/* Extended regular expression matching and search library.
- Copyright (C) 2002, 2003, 2004, 2005, 2006 Free Software Foundation, Inc.
+ Copyright (C) 2002-2014 Free Software Foundation, Inc.
This file is part of the GNU C Library.
Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
- This program is free software; you can redistribute it and/or modify
- it under the terms of the GNU General Public License as published by
- the Free Software Foundation; either version 2, or (at your option)
- any later version.
+ The GNU C Library is free software; you can redistribute it and/or
+ modify it under the terms of the GNU Lesser General Public
+ License as published by the Free Software Foundation; either
+ version 2.1 of the License, or (at your option) any later version.
- This program is distributed in the hope that it will be useful,
+ The GNU C Library is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
- MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- GNU General Public License for more details.
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ Lesser General Public License for more details.
- You should have received a copy of the GNU General Public License along
- with this program; if not, write to the Free Software Foundation,
- Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */
+ 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/>. */
static void re_string_construct_common (const char *str, Idx len,
re_string_t *pstr,
re_string_reconstruct before using the object. */
static reg_errcode_t
-internal_function
+internal_function __attribute_warn_unused_result__
re_string_allocate (re_string_t *pstr, const char *str, Idx len, Idx init_len,
RE_TRANSLATE_TYPE trans, bool icase, const re_dfa_t *dfa)
{
/* This function allocate the buffers, and initialize them. */
static reg_errcode_t
-internal_function
+internal_function __attribute_warn_unused_result__
re_string_construct (re_string_t *pstr, const char *str, Idx len,
RE_TRANSLATE_TYPE trans, bool icase, const re_dfa_t *dfa)
{
/* Helper functions for re_string_allocate, and re_string_construct. */
static reg_errcode_t
-internal_function
+internal_function __attribute_warn_unused_result__
re_string_realloc_buffers (re_string_t *pstr, Idx new_buf_len)
{
#ifdef RE_ENABLE_I18N
{
wint_t *new_wcs;
- /* Avoid overflow. */
- size_t max_object_size = MAX (sizeof (wint_t), sizeof (Idx));
- if (BE (SIZE_MAX / max_object_size < new_buf_len, 0))
+ /* Avoid overflow in realloc. */
+ const size_t max_object_size = MAX (sizeof (wint_t), sizeof (Idx));
+ if (BE (MIN (IDX_MAX, SIZE_MAX / max_object_size) < new_buf_len, 0))
return REG_ESPACE;
new_wcs = re_realloc (pstr->wcs, wint_t, new_buf_len);
}
else
p = (const char *) pstr->raw_mbs + pstr->raw_mbs_idx + byte_idx;
- mbclen = mbrtowc (&wc, p, remain_len, &pstr->cur_state);
- if (BE (mbclen == (size_t) -2, 0))
- {
- /* The buffer doesn't have enough space, finish to build. */
- pstr->cur_state = prev_st;
- break;
- }
- else if (BE (mbclen == (size_t) -1 || mbclen == 0, 0))
+ mbclen = __mbrtowc (&wc, p, remain_len, &pstr->cur_state);
+ if (BE (mbclen == (size_t) -1 || mbclen == 0
+ || (mbclen == (size_t) -2 && pstr->bufs_len >= pstr->len), 0))
