X-Git-Url: http://erislabs.net/gitweb/?a=blobdiff_plain;f=lib%2Fregcomp.c;h=7eff569258d46360497e3778dfc209777486855b;hb=6f730e7ae5225bf9eb36554d94805b97a72784bd;hp=ebb696ac2bf7cd930b1fe17fd6043ffa43eabaf2;hpb=d55324f119e0f203590486bbf4760a1dc25d7735;p=gnulib.git diff --git a/lib/regcomp.c b/lib/regcomp.c index ebb696ac2..7eff56925 100644 --- a/lib/regcomp.c +++ b/lib/regcomp.c @@ -636,7 +636,7 @@ free_dfa_content (re_dfa_t *dfa) re_dfastate_t *state = entry->array[j]; free_state (state); } - re_free (entry->array); + re_free (entry->array); } re_free (dfa->state_table); #ifdef RE_ENABLE_I18N @@ -1021,7 +1021,10 @@ create_initial_state (re_dfa_t *dfa) Idx dest_idx = dfa->edests[node_idx].elems[0]; if (!re_node_set_contains (&init_nodes, dest_idx)) { - re_node_set_merge (&init_nodes, dfa->eclosures + dest_idx); + reg_errcode_t merge_err + = re_node_set_merge (&init_nodes, dfa->eclosures + dest_idx); + if (merge_err != REG_NOERROR) + return merge_err; i = 0; } } @@ -1090,8 +1093,8 @@ optimize_utf8 (re_dfa_t *dfa) } break; case OP_PERIOD: - has_period = true; - break; + has_period = true; + break; case OP_BACK_REF: case OP_ALT: case END_OF_RE: @@ -1192,7 +1195,7 @@ analyze (regex_t *preg) { dfa->inveclosures = re_malloc (re_node_set, dfa->nodes_len); if (BE (dfa->inveclosures == NULL, 0)) - return REG_ESPACE; + return REG_ESPACE; ret = calc_inveclosure (dfa); } @@ -1214,16 +1217,16 @@ postorder (bin_tree_t *root, reg_errcode_t (fn (void *, bin_tree_t *)), if that's the only child). */ while (node->left || node->right) if (node->left) - node = node->left; - else - node = node->right; + node = node->left; + else + node = node->right; do { reg_errcode_t err = fn (extra, node); if (BE (err != REG_NOERROR, 0)) return err; - if (node->parent == NULL) + if (node->parent == NULL) return REG_NOERROR; prev = node; node = node->parent; @@ -1257,7 +1260,7 @@ preorder (bin_tree_t *root, reg_errcode_t (fn (void *, bin_tree_t *)), prev = node; node = node->parent; if (!node) - return REG_NOERROR; + return REG_NOERROR; } node = node->right; } @@ -1280,13 +1283,13 @@ optimize_subexps (void *extra, bin_tree_t *node) } else if (node->token.type == SUBEXP - && node->left && node->left->token.type == SUBEXP) + && node->left && node->left->token.type == SUBEXP) { Idx other_idx = node->left->token.opr.idx; node->left = node->left->left; if (node->left) - node->left->parent = node; + node->left->parent = node; dfa->subexp_map[other_idx] = dfa->subexp_map[node->token.opr.idx]; if (other_idx < BITSET_WORD_BITS) @@ -1371,9 +1374,9 @@ calc_first (void *extra, bin_tree_t *node) node->first = node; node->node_idx = re_dfa_add_node (dfa, node->token); if (BE (node->node_idx == REG_MISSING, 0)) - return REG_ESPACE; + return REG_ESPACE; if (node->token.type == ANCHOR) - dfa->nodes[node->node_idx].constraint = node->token.opr.ctx_type; + dfa->nodes[node->node_idx].constraint = node->token.opr.ctx_type; } return REG_NOERROR; } @@ -1395,7 +1398,7 @@ calc_next (void *extra, bin_tree_t *node) if (node->left) node->left->next = node->next; if (node->right) - node->right->next = node->next; + node->right->next = node->next; break; } return REG_NOERROR; @@ -1446,7 +1449,7 @@ link_nfa_nodes (void *extra, bin_tree_t *node) case OP_BACK_REF: dfa->nexts[idx] = node->next->node_idx; if (node->token.type == OP_BACK_REF) - re_node_set_init_1 (dfa->edests + idx, dfa->nexts[idx]); + err = re_node_set_init_1 (dfa->edests + idx, dfa->nexts[idx]); break; default: @@ -1678,10 +1681,9 @@ calc_eclosure_iter (re_node_set *new_set, re_dfa_t *dfa, Idx node, bool root) { reg_errcode_t err; Idx i; - bool incomplete; - bool ok; re_node_set eclosure; - incomplete = false; + bool ok; + bool incomplete = false; err = re_node_set_alloc (&eclosure, dfa->edests[node].nelem + 1); if (BE (err != REG_NOERROR, 0)) return err; @@ -1726,7 +1728,9 @@ calc_eclosure_iter (re_node_set *new_set, re_dfa_t *dfa, Idx node, bool root) else eclosure_elem = dfa->eclosures[edest]; /* Merge the epsilon closure of `edest'. */ - re_node_set_merge (&eclosure, &eclosure_elem); + err = re_node_set_merge (&eclosure, &eclosure_elem); + if (BE (err != REG_NOERROR, 0)) + return err; /* If the epsilon closure of `edest' is incomplete, the epsilon closure of this node is also incomplete. */ if (dfa->eclosures[edest].nelem == 0) @@ -1736,7 +1740,7 @@ calc_eclosure_iter (re_node_set *new_set, re_dfa_t *dfa, Idx node, bool root) } } - /* Epsilon closures include itself. */ + /* An epsilon closure includes itself. */ ok = re_node_set_insert (&eclosure, node); if (BE (! ok, 0)) return REG_ESPACE; @@ -2323,7 +2327,7 @@ parse_expression (re_string_t *regexp, regex_t *preg, re_token_t *token, && dfa->word_ops_used == 0) init_word_char (dfa); if (token->opr.ctx_type == WORD_DELIM - || token->opr.ctx_type == NOT_WORD_DELIM) + || token->opr.ctx_type == NOT_WORD_DELIM) { bin_tree_t *tree_first, *tree_last; if (token->opr.ctx_type == WORD_DELIM) @@ -2331,13 +2335,13 @@ parse_expression (re_string_t *regexp, regex_t *preg, re_token_t *token, token->opr.ctx_type = WORD_FIRST; tree_first = create_token_tree (dfa, NULL, NULL, token); token->opr.ctx_type = WORD_LAST; - } - else - { + } + else + { token->opr.ctx_type = INSIDE_WORD; tree_first = create_token_tree (dfa, NULL, NULL, token); token->opr.ctx_type = INSIDE_NOTWORD; - } + } tree_last = create_token_tree (dfa, NULL, NULL, token); tree = create_tree (dfa, tree_first, tree_last, OP_ALT); if (BE (tree_first == NULL || tree_last == NULL || tree == NULL, 0)) @@ -2448,7 +2452,7 @@ parse_sub_exp (re_string_t *regexp, regex_t *preg, re_token_t *token, { tree = parse_reg_exp (regexp, preg, token, syntax, nest, err); if (BE (*err == REG_NOERROR && token->type != OP_CLOSE_SUBEXP, 0)) - *err = REG_EPAREN; + *err = REG_EPAREN; if (BE (*err != REG_NOERROR, 0)) return NULL; } @@ -2573,10 +2577,14 @@ parse_dup_op (bin_tree_t *elem, re_string_t *regexp, re_dfa_t *dfa, if (BE (tree == NULL, 0)) goto parse_dup_op_espace; +/* From gnulib's "intprops.h": + True if the arithmetic type T is signed. */ +#define TYPE_SIGNED(t) (! ((t) 0 < (t) -1)) + /* This loop is actually executed only when end != REG_MISSING, to rewrite {0,n} as ((...?)