mirror of
https://sourceware.org/git/glibc.git
synced 2024-11-22 13:00:06 +00:00
* posix/regex_internal.c (re_string_reconstruct): Avoid calling
mbrtowc for very simple UTF-8 case. 2005-09-01 Paul Eggert <eggert@cs.ucla.edu> * posix/regex_internal.c (build_wcs_upper_buffer): Fix portability bugs in int versus size_t comparisons. 2005-09-06 Ulrich Drepper <drepper@redhat.com> * posix/regex_internal.c (re_acquire_state): Make DFA pointer arg a pointer-to-const. (re_acquire_state_context): Likewise. * posix/regex_internal.h: Adjust prototypes. 2005-08-31 Jim Meyering <jim@meyering.net> * posix/regcomp.c (search_duplicated_node): Make first pointer arg a pointer-to-const. * posix/regex_internal.c (create_ci_newstate, create_cd_newstate, register_state): Likewise. * posix/regexec.c (search_cur_bkref_entry, check_dst_limits): (check_dst_limits_calc_pos_1, check_dst_limits_calc_pos): (group_nodes_into_DFAstates): Likewise. * posix/regexec.c (re_search_internal): Simplify update of rm_so and rm_eo by replacing "if (A == B) A += C - B;" with the equivalent of "if (A == B) A = C;". 2005-09-06 Ulrich Drepper <drepper@redhat.com> * posix/regcomp.c (re_compile_internal): Change third parameter type to size_t. (init_dfa): Likewise. Make sure that arithmetic on pat_len doesn't overflow. * posix/regex_internal.h (struct re_dfa_t): Change type of nodes_alloc and nodes_len to size_t. * posix/regex_internal.c (re_dfa_add_node): Use size_t as type for new_nodes_alloc. Check for overflow. 2005-08-31 Paul Eggert <eggert@cs.ucla.edu> * posix/regcomp.c (re_compile_fastmap_iter, init_dfa, init_word_char): (optimize_subexps, lower_subexp): Don't assume 1<<31 has defined behavior on hosts with 32-bit int, since the signed shift might overflow. Use 1u<<31 instead. * posix/regex_internal.h (bitset_set, bitset_clear, bitset_contain): Likewise. * posix/regexec.c (check_dst_limits_calc_pos_1): Likewise. (check_subexp_matching_top): Likewise. * posix/regcomp.c (optimize_subexps, lower_subexp): Use CHAR_BIT rather than 8, for clarity. * posix/regexec.c (check_dst_limits_calc_pos_1): (check_subexp_matching_top): Likewise. * posix/regcomp.c (init_dfa): Make table_size unsigned, so that we don't have to worry about portability issues when shifting it left. Remove no-longer-needed test for table_size > 0. * posix/regcomp.c (parse_sub_exp): Do not shift more bits than there are in a word, as the resulting behavior is undefined. * posix/regexec.c (check_dst_limits_calc_pos_1): Likewise; in one case, a <= should have been an <, and in another case the whole test was missing. * posix/regex_internal.h (BYTE_BITS): Remove. All uses changed to the standard name CHAR_BIT.
This commit is contained in:
Ulrich Drepper
parent
2d87db5b53
commit
01ed6ceb7c
67
ChangeLog
67
ChangeLog
@ -1,5 +1,72 @@
|
||||
2005-09-06 Ulrich Drepper <drepper@redhat.com>
|
||||
|
||||
* posix/regex_internal.c (re_string_reconstruct): Avoid calling
|
||||
mbrtowc for very simple UTF-8 case.
|
||||
|
||||
2005-09-01 Paul Eggert <eggert@cs.ucla.edu>
|
||||
|
||||
* posix/regex_internal.c (build_wcs_upper_buffer): Fix portability
|
||||
bugs in int versus size_t comparisons.
|
||||
|
||||
2005-09-06 Ulrich Drepper <drepper@redhat.com>
|
||||
|
||||
* posix/regex_internal.c (re_acquire_state): Make DFA pointer arg
|
||||
a pointer-to-const.
|
||||
(re_acquire_state_context): Likewise.
|
||||
* posix/regex_internal.h: Adjust prototypes.
|
||||
|
||||
2005-08-31 Jim Meyering <jim@meyering.net>
|
||||
|
||||
* posix/regcomp.c (search_duplicated_node): Make first pointer arg
|
||||
a pointer-to-const.
|
||||
* posix/regex_internal.c (create_ci_newstate, create_cd_newstate,
|
||||
register_state): Likewise.
|
||||
* posix/regexec.c (search_cur_bkref_entry, check_dst_limits):
|
||||
(check_dst_limits_calc_pos_1, check_dst_limits_calc_pos):
|
||||
(group_nodes_into_DFAstates): Likewise.
|
||||
|
||||
2005-08-31 Paul Eggert <eggert@cs.ucla.edu>
|
||||
|
||||
* posix/regexec.c (re_search_internal): Simplify update of
|
||||
rm_so and rm_eo by replacing "if (A == B) A += C - B;"
|
||||
with the equivalent of "if (A == B) A = C;".
