc (position::position): Accept file, line and column as arguments with default values. Always qualify initial line and column literals as unsigned. (location::location): Provide convenience constructors. 2012-03-31 Akim Demaille c++: locations: remove useless "inline". * data/location.cc: "inline" is implicit when defining methods in the class definition. 2012-03-31 Akim Demaille glr: do not use locations when they are not requested When the test suite runs with -O2 and warnings enabled, G++ complains of locations being used, but not initialized. The simplest is to not use locations. * data/glr.c (b4_locuser_formals, b4_locuser_args): New. Use them when locations should not be used. Use b4_locations_if where appropriate. (yyuserAction): Modify the order to the arguments to make it more alike the other routines, and to make use of b4_locuser_args simpler. 2012-03-31 Akim Demaille c++: use nullptr for C++11. C++11 introduces "nullptr" which plays the role of C's NULL, in replacement of "0". Fix the C++ skeletons to avoid warnings about uses of "0" in place of "nullptr", and improve C skeletons to also use this "nullptr" when compiled with a C++11 compiler. * configure.ac: More C++ warnings. * NEWS (2.5.1): Document this. * data/c++.m4, data/c.m4 (b4_null_define): New. (b4_null): Use YY_NULL instead of 0. * data/glr.c, data/lalr1.cc, data/location.cc, data/yacc.c: Call b4_null_define/b4_null where appropriate. Use YY_NULL instead of NULL. * data/location.cc (initialize): Accept a default argument, YY_NULL. * tests/actions.at, tests/calc.at: Adjust. * data/glr.c, lib/libiberty.h, src/system.h (__attribute__): Do not disable it when __STRICT_ANSI__ is defined, as, for instance, it disables the __attribute__((unused)) which protects us from some compiler warnings. This was already done elsewhere in Bison, in 2001, see 4a0d89369599a2cea01f4fbdf791f426a02cb5a3. * tests/regression.at: Adjust output. 2012-03-30 Akim Demaille build: simplify and improve the compiler warnings for tests. * configure.ac (warn_common, warn_c, warn_cxx): New. Use them to compute independently the options supported by the C and C++ compilers. Don't AC_SUBST the variables passed to gl_WARN_ADD: it does it for us. (WARN_CFLAGS_TEST, WARN_CXXFLAGS_TEST): Don't aggregate $WARN_CFLAGS and $WARN_CXXFLAGS in them now, leave it to atlocal.in. (O0CFLAGS, O0CXXFLAGS): Move their definition to... * tests/atlocal.in: here. Be more systematic between C and C++. Reorder to factor between variables. Propagate all of the variables when --compile-c-with-cxx. 2012-03-30 Akim Demaille gnulib: update. 2012-03-30 Akim Demaille maint: formatting changes. * src/system.h: Indent CPP directives using cppi. 2012-03-27 Akim Demaille NEWS: update. * NEWS: Java fixes, more about the doc changes, liby issues. 2012-03-27 Tim Landscheidt Java: Fix syntax error handling without error token. * data/lalr1.java (YYParser::parse): Here. * tests/java.at: Add test case. 2012-03-24 Akim Demaille tests: beware of -pedantic on large #line numbers. * tests/local.at (AT_TEST_TABLES_AND_PARSE): Don't pass -pedantic when compiling large canonical-LR parsers. Reported by Tys Lefering. http://lists.gnu.org/archive/html/bug-bison/2012-03/msg00025.html 2012-03-24 Akim Demaille tests: when using the C++ compiler, use its flags too. * tests/local.at: Go for colors. (--compile-c-with-cxx): New option. We used to pass "CC=$CXX" as command line argument, but it was not possible to adjust CFLAGS accordingly in atlocal, since it is loaded before assignments on the command line are honored (so that the command line takes precedence). * tests/atlocal.in: Implement it. * tests/local.mk: Use it. 2012-03-24 Akim Demaille tests: style changes in the Makefile. * tests/local.mk: Prefer passing variable assignment by the command line, instead of the environment, so that it is reported in the logs. Prefer single quotes for shell literal strings. 2012-03-24 Akim Demaille tests: beware of -pedantic on large #line numbers. * tests/local.at (AT_TEST_TABLES_AND_PARSE): Don't pass -pedantic when compiling large canonical-LR parsers. Reported by Tys Lefering. http://lists.gnu.org/archive/html/bug-bison/2012-03/msg00025.html 2012-03-24 Akim Demaille tests: when using the C++ compiler, use its flags too. * tests/local.at: Go for colors. (--compile-c-with-cxx): New option. We used to pass "CC=$CXX" as command line argument, but it was not possible to adjust CFLAGS accordingly in atlocal, since it is loaded before assignments on the command line are honored (so that the command line takes precedence). * tests/atlocal.in: Implement it. * tests/local.mk: Use it. 2012-03-24 Akim Demaille tests: fix dependencies. * tests/local.mk (check_SCRIPTS): Add atlocal and atconfig so that they are properly updated before running tests. (RUN_TESTSUITE_deps): New. Use it to factor the dependencies of "*-check" targets, especially those that don't bounce to the regular "check-local" target, since then they don't benefit from the proper dependencies (such as atlocal). 2012-03-19 Akim Demaille Merge remote-tracking branch 'fsf/maint' * fsf/maint: (404 commits) doc: update the --verbose report format. doc: spell check. doc: stmt, not stmnt. doc: save width. doc: reformat grammar snippets. doc: use only @example, not @smallexample. doc: style changes. doc: minor fixes to "Understanding" section tests: minor fixes/simplifications tests: be robust to quote style. maint: update gnulib. tests: be robust to POSIXLY_CORRECT being defined. doc: fix environment issues. regen. tests: fix regressions. glr: fix ambiguity reports. doc: stylistic improvements. maint: address sc_prohibit_doubled_word. maint: address sc_prohibit_always-defined_macros. maint: address sc_bindtextdomain, sc_program_name and sc_prohibit_HAVE_MBRTOWC. ... 