You should have received a copy of the GNU Lesser General Public
License along with the GNU C Library; if not, see
- <http://www.gnu.org/licenses/>. */
-
-#include <stdint.h>
+ <https://www.gnu.org/licenses/>. */
#ifdef _LIBC
# include <locale/weight.h>
static reg_errcode_t calc_first (void *extra, bin_tree_t *node);
static reg_errcode_t calc_next (void *extra, bin_tree_t *node);
static reg_errcode_t link_nfa_nodes (void *extra, bin_tree_t *node);
-static int duplicate_node (re_dfa_t *dfa, int org_idx, unsigned int constraint);
-static int search_duplicated_node (const re_dfa_t *dfa, int org_node,
+static Idx duplicate_node (re_dfa_t *dfa, Idx org_idx, unsigned int constraint);
+static Idx search_duplicated_node (const re_dfa_t *dfa, Idx org_node,
unsigned int constraint);
static reg_errcode_t calc_eclosure (re_dfa_t *dfa);
static reg_errcode_t calc_eclosure_iter (re_node_set *new_set, re_dfa_t *dfa,
- int node, int root);
+ Idx node, bool root);
static reg_errcode_t calc_inveclosure (re_dfa_t *dfa);
-static int fetch_number (re_string_t *input, re_token_t *token,
+static Idx fetch_number (re_string_t *input, re_token_t *token,
reg_syntax_t syntax);
static int peek_token (re_token_t *token, re_string_t *input,
reg_syntax_t syntax);
reg_syntax_t syntax, reg_errcode_t *err);
static bin_tree_t *parse_reg_exp (re_string_t *regexp, regex_t *preg,
re_token_t *token, reg_syntax_t syntax,
- int nest, reg_errcode_t *err);
+ Idx nest, reg_errcode_t *err);
static bin_tree_t *parse_branch (re_string_t *regexp, regex_t *preg,
re_token_t *token, reg_syntax_t syntax,
- int nest, reg_errcode_t *err);
+ Idx nest, reg_errcode_t *err);
static bin_tree_t *parse_expression (re_string_t *regexp, regex_t *preg,
re_token_t *token, reg_syntax_t syntax,
- int nest, reg_errcode_t *err);
+ Idx nest, reg_errcode_t *err);
static bin_tree_t *parse_sub_exp (re_string_t *regexp, regex_t *preg,
re_token_t *token, reg_syntax_t syntax,
- int nest, reg_errcode_t *err);
+ Idx nest, reg_errcode_t *err);
static bin_tree_t *parse_dup_op (bin_tree_t *dup_elem, re_string_t *regexp,
re_dfa_t *dfa, re_token_t *token,
reg_syntax_t syntax, reg_errcode_t *err);
re_token_t *token, int token_len,
re_dfa_t *dfa,
reg_syntax_t syntax,
- int accept_hyphen);
+ bool accept_hyphen);
static reg_errcode_t parse_bracket_symbol (bracket_elem_t *elem,
re_string_t *regexp,
re_token_t *token);
#ifdef RE_ENABLE_I18N
static reg_errcode_t build_equiv_class (bitset_t sbcset,
re_charset_t *mbcset,
- int *equiv_class_alloc,
+ Idx *equiv_class_alloc,
const unsigned char *name);
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,
+ Idx *char_class_alloc,
+ const char *class_name,
reg_syntax_t syntax);
#else /* not RE_ENABLE_I18N */
static reg_errcode_t build_equiv_class (bitset_t sbcset,
const unsigned char *name);
static reg_errcode_t build_charclass (RE_TRANSLATE_TYPE trans,
bitset_t sbcset,
- const unsigned char *class_name,
+ const char *class_name,
reg_syntax_t syntax);
#endif /* not RE_ENABLE_I18N */
static bin_tree_t *build_charclass_op (re_dfa_t *dfa,
RE_TRANSLATE_TYPE trans,
- const unsigned char *class_name,
- const unsigned char *extra,
- int non_match, reg_errcode_t *err);
+ const char *class_name,
+ const char *extra,
+ bool non_match, reg_errcode_t *err);
static bin_tree_t *create_tree (re_dfa_t *dfa,
bin_tree_t *left, bin_tree_t *right,
re_token_type_t type);
POSIX doesn't require that we do anything for REG_NOERROR,
but why not be nice? */
-const char __re_error_msgid[] attribute_hidden =
+static const char __re_error_msgid[] =
{
#define REG_NOERROR_IDX 0
gettext_noop ("Success") /* REG_NOERROR */
gettext_noop ("Invalid back reference") /* REG_ESUBREG */
"\0"
#define REG_EBRACK_IDX (REG_ESUBREG_IDX + sizeof "Invalid back reference")
- gettext_noop ("Unmatched [ or [^") /* REG_EBRACK */
+ gettext_noop ("Unmatched [, [^, [:, [., or [=") /* REG_EBRACK */
"\0"
-#define REG_EPAREN_IDX (REG_EBRACK_IDX + sizeof "Unmatched [ or [^")
+#define REG_EPAREN_IDX (REG_EBRACK_IDX + sizeof "Unmatched [, [^, [:, [., or [=")
gettext_noop ("Unmatched ( or \\(") /* REG_EPAREN */
"\0"
#define REG_EBRACE_IDX (REG_EPAREN_IDX + sizeof "Unmatched ( or \\(")
gettext_noop ("Unmatched ) or \\)") /* REG_ERPAREN */
};
-const size_t __re_error_msgid_idx[] attribute_hidden =
+static const size_t __re_error_msgid_idx[] =
{
REG_NOERROR_IDX,
REG_NOMATCH_IDX,
int
re_compile_fastmap (struct re_pattern_buffer *bufp)
{
- re_dfa_t *dfa = (re_dfa_t *) bufp->buffer;
+ re_dfa_t *dfa = bufp->buffer;
char *fastmap = bufp->fastmap;
memset (fastmap, '\0', sizeof (char) * SBC_MAX);
re_compile_fastmap_iter (regex_t *bufp, const re_dfastate_t *init_state,
char *fastmap)
{
- re_dfa_t *dfa = (re_dfa_t *) bufp->buffer;
- int node_cnt;
- int icase = (dfa->mb_cur_max == 1 && (bufp->syntax & RE_ICASE));
+ re_dfa_t *dfa = bufp->buffer;
+ Idx node_cnt;
+ bool icase = (dfa->mb_cur_max == 1 && (bufp->syntax & RE_ICASE));
