2004-01-12  Paolo Bonzini  <bonzini@gnu.org>

	* posix/regcomp.c [_LIBC && !RE_ENABLE_I18N]:
	Drop code to support this, it is never true.
	(build_range_exp) [!_LIBC]: Do not create a range
	in MBCSET for a single-byte character set.
	(build_range_exp) [_LIBC]: Do not create a range
	in MBCSET for a single-byte character set without
	collation elements.
	(init_dfa): Do not conditionalize on _LIBC, it
	just makes the code less clear.
	(parse_bracket_exp): Use NON_MATCH variable in
	addition to "mbcset->non_match", not as an
	alternative.
	(build_charclass_op): rename NOT parameter to
	NON_MATCH, use it instead of declaring a variable.
	(parse_bracket_exp) [!_LIBC]: Pass NULL for MBCSET
	if the character set is single-byte.
This commit is contained in:
Ulrich Drepper 2004-01-14 02:42:20 +00:00
parent 294b6bcc13
commit 10677727e2
7 changed files with 123 additions and 109 deletions

View file

@ -1,3 +1,22 @@
2004-01-12 Paolo Bonzini <bonzini@gnu.org>
* posix/regcomp.c [_LIBC && !RE_ENABLE_I18N]:
Drop code to support this, it is never true.
(build_range_exp) [!_LIBC]: Do not create a range
in MBCSET for a single-byte character set.
(build_range_exp) [_LIBC]: Do not create a range
in MBCSET for a single-byte character set without
collation elements.
(init_dfa): Do not conditionalize on _LIBC, it
just makes the code less clear.
(parse_bracket_exp): Use NON_MATCH variable in
addition to "mbcset->non_match", not as an
alternative.
(build_charclass_op): rename NOT parameter to
NON_MATCH, use it instead of declaring a variable.
(parse_bracket_exp) [!_LIBC]: Pass NULL for MBCSET
if the character set is single-byte.
2004-01-14 Jakub Jelinek <jakub@redhat.com>
* posix/regcomp.c (peek_token_bracket): Check remaining

View file

@ -1,3 +1,7 @@
2003-12-31 Bruno Haible <bruno@clisp.org>
* charmaps/IBM866NAV: Fix the character values.
2004-01-01 Petter Reinholdtsen <pere@hungry.com>
* sort-test.sh: Print the status for both the sorting tests, xfrm

View file

@ -254,16 +254,16 @@ CHARMAP
<U044F> /xef CYRILLIC SMALL LETTER YA
<U0401> /xf0 CYRILLIC CAPITAL LETTER IO
<U0451> /xf1 CYRILLIC SMALL LETTER IO
<U0404> /xf2 CYRILLIC CAPITAL LETTER GHE WITH UPTURN
<U0454> /xf3 CYRILLIC SMALL LETTER GHE WITH UPTURN
<U0407> /xf4 CYRILLIC CAPITAL LETTER UKRAINIAN IE
<U0457> /xf5 CYRILLIC SMALL LETTER UKRAINIAN IE
<U040E> /xf6 CYRILLIC CAPITAL LETTER BYELORUSSIAN-UKRAINIAN I
<U045E> /xf7 CYRILLIC SMALL LETTER BYELORUSSIAN-UKRAINIAN I
<U00B0> /xf8 CYRILLIC CAPITAL LETTER YI (Ukrainian)
<U2219> /xf9 CYRILLIC SMALL LETTER YI (Ukrainian)
<U00B7> /xfa CYRILLIC CAPITAL LETTER SHORT U
<U221A> /xfb CYRILLIC SMALL LETTER SHORT U
<U0490> /xf2 CYRILLIC CAPITAL LETTER GHE WITH UPTURN
<U0491> /xf3 CYRILLIC SMALL LETTER GHE WITH UPTURN
<U0404> /xf4 CYRILLIC CAPITAL LETTER UKRAINIAN IE
<U0454> /xf5 CYRILLIC SMALL LETTER UKRAINIAN IE
<U0406> /xf6 CYRILLIC CAPITAL LETTER BYELORUSSIAN-UKRAINIAN I
<U0456> /xf7 CYRILLIC SMALL LETTER BYELORUSSIAN-UKRAINIAN I
<U0407> /xf8 CYRILLIC CAPITAL LETTER YI
<U0457> /xf9 CYRILLIC SMALL LETTER YI
<U040E> /xfa CYRILLIC CAPITAL LETTER SHORT U
<U045E> /xfb CYRILLIC SMALL LETTER SHORT U
<U2116> /xfc NUMERO SIGN
<U00A4> /xfd CURRENCY SIGN
<U25A0> /xfe BLACK SQUARE