{
/* We treat these cases as a singlebyte character. */
mbclen = 1;
wc = pstr->trans[wc];
pstr->cur_state = prev_st;
}
+ else if (BE (mbclen == (size_t) -2, 0))
+ {
+ /* The buffer doesn't have enough space, finish to build. */
+ pstr->cur_state = prev_st;
+ break;
+ }
/* Write wide character and padding. */
pstr->wcs[byte_idx++] = wc;
but for REG_ICASE. */
static reg_errcode_t
-internal_function
+internal_function __attribute_warn_unused_result__
build_wcs_upper_buffer (re_string_t *pstr)
{
mbstate_t prev_st;
remain_len = end_idx - byte_idx;
prev_st = pstr->cur_state;
- mbclen = mbrtowc (&wc,
- ((const char *) pstr->raw_mbs + pstr->raw_mbs_idx
- + byte_idx), remain_len, &pstr->cur_state);
+ mbclen = __mbrtowc (&wc,
+ ((const char *) pstr->raw_mbs + pstr->raw_mbs_idx
+ + byte_idx), remain_len, &pstr->cur_state);
if (BE (mbclen < (size_t) -2, 1))
{
wchar_t wcu = wc;
for (remain_len = byte_idx + mbclen - 1; byte_idx < remain_len ;)
pstr->wcs[byte_idx++] = WEOF;
}
- else if (mbclen == (size_t) -1 || mbclen == 0)
+ else if (mbclen == (size_t) -1 || mbclen == 0
+ || (mbclen == (size_t) -2 && pstr->bufs_len >= pstr->len))
{
- /* It is an invalid character or '\0'. Just use the byte. */
+ /* It is an invalid character, an incomplete character
+ at the end of the string, or '\0'. Just use the byte. */
int ch = pstr->raw_mbs[pstr->raw_mbs_idx + byte_idx];
pstr->mbs[byte_idx] = ch;
/* And also cast it to wide char. */
}
else
p = (const char *) pstr->raw_mbs + pstr->raw_mbs_idx + src_idx;
- mbclen = mbrtowc (&wc, p, remain_len, &pstr->cur_state);
+ mbclen = __mbrtowc (&wc, p, remain_len, &pstr->cur_state);
if (BE (mbclen < (size_t) -2, 1))
{
wchar_t wcu = wc;
src_idx += mbclen;
continue;
}
- else
- memcpy (pstr->mbs + byte_idx, p, mbclen);
+ else
+ memcpy (pstr->mbs + byte_idx, p, mbclen);
}
else
memcpy (pstr->mbs + byte_idx, p, mbclen);
for (remain_len = byte_idx + mbclen - 1; byte_idx < remain_len ;)
pstr->wcs[byte_idx++] = WEOF;
}
- else if (mbclen == (size_t) -1 || mbclen == 0)
+ else if (mbclen == (size_t) -1 || mbclen == 0
+ || (mbclen == (size_t) -2 && pstr->bufs_len >= pstr->len))
{
/* It is an invalid character or '\0'. Just use the byte. */
int ch = pstr->raw_mbs[pstr->raw_mbs_idx + src_idx];
rawbuf_idx < new_raw_idx;)
{
wchar_t wc2;
- Idx remain_len;
- remain_len = pstr->len - rawbuf_idx;
+ Idx remain_len = pstr->raw_len - rawbuf_idx;
prev_st = pstr->cur_state;
- mbclen = mbrtowc (&wc2, (const char *) pstr->raw_mbs + rawbuf_idx,
- remain_len, &pstr->cur_state);
+ mbclen = __mbrtowc (&wc2, (const char *) pstr->raw_mbs + rawbuf_idx,
+ remain_len, &pstr->cur_state);
if (BE (mbclen == (size_t) -2 || mbclen == (size_t) -1 || mbclen == 0, 0))
{
/* We treat these cases as a single byte character. */
convert to upper case in case of REG_ICASE, apply translation. */
static reg_errcode_t
-internal_function
+internal_function __attribute_warn_unused_result__
re_string_reconstruct (re_string_t *pstr, Idx idx, int eflags)
{
Idx offset;
if (BE (offset != 0, 1))
{
- /* Are the characters which are already checked remain? */
- if (BE (offset < pstr->valid_raw_len, 1)
-#ifdef RE_ENABLE_I18N
- /* Handling this would enlarge the code too much.
