/* * Copyright (c) 2009-2016 Petri Lehtinen * * Jansson is free software; you can redistribute it and/or modify * it under the terms of the MIT license. See LICENSE for details. */ #ifndef _GNU_SOURCE #define _GNU_SOURCE #endif #include "jansson_private.h" #include #include #include #include #include #include #ifdef HAVE_UNISTD_H #include #endif #include "jansson.h" #include "strbuffer.h" #include "utf.h" #define STREAM_STATE_OK 0 #define STREAM_STATE_EOF -1 #define STREAM_STATE_ERROR -2 #define TOKEN_INVALID -1 #define TOKEN_EOF 0 #define TOKEN_STRING 256 #define TOKEN_INTEGER 257 #define TOKEN_REAL 258 #define TOKEN_TRUE 259 #define TOKEN_FALSE 260 #define TOKEN_NULL 261 /* Locale independent versions of isxxx() functions */ #define l_isupper(c) ('A' <= (c) && (c) <= 'Z') #define l_islower(c) ('a' <= (c) && (c) <= 'z') #define l_isalpha(c) (l_isupper(c) || l_islower(c)) #define l_isdigit(c) ('0' <= (c) && (c) <= '9') #define l_isxdigit(c) \ (l_isdigit(c) || ('A' <= (c) && (c) <= 'F') || ('a' <= (c) && (c) <= 'f')) /* Read one byte from stream, convert to unsigned char, then int, and return. return EOF on end of file. This corresponds to the behaviour of fgetc(). */ typedef int (*get_func)(void *data); typedef struct { get_func get; void *data; char buffer[5]; size_t buffer_pos; int state; int line; int column, last_column; size_t position; } stream_t; typedef struct { stream_t stream; strbuffer_t saved_text; size_t flags; size_t depth; int token; union { struct { char *val; size_t len; } string; json_int_t integer; double real; } value; } lex_t; #define stream_to_lex(stream) container_of(stream, lex_t, stream) /*** error reporting ***/ static void error_set(json_error_t *error, const lex_t *lex, enum json_error_code code, const char *msg, ...) { va_list ap; char msg_text[JSON_ERROR_TEXT_LENGTH]; char msg_with_context[JSON_ERROR_TEXT_LENGTH]; int line = -1, col = -1; size_t pos = 0; const char *result = msg_text; if (!error) return; va_start(ap, msg); vsnprintf(msg_text, JSON_ERROR_TEXT_LENGTH, msg, ap); msg_text[JSON_ERROR_TEXT_LENGTH - 1] = '\0'; va_end(ap); if (lex) { const char *saved_text = strbuffer_value(&lex->saved_text); line = lex->stream.line; col = lex->stream.column; pos = lex->stream.position; if (saved_text && saved_text[0]) { if (lex->saved_text.length <= 20) { snprintf(msg_with_context, JSON_ERROR_TEXT_LENGTH, "%s near '%s'", msg_text, saved_text); msg_with_context[JSON_ERROR_TEXT_LENGTH - 1] = '\0'; result = msg_with_context; } } else { if (code == json_error_invalid_syntax) { /* More specific error code for premature end of file. */ code = json_error_premature_end_of_input; } if (lex->stream.state == STREAM_STATE_ERROR) { /* No context for UTF-8 decoding errors */ result = msg_text; } else { snprintf(msg_with_context, JSON_ERROR_TEXT_LENGTH, "%s near end of file", msg_text); msg_with_context[JSON_ERROR_TEXT_LENGTH - 1] = '\0'; result = msg_with_context; } } } jsonp_error_set(error, line, col, pos, code, "%s", result); } /*** lexical analyzer ***/ static void stream_init(stream_t *stream, get_func get, void *data) { stream->get = get; stream->data = data; stream->buffer[0] = '\0'; stream->buffer_pos = 0; stream->state = STREAM_STATE_OK; stream->line = 1; stream->column = 0; stream->position = 0; } static int stream_get(stream_t *stream, json_error_t *error) { int c; if (stream->state != STREAM_STATE_OK) return stream->state; if (!stream->buffer[stream->buffer_pos]) { c = stream->get(stream->data); if (c == EOF) { stream->state = STREAM_STATE_EOF; return STREAM_STATE_EOF; } stream->buffer[0] = c; stream->buffer_pos = 0; if (0x80 <= c && c <= 0xFF) { /* multi-byte UTF-8 sequence */ size_t i, count; count = utf8_check_first(c); if (!count) goto out; assert(count >= 2); for (i = 1; i < count; i++) stream->buffer[i] = stream->get(stream->data); if (!utf8_check_full(stream->buffer, count, NULL)) goto out; stream->buffer[count] = '\0'; } else stream->buffer[1] = '\0'; } c = stream->buffer[stream->buffer_pos++]; stream->position++; if (c == '\n') { stream->line++; stream->last_column = stream->column; stream->column = 0; } else if (utf8_check_first(c)) { /* track the Unicode character column, so increment only if this is the first character of a UTF-8 sequence */ stream->column++; } return c; out: stream->state = STREAM_STATE_ERROR; error_set(error, stream_to_lex(stream), json_error_invalid_utf8, "unable to decode byte 0x%x", c); return STREAM_STATE_ERROR; } static void stream_unget(stream_t *stream, int c) { if (c == STREAM_STATE_EOF || c == STREAM_STATE_ERROR) return; stream->position--; if (c == '\n') { stream->line--; stream->column = stream->last_column; } else if (utf8_check_first(c)) stream->column--; assert(stream->buffer_pos > 0); stream->buffer_pos--; assert(stream->buffer[stream->buffer_pos] == c); } static int lex_get(lex_t *lex, json_error_t *error) { return stream_get(&lex->stream, error); } static void lex_save(lex_t *lex, int c) { strbuffer_append_byte(&lex->saved_text, c); } static int lex_get_save(lex_t *lex, json_error_t *error) { int c = stream_get(&lex->stream, error); if (c != STREAM_STATE_EOF && c != STREAM_STATE_ERROR) lex_save(lex, c); return c; } static void lex_unget(lex_t *lex, int c) { stream_unget(&lex->stream, c); } static void lex_unget_unsave(lex_t *lex, int c) { if (c != STREAM_STATE_EOF && c != STREAM_STATE_ERROR) { /* Since we treat warnings as errors, when assertions are turned * off the "d" variable would be set but never used. Which is * treated as an error by GCC. */ #ifndef NDEBUG char d; #endif stream_unget(&lex->stream, c); #ifndef NDEBUG d = #endif strbuffer_pop(&lex->saved_text); assert(c == d); } } static void lex_save_cached(lex_t *lex) { while (lex->stream.buffer[lex->stream.buffer_pos] != '\0') { lex_save(lex, lex->stream.buffer[lex->stream.buffer_pos]); lex->stream.buffer_pos++; lex->stream.position++; } } static void lex_free_string(lex_t *lex) { jsonp_free(lex->value.string.val); lex->value.string.val = NULL; lex->value.string.len = 0; } /* assumes that str points to 'u' plus at least 4 valid hex digits */ static int32_t decode_unicode_escape(const char *str) { int i; int32_t value = 0; assert(str[0] == 'u'); for (i = 1; i <= 4; i++) { char c = str[i]; value <<= 4; if (l_isdigit(c)) value += c - '0'; else if (l_islower(c)) value += c - 'a' + 10; else if (l_isupper(c)) value += c - 'A' + 10; else return -1; } return value; } static void lex_scan_string(lex_t *lex, json_error_t *error) { int c; const char *p; char *t; int i; lex->value.string.val = NULL; lex->token = TOKEN_INVALID; c = lex_get_save(lex, error); while (c != '"') { if (c == STREAM_STATE_ERROR) goto out; else if (c == STREAM_STATE_EOF) { error_set(error, lex, json_error_premature_end_of_input, "premature end of input"); goto out; } else if (0 <= c && c <= 0x1F) { /* control character */ lex_unget_unsave(lex, c); if (c == '\n') error_set(error, lex, json_error_invalid_syntax, "unexpected newline"); else error_set(error, lex, json_error_invalid_syntax, "control character 0x%x", c); goto out; } else if (c == '\\') { c = lex_get_save(lex, error); if (c == 'u') { c = lex_get_save(lex, error); for (i = 0; i < 