for (node_cnt = 0; node_cnt < init_state->nodes.nelem; ++node_cnt)
{
- int node = init_state->nodes.elems[node_cnt];
+ Idx node = init_state->nodes.elems[node_cnt];
re_token_type_t type = dfa->nodes[node].type;
if (type == CHARACTER)
#ifdef RE_ENABLE_I18N
if ((bufp->syntax & RE_ICASE) && dfa->mb_cur_max > 1)
{
- unsigned char *buf = alloca (dfa->mb_cur_max), *p;
+ unsigned char buf[MB_LEN_MAX];
+ unsigned char *p;
wchar_t wc;
mbstate_t state;
&state) == p - buf
&& (__wcrtomb ((char *) buf, __towlower (wc), &state)
!= (size_t) -1))
- re_set_fastmap (fastmap, 0, buf[0]);
+ re_set_fastmap (fastmap, false, buf[0]);
}
#endif
}
else if (type == COMPLEX_BRACKET)
{
re_charset_t *cset = dfa->nodes[node].opr.mbcset;
- int i;
+ Idx i;
# ifdef _LIBC
/* See if we have to try all bytes which start multiple collation
the return codes and their meanings.) */
int
-regcomp (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
from either regcomp or regexec. We don't use PREG here. */
size_t
-regerror (int errcode, const regex_t *__restrict preg, char *__restrict errbuf,
+regerror (int errcode, const regex_t *_Restrict_ preg, char *_Restrict_ errbuf,
size_t errbuf_size)
{
const char *msg;
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;
static const bitset_t utf8_sb_map =
{
/* Set the first 128 bits. */
+# if defined __GNUC__ && !defined __STRICT_ANSI__
[0 ... 0x80 / BITSET_WORD_BITS - 1] = BITSET_WORD_MAX
+# else
+# if 4 * BITSET_WORD_BITS < ASCII_CHARS
+# error "bitset_word_t is narrower than 32 bits"
+# elif 3 * BITSET_WORD_BITS < ASCII_CHARS
+ BITSET_WORD_MAX, BITSET_WORD_MAX, BITSET_WORD_MAX,
+# elif 2 * BITSET_WORD_BITS < ASCII_CHARS
+ BITSET_WORD_MAX, BITSET_WORD_MAX,
+# elif 1 * BITSET_WORD_BITS < ASCII_CHARS
+ BITSET_WORD_MAX,
+# endif
+ (BITSET_WORD_MAX
+ >> (SBC_MAX % BITSET_WORD_BITS == 0
+ ? 0
+ : BITSET_WORD_BITS - SBC_MAX % BITSET_WORD_BITS))
+# endif
};
#endif
static void
free_dfa_content (re_dfa_t *dfa)
{
- int i, j;
+ Idx i, j;
if (dfa->nodes)
for (i = 0; i < dfa->nodes_len; ++i)
void
regfree (regex_t *preg)
{
- re_dfa_t *dfa = (re_dfa_t *) preg->buffer;
+ re_dfa_t *dfa = preg->buffer;
if (BE (dfa != NULL, 1))
- free_dfa_content (dfa);
+ {
+ lock_fini (dfa->lock);
+ free_dfa_content (dfa);
+ }
preg->buffer = NULL;
preg->allocated = 0;
if (re_comp_buf.fastmap == NULL)
{
- re_comp_buf.fastmap = (char *) malloc (SBC_MAX);
+ re_comp_buf.fastmap = re_malloc (char, SBC_MAX);
if (re_comp_buf.fastmap == NULL)
return (char *) gettext (__re_error_msgid
+ __re_error_msgid_idx[(int) REG_ESPACE]);
if (!ret)
return NULL;
- /* Yes, we're discarding `const' here if !HAVE_LIBINTL. */
+ /* Yes, we're discarding 'const' here if !HAVE_LIBINTL. */
return (char *) gettext (__re_error_msgid + __re_error_msgid_idx[(int) ret]);
}
preg->regs_allocated = REGS_UNALLOCATED;
/* Initialize the dfa. */
- dfa = (re_dfa_t *) preg->buffer;
+ dfa = preg->buffer;
if (BE (preg->allocated < sizeof (re_dfa_t), 0))
{
/* If zero allocated, but buffer is non-null, try to realloc
if (dfa == NULL)
return REG_ESPACE;
preg->allocated = sizeof (re_dfa_t);
- preg->buffer = (unsigned char *) dfa;
+ preg->buffer = dfa;
}
preg->used = sizeof (re_dfa_t);
err = init_dfa (dfa, length);
+ if (BE (err == REG_NOERROR && lock_init (dfa->lock) != 0, 0))
+ err = REG_ESPACE;
if (BE (err != REG_NOERROR, 0))
{
free_dfa_content (dfa);
strncpy (dfa->re_str, pattern, length + 1);
#endif
- __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:
free_workarea_compile (preg);
re_string_destruct (®exp);
+ lock_fini (dfa->lock);
free_dfa_content (dfa);
preg->buffer = NULL;
preg->allocated = 0;
if (BE (err != REG_NOERROR, 0))
{
+ lock_fini (dfa->lock);
free_dfa_content (dfa);
preg->buffer = NULL;
preg->allocated = 0;
static reg_errcode_t
init_dfa (re_dfa_t *dfa, size_t pat_len)
{
- unsigned int table_size;
+ __re_size_t table_size;
#ifndef _LIBC
- char *codeset_name;
+ const char *codeset_name;
+#endif
+#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_i18n_object_size))));
memset (dfa, '\0', sizeof (re_dfa_t));
/* Force allocation of str_tree_storage the first time. */
dfa->str_tree_storage_idx = BIN_TREE_STORAGE_SIZE;
- /* Avoid overflows. */
- if (pat_len == SIZE_MAX)
+ /* Avoid overflows. The extra "/ 2" is for the table_size doubling
+ calculation below, and for similar doubling calculations
+ elsewhere. And it's <= rather than <, because some of the
+ doubling calculations add 1 afterwards. */
+ if (BE (MIN (IDX_MAX, SIZE_MAX / max_object_size) / 2 <= pat_len, 0))
return REG_ESPACE;
dfa->nodes_alloc = pat_len + 1;
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 ((codeset_name[0] == 'U' || codeset_name[0] == 'u')
+ && (codeset_name[1] == 'T' || codeset_name[1] == 't')
+ && (codeset_name[2] == 'F' || codeset_name[2] == 'f')
+ && strcmp (codeset_name + 3 + (codeset_name[3] == '-'), "8") == 0)
dfa->is_utf8 = 1;