- Accept a slowdown in that case. */
- && pstr->offsets_needed == 0
-#endif
- )
+ /* Should the already checked characters be kept? */
+ if (BE (offset < pstr->valid_raw_len, 1))
{
/* Yes, move them to the front of the buffer. */
- pstr->tip_context = re_string_context_at (pstr, offset - 1, eflags);
#ifdef RE_ENABLE_I18N
- if (pstr->mb_cur_max > 1)
- memmove (pstr->wcs, pstr->wcs + offset,
- (pstr->valid_len - offset) * sizeof (wint_t));
+ if (BE (pstr->offsets_needed, 0))
+ {
+ Idx low = 0, high = pstr->valid_len, mid;
+ do
+ {
+ mid = (high + low) / 2;
+ if (pstr->offsets[mid] > offset)
+ high = mid;
+ else if (pstr->offsets[mid] < offset)
+ low = mid + 1;
+ else
+ break;
+ }
+ while (low < high);
+ if (pstr->offsets[mid] < offset)
+ ++mid;
+ pstr->tip_context = re_string_context_at (pstr, mid - 1,
+ eflags);
+ /* This can be quite complicated, so handle specially
+ only the common and easy case where the character with
+ different length representation of lower and upper
+ case is present at or after offset. */
+ if (pstr->valid_len > offset
+ && mid == offset && pstr->offsets[mid] == offset)
+ {
+ memmove (pstr->wcs, pstr->wcs + offset,
+ (pstr->valid_len - offset) * sizeof (wint_t));
+ memmove (pstr->mbs, pstr->mbs + offset, pstr->valid_len - offset);
+ pstr->valid_len -= offset;
+ pstr->valid_raw_len -= offset;
+ for (low = 0; low < pstr->valid_len; low++)
+ pstr->offsets[low] = pstr->offsets[low + offset] - offset;
+ }
+ else
+ {
+ /* Otherwise, just find out how long the partial multibyte
+ character at offset is and fill it with WEOF/255. */
+ pstr->len = pstr->raw_len - idx + offset;
+ pstr->stop = pstr->raw_stop - idx + offset;
+ pstr->offsets_needed = 0;
+ while (mid > 0 && pstr->offsets[mid - 1] == offset)
+ --mid;
+ while (mid < pstr->valid_len)
+ if (pstr->wcs[mid] != WEOF)
+ break;
+ else
+ ++mid;
+ if (mid == pstr->valid_len)
+ pstr->valid_len = 0;
+ else
+ {
+ pstr->valid_len = pstr->offsets[mid] - offset;
+ if (pstr->valid_len)
+ {
+ for (low = 0; low < pstr->valid_len; ++low)
+ pstr->wcs[low] = WEOF;
+ memset (pstr->mbs, 255, pstr->valid_len);
+ }
+ }
+ pstr->valid_raw_len = pstr->valid_len;
+ }
+ }
+ else
+#endif
+ {
+ pstr->tip_context = re_string_context_at (pstr, offset - 1,
+ eflags);
+#ifdef RE_ENABLE_I18N
+ if (pstr->mb_cur_max > 1)
+ memmove (pstr->wcs, pstr->wcs + offset,
+ (pstr->valid_len - offset) * sizeof (wint_t));
#endif /* RE_ENABLE_I18N */
- if (BE (pstr->mbs_allocated, 0))
- memmove (pstr->mbs, pstr->mbs + offset,
- pstr->valid_len - offset);
- pstr->valid_len -= offset;
- pstr->valid_raw_len -= offset;
+ if (BE (pstr->mbs_allocated, 0))
+ memmove (pstr->mbs, pstr->mbs + offset,
+ pstr->valid_len - offset);
+ pstr->valid_len -= offset;
+ pstr->valid_raw_len -= offset;
#if DEBUG
- assert (pstr->valid_len > 0);
+ assert (pstr->valid_len > 0);
#endif
+ }
}
else
{
- /* No, skip all characters until IDX. */
#ifdef RE_ENABLE_I18N
+ /* No, skip all characters until IDX. */
+ Idx prev_valid_len = pstr->valid_len;
+
if (BE (pstr->offsets_needed, 0))
{
pstr->len = pstr->raw_len - idx + offset;
if (pstr->is_utf8)
{
- const unsigned char *raw, *p, *q, *end;
+ const unsigned char *raw, *p, *end;
/* Special case UTF-8. Multi-byte chars start with any
byte other than 0x80 - 0xbf. */
raw = pstr->raw_mbs + pstr->raw_mbs_idx;