2012-03-19 Akim Demaille doc: spell fix. * doc/bison.texinfo: here. Reported by Tim Landscheidt. 2012-03-19 Akim Demaille doc: update the --verbose report format. * doc/bison.texinfo (Understanding): Adjust to match the current format. 2012-03-19 Akim Demaille doc: spell check. * doc/bison.texinfo: here. 2012-03-19 Akim Demaille doc: stmt, not stmnt. * doc/bison.texinfo: s/stmnt/stmt/g. This is a much more common abbreviation for "statement". 2012-03-19 Akim Demaille doc: save width. * doc/bison.texinfo (Language and Grammar): Use the same layout for an example in all the versions, i.e., keep as general case what used to be used only for Info. 2012-03-19 Akim Demaille doc: reformat grammar snippets. * doc/bison.texinfo: Convert the grammar examples to use a narrower style. This helps fitting into the @smallbook constraints. http://lists.gnu.org/archive/html/bison-patches/2012-03/msg00011.html 2012-03-19 Akim Demaille doc: use only @example, not @smallexample. * doc/bison.texinfo: Convert all @smallexamples into @examples. Adjust layout where needed. 2012-03-19 Akim Demaille doc: style changes. * doc/bison.texinfo: Avoid line width issues with TeX. Upgrade ancient messages. Move some comments to better looking places. Add more @group. (Mfcalc Symbol Table): Reduce variable scopes. Prefer size_t for sizes. Prefer declarations with an initial value. Fix a @group environment. 2012-03-19 Paul Eggert doc: minor fixes to "Understanding" section * doc/bison.texinfo (Understanding): Minor wording fixes and improvements. Fixes problems reported in . 2012-03-19 Akim Demaille doc: update the --verbose report format. * doc/bison.texinfo (Understanding): Adjust to match the current format. 2012-03-19 Akim Demaille doc: spell check. * doc/bison.texinfo: here. 2012-03-19 Akim Demaille doc: stmt, not stmnt. * doc/bison.texinfo: s/stmnt/stmt/g. This is a much more common abbreviation for "statement". 2012-03-19 Akim Demaille doc: save width. * doc/bison.texinfo (Language and Grammar): Use the same layout for an example in all the versions, i.e., keep as general case what used to be used only for Info. 2012-03-19 Akim Demaille doc: reformat grammar snippets. * doc/bison.texinfo: Convert the grammar examples to use a narrower style. This helps fitting into the @smallbook constraints. http://lists.gnu.org/archive/html/bison-patches/2012-03/msg00011.html 2012-03-19 Akim Demaille doc: use only @example, not @smallexample. * doc/bison.texinfo: Convert all @smallexamples into @examples. Adjust layout where needed. 2012-03-19 Akim Demaille doc: style changes. * doc/bison.texinfo: Avoid line width issues with TeX. Upgrade ancient messages. Move some comments to better looking places. Add more @group. (Mfcalc Symbol Table): Reduce variable scopes. Prefer size_t for sizes. Prefer declarations with an initial value. Fix a @group environment. 2012-03-19 Akim Demaille c.m4: better newline control with b4_parse_param_use. * data/c.m4: Use m4_ifvaln instead of m4_ifval where applicable. (b4_parse_param_use): Switch order between two nested "if"s to avoid useless empty lines. Adjust callers to avoid useless lines. 2012-03-19 Akim Demaille glr.c: remove (broken) support for YYPRINT. YYPRINT uses yytoknum which glr does not define. Since YYPRINT is considered obsolete, and did not work, don't fix its support, remove it from glr.c. * data/c.m4 (yy_symbol_value_print): Use YYPRINT only for yacc.c. * TODO: Done. 2012-03-17 Paul Eggert doc: minor fixes to "Understanding" section * doc/bison.texinfo (Understanding): Minor wording fixes and improvements. Fixes problems reported in . 2012-03-15 Akim Demaille TODO: update. 2012-03-15 Akim Demaille gnulib: update. 2012-03-13 Akim Demaille tests: minor fixes/simplifications * tests/local.at (AT_BISON_CHECK_NO_XML): Simplify sed programs, quotation, and default value assignments. Ensure a proper value to the numeric variables. Reported by Lie Yan. http://lists.gnu.org/archive/html/bug-bison/2012-03/msg00000.html 2012-03-13 Akim Demaille maint: fix distcheck. * examples/local.mk (MAINTAINERCLEANFILES): Complete, and rename as... (CLEANFILES): this, * examples/calc++/local.mk, examples/mfcalc/local.mk, * examples/rpcalc/local.mk (CLEANFILES): Add the generated files. 2012-03-13 Akim Demaille tests: minor fixes/simplifications * tests/local.at (AT_BISON_CHECK_NO_XML): Simplify sed programs, quotation, and default value assignments. Ensure a proper value to the numeric variables. Reported by Lie Yan. http://lists.gnu.org/archive/html/bug-bison/2012-03/msg00000.html 2012-03-09 Akim Demaille tests: be robust to POSIXLY_CORRECT being defined. * tests/local.at (AT_BISON_CHECK_NO_XML): Check if POSIXLY_CORRECT is defined, not if it is defined to 1. Reported by Lie Yan. http://lists.gnu.org/archive/html/bug-bison/2012-03/msg00000.html 2012-03-09 Akim Demaille tests: be robust to quote style. See . * src/main.c (main): Define the quoting style we use. * tests/atlocal.in: Use ASCII style quotes during the tests. 2012-03-09 Akim Demaille maint: update gnulib. * gnulib: update. * src/scan-gram.l: Don't use the (former version of) STREQ. 2012-03-09 Akim Demaille tests: remove quote magic from the bison test wrapper. Basically, revert 4c4d35394d1bdb4dc3392482ab002bc111a3378f. * tests/bison.in: Leave bison's stderr as is. 2012-03-09 Akim Demaille tests: be robust to quote style. See . * src/main.c (main): Define the quoting style we use. * tests/atlocal.in: Use ASCII style quotes during the tests. 