/* We check exhaustively in the loop below if this charset is a
static void
init_word_char (re_dfa_t *dfa)
{
- dfa->word_ops_used = 1;
int i = 0;
+ int j;
int ch = 0;
+ dfa->word_ops_used = 1;
if (BE (dfa->map_notascii == 0, 1))
{
/* Avoid uint32_t and uint64_t as some non-GCC platforms lack
general_case:
for (; i < BITSET_WORDS; ++i)
- for (int j = 0; j < BITSET_WORD_BITS; ++j, ++ch)
+ for (j = 0; j < BITSET_WORD_BITS; ++j, ++ch)
if (isalnum (ch) || ch == '_')
dfa->word_char[i] |= (bitset_word_t) 1 << j;
}
static void
free_workarea_compile (regex_t *preg)
{
- re_dfa_t *dfa = (re_dfa_t *) preg->buffer;
+ re_dfa_t *dfa = preg->buffer;
bin_tree_storage_t *storage, *next;
for (storage = dfa->str_tree_storage; storage; storage = next)
{
static reg_errcode_t
create_initial_state (re_dfa_t *dfa)
{
- int first, i;
+ Idx first, i;
reg_errcode_t err;
re_node_set init_nodes;
if (dfa->nbackref > 0)
for (i = 0; i < init_nodes.nelem; ++i)
{
- int node_idx = init_nodes.elems[i];
+ Idx node_idx = init_nodes.elems[i];
re_token_type_t type = dfa->nodes[node_idx].type;
- int clexp_idx;
+ Idx clexp_idx;
if (type != OP_BACK_REF)
continue;
for (clexp_idx = 0; clexp_idx < init_nodes.nelem; ++clexp_idx)
if (type == OP_BACK_REF)
{
- int dest_idx = dfa->edests[node_idx].elems[0];
+ Idx dest_idx = dfa->edests[node_idx].elems[0];
if (!re_node_set_contains (&init_nodes, dest_idx))
{
- reg_errcode_t err = re_node_set_merge (&init_nodes,
- dfa->eclosures
- + dest_idx);
- if (err != REG_NOERROR)
- return err;
+ 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;
}
}
static void
optimize_utf8 (re_dfa_t *dfa)
{
- int node, i, mb_chars = 0, has_period = 0;
+ Idx node;
+ int i;
+ bool mb_chars = false;
+ bool has_period = false;
for (node = 0; node < dfa->nodes_len; ++node)
switch (dfa->nodes[node].type)
{
case CHARACTER:
- if (dfa->nodes[node].opr.c >= 0x80)
- mb_chars = 1;
+ if (dfa->nodes[node].opr.c >= ASCII_CHARS)
+ mb_chars = true;
break;
case ANCHOR:
switch (dfa->nodes[node].opr.ctx_type)
}
break;
case OP_PERIOD:
- has_period = 1;
+ has_period = true;
break;
case OP_BACK_REF:
case OP_ALT:
case COMPLEX_BRACKET:
return;
case SIMPLE_BRACKET:
- /* Just double check. The non-ASCII range starts at 0x80. */
- assert (0x80 % BITSET_WORD_BITS == 0);
- for (i = 0x80 / BITSET_WORD_BITS; i < BITSET_WORDS; ++i)
- if (dfa->nodes[node].opr.sbcset[i])
- return;
+ /* Just double check. */
+ {
+ int rshift = (ASCII_CHARS % BITSET_WORD_BITS == 0
+ ? 0
+ : BITSET_WORD_BITS - ASCII_CHARS % BITSET_WORD_BITS);
+ for (i = ASCII_CHARS / BITSET_WORD_BITS; i < BITSET_WORDS; ++i)
+ {
+ if (dfa->nodes[node].opr.sbcset[i] >> rshift != 0)
+ return;
+ rshift = 0;
+ }
+ }
break;
default:
abort ();
for (node = 0; node < dfa->nodes_len; ++node)
{
if (dfa->nodes[node].type == CHARACTER
- && dfa->nodes[node].opr.c >= 0x80)
+ && dfa->nodes[node].opr.c >= ASCII_CHARS)
dfa->nodes[node].mb_partial = 0;
else if (dfa->nodes[node].type == OP_PERIOD)
dfa->nodes[node].type = OP_UTF8_PERIOD;
static reg_errcode_t
analyze (regex_t *preg)
{
- re_dfa_t *dfa = (re_dfa_t *) preg->buffer;
+ re_dfa_t *dfa = preg->buffer;
reg_errcode_t ret;
/* Allocate arrays. */
- dfa->nexts = re_malloc (int, dfa->nodes_alloc);
- dfa->org_indices = re_malloc (int, dfa->nodes_alloc);
+ dfa->nexts = re_malloc (Idx, dfa->nodes_alloc);
+ dfa->org_indices = re_malloc (Idx, dfa->nodes_alloc);
dfa->edests = re_malloc (re_node_set, dfa->nodes_alloc);
dfa->eclosures = re_malloc (re_node_set, dfa->nodes_alloc);
if (BE (dfa->nexts == NULL || dfa->org_indices == NULL || dfa->edests == NULL
|| dfa->eclosures == NULL, 0))
return REG_ESPACE;
- dfa->subexp_map = re_malloc (int, preg->re_nsub);
+ dfa->subexp_map = re_malloc (Idx, preg->re_nsub);
if (dfa->subexp_map != NULL)
{
- int i;
+ Idx i;
for (i = 0; i < preg->re_nsub; i++)
dfa->subexp_map[i] = i;
preorder (dfa->str_tree, optimize_subexps, dfa);
break;
if (i == preg->re_nsub)
{
- free (dfa->subexp_map);
+ re_free (dfa->subexp_map);
dfa->subexp_map = NULL;
}
}
else if (node->token.type == SUBEXP
&& node->left && node->left->token.type == SUBEXP)
{
- int other_idx = node->left->token.opr.idx;
+ Idx other_idx = node->left->token.opr.idx;
node->left = node->left->left;
if (node->left)
dfa->subexp_map[other_idx] = dfa->subexp_map[node->token.opr.idx];
if (other_idx < BITSET_WORD_BITS)
- dfa->used_bkref_map &= ~((bitset_word_t) 1 << other_idx);
+ dfa->used_bkref_map &= ~((bitset_word_t) 1 << other_idx);
}
return REG_NOERROR;
static bin_tree_t *
lower_subexp (reg_errcode_t *err, regex_t *preg, bin_tree_t *node)
{
- re_dfa_t *dfa = (re_dfa_t *) preg->buffer;
+ re_dfa_t *dfa = preg->buffer;
bin_tree_t *body = node->left;
bin_tree_t *op, *cls, *tree1, *tree;
link_nfa_nodes (void *extra, bin_tree_t *node)
{
re_dfa_t *dfa = (re_dfa_t *) extra;
- int idx = node->node_idx;
+ Idx idx = node->node_idx;
reg_errcode_t err = REG_NOERROR;
switch (node->token.type)