|
||||
|
||||
2005-09-06 Ulrich Drepper <drepper@redhat.com>
|
||||
|
||||
* posix/regcomp.c (re_compile_internal): Change third parameter type
|
||||
to size_t.
|
||||
(init_dfa): Likewise. Make sure that arithmetic on pat_len doesn't
|
||||
overflow.
|
||||
* posix/regex_internal.h (struct re_dfa_t): Change type of nodes_alloc
|
||||
and nodes_len to size_t.
|
||||
* posix/regex_internal.c (re_dfa_add_node): Use size_t as type for
|
||||
new_nodes_alloc. Check for overflow.
|
||||
|
||||
2005-08-31 Paul Eggert <eggert@cs.ucla.edu>
|
||||
|
||||
* posix/regcomp.c (re_compile_fastmap_iter, init_dfa, init_word_char):
|
||||
(optimize_subexps, lower_subexp):
|
||||
Don't assume 1<<31 has defined behavior on hosts with 32-bit int,
|
||||
since the signed shift might overflow. Use 1u<<31 instead.
|
||||
* posix/regex_internal.h (bitset_set, bitset_clear, bitset_contain):
|
||||
Likewise.
|
||||
* posix/regexec.c (check_dst_limits_calc_pos_1): Likewise.
|
||||
(check_subexp_matching_top): Likewise.
|
||||
* posix/regcomp.c (optimize_subexps, lower_subexp):
|
||||
Use CHAR_BIT rather than 8, for clarity.
|
||||
* posix/regexec.c (check_dst_limits_calc_pos_1):
|
||||
(check_subexp_matching_top): Likewise.
|
||||
* posix/regcomp.c (init_dfa): Make table_size unsigned, so that we
|
||||
don't have to worry about portability issues when shifting it left.
|
||||
Remove no-longer-needed test for table_size > 0.
|
||||
* posix/regcomp.c (parse_sub_exp): Do not shift more bits than there
|
||||
are in a word, as the resulting behavior is undefined.
|
||||
* posix/regexec.c (check_dst_limits_calc_pos_1): Likewise;
|
||||
in one case, a <= should have been an <, and in another case the
|
||||
whole test was missing.
|
||||
* posix/regex_internal.h (BYTE_BITS): Remove. All uses changed to
|
||||
the standard name CHAR_BIT.
|
||||
|
||||
* posix/regex_internal.h (re_sub_match_top_t): Remove unused member
|
||||
next_last_offset.
|
||||
(struct re_dfa_t): Remove unused member states_alloc.
|
||||
|
||||
@ -19,11 +19,11 @@
|
||||
02111-1307 USA. */
|
||||
|
||||
static reg_errcode_t re_compile_internal (regex_t *preg, const char * pattern,
|
||||
int length, reg_syntax_t syntax);
|
||||
size_t length, reg_syntax_t syntax);
|
||||
static void re_compile_fastmap_iter (regex_t *bufp,
|
||||
const re_dfastate_t *init_state,
|
||||
char *fastmap);
|
||||
static reg_errcode_t init_dfa (re_dfa_t *dfa, int pat_len);
|
||||
static reg_errcode_t init_dfa (re_dfa_t *dfa, size_t pat_len);
|
||||
static void init_word_char (re_dfa_t *dfa);
|
||||
#ifdef RE_ENABLE_I18N
|
||||
static void free_charset (re_charset_t *cset);
|
||||
@ -51,7 +51,7 @@ static reg_errcode_t duplicate_node_closure (re_dfa_t *dfa, int top_org_node,
|
||||
int top_clone_node, int root_node,
|
||||
unsigned int constraint);
|
||||
static int duplicate_node (re_dfa_t *dfa, int org_idx, unsigned int constraint);
|
||||
static int search_duplicated_node (re_dfa_t *dfa, int org_node,
|
||||
static int search_duplicated_node (const re_dfa_t *dfa, int 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,
|
||||
@ -368,7 +368,7 @@ re_compile_fastmap_iter (bufp, init_state, fastmap)
|
||||
int i, j, ch;
|
||||
for (i = 0, ch = 0; i < BITSET_UINTS; ++i)
|
||||
for (j = 0; j < UINT_BITS; ++j, ++ch)
|
||||
if (dfa->nodes[node].opr.sbcset[i] & (1 << j))
|
||||
if (dfa->nodes[node].opr.sbcset[i] & (1u << j))
|
||||
re_set_fastmap (fastmap, icase, ch);
|
||||
}
|
||||
#ifdef RE_ENABLE_I18N
|
||||
@ -740,7 +740,7 @@ static reg_errcode_t
|
||||
re_compile_internal (preg, pattern, length, syntax)
|
||||