2012-03-09 Akim Demaille avoid direct strncmp calls. Before this change, bison would accept either .tab and _tab equivalently, whatever the current platform. Besides, it was not obeying everywhere to the possible definition of TAB_EXT to something else than .tab. For consistency, handle only TAB_EXT (".tab" on non DJGPP platforms). Support for "_tab" is neither documented, nor tested. * src/system.h (STRNCMP_LIT): New. From Jim Meyering. (STRPREFIX_LIT): New. * src/files.c, src/getargs.c: Use it. 2012-03-06 Akim Demaille tests: be robust to POSIXLY_CORRECT being defined. * tests/local.at (AT_BISON_CHECK_NO_XML): Check if POSIXLY_CORRECT is defined, not if it is defined to 1. Reported by Lie Yan. http://lists.gnu.org/archive/html/bug-bison/2012-03/msg00000.html 2012-02-24 Akim Demaille build: comment changes. * Makefile.am, examples/calc++/local.mk, examples/local.mk, * examples/mfcalc/local.mk, examples/rpcalc/local.mk, * lib/local.mk, src/local.mk, tests/local.mk: Make the copyright licenses more uniform. 2012-02-24 Akim Demaille build: fix more example extraction issues. * Makefile.am (dist_TESTS): New. (TESTS, EXTRA_DIST): Run and ship them. * examples/calc++/local.mk: examples/calc++/calc++.stamp no longer exists, don't try to ship it. (.yy.stamp): New recipe. Use it. * examples/calc++/local.mk, examples/mfcalc/local.mk, * examples/rpcalc/local.mk: Don't ship the sources. Adjust the CPPFLAGS: there is nothing left in srcdir. (MAINTAINERCLEANFILES): Remove, now we are in builddir. (TESTS): Rename as... (dist_TESTS): this. 2012-02-24 Akim Demaille maint: fix example extraction issues. calc++: don't rely on Automake to compile a C++ parser. Basically, revert commit 609b3d8096fb0cc1fa4d908b6e1a860ced260bda, Automake 1.11.3 is not safe enough for C++ parser. * examples/calc++/calc++-parser.hh: Remove. * examples/calc++/local.mk (examples/calc++/calc++-parser.stamp): New. examples: factor the extractions into a single step extexi had to be run in the extraction directory. Now, it can be given the files with expected output directory. This allows to use $(*_extracted) variables (before we had to list again their members, but limited to their base names). In turn, this also allows fusing the extraction recipes into a single one. Also, it is currently too hard (or requires too much duplication, since Make wants all the occurrences of the files to be prefixed with $(srcdir)/, which is something Automake cannot support for *_SOURCES) to work in the source tree. So extract, and compile scanners and parsers in the build tree. * examples/extexi (basename): New. (BEGIN): Now "file_wanted" maps base name to extracted file name. * examples/calc++/local.mk, examples/mfcalc/local.mk, * examples/rpcalc/local.mk: Fuse extraction rules into... * examples/local.mk: Here. (extract, extracted): New. 2012-02-23 Akim Demaille maint: use STREQ/STRNEQ. * doc/bison.texinfo: Space change. * src/system.h (STREQ, STRNEQ): New. * src/files.c, src/ielr.c, src/lalr.c, src/muscle-tab.c, * src/output.c, src/print.c, src/print_graph.c, * src/reader.c, src/scan-skel.l, src/tables.c, * src/uniqstr.c: Use them. * src/scan-gram.l: Do not use streq.h, use system.h's STREQ. * cfg.mk: The documentation is an exception. 2012-02-23 Akim Demaille doc: fix environment issues. * doc/bison.texinfo: Do not use @verbatim, in particular when we use @group inside. Use @quotation instead of @display for frequently asked questions, it looks much nicer. 2012-02-23 Akim Demaille doc: fix environment issues. * doc/bison.texinfo: Do not use @verbatim, in particular when we use @group inside. Use @quotation instead of @display for frequently asked questions, it looks much nicer. 2012-02-23 Akim Demaille regen. * src/parse-gram.h, src/parse-gram.c: regen. 2012-02-23 Akim Demaille tests: fix regressions. Exit status 63 is documented for version-mismatch. * bootstrap.conf (gnulib_modules): Remove sysexits. * src/system.h (EX_MISMATCH): Define. * src/parse-gram.y (version_check): Use it instead of EX_CONFIG. Missing includes. * tests/calc.at, tests/named-refs.at: Include assert.h. 2012-02-22 Akim Demaille maint: gitignore. * examples/mfcalc/.gitignore, examples/rpcalc/.gitignore: Fix. 2012-02-21 Akim Demaille regen. * src/parse-gram.c, src/parse-gram.h: regen. 2012-02-21 Akim Demaille tests: fix regressions. Exit status 63 is documented for version-mismatch. * bootstrap.conf (gnulib_modules): Remove sysexits. * src/system.h (EX_MISMATCH): Define. * src/parse-gram.y (version_check): Use it instead of EX_CONFIG. Missing includes. * tests/calc.at, tests/named-refs.at: Include assert.h. 2012-02-21 Akim Demaille tests: post-process stderr to normalize quotes. * tests/bison.in: Save bison's stderr, and convert gettextized quotes to plain ASCII. 2012-02-21 Akim Demaille glr: fix ambiguity reports. * tests/glr-regression.at (Ambiguity reports): New. 2012-02-21 Akim Demaille glr: fix ambiguity reports. Fix a regression introduced in commit 783aa653f4ca70a75919c8516b950494c612cbfe. * tests/glr-regression.at (Ambiguity reports): New. * data/glr.c (yyreportTree): Fix an offset error. 2012-02-19 Akim Demaille doc: stylistic improvements. * doc/bison.texinfo: Prefer "continue" to empty loop bodies. Add some @group/@end group to avoid poor page breaks. 2012-02-19 Akim Demaille maint: address sc_prohibit_doubled_word. * data/yacc.c, doc/bison.texinfo: Reword to avoid having to disable that check. * cfg.mk: No longer skip this test. 2012-02-19 Akim Demaille maint: address sc_prohibit_always-defined_macros. * cfg.mk: No longer skip it, except where EXIT_SUCCESS is used as a witness for stdlib.h. Skip this test when appropriate. * data/yacc.c: Drop a note about why EXIT_SUCCESS is defined here. 2012-02-19 Akim Demaille maint: address sc_bindtextdomain, sc_program_name and sc_prohibit_HAVE_MBRTOWC. * bootstrap.conf (gnulib_modules): Require progname. * src/complain.c, src/getargs.c, src/getargs.h, src/main.c: Use it. * cfg.mk (exclude): New. Use it. Skip lib/main.c for bindtextdomain and set_program_name. 2012-02-19 Akim Demaille maint: remove stray file. * config.hin: Remove. 