end = raw + (offset - pstr->mb_cur_max);
+ if (end < pstr->raw_mbs)
+ end = pstr->raw_mbs;
p = raw + offset - 1;
#ifdef _LIBC
/* We know the wchar_t encoding is UCS4, so for the simple
if (isascii (*p) && BE (pstr->trans == NULL, 1))
{
memset (&pstr->cur_state, '\0', sizeof (mbstate_t));
- pstr->valid_len = 0;
+ /* pstr->valid_len = 0; */
wc = (wchar_t) *p;
}
else
unsigned char buf[6];
size_t mbclen;
- q = p;
+ const unsigned char *pp = p;
if (BE (pstr->trans != NULL, 0))
{
int i = mlen < 6 ? mlen : 6;
while (--i >= 0)
buf[i] = pstr->trans[p[i]];
- q = buf;
+ pp = 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);
+ mbclen = __mbrtowc (&wc2, (const char *) pp, mlen,
+ &cur_state);
if (raw + offset - p <= mbclen
&& mbclen < (size_t) -2)
{
pstr->valid_len = re_string_skip_chars (pstr, idx, &wc) - idx;
if (wc == WEOF)
pstr->tip_context
- = re_string_context_at (pstr, pstr->valid_raw_len - 1, eflags);
+ = re_string_context_at (pstr, prev_valid_len - 1, eflags);
else
pstr->tip_context = ((BE (pstr->word_ops_used != 0, 0)
&& IS_WIDE_WORD_CHAR (wc))
for (wcs_idx = 0; wcs_idx < pstr->valid_len; ++wcs_idx)
pstr->wcs[wcs_idx] = WEOF;
if (pstr->mbs_allocated)
- memset (pstr->mbs, -1, pstr->valid_len);
+ memset (pstr->mbs, 255, pstr->valid_len);
}
pstr->valid_raw_len = pstr->valid_len;
}
}
static unsigned char
-internal_function __attribute ((pure))
+internal_function __attribute__ ((pure))
re_string_peek_byte_case (const re_string_t *pstr, Idx idx)
{
int ch;
}
static unsigned char
-internal_function __attribute ((pure))
+internal_function
re_string_fetch_byte_case (re_string_t *pstr)
{
if (BE (!pstr->mbs_allocated, 1))
/* Functions for set operation. */
static reg_errcode_t
-internal_function
+internal_function __attribute_warn_unused_result__
re_node_set_alloc (re_node_set *set, Idx size)
{
set->alloc = size;
set->nelem = 0;
set->elems = re_malloc (Idx, size);
- if (BE (set->elems == NULL, 0))
+ if (BE (set->elems == NULL, 0) && (MALLOC_0_IS_NONNULL || size != 0))
return REG_ESPACE;
return REG_NOERROR;
}
static reg_errcode_t
-internal_function
+internal_function __attribute_warn_unused_result__
re_node_set_init_1 (re_node_set *set, Idx elem)
{
set->alloc = 1;
}
static reg_errcode_t
-internal_function
+internal_function __attribute_warn_unused_result__
re_node_set_init_2 (re_node_set *set, Idx elem1, Idx elem2)
{
set->alloc = 2;
}
static reg_errcode_t
-internal_function
+internal_function __attribute_warn_unused_result__
re_node_set_init_copy (re_node_set *dest, const re_node_set *src)
{
dest->nelem = src->nelem;
Note: We assume dest->elems is NULL, when dest->alloc is 0. */
static reg_errcode_t
-internal_function
+internal_function __attribute_warn_unused_result__
re_node_set_add_intersect (re_node_set *dest, const re_node_set *src1,
const re_node_set *src2)
{
Idx new_alloc = src1->nelem + src2->nelem + dest->alloc;
Idx *new_elems = re_realloc (dest->elems, Idx, new_alloc);
if (BE (new_elems == NULL, 0))
- return REG_ESPACE;
+ return REG_ESPACE;
dest->elems = new_elems;
dest->alloc = new_alloc;
}
if (delta > 0 && REG_VALID_INDEX (id))
for (;;)
{
- if (dest->elems[is] > dest->elems[id])
- {
- /* Copy from the top. */
- dest->elems[id + delta--] = dest->elems[is--];
- if (delta == 0)
- break;
- }
- else
- {
- /* Slide from the bottom. */
- dest->elems[id + delta] = dest->elems[id];