regex_t *preg;
|
||||
const char * pattern;
|
||||
int length;
|
||||
size_t length;
|
||||
reg_syntax_t syntax;
|
||||
{
|
||||
reg_errcode_t err = REG_NOERROR;
|
||||
@ -781,6 +781,7 @@ re_compile_internal (preg, pattern, length, syntax)
|
||||
return err;
|
||||
}
|
||||
#ifdef DEBUG
|
||||
/* Note: length+1 will not overflow since it is checked in init_dfa. */
|
||||
dfa->re_str = re_malloc (char, length + 1);
|
||||
strncpy (dfa->re_str, pattern, length + 1);
|
||||
#endif
|
||||
@ -840,9 +841,9 @@ re_compile_internal (preg, pattern, length, syntax)
|
||||
static reg_errcode_t
|
||||
init_dfa (dfa, pat_len)
|
||||
re_dfa_t *dfa;
|
||||
int pat_len;
|
||||
size_t pat_len;
|
||||
{
|
||||
int table_size;
|
||||
unsigned int table_size;
|
||||
#ifndef _LIBC
|
||||
char *codeset_name;
|
||||
#endif
|
||||
@ -852,11 +853,15 @@ init_dfa (dfa, pat_len)
|
||||
/* 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)
|
||||
return REG_ESPACE;
|
||||
|
||||
dfa->nodes_alloc = pat_len + 1;
|
||||
dfa->nodes = re_malloc (re_token_t, dfa->nodes_alloc);
|
||||
|
||||
/* table_size = 2 ^ ceil(log pat_len) */
|
||||
for (table_size = 1; table_size > 0; table_size <<= 1)
|
||||
for (table_size = 1; ; table_size <<= 1)
|
||||
if (table_size > pat_len)
|
||||
break;
|
||||
|
||||
@ -916,7 +921,7 @@ init_dfa (dfa, pat_len)
|
||||
{
|
||||
wint_t wch = __btowc (ch);
|
||||
if (wch != WEOF)
|
||||
dfa->sb_char[i] |= 1 << j;
|
||||
dfa->sb_char[i] |= 1u << j;
|
||||
# ifndef _LIBC
|
||||
if (isascii (ch) && wch != ch)
|
||||
dfa->map_notascii = 1;
|
||||
@ -944,7 +949,7 @@ init_word_char (dfa)
|
||||
for (i = 0, ch = 0; i < BITSET_UINTS; ++i)
|
||||
for (j = 0; j < UINT_BITS; ++j, ++ch)
|
||||
if (isalnum (ch) || ch == '_')
|
||||
dfa->word_char[i] |= 1 << j;
|
||||
dfa->word_char[i] |= 1u << j;
|
||||
}
|
||||
|
||||
/* Free the work area which are only used while compiling. */
|
||||
@ -1277,8 +1282,8 @@ optimize_subexps (extra, node)
|
||||
node->left->parent = node;
|
||||
|
||||
dfa->subexp_map[other_idx] = dfa->subexp_map[node->token.opr.idx];
|
||||
if (other_idx < 8 * sizeof (dfa->used_bkref_map))
|
||||
dfa->used_bkref_map &= ~(1 << other_idx);
|
||||
if (other_idx < CHAR_BIT * sizeof dfa->used_bkref_map)
|
||||
dfa->used_bkref_map &= ~(1u << other_idx);
|
||||
}
|
||||
|
||||
return REG_NOERROR;
|
||||
@ -1326,8 +1331,8 @@ lower_subexp (err, preg, node)
|
||||
very common, so we do not lose much. An example that triggers
|
||||
this case is the sed "script" /\(\)/x. */
|
||||
&& node->left != NULL
|
||||
&& (node->token.opr.idx >= 8 * sizeof (dfa->used_bkref_map)
|
||||
|| !(dfa->used_bkref_map & (1 << node->token.opr.idx))))
|
||||
&& (node->token.opr.idx >= CHAR_BIT * sizeof dfa->used_bkref_map
|
||||
|| !(dfa->used_bkref_map & (1u << node->token.opr.idx))))
|
||||
return node->left;
|
||||
|
||||
/* Convert the SUBEXP node to the concatenation of an
|
||||
@ -1574,7 +1579,7 @@ duplicate_node_closure (dfa, top_org_node, top_clone_node, root_node,
|
||||
|
||||
static int
|
||||
search_duplicated_node (dfa, org_node, constraint)
|
||||
re_dfa_t *dfa;
|
||||
const re_dfa_t *dfa;
|
||||
int org_node;
|
||||
unsigned int constraint;
|
||||
{
|
||||
@ -2492,7 +2497,9 @@ parse_sub_exp (regexp, preg, token, syntax, nest, err)
|
||||
if (BE (*err != REG_NOERROR, 0))
|
||||
return NULL;
|
||||
}
|
||||
dfa->completed_bkref_map |= 1 << cur_nsub;
|
||||
|
||||
if (cur_nsub <= '9' - '1')
|
||||
dfa->completed_bkref_map |= 1 << cur_nsub;
|
||||
|
||||
tree = create_tree (dfa, tree, NULL, SUBEXP);
|
||||
if (BE (tree == NULL, 0))
|
||||
|
||||