?)?... We have already created the start+1-th copy. */ - if ((Idx) -1 < 0 || end != REG_MISSING) + if (TYPE_SIGNED (Idx) || end != REG_MISSING) for (i = start + 2; i <= end; ++i) { elem = duplicate_tree (elem, dfa); @@ -2614,11 +2622,17 @@ parse_dup_op (bin_tree_t *elem, re_string_t *regexp, re_dfa_t *dfa, static reg_errcode_t internal_function # ifdef RE_ENABLE_I18N -build_range_exp (bitset_t sbcset, re_charset_t *mbcset, Idx *range_alloc, - bracket_elem_t *start_elem, bracket_elem_t *end_elem) +build_range_exp (const reg_syntax_t syntax, + bitset_t sbcset, + re_charset_t *mbcset, + Idx *range_alloc, + const bracket_elem_t *start_elem, + const bracket_elem_t *end_elem) # else /* not RE_ENABLE_I18N */ -build_range_exp (bitset_t sbcset, bracket_elem_t *start_elem, - bracket_elem_t *end_elem) +build_range_exp (const reg_syntax_t syntax, + bitset_t sbcset, + const bracket_elem_t *start_elem, + const bracket_elem_t *end_elem) # endif /* not RE_ENABLE_I18N */ { unsigned int start_ch, end_ch; @@ -2657,7 +2671,9 @@ build_range_exp (bitset_t sbcset, bracket_elem_t *start_elem, return REG_ECOLLATE; cmp_buf[0] = start_wc; cmp_buf[4] = end_wc; - if (wcscoll (cmp_buf, cmp_buf + 4) > 0) + + if (BE ((syntax & RE_NO_EMPTY_RANGES) + && wcscoll (cmp_buf, cmp_buf + 4) > 0, 0)) return REG_ERANGE; /* Got valid collation sequence values, add them as a new entry. @@ -2667,9 +2683,9 @@ build_range_exp (bitset_t sbcset, bracket_elem_t *start_elem, no MBCSET if dfa->mb_cur_max == 1. */ if (mbcset) { - /* Check the space of the arrays. */ - if (BE (*range_alloc == mbcset->nranges, 0)) - { + /* Check the space of the arrays. */ + if (BE (*range_alloc == mbcset->nranges, 0)) + { /* There is not enough space, need realloc. */ wchar_t *new_array_start, *new_array_end; Idx new_nranges; @@ -2679,9 +2695,9 @@ build_range_exp (bitset_t sbcset, bracket_elem_t *start_elem, /* Use realloc since mbcset->range_starts and mbcset->range_ends are NULL if *range_alloc == 0. */ new_array_start = re_realloc (mbcset->range_starts, wchar_t, - new_nranges); + new_nranges); new_array_end = re_realloc (mbcset->range_ends, wchar_t, - new_nranges); + new_nranges); if (BE (new_array_start == NULL || new_array_end == NULL, 0)) return REG_ESPACE; @@ -2689,10 +2705,10 @@ build_range_exp (bitset_t sbcset, bracket_elem_t *start_elem, mbcset->range_starts = new_array_start; mbcset->range_ends = new_array_end; *range_alloc = new_nranges; - } + } - mbcset->range_starts[mbcset->nranges] = start_wc; - mbcset->range_ends[mbcset->nranges++] = end_wc; + mbcset->range_starts[mbcset->nranges] = start_wc; + mbcset->range_ends[mbcset->nranges++] = end_wc; } /* Build the table for single byte characters. */ @@ -2910,8 +2926,8 @@ parse_bracket_exp (re_string_t *regexp, re_dfa_t *dfa, re_token_t *token, build below suffices. */ if (nrules > 0 || dfa->mb_cur_max > 1) { - /* Check