View file

@ -6,6 +6,11 @@
2004-01-13 Richard Henderson <rth@redhat.com>
* pthread_attr_setstack.c (__old_pthread_attr_setstack): Use
prototype form.
* pthread_attr_setstacksize.c (__old_pthread_attr_setstacksize):
Likewise.
* sysdeps/alpha/Makefile: New file.
* sysdeps/alpha/tcb-offsets.sym: New file.
* sysdeps/unix/sysv/linux/alpha/sysdep-cancel.h (SINGLE_THREAD_P):

View file

@ -59,10 +59,8 @@ versioned_symbol (libpthread, __pthread_attr_setstack, pthread_attr_setstack,
# if SHLIB_COMPAT(libpthread, GLIBC_2_2, GLIBC_2_3_3)
int
__old_pthread_attr_setstack (attr, stackaddr, stacksize)
pthread_attr_t *attr;
void *stackaddr;
size_t stacksize;
__old_pthread_attr_setstack (pthread_attr_t *attr, void *stackaddr,
size_t stacksize)
{
struct pthread_attr *iattr;

View file

@ -52,9 +52,7 @@ versioned_symbol (libpthread, __pthread_attr_setstacksize,
# if SHLIB_COMPAT(libpthread, GLIBC_2_1, GLIBC_2_3_3)
int
__old_pthread_attr_setstacksize (attr, stacksize)
pthread_attr_t *attr;
size_t stacksize;
__old_pthread_attr_setstacksize (pthread_attr_t *attr, size_t stacksize)
{
struct pthread_attr *iattr;

