X-Git-Url: http://erislabs.net/gitweb/?a=blobdiff_plain;f=lib%2Fregcomp.c;h=6472ff6b2eb0d1b7707ddce159c58bd9da81d3fc;hb=5a21606b0d57082670bd060d169492c2f3a410f2;hp=0d6d160bf83904ca65f7c783cafcc847ac988256;hpb=8335a4d6c7b4448cd0bcb6d0bebf1d456bcfdb17;p=gnulib.git diff --git a/lib/regcomp.c b/lib/regcomp.c index 0d6d160bf..6472ff6b2 100644 --- a/lib/regcomp.c +++ b/lib/regcomp.c @@ -1,5 +1,6 @@ /* Extended regular expression matching and search library. - Copyright (C) 2002,2003,2004,2005,2006 Free Software Foundation, Inc. + Copyright (C) 2002,2003,2004,2005,2006,2007,2008,2009 + Free Software Foundation, Inc. This file is part of the GNU C Library. Contributed by Isamu Hasegawa . @@ -333,8 +334,8 @@ re_compile_fastmap_iter (regex_t *bufp, const re_dfastate_t *init_state, && dfa->nodes[node].mb_partial) *p++ = dfa->nodes[node].opr.c; memset (&state, '\0', sizeof (state)); - if (mbrtowc (&wc, (const char *) buf, p - buf, - &state) == p - buf + if (__mbrtowc (&wc, (const char *) buf, p - buf, + &state) == p - buf && (__wcrtomb ((char *) buf, towlower (wc), &state) != (size_t) -1)) re_set_fastmap (fastmap, false, buf[0]); @@ -356,45 +357,65 @@ re_compile_fastmap_iter (regex_t *bufp, const re_dfastate_t *init_state, #ifdef RE_ENABLE_I18N else if (type == COMPLEX_BRACKET) { - Idx i; re_charset_t *cset = dfa->nodes[node].opr.mbcset; - if (cset->non_match || cset->ncoll_syms || cset->nequiv_classes - || cset->nranges || cset->nchar_classes) - { + Idx i; + # ifdef _LIBC - if (_NL_CURRENT_WORD (LC_COLLATE, _NL_COLLATE_NRULES) != 0) + /* See if we have to try all bytes which start multiple collation + elements. + e.g. In da_DK, we want to catch 'a' since "aa" is a valid + collation element, and don't catch 'b' since 'b' is + the only collation element which starts from 'b' (and + it is caught by SIMPLE_BRACKET). */ + if (_NL_CURRENT_WORD (LC_COLLATE, _NL_COLLATE_NRULES) != 0 + && (cset->ncoll_syms || cset->nranges)) { - /* In this case we want to catch the bytes which are - the first byte of any collation elements. - e.g. In da_DK, we want to catch 'a' since "aa" - is a valid collation element, and don't catch - 'b' since 'b' is the only collation element - which starts from 'b'. */ const int32_t *table = (const int32_t *) _NL_CURRENT (LC_COLLATE, _NL_COLLATE_TABLEMB); for (i = 0; i < SBC_MAX; ++i) if (table[i] < 0) re_set_fastmap (fastmap, icase, i); } -# else - if (dfa->mb_cur_max > 1) - for (i = 0; i < SBC_MAX; ++i) - if (__btowc (i) == WEOF) - re_set_fastmap (fastmap, icase, i); -# endif /* not _LIBC */ +# endif /* _LIBC */ + + /* See if we have to start the match at all multibyte characters, + i.e. where we would not find an invalid sequence. This only + applies to multibyte character sets; for single byte character + sets, the SIMPLE_BRACKET again suffices. */ + if (dfa->mb_cur_max > 1 + && (cset->nchar_classes || cset->non_match +# ifdef _LIBC + || cset->nequiv_classes +# endif /* _LIBC */ + )) + { + unsigned char c = 0; + do + { + mbstate_t mbs; + memset (&mbs, 0, sizeof (mbs)); + if (__mbrtowc (NULL, (char *) &c, 1, &mbs) == (size_t) -2) + re_set_fastmap (fastmap, false, (int) c); + } + while (++c != 0); } - for (i = 0; i < cset->nmbchars; ++i) + + else { - char buf[256]; - mbstate_t state; - memset (&state, '\0', sizeof (state)); - if (__wcrtomb (buf, cset->mbchars[i], &state) != (size_t) -1) - re_set_fastmap (fastmap, icase, *(unsigned char *) buf); - if ((bufp->syntax & RE_ICASE) && dfa->mb_cur_max > 1) + /* ... Else catch all bytes which can start the mbchars. */ + for (i = 0; i < cset->nmbchars; ++i) { - if (__wcrtomb (buf, towlower (cset->mbchars[i]), &state) - != (size_t) -1) - re_set_fastmap (fastmap, false, *(unsigned char *) buf); + char buf[256]; + mbstate_t state; + memset (&state, '\0', sizeof (state)); + if (__wcrtomb (buf, cset->mbchars[i], &state) != (size_t) -1) + re_set_fastmap (fastmap, icase, *(unsigned char *) buf); + if ((bufp->syntax & RE_ICASE) && dfa->mb_cur_max > 1) + { + if (__wcrtomb (buf, towlower (cset->mbchars[i]), &state) + != (size_t) -1) + re_set_fastmap (fastmap, false, *(unsigned char *) buf); + } } } } @@ -451,8 +472,8 @@ re_compile_fastmap_iter (regex_t *bufp, const re_dfastate_t *init_state, int regcomp (preg, pattern, cflags) - regex_t *__restrict preg; - const char *__restrict pattern; + regex_t *_Restrict_ preg; + const char *_Restrict_ pattern; int cflags; { reg_errcode_t ret; @@ -515,13 +536,13 @@ weak_alias (__regcomp, regcomp) size_t regerror (errcode, preg, errbuf, errbuf_size) int errcode; - const regex_t *__restrict preg; - char *__restrict errbuf; + const regex_t *_Restrict_ preg; + char *_Restrict_ errbuf; size_t errbuf_size; #else /* size_t might promote */ size_t -regerror (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) #endif { const char *msg; @@ -542,17 +563,13 @@ regerror (int errcode, const regex_t *__restrict preg, if (BE (errbuf_size != 0, 1)) { + size_t cpy_size = msg_size; if (BE (msg_size > errbuf_size, 0)) { -#if defined HAVE_MEMPCPY || defined _LIBC - *((char *) __mempcpy (errbuf, msg, errbuf_size - 1)) = '\0'; -#else - memcpy (errbuf, msg, errbuf_size - 1); - errbuf[errbuf_size - 1] = 0; -#endif + cpy_size = errbuf_size - 1; + errbuf[cpy_size] = '\0'; } - else - memcpy (errbuf, msg, msg_size); + memcpy (errbuf, msg, cpy_size); } return msg_size; @@ -780,7 +797,7 @@ re_compile_internal (regex_t *preg, const char * pattern, size_t length, __libc_lock_init (dfa->lock); err = re_string_construct (®exp, pattern, length, preg->translate, - syntax & RE_ICASE, dfa); + (syntax & RE_ICASE) != 0, dfa); if (BE (err != REG_NOERROR, 0)) { re_compile_internal_free_return: @@ -833,18 +850,17 @@ static reg_errcode_t init_dfa (re_dfa_t *dfa, size_t pat_len) { __re_size_t table_size; -#ifndef _LIBC - char *codeset_name; +#ifdef RE_ENABLE_I18N + size_t max_i18n_object_size = MAX (sizeof (wchar_t), sizeof (wctype_t)); +#else + size_t max_i18n_object_size = 0; #endif size_t max_object_size = MAX (sizeof (struct re_state_table_entry), MAX (sizeof (re_token_t), MAX (sizeof (re_node_set), MAX (sizeof (regmatch_t), - MAX (sizeof (regoff_t), - MAX (sizeof (wchar_t), - MAX (sizeof (wctype_t), - sizeof (Idx)))))))); + max_i18n_object_size)))); memset (dfa, '\0', sizeof (re_dfa_t)); @@ -877,22 +893,7 @@ init_dfa (re_dfa_t *dfa, size_t pat_len) dfa->map_notascii = (_NL_CURRENT_WORD (LC_CTYPE, _NL_CTYPE_MAP_TO_NONASCII) != 0); #else -# ifdef HAVE_LANGINFO_CODESET - codeset_name = nl_langinfo (CODESET); -# else - codeset_name = getenv ("LC_ALL"); - if (codeset_name == NULL || codeset_name[0] == '\0') - codeset_name = getenv ("LC_CTYPE"); - if (codeset_name == NULL || codeset_name[0] == '\0') - codeset_name = getenv ("LANG"); - if (codeset_name == NULL) - codeset_name = ""; - else if (strchr (codeset_name, '.') != NULL) - codeset_name = strchr (codeset_name, '.') + 1; -# endif - - if (strcasecmp (codeset_name, "UTF-8") == 0 - || strcasecmp (codeset_name, "UTF8") == 0) + if (strcmp (locale_charset (), "UTF-8") == 0) dfa->is_utf8 = 1; /* We check exhaustively in the loop below if this charset is a @@ -1069,7 +1070,7 @@ optimize_utf8 (re_dfa_t *dfa) mb_chars = true; break; case ANCHOR: - switch (dfa->nodes[node].opr.idx) + switch (dfa->nodes[node].opr.ctx_type) { case LINE_FIRST: case LINE_LAST: @@ -1077,7 +1078,9 @@ optimize_utf8 (re_dfa_t *dfa) case BUF_LAST: break; default: - /* Word anchors etc. cannot be handled. */ + /* Word anchors etc. cannot be handled. It's okay to test + opr.ctx_type since constraints (for all DFA nodes) are + created by ORing one or more opr.ctx_type values. */ return; } break; @@ -1364,6 +1367,8 @@ calc_first (void *extra, bin_tree_t *node) node->node_idx = re_dfa_add_node (dfa, node->token); if (BE (node->node_idx == REG_MISSING, 0)) return REG_ESPACE; + if (node->token.type == ANCHOR) + dfa->nodes[node->node_idx].constraint = node->token.opr.ctx_type; } return REG_NOERROR; } @@ -1493,21 +1498,18 @@ duplicate_node_closure (re_dfa_t *dfa, Idx top_org_node, Idx top_clone_node, destination. */ org_dest = dfa->edests[org_node].elems[0]; re_node_set_empty (dfa->edests + clone_node); - if (dfa->nodes[org_node].type == ANCHOR) + clone_dest = search_duplicated_node (dfa, org_dest, constraint); + /* If the node is root_node itself, it means the epsilon closure + has a loop. Then tie it to the destination of the root_node. */ + if (org_node == root_node && clone_node != org_node) { - /* In case of the node has another constraint, append it. */ - if (org_node == root_node && clone_node != org_node) - { - /* ...but if the node is root_node itself, it means the - epsilon closure have a loop, then tie it to the - destination of the root_node. */ - ok = re_node_set_insert (dfa->edests + clone_node, org_dest); - if (BE (! ok, 0)) - return REG_ESPACE; - break; - } - constraint |= dfa->nodes[org_node].opr.ctx_type; + ok = re_node_set_insert (dfa->edests + clone_node, org_dest); + if (BE (! ok, 0)) + return REG_ESPACE; + break; } + /* In case the node has another constraint, append it. */ + constraint |= dfa->nodes[org_node].constraint; clone_dest = duplicate_node (dfa, org_dest, constraint); if (BE (clone_dest == REG_MISSING, 0)) return REG_ESPACE; @@ -1525,7 +1527,7 @@ duplicate_node_closure (re_dfa_t *dfa, Idx top_org_node, Idx top_clone_node, clone_dest = search_duplicated_node (dfa, org_dest, constraint); if (clone_dest == REG_MISSING) { - /* There are no such a duplicated node, create a new one. */ + /* There is no such duplicated node, create a new one. */ reg_errcode_t err; clone_dest = duplicate_node (dfa, org_dest, constraint); if (BE (clone_dest == REG_MISSING, 0)) @@ -1540,7 +1542,7 @@ duplicate_node_closure (re_dfa_t *dfa, Idx top_org_node, Idx top_clone_node, } else { - /* There are a duplicated node which satisfy the constraint, + /* There is a duplicated node which satisfy the constraint, use it to avoid infinite loop. */ ok = re_node_set_insert (dfa->edests + clone_node, clone_dest); if (BE (! ok, 0)) @@ -1589,8 +1591,7 @@ duplicate_node (re_dfa_t *dfa, Idx org_idx, unsigned int constraint) if (BE (dup_idx != REG_MISSING, 1)) { dfa->nodes[dup_idx].constraint = constraint; - if (dfa->nodes[org_idx].type == ANCHOR) - dfa->nodes[dup_idx].constraint |= dfa->nodes[org_idx].opr.ctx_type; + dfa->nodes[dup_idx].constraint |= dfa->nodes[org_idx].constraint; dfa->nodes[dup_idx].duplicated = 