2012-02-19 Akim Demaille doc: stylistic improvements. * doc/bison.texinfo: Prefer "continue" to empty loop bodies. Add some @group/@end group to avoid poor page breaks. 2012-02-19 Akim Demaille doc: check the rpcalc. * doc/bison.texinfo: Tag rpcalc.y snippets. Add missing includes. (Rpcalc Rules): Don't issue leading tabs. Complete an Info menu. Use @result. * examples/rpcalc/local.mk: New. * examples/rpcalc/rpcalc.test: New. * examples/local.mk: Use them. * examples/mfcalc/mfcalc.test: Remove dup test. * examples/test: Disable debug traces. 2012-02-19 Akim Demaille regen. * src/parse-gram.c, src/parse-gram.h: Regen. 2012-02-19 Akim Demaille maint: address sc_prohibit_doubled_word. * data/yacc.c, doc/bison.texinfo: Reword to avoid having to disable that check. * cfg.mk: No longer skip this test. 2012-02-19 Akim Demaille maint: address sc_prohibit_always-defined_macros. * cfg.mk: No longer skip it, except where EXIT_SUCCESS is used as a witness for stdlib.h. Skip this test when appropriate. * data/yacc.c: Drop a note about why EXIT_SUCCESS is defined here. 2012-02-19 Akim Demaille maint: address sc_bindtextdomain, sc_program_name and sc_prohibit_HAVE_MBRTOWC. * bootstrap.conf (gnulib_modules): Require progname. * src/complain.c, src/getargs.c, src/getargs.h, src/main.c: Use it. * cfg.mk (exclude): New. Use it. Skip lib/main.c for bindtextdomain and set_program_name. 2012-02-19 Akim Demaille maint: remove stray file. * config.hin: Remove. 2012-02-19 Akim Demaille bitset: fix an incorrect error message. * lib/bitset_stats.c: here. Reported by Stefano Lattarini. 2012-02-19 Akim Demaille maint: address some syntax-issues remaining after cherry-picking from master. * cfg.mk: Skip bison generated files, 2.5 is generating trailing blanks. This is already fixed in master. * tests/conflicts.at, tests/java.at: Fix white space issues. 2012-02-19 Akim Demaille regen. * src/parse-gram.c, src/parse-gram.h: Regen. 2012-02-19 Akim Demaille bitset: fix an incorrect error message. * lib/bitset_stats.c: here. Reported by Stefano Lattarini. 2012-02-19 Jim Meyering maint: reenable sc_m4_quote_check * cfg.mk (local-checks-to-skip): Reenable sc_m4_quote_check. * m4/dmalloc.m4: Add quotes. 2012-02-19 Akim Demaille maint: remove trailing empty lines. * cfg.mk: No longer skip sc_prohibit_empty_lines_at_EOF, except for parse-gram.h (generated). * examples/mfcalc/.gitignore, lib/.gitignore, m4/.gitignore, * po/.gitignore, runtime-po/.gitignore: Remove trailing/leading empty lines. 2012-02-19 Akim Demaille maint: avoid "magic number exit". * cfg.mk (local-checks-to-skip): No longer skip it. * bootstrap.conf (gnulib_modules): Add sysexits. * doc/bison.texinfo, etc/bench.pl.in, src/parse-gram.y, * src/system.h, tests/calc.at, tests/named-refs.at: Use assert where appropriate instead of "if (...) exit". Use symbolic exit status elsewhere. 2012-02-19 Akim Demaille maint: fix some syntax-check issues. * cfg.mk (local-checks-to-skip): Remove sc_prohibit_quotearg_without_use, sc_prohibit_strcmp, sc_unmarked_diagnostics, sc_useless_cpp_parens. (sc_unmarked_diagnostics): Skip DJGPP. * data/yacc.c, src/LR0.c, src/closure.c, * src/flex-scanner.h, src/gram.c, src/lalr.c, * src/print-xml.c, src/print.c, src/print_graph.c, * src/reader.c, src/reduce.c, src/tables.c: Don't use parens with cpp's defined. Remove useless includes. 2012-02-19 Akim Demaille maint: address a couple of syntax-check errors. * cfg.mk (local-checks-to-skip): Remove sc_error_message_period and sc_error_message_uppercase. Address the uncovered issues. * po/POTFILES.in: Add missing files. * src/symtab.c: Remove useless includes. * lib/bitset_stats.c, src/files.c, tests/glr-regression.at: Use conformant error messages. 2012-02-19 Akim Demaille maint: gnulib: upgrade. 2012-02-18 Akim Demaille maint: remove trailing empty lines. * cfg.mk: No longer skip sc_prohibit_empty_lines_at_EOF, except for parse-gram.h (generated). * examples/mfcalc/.gitignore, lib/.gitignore, m4/.gitignore, * po/.gitignore, runtime-po/.gitignore: Remove trailing/leading empty lines. 2012-02-18 Akim Demaille maint: regen. * src/parse-gram.c, src/parse-gram.h: regen. 2012-02-18 Akim Demaille maint: avoid "magic number exit". * cfg.mk (local-checks-to-skip): No longer skip it. * bootstrap.conf (gnulib_modules): Add sysexits. * doc/bison.texinfo, etc/bench.pl.in, src/parse-gram.y, * src/system.h, tests/calc.at, tests/named-refs.at: Use assert where appropriate instead of "if (...) exit". Use symbolic exit status elsewhere. 2012-02-18 Akim Demaille maint: fix some syntax-check issues. * cfg.mk (local-checks-to-skip): Remove sc_prohibit_quotearg_without_use, sc_prohibit_strcmp, sc_unmarked_diagnostics, sc_useless_cpp_parens. (sc_unmarked_diagnostics): Skip DJGPP. * data/yacc.c, src/LR0.c, src/closure.c, * src/flex-scanner.h, src/gram.c, src/lalr.c, * src/print-xml.c, src/print.c, src/print_graph.c, * src/reader.c, src/reduce.c, src/tables.c: Don't use parens with cpp's defined. Remove useless includes. 2012-02-18 Akim Demaille maint: address a couple of syntax-check errors. * cfg.mk (local-checks-to-skip): Remove sc_error_message_period and sc_error_message_uppercase. Address the uncovered issues. * po/POTFILES.in: Add missing files. * src/symtab.c: Remove useless includes. * lib/bitset_stats.c, src/files.c, tests/glr-regression.at: Use conformant error messages. 2012-02-18 Akim Demaille maint: gnulib: upgrade. 2012-02-17 Akim Demaille doc: mfcalc: fix includes. * doc/bison.texinfo: math.h is needed early. 2012-02-17 Akim Demaille examples: factor the test suite. * examples/mfcalc/test, examples/calc++/test: Extract the common bits into... * examples/test: here. (cwd): New. Use it to avoid a race on the temporary directory. Reported by Jim Meyering. * examples/mfcalc/test, examples/calc++/test: Rename into... * examples/mfcalc/mfcalc.test, examples/calc++/calc++.test: these. * examples/calc++/local.mk, examples/mfcalc/local.mk, * examples/local.mk: Adjust. 