From f3f371deb376478176866fd770fbcf9bc0d0609f Mon Sep 17 00:00:00 2001 From: Oliver Kiddle Date: Sat, 13 May 2023 00:56:48 +0200 Subject: [PATCH] 51728: assign pcre named capture groups to a hash --- ChangeLog | 3 +++ Doc/Zsh/mod_pcre.yo | 10 ++++++---- Src/Modules/pcre.c | 43 +++++++++++++++++++++++++++++++++---------- Test/V07pcre.ztst | 14 ++++++++++++++ 4 files changed, 56 insertions(+), 14 deletions(-) # diff --git a/ChangeLog b/ChangeLog # index 285b73b2c..2835a9405 100644 # --- a/ChangeLog # +++ b/ChangeLog # @@ -1,5 +1,8 @@ # 2023-05-13 Oliver Kiddle # + * 51728: Doc/Zsh/mod_pcre.yo, Src/Modules/pcre.c, # + Test/V07pcre.ztst: assign pcre named capture groups to a hash # + # * 51723: Src/Modules/pcre.c, Test/V07pcre.ztst, configure.ac: # migrate pcre module to pcre2 --- a/Doc/Zsh/mod_pcre.yo +++ b/Doc/Zsh/mod_pcre.yo @@ -20,12 +20,12 @@ including those that indicate newline. ) findex(pcre_study) item(tt(pcre_study))( -Studies the previously-compiled PCRE which may result in faster -matching. +Requests JIT compilation for the previously-compiled PCRE which +may result in faster matching. ) findex(pcre_match) item(tt(pcre_match) [ tt(-v) var(var) ] [ tt(-a) var(arr) ] \ -[ tt(-n) var(offset) ] [ tt(-b) ] var(string))( +[ tt(-A) var(assoc) ] [ tt(-n) var(offset) ] [ tt(-b) ] var(string))( Returns successfully if tt(string) matches the previously-compiled PCRE. @@ -36,7 +36,9 @@ substrings, unless the tt(-a) option is case it will set the array var(arr). Similarly, the variable tt(MATCH) will be set to the entire matched portion of the string, unless the tt(-v) option is given, in which case the variable -var(var) will be set. +var(var) will be set. Furthermore, any named captures will +be stored in the associative array tt(.pcre.match) unless an +alternative is given with tt(-A). No variables are altered if there is no successful match. A tt(-n) option starts searching for a match from the byte var(offset) position in var(string). If the tt(-b) option is given, --- a/Src/Modules/pcre.c +++ b/Src/Modules/pcre.c @@ -129,14 +129,17 @@ bin_pcre_study(char *nam, UNUSED(char ** } static int -zpcre_get_substrings(char *arg, pcre2_match_data *mdata, int captured_count, - char *matchvar, char *substravar, int want_offset_pair, - int matchedinarr, int want_begin_end) +zpcre_get_substrings(pcre2_code *pat, char *arg, pcre2_match_data *mdata, + int captured_count, char *matchvar, char *substravar, char *namedassoc, + int want_offset_pair, int matchedinarr, int want_begin_end) { PCRE2_SIZE *ovec; char *match_all, **matches; char offset_all[50]; int capture_start = 1; + int vec_off; + PCRE2_SPTR ntable; /* table of named captures */ + uint32_t ncount, nsize; if (matchedinarr) { /* bash-style ovec[0] entire-matched string in the array */ @@ -174,7 +177,7 @@ zpcre_get_substrings(char *arg, pcre2_ma if (substravar && (!want_begin_end || nelem)) { char **x; - int vec_off, i; + int i; matches = x = (char **) zalloc(sizeof(char *) * (captured_count+1-capture_start)); for (i = capture_start; i < captured_count; i++) { vec_off = 2*i; @@ -184,6 +187,23 @@ zpcre_get_substrings(char *arg, pcre2_ma setaparam(substravar, matches); } + if (!pcre2_pattern_info(pat, PCRE2_INFO_NAMECOUNT, &ncount) && ncount + && !pcre2_pattern_info(pat, PCRE2_INFO_NAMEENTRYSIZE, &nsize) + && !pcre2_pattern_info(pat, PCRE2_INFO_NAMETABLE, &ntable)) + { + char **hash, **hashptr; + uint32_t nidx; + hashptr = hash = (char **)zshcalloc((ncount+1)*2*sizeof(char *)); + for (nidx = 0; nidx < ncount; nidx++) { + vec_off = (ntable[nsize * nidx] << 9) + 2 * ntable[nsize * nidx + 1]; + /* would metafy the key but pcre limits characters in the name */ + *hashptr++ = ztrdup((char *) ntable + nsize * nidx + 2); + *hashptr++ = metafy(arg + ovec[vec_off], + ovec[vec_off+1]-ovec[vec_off], META_DUP); + } + sethparam(namedassoc, hash); + } + if (want_begin_end) { /* * cond-infix rather than builtin; also not bash; so we set a bunch @@ -286,6 +306,7 @@ bin_pcre_match(char *nam, char **args, O char *matched_portion = NULL; char *plaintext = NULL; char *receptacle = NULL; + char *named = ".pcre.match"; int return_value = 1; /* The subject length and offset start are both int values in pcre_exec */ int subject_len; @@ -305,6 +326,9 @@ bin_pcre_match(char *nam, char **args, O if(OPT_HASARG(ops,c='v')) { matched_portion = OPT_ARG(ops,c); } + if (OPT_HASARG(ops, c='A')) { + named = OPT_ARG(ops, c); + } if(OPT_HASARG(ops,c='n')) { /* The offset position to start the search, in bytes. */ if ((offset_start = getposint(OPT_ARG(ops,c), nam)) < 0) return 1; @@ -326,8 +350,8 @@ bin_pcre_match(char *nam, char **args, O if (ret==0) return_value = 0; else if (ret == PCRE2_ERROR_NOMATCH) /* no match */; else if (ret>0) { - zpcre_get_substrings(plaintext, pcre_mdata, ret, matched_portion, receptacle, - want_offset_pair, 0, 0); + zpcre_get_substrings(pcre_pattern, plaintext, pcre_mdata, ret, matched_portion, + receptacle, named, want_offset_pair, 0, 0); return_value = 0; } else { @@ -405,9 +429,8 @@ cond_pcre_match(char **a, int id) break; } else if (r>0) { - zpcre_get_substrings(lhstr_plain, pcre_mdata, r, svar, avar, 0, - isset(BASHREMATCH), - !isset(BASHREMATCH)); + zpcre_get_substrings(pcre_pat, lhstr_plain, pcre_mdata, r, svar, avar, + ".pcre.match", 0, isset(BASHREMATCH), !isset(BASHREMATCH)); return_value = 1; break; } @@ -443,7 +466,7 @@ static struct conddef cotab[] = { static struct builtin bintab[] = { BUILTIN("pcre_compile", 0, bin_pcre_compile, 1, 1, 0, "aimxs", NULL), - BUILTIN("pcre_match", 0, bin_pcre_match, 1, 1, 0, "a:v:n:b", NULL), + BUILTIN("pcre_match", 0, bin_pcre_match, 1, 1, 0, "A:a:v:n:b", NULL), BUILTIN("pcre_study", 0, bin_pcre_study, 0, 0, 0, NULL, NULL) }; --- a/Test/V07pcre.ztst +++ b/Test/V07pcre.ztst @@ -194,3 +194,17 @@ [[ abc =~ 'a(d*)bc' ]] && print "$#MATCH; $#match; ${#match[1]}" 0:empty capture >3; 1; 0 + + [[ category/name-12345 =~ '(?x)^ + (? [^/]* ) / + (? + (? \w+ ) - + (? \d+ ))$' ]] + typeset -p1 .pcre.match +0:named captures +>typeset -g -A .pcre.match=( +> [category]=category +> [name]=name +> [package]=name-12345 +> [version]=12345 +>)