1; /* Store the index of the original node. */ @@ -1672,7 +1673,6 @@ static reg_errcode_t calc_eclosure_iter (re_node_set *new_set, re_dfa_t *dfa, Idx node, bool root) { reg_errcode_t err; - unsigned int constraint; Idx i; bool incomplete; bool ok; @@ -1686,15 +1686,14 @@ calc_eclosure_iter (re_node_set *new_set, re_dfa_t *dfa, Idx node, bool root) We reference this value to avoid infinite loop. */ dfa->eclosures[node].nelem = REG_MISSING; - constraint = ((dfa->nodes[node].type == ANCHOR) - ? dfa->nodes[node].opr.ctx_type : 0); - /* If the current node has constraints, duplicate all nodes. - Since they must inherit the constraints. */ - if (constraint + /* If the current node has constraints, duplicate all nodes + since they must inherit the constraints. */ + if (dfa->nodes[node].constraint && dfa->edests[node].nelem && !dfa->nodes[dfa->edests[node].elems[0]].duplicated) { - err = duplicate_node_closure (dfa, node, node, node, constraint); + err = duplicate_node_closure (dfa, node, node, node, + dfa->nodes[node].constraint); if (BE (err != REG_NOERROR, 0)) return err; } @@ -2175,7 +2174,7 @@ static bin_tree_t * parse_branch (re_string_t *regexp, regex_t *preg, re_token_t *token, reg_syntax_t syntax, Idx nest, reg_errcode_t *err) { - bin_tree_t *tree, *exp; + bin_tree_t *tree, *expr; re_dfa_t *dfa = (re_dfa_t *) preg->buffer; tree = parse_expression (regexp, preg, token, syntax, nest, err); if (BE (*err != REG_NOERROR && tree == NULL, 0)) @@ -2184,14 +2183,14 @@ parse_branch (re_string_t *regexp, regex_t *preg, re_token_t *token, while (token->type != OP_ALT && token->type != END_OF_RE && (nest == 0 || token->type != OP_CLOSE_SUBEXP)) { - exp = parse_expression (regexp, preg, token, syntax, nest, err); - if (BE (*err != REG_NOERROR && exp == NULL, 0)) + expr = parse_expression (regexp, preg, token, syntax, nest, err); + if (BE (*err != REG_NOERROR && expr == NULL, 0)) { return NULL; } - if (tree != NULL && exp != NULL) + if (tree != NULL && expr != NULL) { - tree = create_tree (dfa, tree, exp, CONCAT); + tree = create_tree (dfa, tree, expr, CONCAT); if (tree == NULL) { *err = REG_ESPACE; @@ -2199,8 +2198,8 @@ parse_branch (re_string_t *regexp, regex_t *preg, re_token_t *token, } } else if (tree == NULL) - tree = exp; - /* Otherwise exp == NULL, we don't need to create new tree. */ + tree = expr; + /* Otherwise expr == NULL, we don't need to create new tree. */ } return tree; } @@ -3072,7 +3071,7 @@ parse_bracket_exp (re_string_t *regexp, re_dfa_t *dfa, re_token_t *token, #endif /* not RE_ENABLE_I18N */ non_match = true; if (syntax & RE_HAT_LISTS_NOT_NEWLINE) - bitset_set (sbcset, '\0'); + bitset_set (sbcset, '\n'); re_string_skip_bytes (regexp, token_len); /* Skip a token. */ token_len = peek_token_bracket (token, regexp, syntax); if (BE (token->type == END_OF_RE, 0)) @@ -3529,13 +3528,13 @@ build_charclass (RE_TRANSLATE_TYPE trans, bitset_t sbcset, if (BE (trans != NULL, 0)) \ { \ for (i = 0; i < SBC_MAX; ++i) \ - if (ctype_func (i)) \ + if (ctype_func (i)) \ bitset_set (sbcset, trans[i]); \ } \ else \ { \ for (i = 0; i < SBC_MAX; ++i) \ - if (ctype_func (i)) \ + if (ctype_func (i)) \ bitset_set (sbcset, i); \ } \ } while (0) @@ -3603,10 +3602,6 @@ build_charclass_op (re_dfa_t *dfa, RE_TRANSLATE_TYPE trans, if (non_match) { #ifdef RE_ENABLE_I18N - /* - if (syntax & RE_HAT_LISTS_NOT_NEWLINE) - bitset_set(cset->sbcset, '\0'); - */ mbcset->non_match = 1; #endif /* not RE_ENABLE_I18N */ }