4; i++) { if (!l_isxdigit(c)) { error_set(error, lex, json_error_invalid_syntax, "invalid escape"); goto out; } c = lex_get_save(lex, error); } } else if (c == '"' || c == '\\' || c == '/' || c == 'b' || c == 'f' || c == 'n' || c == 'r' || c == 't') c = lex_get_save(lex, error); else { error_set(error, lex, json_error_invalid_syntax, "invalid escape"); goto out; } } else c = lex_get_save(lex, error); } /* the actual value is at most of the same length as the source string, because: - shortcut escapes (e.g. "\t") (length 2) are converted to 1 byte - a single \uXXXX escape (length 6) is converted to at most 3 bytes - two \uXXXX escapes (length 12) forming an UTF-16 surrogate pair are converted to 4 bytes */ t = jsonp_malloc(lex->saved_text.length + 1); if (!t) { /* this is not very nice, since TOKEN_INVALID is returned */ goto out; } lex->value.string.val = t; /* + 1 to skip the " */ p = strbuffer_value(&lex->saved_text) + 1; while (*p != '"') { if (*p == '\\') { p++; if (*p == 'u') { size_t length; int32_t value; value = decode_unicode_escape(p); if (value < 0) { error_set(error, lex, json_error_invalid_syntax, "invalid Unicode escape '%.6s'", p - 1); goto out; } p += 5; if (0xD800 <= value && value <= 0xDBFF) { /* surrogate pair */ if (*p == '\\' && *(p + 1) == 'u') { int32_t value2 = decode_unicode_escape(++p); if (value2 < 0) { error_set(error, lex, json_error_invalid_syntax, "invalid Unicode escape '%.6s'", p - 1); goto out; } p += 5; if (0xDC00 <= value2 && value2 <= 0xDFFF) { /* valid second surrogate */ value = ((value - 0xD800) << 10) + (value2 - 0xDC00) + 0x10000; } else { /* invalid second surrogate */ error_set(error, lex, json_error_invalid_syntax, "invalid Unicode '\\u%04X\\u%04X'", value, value2); goto out; } } else { /* no second surrogate */ error_set(error, lex, json_error_invalid_syntax, "invalid Unicode '\\u%04X'", value); goto out; } } else if (0xDC00 <= value && value <= 0xDFFF) { error_set(error, lex, json_error_invalid_syntax, "invalid Unicode '\\u%04X'", value); goto out; } if (utf8_encode(value, t, &length)) assert(0); t += length; } else { switch (*p) { case '"': case '\\': case '/': *t = *p; break; case 'b': *t = '\b'; break; case 'f': *t = '\f'; break; case 'n': *t = '\n'; break; case 'r': *t = '\r'; break; case 't': *t = '\t'; break; default: assert(0); } t++; p++; } } else *(t++) = *(p++); } *t = '\0'; lex->value.string.len = t - lex->value.string.val; lex->token = TOKEN_STRING; return; out: lex_free_string(lex); } #ifndef JANSSON_USING_CMAKE /* disabled if using cmake */ #if JSON_INTEGER_IS_LONG_LONG #ifdef _MSC_VER /* Microsoft Visual Studio */ #define json_strtoint _strtoi64 #else #define json_strtoint strtoll #endif #else #define json_strtoint strtol #endif #endif static int lex_scan_number(lex_t *lex, int c, json_error_t *error) { const char *saved_text; char *end; double doubleval; lex->token = TOKEN_INVALID; if (c == '-') c = lex_get_save(lex, error); if (c == '0') { c = lex_get_save(lex, error); if (l_isdigit(c)) { lex_unget_unsave(lex, c); goto out; } } else if (l_isdigit(c)) { do c = lex_get_save(lex, error); while (l_isdigit(c)); } else { lex_unget_unsave(lex, c); goto out; } if (!