@ -26,12 +26,13 @@ static void re_string_construct_common (const char *str, int len,
|
||||
static int re_string_skip_chars (re_string_t *pstr, int new_raw_idx,
|
||||
wint_t *last_wc) internal_function;
|
||||
#endif /* RE_ENABLE_I18N */
|
||||
static reg_errcode_t register_state (re_dfa_t *dfa, re_dfastate_t *newstate,
|
||||
static reg_errcode_t register_state (const re_dfa_t *dfa,
|
||||
re_dfastate_t *newstate,
|
||||
unsigned int hash) internal_function;
|
||||
static re_dfastate_t *create_ci_newstate (re_dfa_t *dfa,
|
||||
static re_dfastate_t *create_ci_newstate (const re_dfa_t *dfa,
|
||||
const re_node_set *nodes,
|
||||
unsigned int hash) internal_function;
|
||||
static re_dfastate_t *create_cd_newstate (re_dfa_t *dfa,
|
||||
static re_dfastate_t *create_cd_newstate (const re_dfa_t *dfa,
|
||||
const re_node_set *nodes,
|
||||
unsigned int context,
|
||||
unsigned int hash) internal_function;
|
||||
@ -654,37 +655,50 @@ re_string_reconstruct (pstr, idx, eflags)
|
||||
byte other than 0x80 - 0xbf. */
|
||||
raw = pstr->raw_mbs + pstr->raw_mbs_idx;
|
||||
end = raw + (offset - pstr->mb_cur_max);
|
||||
for (p = raw + offset - 1; p >= end; --p)
|
||||
if ((*p & 0xc0) != 0x80)
|
||||
{
|
||||
mbstate_t cur_state;
|
||||
wchar_t wc2;
|
||||
int mlen = raw + pstr->len - p;
|
||||
unsigned char buf[6];
|
||||
p = raw + offset - 1;
|
||||
#ifdef _LIBC
|
||||
/* We know the wchar_t encoding is UCS4, so for the simple
|
||||
case, ASCII characters, skip the conversion step. */
|
||||
if (isascii (*p) && BE (pstr->trans == NULL, 1))
|
||||
{
|
||||
memset (&pstr->cur_state, '\0', sizeof (mbstate_t));
|
||||
pstr->valid_len = 0;
|
||||
wc = (wchar_t) *p;
|
||||
}
|
||||
else
|
||||
#endif
|
||||
for (; p >= end; --p)
|
||||
if ((*p & 0xc0) != 0x80)
|
||||
{
|
||||
mbstate_t cur_state;
|
||||
wchar_t wc2;
|
||||
int mlen = raw + pstr->len - p;
|
||||
unsigned char buf[6];
|
||||
size_t mbclen;
|
||||
|
||||
q = p;
|
||||
if (BE (pstr->trans != NULL, 0))
|
||||
{
|
||||
int i = mlen < 6 ? mlen : 6;
|
||||
while (--i >= 0)
|
||||
buf[i] = pstr->trans[p[i]];
|
||||
q = buf;
|
||||
}
|
||||
/* XXX Don't use mbrtowc, we know which conversion
|
||||
to use (UTF-8 -> UCS4). */
|
||||
memset (&cur_state, 0, sizeof (cur_state));
|
||||
mlen = (mbrtowc (&wc2, (const char *) p, mlen,
|
||||
&cur_state)
|
||||
- (raw + offset - p));
|
||||
if (mlen >= 0)
|
||||
{
|
||||
memset (&pstr->cur_state, '\0',
|
||||
sizeof (mbstate_t));
|
||||
pstr->valid_len = mlen;
|
||||
wc = wc2;
|
||||
}
|
||||
break;
|
||||
}
|
||||
q = p;
|
||||
if (BE (pstr->trans != NULL, 0))
|
||||
{
|
||||
int i = mlen < 6 ? mlen : 6;
|
||||
while (--i >= 0)
|
||||
buf[i] = pstr->trans[p[i]];
|
||||
q = buf;
|
||||
}
|
||||
/* XXX Don't use mbrtowc, we know which conversion
|
||||
to use (UTF-8 -> UCS4). */
|
||||
memset (&cur_state, 0, sizeof (cur_state));
|
||||
mbclen = mbrtowc (&wc2, (const char *) p, mlen,
|
||||
&cur_state);
|
||||
if (raw + offset - p <= mbclen
|
||||
&& mbclen < (size_t) -2)
|
||||
{
|
||||
memset (&pstr->cur_state, '\0',
|
||||
sizeof (mbstate_t));
|
||||
pstr->valid_len = mbclen - (raw + offset - p);
|
||||
wc = wc2;
|
||||
}
|
||||
break;
|
||||
}
|
||||
}
|
||||
|
||||
if (wc == WEOF)
|
||||
@ -738,15 +752,15 @@ re_string_reconstruct (pstr, idx, eflags)
|
||||
}
|
||||
else
|
||||
#endif /* RE_ENABLE_I18N */
|
||||
if (BE (pstr->mbs_allocated, 0))
|
||||
{
|
||||
if (pstr->icase)
|
||||
build_upper_buffer (pstr);
|
||||
else if (pstr->trans != NULL)
|
||||
re_string_translate_buffer (pstr);
|
||||
}
|
||||
else
|
||||
pstr->valid_len = pstr->len;
|
||||