case OP_DUP_ASTERISK:
case OP_ALT:
{
- int left, right;
+ Idx left, right;
dfa->has_plural_match = 1;
if (node->left != NULL)
left = node->left->first->node_idx;
to their own constraint. */
static reg_errcode_t
-duplicate_node_closure (re_dfa_t *dfa, int top_org_node, int top_clone_node,
- int root_node, unsigned int init_constraint)
+duplicate_node_closure (re_dfa_t *dfa, Idx top_org_node, Idx top_clone_node,
+ Idx root_node, unsigned int init_constraint)
{
- int org_node, clone_node, ret;
+ Idx org_node, clone_node;
+ bool ok;
unsigned int constraint = init_constraint;
for (org_node = top_org_node, clone_node = top_clone_node;;)
{
- int org_dest, clone_dest;
+ Idx org_dest, clone_dest;
if (dfa->nodes[org_node].type == OP_BACK_REF)
{
/* If the back reference epsilon-transit, its destination must
if (BE (clone_dest == -1, 0))
return REG_ESPACE;
dfa->nexts[clone_node] = dfa->nexts[org_node];
- ret = re_node_set_insert (dfa->edests + clone_node, clone_dest);
- if (BE (ret < 0, 0))
+ ok = re_node_set_insert (dfa->edests + clone_node, clone_dest);
+ if (BE (! ok, 0))
return REG_ESPACE;
}
else if (dfa->edests[org_node].nelem == 0)
org_dest = dfa->edests[org_node].elems[0];
re_node_set_empty (dfa->edests + clone_node);
/* 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. */
+ has a loop. Then tie it to the destination of the root_node. */
if (org_node == root_node && clone_node != org_node)
{
- ret = re_node_set_insert (dfa->edests + clone_node, org_dest);
- if (BE (ret < 0, 0))
- return REG_ESPACE;
+ 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. */
clone_dest = duplicate_node (dfa, org_dest, constraint);
if (BE (clone_dest == -1, 0))
return REG_ESPACE;
- ret = re_node_set_insert (dfa->edests + clone_node, clone_dest);
- if (BE (ret < 0, 0))
+ ok = re_node_set_insert (dfa->edests + clone_node, clone_dest);
+ if (BE (! ok, 0))
return REG_ESPACE;
}
else /* dfa->edests[org_node].nelem == 2 */
clone_dest = duplicate_node (dfa, org_dest, constraint);
if (BE (clone_dest == -1, 0))
return REG_ESPACE;
- ret = re_node_set_insert (dfa->edests + clone_node, clone_dest);
- if (BE (ret < 0, 0))
+ ok = re_node_set_insert (dfa->edests + clone_node, clone_dest);
+ if (BE (! ok, 0))
return REG_ESPACE;
err = duplicate_node_closure (dfa, org_dest, clone_dest,
root_node, constraint);
{
/* There is a duplicated node which satisfies the constraint,
use it to avoid infinite loop. */
- ret = re_node_set_insert (dfa->edests + clone_node, clone_dest);
- if (BE (ret < 0, 0))
+ ok = re_node_set_insert (dfa->edests + clone_node, clone_dest);
+ if (BE (! ok, 0))
return REG_ESPACE;
}
clone_dest = duplicate_node (dfa, org_dest, constraint);
if (BE (clone_dest == -1, 0))
return REG_ESPACE;
- ret = re_node_set_insert (dfa->edests + clone_node, clone_dest);
- if (BE (ret < 0, 0))
+ ok = re_node_set_insert (dfa->edests + clone_node, clone_dest);
+ if (BE (! ok, 0))
return REG_ESPACE;
}
org_node = org_dest;
/* Search for a node which is duplicated from the node ORG_NODE, and
satisfies the constraint CONSTRAINT. */
-static int
-search_duplicated_node (const re_dfa_t *dfa, int org_node,
+static Idx
+search_duplicated_node (const re_dfa_t *dfa, Idx org_node,
unsigned int constraint)
{
- int idx;
+ Idx idx;
for (idx = dfa->nodes_len - 1; dfa->nodes[idx].duplicated && idx > 0; --idx)
{
if (org_node == dfa->org_indices[idx]
Return the index of the new node, or -1 if insufficient storage is
available. */
-static int
-duplicate_node (re_dfa_t *dfa, int org_idx, unsigned int constraint)
+static Idx
+duplicate_node (re_dfa_t *dfa, Idx org_idx, unsigned int constraint)
{
- int dup_idx = re_dfa_add_node (dfa, dfa->nodes[org_idx]);
+ Idx dup_idx = re_dfa_add_node (dfa, dfa->nodes[org_idx]);
if (BE (dup_idx != -1, 1))
{
dfa->nodes[dup_idx].constraint = constraint;
static reg_errcode_t
calc_inveclosure (re_dfa_t *dfa)
{
- int src, idx, ret;
+ Idx src, idx;
+ bool ok;
for (idx = 0; idx < dfa->nodes_len; ++idx)
re_node_set_init_empty (dfa->inveclosures + idx);
for (src = 0; src < dfa->nodes_len; ++src)
{
- int *elems = dfa->eclosures[src].elems;
+ Idx *elems = dfa->eclosures[src].elems;
for (idx = 0; idx < dfa->eclosures[src].nelem; ++idx)
{
- ret = re_node_set_insert_last (dfa->inveclosures + elems[idx], src);
- if (BE (ret == -1, 0))
+ ok = re_node_set_insert_last (dfa->inveclosures + elems[idx], src);
+ if (BE (! ok, 0))
return REG_ESPACE;
}
}
static reg_errcode_t
calc_eclosure (re_dfa_t *dfa)
{
- int node_idx, incomplete;
+ Idx node_idx;
+ bool incomplete;
#ifdef DEBUG
assert (dfa->nodes_len > 0);
#endif
- incomplete = 0;
+ incomplete = false;
/* For each nodes, calculate epsilon closure. */
for (node_idx = 0; ; ++node_idx)
{
{
if (!incomplete)
break;
- incomplete = 0;
+ incomplete = false;
node_idx = 0;
}
if (dfa->eclosures[node_idx].nelem != 0)
continue;
/* Calculate epsilon closure of 'node_idx'. */