the space of the arrays. */ - if (BE (*range_alloc == mbcset->nranges, 0)) + /* Check the space of the arrays. */ + if (BE (*range_alloc == mbcset->nranges, 0)) { /* There is not enough space, need realloc. */ uint32_t *new_array_start; @@ -2923,18 +2939,18 @@ parse_bracket_exp (re_string_t *regexp, re_dfa_t *dfa, re_token_t *token, new_array_start = re_realloc (mbcset->range_starts, uint32_t, new_nranges); new_array_end = re_realloc (mbcset->range_ends, uint32_t, - new_nranges); + new_nranges); if (BE (new_array_start == NULL || new_array_end == NULL, 0)) - return REG_ESPACE; + return REG_ESPACE; mbcset->range_starts = new_array_start; mbcset->range_ends = new_array_end; *range_alloc = new_nranges; } - mbcset->range_starts[mbcset->nranges] = start_collseq; - mbcset->range_ends[mbcset->nranges++] = end_collseq; + mbcset->range_starts[mbcset->nranges] = start_collseq; + mbcset->range_ends[mbcset->nranges++] = end_collseq; } /* Build the table for single byte characters. */ @@ -3160,11 +3176,11 @@ parse_bracket_exp (re_string_t *regexp, re_dfa_t *dfa, re_token_t *token, &start_elem, &end_elem); #else # ifdef RE_ENABLE_I18N - *err = build_range_exp (sbcset, + *err = build_range_exp (syntax, sbcset, dfa->mb_cur_max > 1 ? mbcset : NULL, &range_alloc, &start_elem, &end_elem); # else - *err = build_range_exp (sbcset, &start_elem, &end_elem); + *err = build_range_exp (syntax, sbcset, &start_elem, &end_elem); # endif #endif /* RE_ENABLE_I18N */ if (BE (*err != REG_NOERROR, 0)) @@ -3268,17 +3284,17 @@ parse_bracket_exp (re_string_t *regexp, re_dfa_t *dfa, re_token_t *token, of having both SIMPLE_BRACKET and COMPLEX_BRACKET. */ if (sbc_idx < BITSET_WORDS) { - /* Build a tree for simple bracket. */ - br_token.type = SIMPLE_BRACKET; - br_token.opr.sbcset = sbcset; - work_tree = create_token_tree (dfa, NULL, NULL, &br_token); - if (BE (work_tree == NULL, 0)) - goto parse_bracket_exp_espace; + /* Build a tree for simple bracket. */ + br_token.type = SIMPLE_BRACKET; + br_token.opr.sbcset = sbcset; + work_tree = create_token_tree (dfa, NULL, NULL, &br_token); + if (BE (work_tree == NULL, 0)) + goto parse_bracket_exp_espace; - /* Then join them by ALT node. */ - work_tree = create_tree (dfa, work_tree, mbc_tree, OP_ALT); - if (BE (work_tree == NULL, 0)) - goto parse_bracket_exp_espace; + /* Then join them by ALT node. */ + work_tree = create_tree (dfa, work_tree, mbc_tree, OP_ALT); + if (BE (work_tree == NULL, 0)) + goto parse_bracket_exp_espace; } else { @@ -3297,7 +3313,7 @@ parse_bracket_exp (re_string_t *regexp, re_dfa_t *dfa, re_token_t *token, br_token.opr.sbcset = sbcset; work_tree = create_token_tree (dfa, NULL, NULL, &br_token); if (BE (work_tree == NULL, 0)) - goto parse_bracket_exp_espace; + goto parse_bracket_exp_espace; } return work_tree; @@ -3852,7 +3868,7 @@ duplicate_tree (const bin_tree_t *root, re_dfa_t *dfa) node = node->parent; dup_node = dup_node->parent; if (!node) - return dup_root; + return dup_root; } node = node->right; p_new = &dup_node->right;