if (BE (pstr->mbs_allocated, 0))
|
||||
{
|
||||
if (pstr->icase)
|
||||
build_upper_buffer (pstr);
|
||||
else if (pstr->trans != NULL)
|
||||
re_string_translate_buffer (pstr);
|
||||
}
|
||||
else
|
||||
pstr->valid_len = pstr->len;
|
||||
|
||||
pstr->cur_idx = 0;
|
||||
return REG_NOERROR;
|
||||
@ -1345,12 +1359,16 @@ re_dfa_add_node (dfa, token)
|
||||
int type = token.type;
|
||||
if (BE (dfa->nodes_len >= dfa->nodes_alloc, 0))
|
||||
{
|
||||
int new_nodes_alloc = dfa->nodes_alloc * 2;
|
||||
size_t new_nodes_alloc = dfa->nodes_alloc * 2;
|
||||
int *new_nexts, *new_indices;
|
||||
re_node_set *new_edests, *new_eclosures;
|
||||
re_token_t *new_nodes;
|
||||
|
||||
re_token_t *new_nodes = re_realloc (dfa->nodes, re_token_t,
|
||||
new_nodes_alloc);
|
||||
/* Avoid overflows. */
|
||||
if (BE (new_nodes_alloc < dfa->nodes_alloc, 0))
|
||||
return -1;
|
||||
|
||||
new_nodes = re_realloc (dfa->nodes, re_token_t, new_nodes_alloc);
|
||||
if (BE (new_nodes == NULL, 0))
|
||||
return -1;
|
||||
dfa->nodes = new_nodes;
|
||||
@ -1403,7 +1421,7 @@ calc_state_hash (nodes, context)
|
||||
static re_dfastate_t*
|
||||
re_acquire_state (err, dfa, nodes)
|
||||
reg_errcode_t *err;
|
||||
re_dfa_t *dfa;
|
||||
const re_dfa_t *dfa;
|
||||
const re_node_set *nodes;
|
||||
{
|
||||
unsigned int hash;
|
||||
@ -1448,7 +1466,7 @@ re_acquire_state (err, dfa, nodes)
|
||||
static re_dfastate_t*
|
||||
re_acquire_state_context (err, dfa, nodes, context)
|
||||
reg_errcode_t *err;
|
||||
re_dfa_t *dfa;
|
||||
const re_dfa_t *dfa;
|
||||
const re_node_set *nodes;
|
||||
unsigned int context;
|
||||
{
|
||||
@ -1486,7 +1504,7 @@ re_acquire_state_context (err, dfa, nodes, context)
|
||||
|
||||
static reg_errcode_t
|
||||
register_state (dfa, newstate, hash)
|
||||
re_dfa_t *dfa;
|
||||
const re_dfa_t *dfa;
|
||||
re_dfastate_t *newstate;
|
||||
unsigned int hash;
|
||||
{
|
||||
@ -1525,7 +1543,7 @@ register_state (dfa, newstate, hash)
|
||||
|
||||
static re_dfastate_t *
|
||||
create_ci_newstate (dfa, nodes, hash)
|
||||
re_dfa_t *dfa;
|
||||
const re_dfa_t *dfa;
|
||||
const re_node_set *nodes;
|
||||
unsigned int hash;
|
||||
{
|
||||
@ -1576,7 +1594,7 @@ create_ci_newstate (dfa, nodes, hash)
|
||||
|
||||
static re_dfastate_t *
|
||||
create_cd_newstate (dfa, nodes, context, hash)
|
||||
re_dfa_t *dfa;
|
||||
const re_dfa_t *dfa;
|
||||
const re_node_set *nodes;
|
||||
unsigned int context, hash;
|
||||
{
|
||||
|
||||
@ -91,8 +91,6 @@
|
||||
# define inline
|
||||
#endif
|
||||
|
||||
/* Number of bits in a byte. */
|
||||
#define BYTE_BITS 8
|
||||
/* Number of single byte character. */
|
||||
#define SBC_MAX 256
|
||||
|
||||
@ -123,16 +121,16 @@ extern const char __re_error_msgid[] attribute_hidden;
|
||||
extern const size_t __re_error_msgid_idx[] attribute_hidden;
|
||||
|
||||
/* Number of bits in an unsinged int. */
|
||||
#define UINT_BITS (sizeof (unsigned int) * BYTE_BITS)
|
||||
#define UINT_BITS (sizeof (unsigned int) * CHAR_BIT)
|
||||
/* Number of unsigned int in an bit_set. */
|
||||
#define BITSET_UINTS ((SBC_MAX + UINT_BITS - 1) / UINT_BITS)
|
||||
typedef unsigned int bitset[BITSET_UINTS];
|
||||
typedef unsigned int *re_bitset_ptr_t;
|
||||
typedef const unsigned int *re_const_bitset_ptr_t;
|
||||
|
||||
#define bitset_set(set,i) (set[i / UINT_BITS] |= 1 << i % UINT_BITS)
|
||||
#define bitset_clear(set,i) (set[i / UINT_BITS] &= ~(1 << i % UINT_BITS))
|
||||
#define bitset_contain(set,i) (set[i / UINT_BITS] & (1 << i % UINT_BITS))
|
||||
#define bitset_set(set,i) (set[i / UINT_BITS] |= 1u << i % UINT_BITS)
|
||||