2012-02-17 Akim Demaille examples: fix the test suites. * examples/calc++/test, examples/mfcalc/test (me): Be more meaningfull: include the example name. (prog): Factor. (run): Avoid printf, use echo. Add missing parens. (cleanup): New. Call it on trap. Remove the previous "rm" that did the cleanup. Move into a private directory to avoid concurrency issues. Reported by Jim Meyering. 2012-02-17 Jim Meyering examples: link mfcalc with -lm for uses of pow, cos, atan, etc. * examples/mfcalc/local.mk (examples_mfcalc_mfcalc_LDADD): Define. 2012-02-16 Akim Demaille mfcalc: extract and exercise. * examples/mfcalc/local.mk, examples/mfcalc/test: New, based on calc++'s ones. * examples/local.mk: Include mfcalc/local.mk. 2012-02-16 Akim Demaille calc++: factor for other extracted tests. * Makefile.am (TESTS, check_PROGRAMS): Initialize here. * examples/local.mk (doc, extexi): Define here. * examples/calc++/local.mk: Adjust accordingly. * configure.ac: Ask for parallel-tests (for the way the logs are handled). * examples/calc++/test: As a consequence, always be verbose. ($prog): New. (run): Use it. Sort the tests in a more natural order (simplest first). 2012-02-16 Akim Demaille doc: mfcalc: send errors to stderr. * doc/bison.texinfo (Mfcalc Lexer): New. (Mfcalc Main): Move the definition of main and yyerror here, for clarity. Let yyerror report on stderr. 2012-02-16 Akim Demaille doc: fix mfcalc code. * doc/bison.texinfo (Multi-function Calc): Add missing includes. Fix the rendering of the result: use @result and remove the initial tabulation in the actual code. Fix stylistic issues: avoid the , operator. Add extexi mark-up. * examples/extexi: Also support @smallexample. 2012-02-16 Akim Demaille tests: c++: stylistic changes. * tests/c++.at: Don't use void for incoming arguments. Prefer cstdlib to stdlib.h. 2012-02-16 Jim Meyering tests: avoid c++ failure due to lack of getenv decl * tests/c++.at (Syntax error as exception): Avoid spurious failure at least when compiling with g++-4.7.x due to lack of declaration of getenv. Include . 2012-02-15 Akim Demaille maint: rely on Automake for parsers. * Makefile.am (AM_YFLAGS): Automake looks for "-d" alone. Move other options in here. (BISON): New. (YACC): Use it. (bison_SOURCES): Now that automake can see `-d' in AM_YFLAGS, we can rely on it to compile and ship the parser header files. Based on commit 737406a32c201471699bfa0843d1f432f3ec29ab and commit 3d6ca339083c278d907c9f030f4ba6bc5ecb07f2. 2012-02-15 Akim Demaille maint: trust Automake for parser headers. * examples/calc++/local.mk, src/local.mk: Now that automake can see `-d' in AM_YFLAGS, we can rely on it to compile and ship the parser header files. 2012-02-15 Akim Demaille maint: help Automake reading Yacc flags. * Makefile.am (AM_YFLAGS): Automake looks for "-d" alone. 2012-02-15 Akim Demaille calc++: rely on Automake. Rely on $(YACC) being the bison being built, and let Automake do the rest. It turned out to be much more difficult than anticipated, for various reasons, including some bad behavior from Automake 1.11.2 which (i) generates calc++-parser.h instead of calc++-parser.hh, and (ii) leaves an #include "y.tab.h" in the generated parser instead of #include "calc++-parser.h". The authors of Automake appear to be aware of the problem, http://lists.gnu.org/archive/html/automake/2011-05/msg00008.html so a simple work around will suffice for the time being. * examples/calc++/y.tab.h, examples/calc++/calc++-parser.hh: New. To work around Automake 1.11.2 issues. * examples/calc++/local.mk: Remove all the rules for compilation with bison, leave them to Automake. So provide it with "calc++-parse.yy" as a source file. (calc_sources_generated, calc_sources_extracted): Rename as. (calc_generated, calc_extracted): these. (calc_sources): New. Fix them. 2012-02-14 Akim Demaille maint: tidy the Makefile a bit. * src/local.mk: Put yacc related variables together. (AUTOMAKE_OPTIONS): Move to... * Makefile.am: here. Remove an old Emacs mode request which disables Automake support. * src/local.mk (YACC, AM_YFLAGS): Move to... * Makefile.am: here, as they will be used by other local.mks. 2012-02-14 Akim Demaille maint: de-recurse the handling of examples The directory was still using a local Makefile.am because it provides "scoped" Make variables: these examples are not meant to use the same CPPFLAGS etc. If we were to use the same -I set, we'd pick up gnulib's stdio.h for instance, which we do not want for these simple examples. Yet, as a result, the dependencies are less accurate, there is code duplication, etc. This is especially perceptible when trying to extract more examples from the documentation, as will be done in forthcoming changes. In order to make the tuning of CPPFLAGS easier, discard the predefined -I from Automake. * examples/calc++/Makefile.am: Rename as... * examples/calc++/local.mk: this. Adjust the paths which are now rooted in top_srcdir/top_builddir. Handle BISON_CXX_WORKS here, instead of the too crude previous approach that completely discarded the whole directory. ($(BISON)): Remove now useless bouncing recipe. (calc___CPPFLAGS): New. Stay away from -Ilib. * Makefile.am, configure.ac, examples/local.mk, * examples/calc++/test: Adjust. * configure.ac: Pass nostdinc to Automake. * src/local.mk, lib/local.mk (AM_CPPFLAGS): Move to... * Makefile.am: here. * src/local.mk, examples/calc++/Makefile.am (BISON, BISON_IN): Factor to... * Makefile.am: here. * tests/local.mk: Use it. 2012-02-14 Akim Demaille variant: fix the example. * examples/variant.yy: Adjust to "assert" being now "parse.assert". 