From 63c18a8309ee4c333923de904a29dd7e81d2cdc7 Mon Sep 17 00:00:00 2001 From: Denis Vlasenko Date: Fri, 26 Dec 2008 16:09:46 +0000 Subject: regex: remove useless casts on allocations; remove old-style-C function params declarations; change re_comp_buf from struct to pointer (more static build friendly) text data bss dec hex filename - 514952 2731 15424 533107 82273 lib/libuClibc-0.9.30-svn.so + 515011 2731 15396 533138 82292 lib/libuClibc-0.9.30-svn.so --- libc/misc/regex/regcomp.c | 158 ++++++++++++++++++++------------------- libc/misc/regex/regex_internal.c | 4 +- libc/misc/regex/regex_old.c | 7 +- libc/misc/regex/regexec.c | 19 ++--- 4 files changed, 96 insertions(+), 92 deletions(-) (limited to 'libc/misc') diff --git a/libc/misc/regex/regcomp.c b/libc/misc/regex/regcomp.c index 8aa152df9..bacba36c6 100644 --- a/libc/misc/regex/regcomp.c +++ b/libc/misc/regex/regcomp.c @@ -26,7 +26,7 @@ static void re_compile_fastmap_iter (regex_t *bufp, static reg_errcode_t init_dfa (re_dfa_t *dfa, size_t pat_len); #ifdef RE_ENABLE_I18N static void free_charset (re_charset_t *cset); -#endif /* RE_ENABLE_I18N */ +#endif static void free_workarea_compile (regex_t *preg); static reg_errcode_t create_initial_state (re_dfa_t *dfa); #ifdef RE_ENABLE_I18N @@ -211,10 +211,9 @@ static const uint16_t __re_error_msgid_idx[] = are set in BUFP on entry. */ const char * -re_compile_pattern (pattern, length, bufp) - const char *pattern; - size_t length; - struct re_pattern_buffer *bufp; +re_compile_pattern (const char *pattern, + size_t length, + struct re_pattern_buffer *bufp) { reg_errcode_t ret; @@ -249,8 +248,7 @@ reg_syntax_t re_syntax_options; defined in regex.h. We return the old syntax. */ reg_syntax_t -re_set_syntax (syntax) - reg_syntax_t syntax; +re_set_syntax (reg_syntax_t syntax) { reg_syntax_t ret = re_syntax_options; @@ -259,8 +257,7 @@ re_set_syntax (syntax) } int -re_compile_fastmap (bufp) - struct re_pattern_buffer *bufp; +re_compile_fastmap (struct re_pattern_buffer *bufp) { re_dfa_t *dfa = (re_dfa_t *) bufp->buffer; char *fastmap = bufp->fastmap; @@ -388,7 +385,7 @@ re_compile_fastmap_iter (regex_t *bufp, const re_dfastate_t *init_state, else if (type == OP_PERIOD #ifdef RE_ENABLE_I18N || type == OP_UTF8_PERIOD -#endif /* RE_ENABLE_I18N */ +#endif || type == END_OF_RE) { memset (fastmap, '\1', sizeof (char) * SBC_MAX); @@ -436,10 +433,9 @@ re_compile_fastmap_iter (regex_t *bufp, const re_dfastate_t *init_state, the return codes and their meanings.) */ int -regcomp (preg, pattern, cflags) - regex_t *__restrict preg; - const char *__restrict pattern; - int cflags; +regcomp (regex_t *__restrict preg, + const char *__restrict pattern, + int cflags) { reg_errcode_t ret; reg_syntax_t syntax = ((cflags & REG_EXTENDED) ? RE_SYNTAX_POSIX_EXTENDED @@ -495,11 +491,10 @@ regcomp (preg, pattern, cflags) from either regcomp or regexec. We don't use PREG here. */ size_t -regerror (errcode, preg, errbuf, errbuf_size) - int errcode; - const regex_t *__restrict preg; - char *__restrict errbuf; - size_t errbuf_size; +regerror (int errcode, + const regex_t *__restrict preg, + char *__restrict errbuf, + size_t errbuf_size) { const char *msg; size_t msg_size; @@ -596,8 +591,7 @@ free_dfa_content (re_dfa_t *dfa) /* Free dynamically