#define bitset_clear(set,i) (set[i / UINT_BITS] &= ~(1u << i % UINT_BITS))
|
||||
#define bitset_contain(set,i) (set[i / UINT_BITS] & (1u << i % UINT_BITS))
|
||||
#define bitset_empty(set) memset (set, 0, sizeof (unsigned int) * BITSET_UINTS)
|
||||
#define bitset_set_all(set) \
|
||||
memset (set, 255, sizeof (unsigned int) * BITSET_UINTS)
|
||||
@ -627,8 +625,8 @@ struct re_fail_stack_t
|
||||
struct re_dfa_t
|
||||
{
|
||||
re_token_t *nodes;
|
||||
int nodes_alloc;
|
||||
int nodes_len;
|
||||
size_t nodes_alloc;
|
||||
size_t nodes_len;
|
||||
int *nexts;
|
||||
int *org_indices;
|
||||
re_node_set *edests;
|
||||
@ -701,10 +699,12 @@ static void re_node_set_remove_at (re_node_set *set, int idx) internal_function;
|
||||
#define re_node_set_empty(p) ((p)->nelem = 0)
|
||||
#define re_node_set_free(set) re_free ((set)->elems)
|
||||
static int re_dfa_add_node (re_dfa_t *dfa, re_token_t token) internal_function;
|
||||
static re_dfastate_t *re_acquire_state (reg_errcode_t *err, re_dfa_t *dfa,
|
||||
const re_node_set *nodes) internal_function;
|
||||
static re_dfastate_t *re_acquire_state (reg_errcode_t *err, const
|
||||
re_dfa_t *dfa,
|
||||
const re_node_set *nodes)
|
||||
internal_function;
|
||||
static re_dfastate_t *re_acquire_state_context (reg_errcode_t *err,
|
||||
re_dfa_t *dfa,
|
||||
const re_dfa_t *dfa,
|
||||
const re_node_set *nodes,
|
||||
unsigned int context) internal_function;
|
||||
static void free_state (re_dfastate_t *state) internal_function;
|
||||
|
||||
@ -25,7 +25,7 @@ static void match_ctx_free (re_match_context_t *cache) internal_function;
|
||||
static reg_errcode_t match_ctx_add_entry (re_match_context_t *cache, int node,
|
||||
int str_idx, int from, int to)
|
||||
internal_function;
|
||||
static int search_cur_bkref_entry (re_match_context_t *mctx, int str_idx)
|
||||
static int search_cur_bkref_entry (const re_match_context_t *mctx, int str_idx)
|
||||
internal_function;
|
||||
static reg_errcode_t match_ctx_add_subtop (re_match_context_t *mctx, int node,
|
||||
int str_idx) internal_function;
|
||||
@ -104,13 +104,14 @@ static reg_errcode_t add_epsilon_src_nodes (re_dfa_t *dfa,
|
||||
static reg_errcode_t sub_epsilon_src_nodes (re_dfa_t *dfa, int node,
|
||||
re_node_set *dest_nodes,
|
||||
const re_node_set *and_nodes) internal_function;
|
||||
static int check_dst_limits (re_match_context_t *mctx, re_node_set *limits,
|
||||
static int check_dst_limits (const re_match_context_t *mctx,
|
||||
re_node_set *limits,
|
||||
int dst_node, int dst_idx, int src_node,
|
||||
int src_idx) internal_function;
|
||||
static int check_dst_limits_calc_pos_1 (re_match_context_t *mctx,
|
||||
static int check_dst_limits_calc_pos_1 (const re_match_context_t *mctx,
|
||||
int boundaries, int subexp_idx,
|
||||
int from_node, int bkref_idx) internal_function;
|
||||
static int check_dst_limits_calc_pos (re_match_context_t *mctx,
|
||||
static int check_dst_limits_calc_pos (const re_match_context_t *mctx,
|
||||
int limit, int subexp_idx,
|
||||
int node, int str_idx,
|
||||
int bkref_idx) internal_function;
|
||||
@ -185,7 +186,7 @@ static unsigned int find_collation_sequence_value (const unsigned char *mbs,
|
||||
size_t name_len) internal_function;
|
||||
# endif /* _LIBC */
|
||||
#endif /* RE_ENABLE_I18N */
|
||||
static int group_nodes_into_DFAstates (re_dfa_t *dfa,
|
||||
static int group_nodes_into_DFAstates (const re_dfa_t *dfa,
|
||||
const re_dfastate_t *state,
|
||||
re_node_set *states_node,
|
||||
bitset *states_ch) internal_function;
|
||||
@ -883,14 +884,14 @@ re_search_internal (preg, string, length, start, range, stop, nmatch, pmatch,