2012-02-14 Akim Demaille maint: more authors. * AUTHORS: here. Suggested by Tys Lefering. 2012-02-14 Akim Demaille maint: add license headers. * examples/calc++/test, examples/variant.yy, AUTHORS, THANKS, * tests/atlocal.in, tests/bison.in: Add license headers. Reported by Tys Lefering. 2012-02-14 Akim Demaille maint: remove obsolete file. * etc/make-ChangeLogs: Remove (used for rcs to cvs migration!). Reported by Tys Lefering. 2012-02-14 Akim Demaille maint: more authors. * AUTHORS: here. Suggested by Tys Lefering. 2012-02-14 Akim Demaille maint: add license headers. * examples/calc++/test, examples/variant.yy, AUTHORS, THANKS, * tests/atlocal.in, tests/bison.in: Add license headers. Reported by Tys Lefering. 2012-02-14 Akim Demaille maint: remove obsolete file. * etc/make-ChangeLogs: Remove (used for rcs to cvs migration!). Reported by Tys Lefering. 2012-02-10 Akim Demaille lalr1.cc: also handle syntax_error when calling yylex. * data/lalr1.cc (parse): Catch syntax_error around yylex and forward them to errlab1. * tests/c++.at (Syntax error as exception): Check support for syntax exceptions raised by the scanner. * NEWS, doc/bison.texinfo: Document it. 2012-02-10 Akim Demaille tests: lalr1.cc: check syntax_error. * tests/c++.at (Syntax error as exception): New. 2012-02-10 Akim Demaille tests: don't require locations uselessly. * tests/c++.at (Syntax error discarding no lookahead): Contrary to 2.5, C++ parsers can work without locations. 2012-02-10 Akim Demaille maint: more silent rules. * tests/local.mk (TESTSUITE_AT): Include plackage.m4. Adjust dependencies. Make testsuite.at its first argument. (package.m4): Be silent. (testsuite): Be silent. Use $<. 2012-02-10 Akim Demaille skeletons: simplify the protections against "unused" warnings. * data/c.m4 (b4_parse_param_use): Also accept optional arguments to "use". Simplify callers. * data/glr.c (yyuserAction): Simplify use of b4_parse_param_use. (yy_reduce_print): Don't use b4_parse_param_use, as all the arguments _are_ used. * data/lalr1.cc (YY_SYMBOL_PRINT): Even when disabled, "use" the symbol argument. This neutralizes a warning in yypush_ when there are no symbols with a semantic values. (yy_destroy_): Remove useless "use" of yymsg. 2012-02-10 Akim Demaille glr: formatting changes. * data/glr.c: Split long strings. 2012-02-08 Akim Demaille use a more consistent quoting style. See . Use quotearg as often as possible instead of leaving the choice of the quotes to the translators. Use shorter messages. Factor similar messages to a single format, to make localization easier. * src/files.c, src/getargs.c, src/muscle-tab.c, src/reader.c * src/scan-code.l, src/scan-gram.l, src/symtab.c: Use quote() or quotearg_colon() on printf arguments instead of quotes in the format string. * data/bison.m4: Keep sync with the changes in muscle-tab.c. * tests/skeletons.at, tests/input.at, tests/regression.at: Adjust expected messages. 2012-02-08 Akim Demaille use a more consistent quoting style. See . Use quotearg as often as possible instead of leaving the choice of the quotes to the translators. Use shorter messages. Factor similar messages to a single format, to make localization easier. * src/files.c, src/getargs.c, src/muscle-tab.c, src/reader.c * src/scan-code.l, src/scan-gram.l, src/symtab.c: Use quote() or quotearg_colon() on printf arguments instead of quotes in the format string. * data/bison.m4: Keep sync with the changes in muscle-tab.c. * tests/skeletons.at, tests/input.at, tests/regression.at: Adjust expected messages. 2012-01-31 Jim Meyering maint: reenable sc_m4_quote_check * cfg.mk (local-checks-to-skip): Reenable sc_m4_quote_check. * m4/dmalloc.m4: Add quotes. 2012-01-31 Jim Meyering maint: force "make syntax-check" to pass by skipping failing tests * cfg.mk (local-checks-to-skip): Skip all currently-failing tests. Remove changelog-check; it's long gone. 2012-01-31 Akim Demaille maint: remove stray debug code. * src/Makefile.am (echo): Remove. 2012-01-31 Akim Demaille maint: space changes. * src/Makefile.am: Use 2 leading spaces for variable definition spreading over several lines. 2012-01-31 Akim Demaille maint: more silent-rules. * doc/local.mk, src/local.mk, examples/calc++/Makefile.am: Use $(AM_V_GEN) and $(AM_V_at) where appropriate. 2012-01-31 Jim Meyering do not ignore errors like ENOSPC,EIO when writing to stdout Standard output was never explicitly closed, so we could not detect failure. Thus, bison would ignore the errors of writing to a full file system and getting an I/O error on write, but only for standard output, e.g., for --print-localedir, --print-datadir, --help and some verbose output. Now, "bison --print-datadir > /dev/full" reports the write failure: bison: write error: No space left on device Before, it would exit 0 with no diagnostic, implying success. This is not an issue for "--output=-" or the other FILE-accepting command-line options, because unlike most other GNU programs, an output file argument of "-" is treated as the literal "./-", rather than standard output. * bootstrap.conf (gnulib_modules): Add closeout. * src/main.c: Include "closeout.h". Use atexit to ensure we close stdout. * .gitignore: Ignore new files pulled in via gnulib-tool. 2012-01-31 Akim Demaille maint: more silent-rules. * doc/local.mk, src/local.mk, examples/calc++/Makefile.am: Use $(AM_V_GEN) and $(AM_V_at) where appropriate. 