(lex->flags & JSON_DECODE_INT_AS_REAL) && c != '.' && c != 'E' && c != 'e') { json_int_t intval; lex_unget_unsave(lex, c); saved_text = strbuffer_value(&lex->saved_text); errno = 0; intval = json_strtoint(saved_text, &end, 10); if (errno == ERANGE) { if (intval < 0) error_set(error, lex, json_error_numeric_overflow, "too big negative integer"); else error_set(error, lex, json_error_numeric_overflow, "too big integer"); goto out; } assert(end == saved_text + lex->saved_text.length); lex->token = TOKEN_INTEGER; lex->value.integer = intval; return 0; } if (c == '.') { c = lex_get(lex, error); if (!l_isdigit(c)) { lex_unget(lex, c); goto out; } lex_save(lex, c); do c = lex_get_save(lex, error); while (l_isdigit(c)); } if (c == 'E' || c == 'e') { c = lex_get_save(lex, error); if (c == '+' || c == '-') c = lex_get_save(lex, error); if (!l_isdigit(c)) { lex_unget_unsave(lex, c); goto out; } do c = lex_get_save(lex, error); while (l_isdigit(c)); } lex_unget_unsave(lex, c); if (jsonp_strtod(&lex->saved_text, &doubleval)) { error_set(error, lex, json_error_numeric_overflow, "real number overflow"); goto out; } lex->token = TOKEN_REAL; lex->value.real = doubleval; return 0; out: return -1; } static int lex_scan(lex_t *lex, json_error_t *error) { int c; strbuffer_clear(&lex->saved_text); if (lex->token == TOKEN_STRING) lex_free_string(lex); do c = lex_get(lex, error); while (c == ' ' || c == '\t' || c == '\n' || c == '\r'); if (c == STREAM_STATE_EOF) { lex->token = TOKEN_EOF; goto out; } if (c == STREAM_STATE_ERROR) { lex->token = TOKEN_INVALID; goto out; } lex_save(lex, c); if (c == '{' || c == '}' || c == '[' || c == ']' || c == ':' || c == ',') lex->token = c; else if (c == '"') lex_scan_string(lex, error); else if (l_isdigit(c) || c == '-') { if (lex_scan_number(lex, c, error)) goto out; } else if (l_isalpha(c)) { /* eat up the whole identifier for clearer error messages */ const char *saved_text; do c = lex_get_save(lex, error); while (l_isalpha(c)); lex_unget_unsave(lex, c); saved_text = strbuffer_value(&lex->saved_text); if (strcmp(saved_text, "true") == 0) lex->token = TOKEN_TRUE; else if (strcmp(saved_text, "false") == 0) lex->token = TOKEN_FALSE; else if (strcmp(saved_text, "null") == 0) lex->token = TOKEN_NULL; else lex->token = TOKEN_INVALID; } else { /* save the rest of the input UTF-8 sequence to get an error message of valid UTF-8 */ lex_save_cached(lex); lex->token = TOKEN_INVALID; } out: return lex->token; } static char *lex_steal_string(lex_t *lex, size_t *out_len) { char *result = NULL; if (lex->token == TOKEN_STRING) { result = lex->value.string.val; *out_len = lex->value.string.len; lex->value.string.val = NULL; lex->value.string.len = 0; } return result; } static int lex_init(lex_t *lex, get_func get, size_t flags, void *data) { stream_init(&lex->stream, get, data); if (strbuffer_init(&lex->saved_text)) return -1; lex->flags = flags; lex->token = TOKEN_INVALID; return 0; } static void lex_close(lex_t *lex) { if (lex->token == TOKEN_STRING) lex_free_string(lex); strbuffer_close(&lex->saved_text); } /*** parser ***/ static json_t *parse_value(lex_t *lex, size_t flags, json_error_t *error); static json_t *parse_object(lex_t *lex, size_t flags, json_error_t *error) { json_t *object = json_object(); if (!object) return NULL; lex_scan(lex, error); if (lex->token == '}') return object; while (1) { char *key; size_t len; json_t *value; if (lex->token != TOKEN_STRING) { error_set(error, lex, json_error_invalid_syntax, "string or '}' expected"); goto error; } key = lex_steal_string(lex, &len); if (!key) return NULL; if (memchr(key, '\0', len)) { jsonp_free(key); error_set(error, lex, json_error_null_byte_in_key, "NUL byte in object key not supported"); goto error; } if (flags & JSON_REJECT_DUPLICATES) { if (json_object_getn(object, key, len)) { jsonp_free(key); error_set(error, lex, json_error_duplicate_key, "duplicate object key"); goto error; } } lex_scan(lex, error); if (lex->token != ':') { jsonp_free(key); error_set(error, lex, json_error_invalid_syntax, "':' expected"); goto error; } lex_scan(lex, error); value = parse_value(lex, flags, error); if (!value) { jsonp_free(key); goto error; } if (json_object_setn_new_nocheck(object, key, len, value)) { jsonp_free(key); goto error; } jsonp_free(key); lex_scan(lex, error); if (lex->token != ',') break; lex_scan(lex, error); } if (lex->token != '}') { error_set(error, lex, json_error_invalid_syntax, "'}' expected"); goto error; } return object; error: json_decref(object); return NULL; } static json_t *parse_array(lex_t *lex, size_t flags, json_error_t *error) { json_t *array = json_array(); if (!array) return NULL; lex_scan(lex, error); if (lex->token == ']') return array; while (lex->token) { json_t *elem = parse_value(lex, flags, error); if (!elem) goto error; if (json_array_append_new(array, elem)) { goto error; } lex_scan(lex, error); if (lex->token != ',') break; lex_scan(lex, error); } if (lex->token != ']') { error_set(error, lex, json_error_invalid_syntax, "']' expected"); goto error; } return array; error: json_decref(array); return NULL; } static json_t *parse_value(lex_t *lex, size_t flags, json_error_t *error) { json_t *json; lex->depth++; if (lex->depth > JSON_PARSER_MAX_DEPTH) { error_set(error, lex, json_error_stack_overflow, "maximum parsing depth reached"); return NULL; } switch (lex->token) { case TOKEN_STRING: { const char *value = lex->value.string.val; size_t len = lex->value.string.len; if (!(flags & JSON_ALLOW_NUL)) { if (memchr(value, '\0', len)) { error_set(error, lex, json_error_null_character, "\\u0000 is not allowed without JSON_ALLOW_NUL"); return NULL; } } json = jsonp_stringn_nocheck_own(value, len); lex->value.string.val = NULL; lex->value.string.len = 0; break; } case TOKEN_INTEGER: { json = json_integer(lex->value.integer); break; } case TOKEN_REAL: { json = json_real(lex->value.real); break; } case TOKEN_TRUE: json = json_true(); break; case TOKEN_FALSE: json = json_false(); break; case TOKEN_NULL: json = json_null(); break; case '{': json = parse_object(lex, flags, error); break; case '[': json = parse_array(lex, flags, error); break; case TOKEN_INVALID: error_set(error, lex, json_error_invalid_syntax, "invalid token"); return NULL; default: error_set(error, lex, json_error_invalid_syntax, "unexpected token"); return NULL; } if (!json) return NULL; lex->depth--; return json; } static json_t *parse_json(lex_t *lex, size_t flags, json_error_t *error) { json_t *result; lex->depth = 0; lex_scan(lex, error); if (!(flags & JSON_DECODE_ANY)) { if (lex->token != '[' && lex->token != '{') { error_set(error, lex, json_error_invalid_syntax, "'[' or '{' expected"); return NULL; } } result = parse_value(lex, flags, error); if (!result) return NULL; if (!(flags & JSON_DISABLE_EOF_CHECK)) { lex_scan(lex, error); if (lex->token != TOKEN_EOF) { error_set(error, lex, json_error_end_of_input_expected, "end of file expected"); json_decref(result); return NULL; } } if (error) { /* Save the position even though there was no error */ error->position = (int)lex->stream.position; } return result; } typedef struct { const char *data; size_t pos; } string_data_t; static int string_get(void *data) { char c; string_data_t *stream = (string_data_t *)data; c = stream->data[stream->pos]; if (c == '\0') return EOF; else { stream->pos++; return (unsigned char)c; } } json_t *json_loads(const char *string, size_t flags, json_error_t *error) { lex_t lex; json_t *result; string_data_t stream_data; jsonp_error_init(error, ""); if (string == NULL) { error_set(error, NULL, json_error_invalid_argument, "wrong