- err = calc_eclosure_iter (&eclosure_elem, dfa, node_idx, 1);
+ err = calc_eclosure_iter (&eclosure_elem, dfa, node_idx, true);
if (BE (err != REG_NOERROR, 0))
return err;
if (dfa->eclosures[node_idx].nelem == 0)
{
- incomplete = 1;
+ incomplete = true;
re_node_set_free (&eclosure_elem);
}
}
/* Calculate epsilon closure of NODE. */
static reg_errcode_t
-calc_eclosure_iter (re_node_set *new_set, re_dfa_t *dfa, int node, int root)
+calc_eclosure_iter (re_node_set *new_set, re_dfa_t *dfa, Idx node, bool root)
{
reg_errcode_t err;
- int i;
+ Idx i;
re_node_set eclosure;
- int ret;
- int incomplete = 0;
+ bool ok;
+ bool incomplete = false;
err = re_node_set_alloc (&eclosure, dfa->edests[node].nelem + 1);
if (BE (err != REG_NOERROR, 0))
return err;
for (i = 0; i < dfa->edests[node].nelem; ++i)
{
re_node_set eclosure_elem;
- int edest = dfa->edests[node].elems[i];
- /* If calculating the epsilon closure of `edest' is in progress,
+ Idx edest = dfa->edests[node].elems[i];
+ /* If calculating the epsilon closure of 'edest' is in progress,
return intermediate result. */
if (dfa->eclosures[edest].nelem == -1)
{
- incomplete = 1;
+ incomplete = true;
continue;
}
- /* If we haven't calculated the epsilon closure of `edest' yet,
+ /* If we haven't calculated the epsilon closure of 'edest' yet,
calculate now. Otherwise use calculated epsilon closure. */
if (dfa->eclosures[edest].nelem == 0)
{
- err = calc_eclosure_iter (&eclosure_elem, dfa, edest, 0);
+ err = calc_eclosure_iter (&eclosure_elem, dfa, edest, false);
if (BE (err != REG_NOERROR, 0))
return err;
}
the epsilon closure of this node is also incomplete. */
if (dfa->eclosures[edest].nelem == 0)
{
- incomplete = 1;
+ incomplete = true;
re_node_set_free (&eclosure_elem);
}
}
/* An epsilon closure includes itself. */
- ret = re_node_set_insert (&eclosure, node);
- if (BE (ret < 0, 0))
+ ok = re_node_set_insert (&eclosure, node);
+ if (BE (! ok, 0))
return REG_ESPACE;
if (incomplete && !root)
dfa->eclosures[node].nelem = 0;
case '.':
token->type = OP_OPEN_COLL_ELEM;
break;
+
case '=':
token->type = OP_OPEN_EQUIV_CLASS;
break;
+
case ':':
if (syntax & RE_CHAR_CLASSES)
{
token->type = OP_OPEN_CHAR_CLASS;
break;
}
- /* else fall through. */
+ FALLTHROUGH;
default:
token->type = CHARACTER;
token->opr.c = c;
parse (re_string_t *regexp, regex_t *preg, reg_syntax_t syntax,
reg_errcode_t *err)
{
- re_dfa_t *dfa = (re_dfa_t *) preg->buffer;
+ re_dfa_t *dfa = preg->buffer;
bin_tree_t *tree, *eor, *root;
re_token_t current_token;
dfa->syntax = syntax;
static bin_tree_t *
parse_reg_exp (re_string_t *regexp, regex_t *preg, re_token_t *token,
- reg_syntax_t syntax, int nest, reg_errcode_t *err)
+ reg_syntax_t syntax, Idx nest, reg_errcode_t *err)
{
- re_dfa_t *dfa = (re_dfa_t *) preg->buffer;
+ re_dfa_t *dfa = preg->buffer;
bin_tree_t *tree, *branch = NULL;
+ bitset_word_t initial_bkref_map = dfa->completed_bkref_map;
tree = parse_branch (regexp, preg, token, syntax, nest, err);
if (BE (*err != REG_NOERROR && tree == NULL, 0))
return NULL;
if (token->type != OP_ALT && token->type != END_OF_RE
&& (nest == 0 || token->type != OP_CLOSE_SUBEXP))
{
+ bitset_word_t accumulated_bkref_map = dfa->completed_bkref_map;
+ dfa->completed_bkref_map = initial_bkref_map;
branch = parse_branch (regexp, preg, token, syntax, nest, err);
if (BE (*err != REG_NOERROR && branch == NULL, 0))
{
postorder (tree, free_tree, NULL);
return NULL;
}
+ dfa->completed_bkref_map |= accumulated_bkref_map;
}
else
branch = NULL;
static bin_tree_t *
parse_branch (re_string_t *regexp, regex_t *preg, re_token_t *token,
- reg_syntax_t syntax, int nest, reg_errcode_t *err)
+ reg_syntax_t syntax, Idx nest, reg_errcode_t *err)
{
- bin_tree_t *tree, *exp;
- re_dfa_t *dfa = (re_dfa_t *) preg->buffer;
+ bin_tree_t *tree, *expr;
+ re_dfa_t *dfa = preg->buffer;
tree = parse_expression (regexp, preg, token, syntax, nest, err);
if (BE (*err != REG_NOERROR && tree == NULL, 0))
return NULL;
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))
{
if (tree != NULL)
postorder (tree, free_tree, NULL);
return NULL;
}
- if (tree != NULL && exp != NULL)
+ if (tree != NULL && expr != NULL)
{
- bin_tree_t *newtree = create_tree (dfa, tree, exp, CONCAT);
+ bin_tree_t *newtree = create_tree (dfa, tree, expr, CONCAT);
if (newtree == NULL)
{
- postorder (exp, free_tree, NULL);
+ postorder (expr, free_tree, NULL);
postorder (tree, free_tree, NULL);
*err = REG_ESPACE;
return NULL;
tree = newtree;
}
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;
}
static bin_tree_t *
parse_expression (re_string_t *regexp, regex_t *preg, re_token_t *token,
- reg_syntax_t syntax, int nest, reg_errcode_t *err)
+ reg_syntax_t syntax, Idx nest, reg_errcode_t *err)
{
- re_dfa_t *dfa = (re_dfa_t *) preg->buffer;
+ re_dfa_t *dfa = preg->buffer;
bin_tree_t *tree;
switch (token->type)
{
}
#endif
break;
+
case OP_OPEN_SUBEXP:
tree = parse_sub_exp (regexp, preg, token, syntax, nest + 1, err);