|
||||
#ifdef RE_ENABLE_I18N
|
||||
if (BE (mctx.input.offsets_needed != 0, 0))
|
||||
{
|
||||
if (pmatch[reg_idx].rm_so == mctx.input.valid_len)
|
||||
pmatch[reg_idx].rm_so += mctx.input.valid_raw_len - mctx.input.valid_len;
|
||||
else
|
||||
pmatch[reg_idx].rm_so = mctx.input.offsets[pmatch[reg_idx].rm_so];
|
||||
if (pmatch[reg_idx].rm_eo == mctx.input.valid_len)
|
||||
pmatch[reg_idx].rm_eo += mctx.input.valid_raw_len - mctx.input.valid_len;
|
||||
else
|
||||
pmatch[reg_idx].rm_eo = mctx.input.offsets[pmatch[reg_idx].rm_eo];
|
||||
pmatch[reg_idx].rm_so =
|
||||
(pmatch[reg_idx].rm_so == mctx.input.valid_len
|
||||
? mctx.input.valid_raw_len
|
||||
: mctx.input.offsets[pmatch[reg_idx].rm_so]);
|
||||
pmatch[reg_idx].rm_eo =
|
||||
(pmatch[reg_idx].rm_eo == mctx.input.valid_len
|
||||
? mctx.input.valid_raw_len
|
||||
: mctx.input.offsets[pmatch[reg_idx].rm_eo]);
|
||||
}
|
||||
#else
|
||||
assert (mctx.input.offsets_needed == 0);
|
||||
@ -1887,7 +1888,7 @@ sub_epsilon_src_nodes (dfa, node, dest_nodes, candidates)
|
||||
|
||||
static int
|
||||
check_dst_limits (mctx, limits, dst_node, dst_idx, src_node, src_idx)
|
||||
re_match_context_t *mctx;
|
||||
const re_match_context_t *mctx;
|
||||
re_node_set *limits;
|
||||
int dst_node, dst_idx, src_node, src_idx;
|
||||
{
|
||||
@ -1924,7 +1925,7 @@ check_dst_limits (mctx, limits, dst_node, dst_idx, src_node, src_idx)
|
||||
|
||||
static int
|
||||
check_dst_limits_calc_pos_1 (mctx, boundaries, subexp_idx, from_node, bkref_idx)
|
||||
re_match_context_t *mctx;
|
||||
const re_match_context_t *mctx;
|
||||
int boundaries, subexp_idx, from_node, bkref_idx;
|
||||
{
|
||||
re_dfa_t *const dfa = mctx->dfa;
|
||||
@ -1949,8 +1950,9 @@ check_dst_limits_calc_pos_1 (mctx, boundaries, subexp_idx, from_node, bkref_idx)
|
||||
if (ent->node != node)
|
||||
continue;
|
||||
|
||||
if (subexp_idx <= 8 * sizeof (ent->eps_reachable_subexps_map)
|
||||
&& !(ent->eps_reachable_subexps_map & (1 << subexp_idx)))
|
||||
if (subexp_idx
|
||||
< CHAR_BIT * sizeof ent->eps_reachable_subexps_map
|
||||
&& !(ent->eps_reachable_subexps_map & (1u << subexp_idx)))
|
||||
continue;
|
||||
|
||||
/* Recurse trying to reach the OP_OPEN_SUBEXP and
|
||||
@ -1976,7 +1978,9 @@ check_dst_limits_calc_pos_1 (mctx, boundaries, subexp_idx, from_node, bkref_idx)
|
||||
if (cpos == 0 && (boundaries & 2))
|
||||
return 0;
|
||||
|
||||
ent->eps_reachable_subexps_map &= ~(1 << subexp_idx);
|
||||
if (subexp_idx
|
||||
< CHAR_BIT * sizeof ent->eps_reachable_subexps_map)
|
||||
ent->eps_reachable_subexps_map &= ~(1u << subexp_idx);
|
||||
}
|
||||
while (ent++->more);
|
||||
}
|
||||
@ -2002,7 +2006,7 @@ check_dst_limits_calc_pos_1 (mctx, boundaries, subexp_idx, from_node, bkref_idx)
|
||||
|
||||
static int
|
||||
check_dst_limits_calc_pos (mctx, limit, subexp_idx, from_node, str_idx, bkref_idx)
|
||||
re_match_context_t *mctx;
|
||||
const re_match_context_t *mctx;
|
||||
int limit, subexp_idx, from_node, str_idx, bkref_idx;
|
||||
{
|
||||
struct re_backref_cache_entry *lim = mctx->bkref_ents + limit;
|
||||
@ -2443,8 +2447,8 @@ check_subexp_matching_top (mctx, cur_nodes, str_idx)
|
||||
{
|
||||
int node = cur_nodes->elems[node_idx];
|
||||
if (dfa->nodes[node].type == OP_OPEN_SUBEXP
|
||||
&& dfa->nodes[node].opr.idx < (8 * sizeof (dfa->used_bkref_map))
|
||||
&& dfa->used_bkref_map & (1 << dfa->nodes[node].opr.idx))
|
||||
&& dfa->nodes[node].opr.idx < CHAR_BIT * sizeof dfa->used_bkref_map
|
||||
&& dfa->used_bkref_map & (1u << dfa->nodes[node].opr.idx))
|
||||
{
|
||||