arguments"); return NULL; } stream_data.data = string; stream_data.pos = 0; if (lex_init(&lex, string_get, flags, (void *)&stream_data)) return NULL; result = parse_json(&lex, flags, error); lex_close(&lex); return result; } typedef struct { const char *data; size_t len; size_t pos; } buffer_data_t; static int buffer_get(void *data) { char c; buffer_data_t *stream = data; if (stream->pos >= stream->len) return EOF; c = stream->data[stream->pos]; stream->pos++; return (unsigned char)c; } json_t *json_loadb(const char *buffer, size_t buflen, size_t flags, json_error_t *error) { lex_t lex; json_t *result; buffer_data_t stream_data; jsonp_error_init(error, ""); if (buffer == NULL) { error_set(error, NULL, json_error_invalid_argument, "wrong arguments"); return NULL; } stream_data.data = buffer; stream_data.pos = 0; stream_data.len = buflen; if (lex_init(&lex, buffer_get, flags, (void *)&stream_data)) return NULL; result = parse_json(&lex, flags, error); lex_close(&lex); return result; } json_t *json_loadf(FILE *input, size_t flags, json_error_t *error) { lex_t lex; const char *source; json_t *result; if (input == stdin) source = ""; else source = ""; jsonp_error_init(error, source); if (input == NULL) { error_set(error, NULL, json_error_invalid_argument, "wrong arguments"); return NULL; } if (lex_init(&lex, (get_func)fgetc, flags, input)) return NULL; result = parse_json(&lex, flags, error); lex_close(&lex); return result; } static int fd_get_func(int *fd) { #ifdef HAVE_UNISTD_H uint8_t c; if (read(*fd, &c, 1) == 1) return c; #endif return EOF; } json_t *json_loadfd(int input, size_t flags, json_error_t *error) { lex_t lex; const char *source; json_t *result; #ifdef HAVE_UNISTD_H if (input == STDIN_FILENO) source = ""; else #endif source = ""; jsonp_error_init(error, source); if (input < 0) { error_set(error, NULL, json_error_invalid_argument, "wrong arguments"); return NULL; } if (lex_init(&lex, (get_func)fd_get_func, flags, &input)) return NULL; result = parse_json(&lex, flags, error); lex_close(&lex); return result; } json_t *json_load_file(const char *path, size_t flags, json_error_t *error) { json_t *result; FILE *fp; jsonp_error_init(error, path); if (path == NULL) { error_set(error, NULL, json_error_invalid_argument, "wrong arguments"); return NULL; } fp = fopen(path, "rb"); if (!fp) { error_set(error, NULL, json_error_cannot_open_file, "unable to open %s: %s", path, strerror(errno)); return NULL; } result = json_loadf(fp, flags, error); fclose(fp); return result; } #define MAX_BUF_LEN 1024 typedef struct { char data[MAX_BUF_LEN]; size_t len; size_t pos; json_load_callback_t callback; void *arg; } callback_data_t; static int callback_get(void *data) { char c; callback_data_t *stream = data; if (stream->pos >= stream->len) { stream->pos = 0; stream->len = stream->callback(stream->data, MAX_BUF_LEN, stream->arg); if (stream->len == 0 || stream->len == (size_t)-1) return EOF; } c = stream->data[stream->pos]; stream->pos++; return (unsigned char)c; } json_t *json_load_callback(json_load_callback_t callback, void *arg, size_t flags, json_error_t *error) { lex_t lex; json_t *result; callback_data_t stream_data; memset(&stream_data, 0, sizeof(stream_data)); stream_data.callback = callback; stream_data.arg = arg; jsonp_error_init(error, ""); if (callback == NULL) { error_set(error, NULL, json_error_invalid_argument, "wrong arguments"); return NULL; } if (lex_init(&lex, (get_func)callback_get, flags, &stream_data)) return NULL; result = parse_json(&lex, flags, error); lex_close(&lex); return result; }