- if (! REG_VALID_INDEX (--id))
- break;
- }
+ if (dest->elems[is] > dest->elems[id])
+ {
+ /* Copy from the top. */
+ dest->elems[id + delta--] = dest->elems[is--];
+ if (delta == 0)
+ break;
+ }
+ else
+ {
+ /* Slide from the bottom. */
+ dest->elems[id + delta] = dest->elems[id];
+ if (! REG_VALID_INDEX (--id))
+ break;
+ }
}
/* Copy remaining SRC elements. */
DEST. Return value indicate the error code or REG_NOERROR if succeeded. */
static reg_errcode_t
-internal_function
+internal_function __attribute_warn_unused_result__
re_node_set_init_union (re_node_set *dest, const re_node_set *src1,
const re_node_set *src2)
{
DEST. Return value indicate the error code or REG_NOERROR if succeeded. */
static reg_errcode_t
-internal_function
+internal_function __attribute_warn_unused_result__
re_node_set_merge (re_node_set *dest, const re_node_set *src)
{
Idx is, id, sbase, delta;
REG_VALID_INDEX (is) && REG_VALID_INDEX (id); )
{
if (dest->elems[id] == src->elems[is])
- is--, id--;
+ is--, id--;
else if (dest->elems[id] < src->elems[is])
- dest->elems[--sbase] = src->elems[is--];
+ dest->elems[--sbase] = src->elems[is--];
else /* if (dest->elems[id] > src->elems[is]) */
- --id;
+ --id;
}
if (REG_VALID_INDEX (is))
for (;;)
{
if (dest->elems[is] > dest->elems[id])
- {
+ {
/* Copy from the top. */
- dest->elems[id + delta--] = dest->elems[is--];
+ dest->elems[id + delta--] = dest->elems[is--];
if (delta == 0)
break;
}
else
- {
- /* Slide from the bottom. */
- dest->elems[id + delta] = dest->elems[id];
+ {
+ /* Slide from the bottom. */
+ dest->elems[id + delta] = dest->elems[id];
if (! REG_VALID_INDEX (--id))
{
/* Copy remaining SRC elements. */
memcpy (dest->elems, dest->elems + sbase,
- delta * sizeof (Idx));
+ delta * sizeof (Idx));
break;
}
}
Return true if successful. */
static bool
-internal_function
+internal_function __attribute_warn_unused_result__
re_node_set_insert (re_node_set *set, Idx elem)
{
Idx idx;
{
idx = 0;
for (idx = set->nelem; idx > 0; idx--)
- set->elems[idx] = set->elems[idx - 1];
+ set->elems[idx] = set->elems[idx - 1];
}
else
{
for (idx = set->nelem; set->elems[idx - 1] > elem; idx--)
- set->elems[idx] = set->elems[idx - 1];
+ set->elems[idx] = set->elems[idx - 1];
}
/* Insert the new element. */
Return true if successful. */
static bool
-internal_function
+internal_function __attribute_warn_unused_result__
re_node_set_insert_last (re_node_set *set, Idx elem)
{
/* Realloc if we need. */
Return true if SET1 and SET2 are equivalent. */
static bool
-internal_function __attribute ((pure))
+internal_function __attribute__ ((pure))
re_node_set_compare (const re_node_set *set1, const re_node_set *set2)
{
Idx i;
/* Return (idx + 1) if SET contains the element ELEM, return 0 otherwise. */
static Idx
-internal_function __attribute ((pure))
+internal_function __attribute__ ((pure))
re_node_set_contains (const re_node_set *set, Idx elem)
{
__re_size_t idx, right, mid;
Idx *new_nexts, *new_indices;
re_node_set *new_edests, *new_eclosures;
re_token_t *new_nodes;
- size_t max_object_size =
- MAX (sizeof (re_token_t),
- MAX (sizeof (re_node_set),
- sizeof (Idx)));
- /* Avoid overflows. */
- if (BE (SIZE_MAX / 2 / max_object_size < dfa->nodes_alloc, 0))
+ /* Avoid overflows in realloc. */
+ const size_t max_object_size = MAX (sizeof (re_token_t),
+ MAX (sizeof (re_node_set),