2012-01-29 Jim Meyering do not ignore errors like ENOSPC,EIO when writing to stdout Standard output was never explicitly closed, so we could not detect failure. Thus, bison would ignore the errors of writing to a full file system and getting an I/O error on write, but only for standard output, e.g., for --print-localedir, --print-datadir, --help and some verbose output. Now, "bison --print-datadir > /dev/full" reports the write failure: bison: write error: No space left on device Before, it would exit 0 with no diagnostic, implying success. This is not an issue for "--output=-" or the other FILE-accepting command-line options, because unlike most other GNU programs, an output file argument of "-" is treated as the literal "./-", rather than standard output. * bootstrap.conf (gnulib_modules): Add closeout. * src/main.c: Include "closeout.h". Use atexit to ensure we close stdout. * .gitignore: Ignore new files pulled in via gnulib-tool. 2012-01-26 Akim Demaille tests: fix expected output. * tests/actions.at (YYBACKUP): here. 2012-01-26 Akim Demaille tests: fix expected output. * tests/actions.at (YYBACKUP): here. 2012-01-26 Akim Demaille maint: fix configure.ac Fix commit 1890a2a816dab86c23cc1d0af8fac3986335deb7. * configure.ac: Fix variable assignment. 2012-01-26 Akim Demaille yacc: fix YYBACKUP. Reported by David Kastrup: https://lists.gnu.org/archive/html/bug-bison/2011-10/msg00002.html. * data/yacc.c (YYBACKUP): Accept rhs size. Restore the proper state value. * TODO (YYBACKUP): Make it... * tests/actions.at: a new test case. * NEWS, THANKS: Update. 2012-01-26 Akim Demaille maint: update TODO. * TODO (Labeling the symbols): Remove, it's done ("Name references"). 2012-01-26 Akim Demaille maint: update THANKS. * THANKS: Update Tys's address, on his request. 2012-01-26 Akim Demaille maint: fix --gcc-warnings support. * configure.ac: Use enable_gcc_warnings instead of enableval, which is valid only with AC_ARG_ENABLE. 2012-01-26 Akim Demaille maint: silent-rules. * configure.ac: Ask for silent-rules support. Enable it by default. 2012-01-26 Akim Demaille maint: remove trailing blanks. * src/scan-code.l: Here. 2012-01-26 Akim Demaille yacc: fix YYBACKUP. Reported by David Kastrup: https://lists.gnu.org/archive/html/bug-bison/2011-10/msg00002.html. * data/yacc.c (YYBACKUP): Accept rhs size. Restore the proper state value. * TODO (YYBACKUP): Make it... * tests/actions.at: a new test case. * NEWS, THANKS: Update. 2012-01-25 Akim Demaille maint: update TODO. * TODO (Labeling the symbols): Remove, it's done ("Name references"). 2012-01-25 Akim Demaille maint: update THANKS. * THANKS: Update Tys's address, on his request. 2012-01-25 Akim Demaille maint: fix --gcc-warnings support. * configure.ac: Use enable_gcc_warnings instead of enableval, which is valid only with AC_ARG_ENABLE. 2012-01-25 Akim Demaille maint: silent-rules. * configure.ac: Ask for silent-rules support. Enable it by default. 2012-01-25 Paul Eggert tests: port to Solaris 10 'diff -u' * tests/regression.at (parse-gram.y: LALR = IELR): Port to Solaris 10, where "diff -u X X" outputs "No differences encountered" instead of outputting nothing. Reported by Tomohiro Suzuki in . 2012-01-25 Jim Meyering build: avoid possibly-replaced fprintf in liby-source, yyerror.c * lib/yyerror.c (yyerror): Use fputs and fputc rather than fprintf with a mere "%s\n" format. Always return 0 now, on the assumption that the return value was never used anyway. Don't include after all. This avoids a problem reported by Thiru Ramakrishnan in http://lists.gnu.org/archive/html/help-bison/2011-11/msg00000.html * cfg.mk: Exempt lib/yyerror.c from the sc_require_config_h_first test. * THANKS: Update. 2012-01-24 Paul Eggert tests: port to Solaris 10 'diff -u' * tests/regression.at (parse-gram.y: LALR = IELR): Port to Solaris 10, where "diff -u X X" outputs "No differences encountered" instead of outputting nothing. Reported by Tomohiro Suzuki in . 2012-01-24 Jim Meyering maint: generate ChangeLog from git log * Makefile.am (gen-ChangeLog): New rule. (dist-hook): Depend on it. (EXTRA_DIST): Distribute the two ChangeLog-* files. * bootstrap.conf (gnulib_modules): Add gitlog-to-changelog. (bootstrap_post_import_hook): Ensure that ChangeLog exists. * build-aux/git-log-fix: New file. * ChangeLog-2012: Renamed ... * ChangeLog: ... from this. * ChangeLog-1998: Renamed ... * OChangeLog: ...from this * .gitignore: Add ChangeLog. 2012-01-24 Jim Meyering change more quotes in source, and adjust tests to match Run this command to change each `%s' to '%s' in source directories: git grep -l '`%s'\' src djgpp data \ |xargs perl -pi -e '$q="'\''";s/`%s$q/$q%s$q/g' * data/bison.m4: Affected per the above. * djgpp/subpipe.c: Likewise. * src/files.c: Likewise. * src/getargs.c: Likewise. * src/muscle-tab.c: Likewise. * src/reader.c: Likewise. * tests/glr-regression.at: Adjust to match. * tests/input.at: Likewise. * tests/push.at: Likewise. * tests/skeletons.at: Likewise. 2012-01-23 Jim Meyering quote consistently and make tests pass with new quoting from gnulib Updating to gnulib pulled in new quote and quotarg modules, by which quoting is now done like 'this' rather than `this'. That change induces many "make check" test failures. This change adapts code and tests so that "make check" passes once again. * src/scan-code.l: Quote like 'this', not like `this'. * src/scan-gram.l: Likewise. * src/symtab.c: Likewise. * tests/actions.at: Adjust tests to match. * tests/input.at: Likewise. * tests/named-refs.at: Likewise. * tests/output.at: Likewise. * tests/regression.at: Likewise. * lib/.gitignore: Regenerate. * m4/.gitignore: Likewise. 2012-01-23 Jim Meyering build: avoid possibly-replaced fprintf in liby-source, yyerror.c * lib/yyerror.c (yyerror): Use fputs and fputc rather than fprintf with a mere "%s\n" format. Always return 0 now, on the assumption that the return value was never used anyway. Don't include after all. This avoids a problem reported by Thiru Ramakrishnan in http://lists.gnu.org/archive/html/help-bison/2011-11/msg00000.html * cfg.mk: Exempt lib/yyerror.c from the sc_require_config_h_first test. * THANKS: Update. 