if (BE (*err != REG_NOERROR && tree == NULL, 0))
return NULL;
break;
+
case OP_OPEN_BRACKET:
tree = parse_bracket_exp (regexp, dfa, token, syntax, err);
if (BE (*err != REG_NOERROR && tree == NULL, 0))
return NULL;
break;
+
case OP_BACK_REF:
if (!BE (dfa->completed_bkref_map & (1 << token->opr.idx), 1))
{
++dfa->nbackref;
dfa->has_mb_node = 1;
break;
+
case OP_OPEN_DUP_NUM:
if (syntax & RE_CONTEXT_INVALID_DUP)
{
*err = REG_BADRPT;
return NULL;
}
- /* FALLTHROUGH */
+ FALLTHROUGH;
case OP_DUP_ASTERISK:
case OP_DUP_PLUS:
case OP_DUP_QUESTION:
fetch_token (token, regexp, syntax);
return parse_expression (regexp, preg, token, syntax, nest, err);
}
- /* else fall through */
+ FALLTHROUGH;
case OP_CLOSE_SUBEXP:
if ((token->type == OP_CLOSE_SUBEXP) &&
!(syntax & RE_UNMATCHED_RIGHT_PAREN_ORD))
*err = REG_ERPAREN;
return NULL;
}
- /* else fall through */
+ FALLTHROUGH;
case OP_CLOSE_DUP_NUM:
/* We treat it as a normal character. */
return NULL;
}
break;
+
case ANCHOR:
if ((token->opr.ctx_type
& (WORD_DELIM | NOT_WORD_DELIM | WORD_FIRST | WORD_LAST))
it must not be "<ANCHOR(^)><REPEAT(*)>". */
fetch_token (token, regexp, syntax);
return tree;
+
case OP_PERIOD:
tree = create_token_tree (dfa, NULL, NULL, token);
if (BE (tree == NULL, 0))
if (dfa->mb_cur_max > 1)
dfa->has_mb_node = 1;
break;
+
case OP_WORD:
case OP_NOTWORD:
tree = build_charclass_op (dfa, regexp->trans,
- (const unsigned char *) "alnum",
- (const unsigned char *) "_",
+ "alnum",
+ "_",
token->type == OP_NOTWORD, err);
if (BE (*err != REG_NOERROR && tree == NULL, 0))
return NULL;
break;
+
case OP_SPACE:
case OP_NOTSPACE:
tree = build_charclass_op (dfa, regexp->trans,
- (const unsigned char *) "space",
- (const unsigned char *) "",
+ "space",
+ "",
token->type == OP_NOTSPACE, err);
if (BE (*err != REG_NOERROR && tree == NULL, 0))
return NULL;
break;
+
case OP_ALT:
case END_OF_RE:
return NULL;
+
case BACK_SLASH:
*err = REG_EESCAPE;
return NULL;
+
default:
/* Must not happen? */
#ifdef DEBUG
while (token->type == OP_DUP_ASTERISK || token->type == OP_DUP_PLUS
|| token->type == OP_DUP_QUESTION || token->type == OP_OPEN_DUP_NUM)
{
- bin_tree_t *dup_tree = parse_dup_op (tree, regexp, dfa, token, syntax, err);
+ bin_tree_t *dup_tree = parse_dup_op (tree, regexp, dfa, token,
+ syntax, err);
if (BE (*err != REG_NOERROR && dup_tree == NULL, 0))
{
if (tree != NULL)
static bin_tree_t *
parse_sub_exp (re_string_t *regexp, regex_t *preg, re_token_t *token,
- reg_syntax_t syntax, int nest, reg_errcode_t *err)
+ reg_syntax_t syntax, Idx nest, reg_errcode_t *err)
{
- re_dfa_t *dfa = (re_dfa_t *) preg->buffer;
+ re_dfa_t *dfa = preg->buffer;
bin_tree_t *tree;
size_t cur_nsub;
cur_nsub = preg->re_nsub++;
re_token_t *token, reg_syntax_t syntax, reg_errcode_t *err)
{
bin_tree_t *tree = NULL, *old_tree = NULL;
- int i, start, end, start_idx = re_string_cur_idx (regexp);
+ Idx i, start, end, start_idx = re_string_cur_idx (regexp);
re_token_t start_token = *token;
if (token->type == OP_OPEN_DUP_NUM)
return elem;
}
- if (BE ((end != -1 && start > end) || token->type != OP_CLOSE_DUP_NUM, 0))
+ if (BE ((end != -1 && start > end)
+ || token->type != OP_CLOSE_DUP_NUM, 0))
{
/* First number greater than second. */
*err = REG_BADBR;
return NULL;
}
+
+ if (BE (RE_DUP_MAX < (end == -1 ? start : end), 0))
+ {
+ *err = REG_ESIZE;
+ return NULL;
+ }
}
else
{
old_tree = NULL;
if (elem->token.type == SUBEXP)
- postorder (elem, mark_opt_subexp, (void *) (long) elem->token.opr.idx);
+ {
+ uintptr_t subidx = elem->token.opr.idx;
+ postorder (elem, mark_opt_subexp, (void *) subidx);
+ }
- tree = create_tree (dfa, elem, NULL, (end == -1 ? OP_DUP_ASTERISK : OP_ALT));
+ tree = create_tree (dfa, elem, NULL,
+ (end == -1 ? OP_DUP_ASTERISK : OP_ALT));
if (BE (tree == NULL, 0))
goto parse_dup_op_espace;
/* This loop is actually executed only when end != -1,
to rewrite <re>{0,n} as (<re>(<re>...<re>?)?)?... We have
already created the start+1-th copy. */
- for (i = start + 2; i <= end; ++i)
- {
- elem = duplicate_tree (elem, dfa);
- tree = create_tree (dfa, tree, elem, CONCAT);
- if (BE (elem == NULL || tree == NULL, 0))
- goto parse_dup_op_espace;
-
- tree = create_tree (dfa, tree, NULL, OP_ALT);
- if (BE (tree == NULL, 0))
- goto parse_dup_op_espace;
- }
+ if (TYPE_SIGNED (Idx) || end != -1)
+ for (i = start + 2; i <= end; ++i)
+ {
+ elem = duplicate_tree (elem, dfa);
+ tree = create_tree (dfa, tree, elem, CONCAT);
+ if (BE (elem == NULL || tree == NULL, 0))
+ goto parse_dup_op_espace;
+
+ tree = create_tree (dfa, tree, NULL, OP_ALT);
+ if (BE (tree == NULL, 0))
+ goto parse_dup_op_espace;
+ }
if (old_tree)
tree = create_tree (dfa, old_tree, tree, CONCAT);
#define BRACKET_NAME_BUF_SIZE 32
#ifndef _LIBC
+
+# ifdef RE_ENABLE_I18N
+/* Convert the byte B to the corresponding wide character. In a
+ unibyte locale, treat B as itself if it is an encoding error.