allocated space used by PREG. */ void -regfree (preg) - regex_t *preg; +regfree (regex_t *preg) { re_dfa_t *dfa = (re_dfa_t *) preg->buffer; if (BE (dfa != NULL, 1)) @@ -619,7 +613,7 @@ libc_hidden_def(regfree) #if defined _REGEX_RE_COMP || defined _LIBC || defined __UCLIBC__ /* BSD has one and only one pattern buffer. */ -static struct re_pattern_buffer re_comp_buf; +static struct re_pattern_buffer *re_comp_buf; char * # if defined _LIBC || defined __UCLIBC__ @@ -628,47 +622,59 @@ char * regcomp/regexec above without link errors. */ weak_function # endif -re_comp (s) - const char *s; +re_comp (const char *s) { reg_errcode_t ret; - char *fastmap; - if (!s) + /* "If re_comp() is passed NULL or a null string, it returns + * without changing the currently compiled regular expression." */ + if (!s || !s[0]) { - if (!re_comp_buf.buffer) + if (!re_comp_buf) return gettext ("No previous regular expression"); - return 0; + return NULL; + } + + if (!re_comp_buf) + { + re_comp_buf = calloc (1, sizeof(*re_comp_buf)); + if (!re_comp_buf) + { + ret = REG_ESPACE; + goto err; + } } - if (re_comp_buf.buffer) + if (re_comp_buf->buffer) { - fastmap = re_comp_buf.fastmap; - re_comp_buf.fastmap = NULL; - regfree (&re_comp_buf); - memset (&re_comp_buf, '\0', sizeof (re_comp_buf)); - re_comp_buf.fastmap = fastmap; + regfree (re_comp_buf); + memset (re_comp_buf, '\0', sizeof(*re_comp_buf)); } - if (re_comp_buf.fastmap == NULL) + if (re_comp_buf->fastmap == NULL) { - re_comp_buf.fastmap = (char *) malloc (SBC_MAX); - if (re_comp_buf.fastmap == NULL) - return (char *) gettext (__re_error_msgid - + __re_error_msgid_idx[(int) REG_ESPACE]); + re_comp_buf->fastmap = malloc (SBC_MAX); + if (re_comp_buf->fastmap == NULL) + { + ret = REG_ESPACE; + goto err; + } } /* Since `re_exec' always passes NULL for the `regs' argument, we don't need to initialize the pattern buffer fields which affect it. */ /* Match anchors at newlines. */ - re_comp_buf.newline_anchor = 1; + re_comp_buf->newline_anchor = 1; - ret = re_compile_internal (&re_comp_buf, s, strlen (s), re_syntax_options); + ret = re_compile_internal (re_comp_buf, s, strlen (s), re_syntax_options); if (!ret) return NULL; + free (re_comp_buf); + re_comp_buf = NULL; + err: /* Yes, we're discarding `const' here if !HAVE_LIBINTL. */ return (char *) gettext (__re_error_msgid + __re_error_msgid_idx[(int) ret]); } @@ -676,7 +682,9 @@ re_comp (s) #ifdef _LIBC libc_freeres_fn (free_mem) { - regfree (&re_comp_buf); + regfree (re_comp_buf); + free (re_comp_buf); + re_comp_buf = NULL; } #endif @@ -861,7 +869,7 @@ init_dfa (re_dfa_t *dfa, size_t pat_len) { int i, j, ch; - dfa->sb_char = (re_bitset_ptr_t) calloc (sizeof (bitset_t), 1); + dfa->sb_char = calloc (sizeof (bitset_t), 1); if (BE (dfa->sb_char == NULL, 0)) return REG_ESPACE; @@ -1955,7 +1963,7 @@ peek_token_bracket (re_token_t *token, re_string_t *input, reg_syntax_t syntax) token->type = CHARACTER; return 1; } -#endif /* RE_ENABLE_I18N */ +#endif if (c == '\\' && (syntax & RE_BACKSLASH_ESCAPE_IN_LISTS) && re_string_cur_idx (input) + 1 < re_string_length (input)) @@ -2546,10 +2554,10 @@ internal_function # ifdef RE_ENABLE_I18N build_range_exp (bitset_t sbcset, re_charset_t *mbcset, int *range_alloc, bracket_elem_t *start_elem, bracket_elem_t *end_elem) -# else /* not RE_ENABLE_I18N */ +# else build_range_exp (bitset_t sbcset, bracket_elem_t *start_elem, bracket_elem_t *end_elem) -# endif /* not RE_ENABLE_I18N */ +# endif { unsigned int start_ch, end_ch; /* Equivalence Classes and Character Classes can't be a range start/end. */ @@ -2667,9 +2675,9 @@ internal_function # ifdef RE_ENABLE_I18N build_collating_symbol (bitset_t sbcset, re_charset_t *mbcset, int *coll_sym_alloc, const unsigned char *name) -# else /* not RE_ENABLE_I18N */ +# else build_collating_symbol (bitset_t sbcset, const unsigned char *name) -# endif /* not RE_ENABLE_I18N */ +# endif { size_t name_len = strlen ((const char *) name); if (BE (name_len != 1, 0)) @@ -2958,7 +2966,7 @@ parse_bracket_exp (re_string_t *regexp, re_dfa_t *dfa, re_token_t *token, re_charset_t *mbcset; int coll_sym_alloc = 0, range_alloc = 0, mbchar_alloc = 0; int equiv_class_alloc = 0, char_class_alloc = 0; -#endif /* not RE_ENABLE_I18N */ +#endif int non_match = 0; bin_tree_t *work_tree; int token_len; @@ -2980,15 +2988,15 @@ parse_bracket_exp (re_string_t *regexp, re_dfa_t *dfa, re_token_t *token, _NL_COLLATE_SYMB_EXTRAMB); } #endif - sbcset = (re_bitset_ptr_t) calloc (sizeof (bitset_t), 1); + sbcset = calloc (sizeof (bitset_t), 1); #ifdef RE_ENABLE_I18N - mbcset = (re_charset_t *) calloc (sizeof (re_charset_t), 1); -#endif /* RE_ENABLE_I18N */ + mbcset = calloc (sizeof (re_charset_t), 1); +#endif #ifdef RE_ENABLE_I18N if (BE (sbcset == NULL || mbcset == NULL, 0)) #else if (BE (sbcset == NULL, 0)) -#endif /* RE_ENABLE_I18N */ +#endif { *err = REG_ESPACE; return NULL; @@ -3004,7 +3012,7 @@ parse_bracket_exp (re_string_t *regexp, re_dfa_t *dfa, re_token_t *token, { #ifdef RE_ENABLE_I18N mbcset->non_match = 1; -#endif /* not RE_ENABLE_I18N */ +#endif non_match = 1; if (syntax & RE_HAT_LISTS_NOT_NEWLINE) bitset_set (sbcset, '\0'); @@ -3095,7 +3103,7 @@ parse_bracket_exp (re_string_t *regexp, re_dfa_t *dfa, re_token_t *token, # else *err = build_range_exp (sbcset, &start_elem, &end_elem); # endif -#endif /* RE_ENABLE_I18N */ +#endif if (BE (*err != REG_NOERROR, 0)) goto parse_bracket_exp_free_return; } @@ -3129,7 +3137,7 @@ parse_bracket_exp (re_string_t *regexp, re_dfa_t *dfa, re_token_t *token, *err = build_equiv_class (sbcset, #ifdef RE_ENABLE_I18N mbcset, &equiv_class_alloc, -#endif /* RE_ENABLE_I18N */ +#endif start_elem.opr.name); if (BE (*err != REG_NOERROR, 0)) goto parse_bracket_exp_free_return; @@ -3138,7 +3146,7 @@ parse_bracket_exp (re_string_t *regexp, re_dfa_t *dfa, re_token_t *token, *err = build_collating_symbol (sbcset, #ifdef RE_ENABLE_I18N mbcset, &coll_sym_alloc, -#endif /* RE_ENABLE_I18N */ +#endif start_elem.opr.name); if (BE (*err != REG_NOERROR, 0)) goto parse_bracket_exp_free_return; @@ -3147,7 +3155,7 @@ parse_bracket_exp (re_string_t *regexp, re_dfa_t *dfa, re_token_t *token, *err = build_charclass (regexp->trans, sbcset, #ifdef RE_ENABLE_I18N mbcset, &char_class_alloc, -#endif /* RE_ENABLE_I18N */ +#endif start_elem.opr.name, syntax); if (BE (*err != REG_NOERROR, 0)) goto parse_bracket_exp_free_return; @@ -3236,7 +3244,7 @@ parse_bracket_exp (re_string_t *regexp, re_dfa_t *dfa, re_token_t *token, re_free (sbcset); #ifdef RE_ENABLE_I18N free_charset (mbcset); -#endif /* RE_ENABLE_I18N */ +#endif return NULL; } @@ -3333,9 +3341,9 @@ static reg_errcode_t #ifdef RE_ENABLE_I18N build_equiv_class (bitset_t sbcset, re_charset_t *mbcset, int *equiv_class_alloc, const unsigned char *name) -#else /* not RE_ENABLE_I18N */ +#else build_equiv_class (bitset_t sbcset, const unsigned char *name) -#endif /* not RE_ENABLE_I18N */ +#endif { #ifdef _LIBC uint32_t nrules = _NL_CURRENT_WORD (LC_COLLATE, _NL_COLLATE_NRULES); @@ -3426,10 +3434,10 @@ static reg_errcode_t build_charclass (RE_TRANSLATE_TYPE trans, bitset_t sbcset, re_charset_t *mbcset, int *char_class_alloc, const unsigned char *class_name, reg_syntax_t syntax) -#else /* not RE_ENABLE_I18N */ +#else build_charclass (RE_TRANSLATE_TYPE trans, bitset_t sbcset, const unsigned char *class_name, reg_syntax_t syntax) -#endif /* not RE_ENABLE_I18N */ +#endif { int i; const char *name = (const char *) class_name; @@ -3514,21 +3522,21 @@ build_charclass_op (re_dfa_t *dfa, RE_TRANSLATE_TYPE trans, #ifdef RE_ENABLE_I18N re_charset_t *mbcset; int alloc = 0; -#endif /* not RE_ENABLE_I18N */ +#endif reg_errcode_t ret; re_token_t br_token; bin_tree_t *tree; - sbcset = (re_bitset_ptr_t) calloc (sizeof (bitset_t), 1); + sbcset = calloc (sizeof (bitset_t), 1); #ifdef RE_ENABLE_I18N - mbcset = (re_charset_t *) calloc (sizeof (re_charset_t), 1); -#endif /* RE_ENABLE_I18N */ + mbcset = calloc (sizeof (re_charset_t), 1); +#endif #ifdef RE_ENABLE_I18N if (BE (sbcset == NULL || mbcset == NULL, 0)) -#else /* not RE_ENABLE_I18N */ +#else if (BE (sbcset == NULL, 0)) -#endif /* not RE_ENABLE_I18N */ +#endif { *err = REG_ESPACE; return NULL; @@ -3542,14 +3550,14 @@ build_charclass_op (re_dfa_t *dfa, RE_TRANSLATE_TYPE trans, bitset_set(cset->sbcset, '\0'); */ mbcset->non_match = 1; -#endif /* not RE_ENABLE_I18N */ +#endif } /* We don't care the syntax in this case. */ ret = build_charclass (trans, sbcset, #ifdef RE_ENABLE_I18N mbcset, &alloc, -#endif /* RE_ENABLE_I18N */ +#endif class_name, 0); if (BE (ret != REG_NOERROR, 0)) @@ -3557,7 +3565,7 @@ build_charclass_op (re_dfa_t *dfa, RE_TRANSLATE_TYPE trans, re_free (sbcset); #ifdef RE_ENABLE_I18N free_charset (mbcset); -#endif /* RE_ENABLE_I18N */ +#endif *err = ret; return NULL; } @@ -3605,13 +3613,13 @@ build_charclass_op (re_dfa_t *dfa, RE_TRANSLATE_TYPE trans, } #else /* not RE_ENABLE_I18N */ return tree; -#endif /* not RE_ENABLE_I18N */ +#endif build_word_op_espace: re_free (sbcset); #ifdef RE_ENABLE_I18N free_charset (mbcset); -#endif /* RE_ENABLE_I18N */ +#endif *err = REG_ESPACE; return NULL; } @@ -3726,7 +3734,7 @@ free_token (re_token_t *node) if (node->type == COMPLEX_BRACKET && node->duplicated == 0) free_charset (node->opr.mbcset); else -#endif /* RE_ENABLE_I18N */ +#endif if (node->type == SIMPLE_BRACKET && node->duplicated == 0) re_free (node->opr.sbcset); } diff --git a/libc/misc/regex/regex_internal.c b/libc/misc/regex/regex_internal.c index 11b1be7c2..1c94c979a 100644 --- a/libc/misc/regex/regex_internal.c +++ b/libc/misc/regex/regex_internal.c @@ -1529,7 +1529,7 @@ create_ci_newstate (const re_dfa_t *dfa, const re_node_set *nodes, reg_errcode_t err; re_dfastate_t *newstate; - newstate = (re_dfastate_t *) calloc (sizeof (re_dfastate_t), 1); + newstate = calloc (sizeof (re_dfastate_t), 1); if (BE (newstate == NULL, 0)) return NULL; err = re_node_set_init_copy (&newstate->nodes, nodes); @@ -1579,7 +1579,7 @@ create_cd_newstate (const re_dfa_t *dfa, const re_node_set *nodes, reg_errcode_t err; re_dfastate_t *newstate; - newstate = (re_dfastate_t *) calloc (sizeof (re_dfastate_t), 1); + newstate = calloc (sizeof (re_dfastate_t), 1); if (BE (newstate == NULL, 0)) return NULL; err = re_node_set_init_copy (&newstate->nodes, nodes); diff --git a/libc/misc/regex/regex_old.c b/libc/misc/regex/regex_old.c index be93cc028..af1af50a6 100644 --- a/libc/misc/regex/regex_old.c +++ b/libc/misc/regex/regex_old.c @@ -7832,10 +7832,9 @@ PREFIX(bcmp_translate) ( We call regex_compile to do the actual compilation. */ const char * -re_compile_pattern ( - const char *pattern, - size_t length, - struct re_pattern_buffer *bufp) +re_compile_pattern (const char *pattern, + size_t length, + struct re_pattern_buffer *bufp) { reg_errcode_t ret; diff --git a/libc/misc/regex/regexec.c b/libc/misc/regex/regexec.c index 17d58a602..1e7e1c3b7 100644 --- a/libc/misc/regex/regexec.c +++ b/libc/misc/regex/regexec.c @@ -79,7 +79,7 @@ static int sift_states_iter_mb (const re_match_context_t *mctx, re_sift_context_t *sctx, int node_idx, int str_idx, int max_str_idx) internal_function; -#endif /* RE_ENABLE_I18N */ +#endif static reg_errcode_t sift_states_backward (const re_match_context_t *mctx, re_sift_context_t *sctx) internal_function; @@ -144,7 +144,7 @@ static re_dfastate_t *transit_state_sb (reg_errcode_t *err, static reg_errcode_t transit_state_mb (re_match_context_t *mctx, re_dfastate_t *pstate) internal_function; -#endif /* RE_ENABLE_I18N */ +#endif static reg_errcode_t transit_state_bkref (re_match_context_t *mctx, const re_node_set *nodes) internal_function; @@ -189,8 +189,8 @@ static int check_node_accept_bytes (const re_dfa_t *dfa, int node_idx, static unsigned int find_collation_sequence_value (const unsigned char *mbs, size_t name_len) internal_function; -# endif /* _LIBC */ -#endif /* RE_ENABLE_I18N */ +# endif +#endif static int group_nodes_into_DFAstates (const re_dfa_t *dfa, const re_dfastate_t *state, re_node_set *states_node, @@ -564,10 +564,9 @@ int # if defined _LIBC || defined __UCLIBC__ weak_function # endif -re_exec (s) - const char *s; +re_exec (const char *s) { - return 0 == regexec (&re_comp_buf, s, 0, NULL, 0); + return 0 == regexec (re_comp_buf, s, 0, NULL, 0); } #endif /* _REGEX_RE_COMP */ @@ -3395,8 +3394,7 @@ out_free: character, or we are in a single-byte character set so we can discern by looking at the character code: allocate a 256-entry transition table. */ - trtable = state->trtable = - (re_dfastate_t **) calloc (sizeof (re_dfastate_t *), SBC_MAX); + trtable = state->trtable = calloc (sizeof (re_dfastate_t *), SBC_MAX); if (BE (trtable == NULL, 0)) goto out_free; @@ -3426,8 +3424,7 @@ out_free: by looking at the character code: build two 256-entry transition tables, one starting at trtable[0] and one starting at trtable[SBC_MAX]. */ - trtable = state->word_trtable = - (re_dfastate_t **) calloc (sizeof (re_dfastate_t *), 2 * SBC_MAX); + trtable = state->word_trtable = calloc (sizeof (re_dfastate_t *), 2 * SBC_MAX); if (BE (trtable == NULL, 0)) goto out_free; -- cgit v1.2.3