2012-01-23 Jim Meyering build: update gnulib and autoconf submodules to latest (cherry picked from commit 728415f885e5cb8e518c8576fa6e1f541e384130) 2012-01-23 Jim Meyering build: manually update bootstrap from gnulib, and adapt Updating to the latest bootstrap from gnulib involves more of a change than usual, and updating to the latest gnulib would involve its own set of challenges with the upcoming quoting changes, so we update bootstrap manually and separately. * bootstrap: Update from gnulib. * lib/Makefile.am: Initialize more variables to empty, so that gnulib.mk can append to them with "+=". * bootstrap.conf (gnulib_mk_hook): Remove. No longer honored. (gnulib_tool_option_extras): Generate gnulib.mk. 2012-01-23 Jim Meyering maint: include first * cfg.mk (exclude_file_name_regexp--sc_require_config_h_first): Exempt data/glr.c and data/yacc.c from the include-config.h-first requirement. 2012-01-23 Jim Meyering build: include from lib/yyerror.c * lib/yyerror.c: Include . 2012-01-23 Jim Meyering maint: list djgpp/subpipe.c in po/POTFILES.in * po/POTFILES.in: Add djgpp/subpipe.c. 2012-01-23 Jim Meyering maint: placate the space-TAB syntax-check * cfg.mk (exclude_file_name_regexp--sc_space_tab): Exempt tests/input.at and tests/c++.at, since they appear to use SP-TAB sequences deliberately. * OChangeLog: Remove space-before-TAB. 2012-01-23 Jim Meyering doc: correct typo: s/can not/cannot/ * doc/bison.texinfo (Bug Reports): s/can not/cannot/ And remove trailing blanks. 2012-01-23 Jim Meyering build: generalize etc/prefix-gnulib-mk This script hard-coded "libbison" and lib/gnulib.mk. Adjust the script to require a --lib-name=$gnulib_name option and a FILE argument like lib/$gnulib_mk. Also add support for --help and --version. * etc/prefix-gnulib-mk: Generalize. * bootstrap.conf (bootstrap_post_import_hook): Update its invocation. 2012-01-22 Jim Meyering maint: get gpl-3.0 from gnulib * bootstrap.conf (gnulib_modules): Add gpl-3.0. * doc/gpl-3.0.texi: Remove from version control, now that we get it via gnulib. * doc/.gitignore: Ignore it. 2012-01-20 Akim Demaille maint: be more robust to gnulib's FOO_H variables. * configure.ac: Instead of listing gnulib's variables, look for them among AC_SUBST variables. 2012-01-20 Jim Meyering maint: generate ChangeLog from git log * Makefile.am (gen-ChangeLog): New rule. (dist-hook): Depend on it. (EXTRA_DIST): Distribute the two ChangeLog-* files. * bootstrap.conf (gnulib_modules): Add gitlog-to-changelog. (bootstrap_post_import_hook): Ensure that ChangeLog exists. * build-aux/git-log-fix: New file. * ChangeLog-2012: Renamed ... * ChangeLog: ... from this. * ChangeLog-1998: Renamed ... * OChangeLog: ...from this * .gitignore: Add ChangeLog. 2012-01-19 Jim Meyering change more quotes in source, and adjust tests to match Run this command to change each `%s' to '%s' in source directories: git grep -l '`%s'\' src djgpp data \ |xargs perl -pi -e '$q="'\''";s/`%s$q/$q%s$q/g' * data/bison.m4: Affected per the above. * djgpp/subpipe.c: Likewise. * src/files.c: Likewise. * src/getargs.c: Likewise. * src/muscle-tab.c: Likewise. * src/reader.c: Likewise. * tests/glr-regression.at: Adjust to match. * tests/input.at: Likewise. * tests/push.at: Likewise. * tests/skeletons.at: Likewise. 2012-01-19 Jim Meyering quote consistently and make tests pass with new quoting from gnulib Updating to gnulib pulled in new quote and quotarg modules, by which quoting is now done like 'this' rather than `this'. That change induces many "make check" test failures. This change adapts code and tests so that "make check" passes once again. * src/scan-code.l: Quote like 'this', not like `this'. * src/scan-gram.l: Likewise. * src/symtab.c: Likewise. * tests/actions.at: Adjust tests to match. * tests/input.at: Likewise. * tests/named-refs.at: Likewise. * tests/output.at: Likewise. * tests/regression.at: Likewise. * lib/.gitignore: Regenerate. * m4/.gitignore: Likewise. 2012-01-19 Jim Meyering build: update gnulib and autoconf submodules to latest 2012-01-19 Jim Meyering build: manually update bootstrap from gnulib, and adapt Updating to the latest bootstrap from gnulib involves more of a change than usual, and updating to the latest gnulib would involve its own set of challenges with the upcoming quoting changes, so we update bootstrap manually and separately. * bootstrap: Update from gnulib. * Makefile.am: Initialize more variables to empty, so that gnulib.mk can append to them with "+=". * bootstrap.conf (gnulib_mk_hook): Remove. No longer honored. (bootstrap_post_import_hook): Instead, run the same command, etc/prefix-gnulib-mk lib/$gnulib_mk, via slightly different API. Temporarily disable "bootstrap_sync=true". * etc/prefix-gnulib-mk: Don't prepend "lib/" to tokens like -I$(... or "\". 2012-01-19 Jim Meyering maint: include first * cfg.mk (exclude_file_name_regexp--sc_require_config_h_first): Exempt data/glr.c and data/yacc.c from the include-config.h-first requirement. 2012-01-19 Jim Meyering build: include from lib/yyerror.c * lib/yyerror.c: Include . 2012-01-19 Jim Meyering maint: list djgpp/subpipe.c in po/POTFILES.in * po/POTFILES.in: Add djgpp/subpipe.c. 2012-01-19 Jim Meyering maint: placate the space-TAB syntax-check * cfg.mk (exclude_file_name_regexp--sc_space_tab): Exempt tests/input.at and tests/c++.at, since they appear to use SP-TAB sequences deliberately. * OChangeLog: Remove space-before-TAB. 2012-01-19 Jim Meyering maint: remove final trailing space * src/scan-gram.l (%): Remove single space at end of line. 2012-01-19 Jim Meyering maint: get gpl-3.0 from gnulib * bootstrap.conf (gnulib_modules): Add gpl-3.0. * doc/gpl-3.0.texi: Remove from version control, now that we get it via gnulib. * doc/.gitignore: Ignore it. 2012-01-19 Jim Meyering doc: correct typo: s/can not/cannot/ * doc/bison.texinfo (Bug Reports): s/can not/cannot/ And remove trailing blanks.