+ In a multibyte locale, return WEOF if B is an encoding error. */
+static wint_t
+parse_byte (unsigned char b, re_charset_t *mbcset)
+{
+ wint_t wc = __btowc (b);
+ return wc == WEOF && !mbcset ? b : wc;
+}
+#endif
+
/* Local function for parse_bracket_exp only used in case of NOT _LIBC.
Build the range expression which starts from START_ELEM, and ends
at END_ELEM. The result are written to MBCSET and SBCSET.
static reg_errcode_t
# 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)
+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;
wchar_t wc;
wint_t start_wc;
wint_t end_wc;
- wchar_t cmp_buf[6] = {L'\0', L'\0', L'\0', L'\0', L'\0', L'\0'};
start_ch = ((start_elem->type == SB_CHAR) ? start_elem->opr.ch
: ((start_elem->type == COLL_SYM) ? start_elem->opr.name[0]
: ((end_elem->type == COLL_SYM) ? end_elem->opr.name[0]
: 0));
start_wc = ((start_elem->type == SB_CHAR || start_elem->type == COLL_SYM)
- ? __btowc (start_ch) : start_elem->opr.wch);
+ ? parse_byte (start_ch, mbcset) : start_elem->opr.wch);
end_wc = ((end_elem->type == SB_CHAR || end_elem->type == COLL_SYM)
- ? __btowc (end_ch) : end_elem->opr.wch);
+ ? parse_byte (end_ch, mbcset) : end_elem->opr.wch);
if (start_wc == WEOF || end_wc == WEOF)
return REG_ECOLLATE;
- cmp_buf[0] = start_wc;
- cmp_buf[4] = end_wc;
- if (__wcscoll (cmp_buf, cmp_buf + 4) > 0)
+ else if (BE ((syntax & RE_NO_EMPTY_RANGES) && start_wc > end_wc, 0))
return REG_ERANGE;
/* Got valid collation sequence values, add them as a new entry.
{
/* There is not enough space, need realloc. */
wchar_t *new_array_start, *new_array_end;
- int new_nranges;
+ Idx new_nranges;
/* +1 in case of mbcset->nranges is 0. */
new_nranges = 2 * mbcset->nranges + 1;
new_nranges);
if (BE (new_array_start == NULL || new_array_end == NULL, 0))
- return REG_ESPACE;
+ {
+ re_free (new_array_start);
+ re_free (new_array_end);
+ return REG_ESPACE;
+ }
mbcset->range_starts = new_array_start;
mbcset->range_ends = new_array_end;
/* Build the table for single byte characters. */
for (wc = 0; wc < SBC_MAX; ++wc)
{
- cmp_buf[2] = wc;
- if (__wcscoll (cmp_buf, cmp_buf + 2) <= 0
- && __wcscoll (cmp_buf + 2, cmp_buf + 4) <= 0)
+ if (start_wc <= wc && wc <= end_wc)
bitset_set (sbcset, wc);
}
}
static reg_errcode_t
# ifdef RE_ENABLE_I18N
build_collating_symbol (bitset_t sbcset, re_charset_t *mbcset,
- int *coll_sym_alloc, const unsigned char *name)
+ Idx *coll_sym_alloc, const unsigned char *name)
# else /* not RE_ENABLE_I18N */
build_collating_symbol (bitset_t sbcset, const unsigned char *name)
# endif /* not RE_ENABLE_I18N */
0))
return REG_ERANGE;
+ /* FIXME: Implement rational ranges here, too. */
start_collseq = lookup_collation_sequence_value (start_elem);
end_collseq = lookup_collation_sequence_value (end_elem);
/* Check start/end collation sequence values. */
/* There is not enough space, need realloc. */
uint32_t *new_array_start;
uint32_t *new_array_end;
- int new_nranges;
+ Idx new_nranges;
/* +1 in case of mbcset->nranges is 0. */
new_nranges = 2 * mbcset->nranges + 1;
auto inline reg_errcode_t
__attribute__ ((always_inline))
build_collating_symbol (bitset_t sbcset, re_charset_t *mbcset,
- int *coll_sym_alloc, const unsigned char *name)
+ Idx *coll_sym_alloc, const unsigned char *name)
{
int32_t elem, idx;
size_t name_len = strlen ((const char *) name);
{
/* Not enough, realloc it. */
/* +1 in case of mbcset->ncoll_syms is 0. */
- int new_coll_sym_alloc = 2 * mbcset->ncoll_syms + 1;
+ Idx new_coll_sym_alloc = 2 * mbcset->ncoll_syms + 1;
/* Use realloc since mbcset->coll_syms is NULL
if *alloc == 0. */
int32_t *new_coll_syms = re_realloc (mbcset->coll_syms, int32_t,
re_bitset_ptr_t sbcset;
#ifdef RE_ENABLE_I18N
re_charset_t *mbcset;
- int coll_sym_alloc = 0, range_alloc = 0, mbchar_alloc = 0;
- int equiv_class_alloc = 0, char_class_alloc = 0;
+ Idx coll_sym_alloc = 0, range_alloc = 0, mbchar_alloc = 0;
+ Idx equiv_class_alloc = 0, char_class_alloc = 0;
#endif /* not RE_ENABLE_I18N */
- int non_match = 0;
+ bool non_match = false;
bin_tree_t *work_tree;
int token_len;
- int first_round = 1;
+ bool first_round = true;
#ifdef _LIBC
collseqmb = (const unsigned char *)
_NL_CURRENT (LC_COLLATE, _NL_COLLATE_COLLSEQMB);
#ifdef RE_ENABLE_I18N
mbcset->non_match = 1;
#endif /* not RE_ENABLE_I18N */
- non_match = 1;
+ non_match = true;
if (syntax & RE_HAT_LISTS_NOT_NEWLINE)
bitset_set (sbcset, '\n');
re_string_skip_bytes (regexp, token_len); /* Skip a token. */
unsigned char start_name_buf[BRACKET_NAME_BUF_SIZE];
unsigned char end_name_buf[BRACKET_NAME_BUF_SIZE];
reg_errcode_t ret;
- int token_len2 = 0, is_range_exp = 0;
+ int token_len2 = 0;
+ bool is_range_exp = false;
re_token_t token2;
start_elem.opr.name = start_name_buf;
*err = ret;
goto parse_bracket_exp_free_return;
}
- first_round = 0;
+ first_round = false;
/* Get information about the next token. We need it in any case. */
token_len = peek_token_bracket (token, regexp, syntax);
token->type = CHARACTER;
}
else
- is_range_exp = 1;
+ is_range_exp = true;
}
}
- if (is_range_exp == 1)
+ if (is_range_exp == true)
{
end_elem.opr.name = end_name_buf;