err = match_ctx_add_subtop (mctx, node, str_idx);
|
||||
if (BE (err != REG_NOERROR, 0))
|
||||
@ -2557,7 +2561,8 @@ transit_state_mb (mctx, pstate)
|
||||
if (BE (err != REG_NOERROR, 0))
|
||||
return err;
|
||||
}
|
||||
context = re_string_context_at (&mctx->input, dest_idx - 1, mctx->eflags);
|
||||
context = re_string_context_at (&mctx->input, dest_idx - 1,
|
||||
mctx->eflags);
|
||||
mctx->state_log[dest_idx]
|
||||
= re_acquire_state_context (&err, dfa, &dest_nodes, context);
|
||||
if (dest_state != NULL)
|
||||
@ -2696,7 +2701,8 @@ get_subexp (mctx, bkref_node, bkref_str_idx)
|
||||
int cache_idx = search_cur_bkref_entry (mctx, bkref_str_idx);
|
||||
if (cache_idx != -1)
|
||||
{
|
||||
const struct re_backref_cache_entry *entry = mctx->bkref_ents + cache_idx;
|
||||
const struct re_backref_cache_entry *entry
|
||||
= mctx->bkref_ents + cache_idx;
|
||||
do
|
||||
if (entry->node == bkref_node)
|
||||
return REG_NOERROR; /* We already checked it. */
|
||||
@ -2743,7 +2749,8 @@ get_subexp (mctx, bkref_node, bkref_str_idx)
|
||||
buf = (const char *) re_string_get_buffer (&mctx->input);
|
||||
}
|
||||
if (memcmp (buf + bkref_str_off, buf + sl_str, sl_str_diff) != 0)
|
||||
break; /* We don't need to search this sub expression any more. */
|
||||
/* We don't need to search this sub expression any more. */
|
||||
break;
|
||||
}
|
||||
bkref_str_off += sl_str_diff;
|
||||
sl_str += sl_str_diff;
|
||||
@ -2794,7 +2801,8 @@ get_subexp (mctx, bkref_node, bkref_str_idx)
|
||||
continue;
|
||||
/* Does this state have a ')' of the sub expression? */
|
||||
nodes = &mctx->state_log[sl_str]->nodes;
|
||||
cls_node = find_subexp_node (dfa, nodes, subexp_num, OP_CLOSE_SUBEXP);
|
||||
cls_node = find_subexp_node (dfa, nodes, subexp_num,
|
||||
OP_CLOSE_SUBEXP);
|
||||
if (cls_node == -1)
|
||||
continue; /* No. */
|
||||
if (sub_top->path == NULL)
|
||||
@ -2807,7 +2815,8 @@ get_subexp (mctx, bkref_node, bkref_str_idx)
|
||||
/* Can the OP_OPEN_SUBEXP node arrive the OP_CLOSE_SUBEXP node
|
||||
in the current context? */
|
||||
err = check_arrival (mctx, sub_top->path, sub_top->node,
|
||||
sub_top->str_idx, cls_node, sl_str, OP_CLOSE_SUBEXP);
|
||||
sub_top->str_idx, cls_node, sl_str,
|
||||
OP_CLOSE_SUBEXP);
|
||||
if (err == REG_NOMATCH)
|
||||
continue;
|
||||
if (BE (err != REG_NOERROR, 0))
|
||||
@ -2841,7 +2850,8 @@ get_subexp_sub (mctx, sub_top, sub_last, bkref_node, bkref_str)
|
||||
int to_idx;
|
||||
/* Can the subexpression arrive the back reference? */
|
||||
err = check_arrival (mctx, &sub_last->path, sub_last->node,
|
||||
sub_last->str_idx, bkref_node, bkref_str, OP_OPEN_SUBEXP);
|
||||
sub_last->str_idx, bkref_node, bkref_str,
|
||||
OP_OPEN_SUBEXP);
|
||||
if (err != REG_NOERROR)
|
||||
return err;
|
||||
err = match_ctx_add_entry (mctx, bkref_node, bkref_str, sub_top->str_idx,
|
||||
@ -3539,10 +3549,10 @@ out_free:
|
||||
|
||||
static int
|
||||
group_nodes_into_DFAstates (dfa, state, dests_node, dests_ch)
|
||||
re_dfa_t *dfa;
|
||||
const re_dfastate_t *state;
|
||||
re_node_set *dests_node;
|
||||
bitset *dests_ch;
|
||||
const re_dfa_t *dfa;
|
||||
const re_dfastate_t *state;
|
||||
re_node_set *dests_node;
|
||||
bitset *dests_ch;
|
||||
{
|
||||
reg_errcode_t err;
|
||||
int result;
|
||||
@ -4265,7 +4275,7 @@ match_ctx_add_entry (mctx, node, str_idx, from, to)
|
||||
|
||||
static int
|
||||
search_cur_bkref_entry (mctx, str_idx)
|
||||
re_match_context_t *mctx;
|
||||
const re_match_context_t *mctx;
|
||||
int str_idx;
|
||||
{
|
||||
int left, right, mid, last;
|
||||
|
||||
Loading…
Reference in New Issue
Block a user