+ sizeof (Idx)));
+ if (BE (MIN (IDX_MAX, SIZE_MAX / max_object_size) < new_nodes_alloc, 0))
return REG_MISSING;
new_nodes = re_realloc (dfa->nodes, re_token_t, new_nodes_alloc);
dfa->nodes[dfa->nodes_len] = token;
dfa->nodes[dfa->nodes_len].constraint = 0;
#ifdef RE_ENABLE_I18N
- {
- int type = token.type;
dfa->nodes[dfa->nodes_len].accept_mb =
- (type == OP_PERIOD && dfa->mb_cur_max > 1) || type == COMPLEX_BRACKET;
- }
+ ((token.type == OP_PERIOD && dfa->mb_cur_max > 1)
+ || token.type == COMPLEX_BRACKET);
#endif
dfa->nexts[dfa->nodes_len] = REG_MISSING;
re_node_set_init_empty (dfa->edests + dfa->nodes_len);
return dfa->nodes_len++;
}
-static inline re_hashval_t
+static re_hashval_t
internal_function
calc_state_hash (const re_node_set *nodes, unsigned int context)
{
optimization. */
static re_dfastate_t *
-internal_function
+internal_function __attribute_warn_unused_result__
re_acquire_state (reg_errcode_t *err, const re_dfa_t *dfa,
const re_node_set *nodes)
{
optimization. */
static re_dfastate_t *
-internal_function
+internal_function __attribute_warn_unused_result__
re_acquire_state_context (reg_errcode_t *err, const re_dfa_t *dfa,
const re_node_set *nodes, unsigned int context)
{
&& re_node_set_compare (state->entrance_nodes, nodes))
return state;
}
- /* There are no appropriate state in `dfa', create the new one. */
+ /* There are no appropriate state in 'dfa', create the new one. */
new_state = create_cd_newstate (dfa, nodes, context, hash);
if (BE (new_state == NULL, 0))
*err = REG_ESPACE;
indicates the error code if failed. */
static reg_errcode_t
+__attribute_warn_unused_result__
register_state (const re_dfa_t *dfa, re_dfastate_t *newstate,
re_hashval_t hash)
{
{
Idx elem = newstate->nodes.elems[i];
if (!IS_EPSILON_NODE (dfa->nodes[elem].type))
- if (BE (! re_node_set_insert_last (&newstate->non_eps_nodes, elem), 0))
+ if (! re_node_set_insert_last (&newstate->non_eps_nodes, elem))
return REG_ESPACE;
}
re_free (state);
}
-/* Create the new state which is independ of contexts.
+/* Create the new state which is independent of contexts.
Return the new state if succeeded, otherwise return NULL. */
static re_dfastate_t *
-internal_function
+internal_function __attribute_warn_unused_result__
create_ci_newstate (const re_dfa_t *dfa, const re_node_set *nodes,
re_hashval_t hash)
{
Return the new state if succeeded, otherwise return NULL. */
static re_dfastate_t *
-internal_function
+internal_function __attribute_warn_unused_result__
create_cd_newstate (const re_dfa_t *dfa, const re_node_set *nodes,
unsigned int context, re_hashval_t hash)
{
for (i = 0 ; i < nodes->nelem ; i++)
{
- unsigned int constraint = 0;
re_token_t *node = dfa->nodes + nodes->elems[i];
re_token_type_t type = node->type;
- if (node->constraint)
- constraint = node->constraint;
+ unsigned int constraint = node->constraint;
if (type == CHARACTER && !constraint)
continue;
newstate->halt = 1;
else if (type == OP_BACK_REF)
newstate->has_backref = 1;
- else if (type == ANCHOR)
- constraint = node->opr.ctx_type;
if (constraint)
{
free_state (newstate);
return NULL;
}
- re_node_set_init_copy (newstate->entrance_nodes, nodes);
+ if (re_node_set_init_copy (newstate->entrance_nodes, nodes)
+ != REG_NOERROR)
+ return NULL;
nctx_nodes = 0;
newstate->has_constraint = 1;
}
projects / gnulib.git / blobdiff