end_elem.type = COLL_SYM;
ret = parse_bracket_element (&end_elem, regexp, &token2, token_len2,
- dfa, syntax, 1);
+ dfa, syntax, true);
if (BE (ret != REG_NOERROR, 0))
{
*err = ret;
&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))
#ifdef RE_ENABLE_I18N
mbcset, &char_class_alloc,
#endif /* RE_ENABLE_I18N */
- start_elem.opr.name, syntax);
+ (const char *) start_elem.opr.name,
+ syntax);
if (BE (*err != REG_NOERROR, 0))
goto parse_bracket_exp_free_return;
break;
static reg_errcode_t
parse_bracket_element (bracket_elem_t *elem, re_string_t *regexp,
re_token_t *token, int token_len, re_dfa_t *dfa,
- reg_syntax_t syntax, int accept_hyphen)
+ reg_syntax_t syntax, bool accept_hyphen)
{
#ifdef RE_ENABLE_I18N
int cur_char_size;
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)
+ Idx *equiv_class_alloc, const unsigned char *name)
#else /* not RE_ENABLE_I18N */
build_equiv_class (bitset_t sbcset, const unsigned char *name)
#endif /* not RE_ENABLE_I18N */
{
/* Not enough, realloc it. */
/* +1 in case of mbcset->nequiv_classes is 0. */
- int new_equiv_class_alloc = 2 * mbcset->nequiv_classes + 1;
+ Idx new_equiv_class_alloc = 2 * mbcset->nequiv_classes + 1;
/* Use realloc since the array is NULL if *alloc == 0. */
int32_t *new_equiv_classes = re_realloc (mbcset->equiv_classes,
int32_t,
static reg_errcode_t
#ifdef RE_ENABLE_I18N
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)
+ re_charset_t *mbcset, Idx *char_class_alloc,
+ const char *class_name, reg_syntax_t syntax)
#else /* not RE_ENABLE_I18N */
build_charclass (RE_TRANSLATE_TYPE trans, bitset_t sbcset,
- const unsigned char *class_name, reg_syntax_t syntax)
+ const char *class_name, reg_syntax_t syntax)
#endif /* not RE_ENABLE_I18N */
{
int i;
- const char *name = (const char *) class_name;
+ const char *name = class_name;
/* In case of REG_ICASE "upper" and "lower" match the both of
upper and lower cases. */
{
/* Not enough, realloc it. */
/* +1 in case of mbcset->nchar_classes is 0. */
- int new_char_class_alloc = 2 * mbcset->nchar_classes + 1;
+ Idx new_char_class_alloc = 2 * mbcset->nchar_classes + 1;
/* Use realloc since array is NULL if *alloc == 0. */
wctype_t *new_char_classes = re_realloc (mbcset->char_classes, wctype_t,
new_char_class_alloc);
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)
static bin_tree_t *
build_charclass_op (re_dfa_t *dfa, RE_TRANSLATE_TYPE trans,
- const unsigned char *class_name,
- const unsigned char *extra, int non_match,
+ const char *class_name,
+ const char *extra, bool non_match,
reg_errcode_t *err)
{
re_bitset_ptr_t sbcset;
#ifdef RE_ENABLE_I18N
re_charset_t *mbcset;
- int alloc = 0;
+ Idx alloc = 0;
#endif /* not RE_ENABLE_I18N */
reg_errcode_t ret;
re_token_t br_token;
bin_tree_t *tree;
sbcset = (re_bitset_ptr_t) calloc (sizeof (bitset_t), 1);
-#ifdef RE_ENABLE_I18N
- mbcset = (re_charset_t *) calloc (sizeof (re_charset_t), 1);
-#endif /* RE_ENABLE_I18N */
-
-#ifdef RE_ENABLE_I18N
- if (BE (sbcset == NULL || mbcset == NULL, 0))
-#else /* not RE_ENABLE_I18N */
if (BE (sbcset == NULL, 0))
-#endif /* not RE_ENABLE_I18N */
{
*err = REG_ESPACE;
return NULL;
}
-
- if (non_match)
- {
#ifdef RE_ENABLE_I18N
- mbcset->non_match = 1;
-#endif /* not RE_ENABLE_I18N */
+ mbcset = (re_charset_t *) calloc (sizeof (re_charset_t), 1);
+ if (BE (mbcset == NULL, 0))
+ {
+ re_free (sbcset);
+ *err = REG_ESPACE;
+ return NULL;
}
+ mbcset->non_match = non_match;
+#endif /* RE_ENABLE_I18N */
/* We don't care the syntax in this case. */
ret = build_charclass (trans, sbcset,
#endif
/* Build a tree for simple bracket. */
+#if defined GCC_LINT || defined lint
+ memset (&br_token, 0, sizeof br_token);
+#endif
br_token.type = SIMPLE_BRACKET;
br_token.opr.sbcset = sbcset;
tree = create_token_tree (dfa, NULL, NULL, &br_token);
}
/* This is intended for the expressions like "a{1,3}".
- Fetch a number from `input', and return the number.
- Return -1, if the number field is empty like "{,1}".
- Return -2, If an error is occured. */
+ Fetch a number from 'input', and return the number.
+ Return -1 if the number field is empty like "{,1}".
+ Return RE_DUP_MAX + 1 if the number field is too large.
+ Return -2 if an error occurred. */
-static int
+static Idx
fetch_number (re_string_t *input, re_token_t *token, reg_syntax_t syntax)
{
- int num = -1;
+ Idx num = -1;
unsigned char c;
while (1)
{
if (token->type == OP_CLOSE_DUP_NUM || c == ',')
break;
num = ((token->type != CHARACTER || c < '0' || '9' < c || num == -2)
- ? -2 : ((num == -1) ? c - '0' : num * 10 + c - '0'));
- num = (num > RE_DUP_MAX) ? -2 : num;
+ ? -2
+ : num == -1
+ ? c - '0'
+ : MIN (RE_DUP_MAX + 1, num * 10 + c - '0'));
}
return num;
}
re_token_type_t type)
{
re_token_t t;
+#if defined GCC_LINT || defined lint
+ memset (&t, 0, sizeof t);
+#endif
t.type = type;
return create_token_tree (dfa, left, right, &t);
}
static reg_errcode_t
mark_opt_subexp (void *extra, bin_tree_t *node)
{
- int idx = (int) (long) extra;
+ Idx idx = (uintptr_t) extra;
if (node->token.type == SUBEXP && node->token.opr.idx == idx)
node->token.opt_subexp = 1;
projects / thirdparty / glibc.git / blobdiff