View file

@ -126,8 +126,8 @@ static reg_errcode_t build_charclass (unsigned RE_TRANSLATE_TYPE trans,
static bin_tree_t *build_charclass_op (re_dfa_t *dfa,
unsigned RE_TRANSLATE_TYPE trans,
const unsigned char *class_name,
const unsigned char *extra, int not,
reg_errcode_t *err);
const unsigned char *extra,
int 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, int index);
@ -862,11 +862,9 @@ init_dfa (dfa, pat_len)
dfa->sb_char = (re_bitset_ptr_t) calloc (sizeof (bitset), 1);
if (BE (dfa->sb_char == NULL, 0))
return REG_ESPACE;
#ifdef _LIBC
if (dfa->is_utf8)
memset (dfa->sb_char, 255, sizeof (unsigned int) * BITSET_UINTS / 2);
else
#endif
for (i = 0, ch = 0; i < BITSET_UINTS; ++i)
for (j = 0; j < UINT_BITS; ++j, ++ch)
if (btowc (ch) != WEOF)
@ -2567,33 +2565,41 @@ build_range_exp (sbcset, start_elem, end_elem)
if (wcscoll (cmp_buf, cmp_buf + 4) > 0)
return REG_ERANGE;
/* Check the space of the arrays. */
if (BE (*range_alloc == mbcset->nranges, 0))
/* Got valid collation sequence values, add them as a new entry.
However, for !_LIBC we have no collation elements: if the
character set is single byte, the single byte character set
that we build below suffices. parse_bracket_exp passes
no MBCSET if dfa->mb_cur_max == 1. */
if (mbcset)
{
/* There are not enough space, need realloc. */
wchar_t *new_array_start, *new_array_end;
int new_nranges;
/* Check the space of the arrays. */
if (BE (*range_alloc == mbcset->nranges, 0))
{
/* There is not enough space, need realloc. */
wchar_t *new_array_start, *new_array_end;
int new_nranges;
/* +1 in case of mbcset->nranges is 0. */
new_nranges = 2 * mbcset->nranges + 1;
/* Use realloc since mbcset->range_starts and mbcset->range_ends
are NULL if *range_alloc == 0. */
new_array_start = re_realloc (mbcset->range_starts, wchar_t,
new_nranges);
new_array_end = re_realloc (mbcset->range_ends, wchar_t,
new_nranges);
/* +1 in case of mbcset->nranges is 0. */
new_nranges = 2 * mbcset->nranges + 1;
/* Use realloc since mbcset->range_starts and mbcset->range_ends
are NULL if *range_alloc == 0. */
new_array_start = re_realloc (mbcset->range_starts, wchar_t,
new_nranges);
new_array_end = re_realloc (mbcset->range_ends, wchar_t,
new_nranges);
if (BE (new_array_start == NULL || new_array_end == NULL, 0))
return REG_ESPACE;
if (BE (new_array_start == NULL || new_array_end == NULL, 0))
return REG_ESPACE;
mbcset->range_starts = new_array_start;
mbcset->range_ends = new_array_end;
*range_alloc = new_nranges;
mbcset->range_starts = new_array_start;
mbcset->range_ends = new_array_end;
*range_alloc = new_nranges;
}
mbcset->range_starts[mbcset->nranges] = start_wc;
mbcset->range_ends[mbcset->nranges++] = end_wc;
}
mbcset->range_starts[mbcset->nranges] = start_wc;
mbcset->range_ends[mbcset->nranges++] = end_wc;
/* Build the table for single byte characters. */
for (wc = 0; wc <= SBC_MAX; ++wc)
{
@ -2779,13 +2785,9 @@ parse_bracket_exp (regexp, dfa, token, syntax, err)
static inline reg_errcode_t
__attribute ((always_inline))
# ifdef RE_ENABLE_I18N
build_range_exp (sbcset, mbcset, range_alloc, start_elem, end_elem)
re_charset_t *mbcset;
int *range_alloc;
# else /* not RE_ENABLE_I18N */
build_range_exp (sbcset, start_elem, end_elem)
# endif /* not RE_ENABLE_I18N */
re_bitset_ptr_t sbcset;
bracket_elem_t *start_elem, *end_elem;
{
@ -2793,33 +2795,6 @@ parse_bracket_exp (regexp, dfa, token, syntax, err)
uint32_t start_collseq;
uint32_t end_collseq;
# ifdef RE_ENABLE_I18N
/* Check the space of the arrays. */
if (BE (*range_alloc == mbcset->nranges, 0))
{
/* There are not enough space, need realloc. */
uint32_t *new_array_start;
uint32_t *new_array_end;
int new_nranges;
/* +1 in case of mbcset->nranges is 0. */
new_nranges = 2 * mbcset->nranges + 1;
/* Use realloc since mbcset->range_starts and mbcset->range_ends
are NULL if *range_alloc == 0. */
new_array_start = re_realloc (mbcset->range_starts, uint32_t,
new_nranges);
new_array_end = re_realloc (mbcset->range_ends, uint32_t,
new_nranges);
if (BE (new_array_start == NULL || new_array_end == NULL, 0))
return REG_ESPACE;
mbcset->range_starts = new_array_start;
mbcset->range_ends = new_array_end;
*range_alloc = new_nranges;
}
# endif /* RE_ENABLE_I18N */
/* Equivalence Classes and Character Classes can't be a range
start/end. */
if (BE (start_elem->type == EQUIV_CLASS || start_elem->type == CHAR_CLASS
@ -2835,11 +2810,38 @@ parse_bracket_exp (regexp, dfa, token, syntax, err)
if (BE ((syntax & RE_NO_EMPTY_RANGES) && start_collseq > end_collseq, 0))
return REG_ERANGE;
# ifdef RE_ENABLE_I18N
/* Got valid collation sequence values, add them as a new entry. */
mbcset->range_starts[mbcset->nranges] = start_collseq;
mbcset->range_ends[mbcset->nranges++] = end_collseq;
# endif /* RE_ENABLE_I18N */
/* Got valid collation sequence values, add them as a new entry.
However, if we have no collation elements, and the character set
is single byte, the single byte character set that we
build below suffices. */
if (nrules > 0 || dfa->mb_cur_max > 1)
{
/* Check the space of the arrays. */
if (BE (*range_alloc == mbcset->nranges, 0))
{
/* There is not enough space, need realloc. */
uint32_t *new_array_start;
uint32_t *new_array_end;
int new_nranges;
/* +1 in case of mbcset->nranges is 0. */
new_nranges = 2 * mbcset->nranges + 1;
new_array_start = re_realloc (mbcset->range_starts, uint32_t,
new_nranges);
new_array_end = re_realloc (mbcset->range_ends, uint32_t,
new_nranges);
if (BE (new_array_start == NULL || new_array_end == NULL, 0))
return REG_ESPACE;
mbcset->range_starts = new_array_start;
mbcset->range_ends = new_array_end;
*range_alloc = new_nranges;
}
mbcset->range_starts[mbcset->nranges] = start_collseq;
mbcset->range_ends[mbcset->nranges++] = end_collseq;
}
/* Build the table for single byte characters. */
for (ch = 0; ch <= SBC_MAX; ch++)
@ -2866,13 +2868,9 @@ parse_bracket_exp (regexp, dfa, token, syntax, err)
static inline reg_errcode_t
__attribute ((always_inline))
# ifdef RE_ENABLE_I18N
build_collating_symbol (sbcset, mbcset, coll_sym_alloc, name)
re_charset_t *mbcset;
int *coll_sym_alloc;
# else /* not RE_ENABLE_I18N */
build_collating_symbol (sbcset, name)
# endif /* not RE_ENABLE_I18N */
re_bitset_ptr_t sbcset;
const unsigned char *name;
{
@ -2898,7 +2896,6 @@ parse_bracket_exp (regexp, dfa, token, syntax, err)
else
return REG_ECOLLATE;
# ifdef RE_ENABLE_I18N
/* Got valid collation sequence, add it as a new entry. */
/* Check the space of the arrays. */
if (BE (*coll_sym_alloc == mbcset->ncoll_syms, 0))
@ -2916,7 +2913,6 @@ parse_bracket_exp (regexp, dfa, token, syntax, err)
*coll_sym_alloc = new_coll_sym_alloc;
}
mbcset->coll_syms[mbcset->ncoll_syms++] = idx;
# endif /* RE_ENABLE_I18N */
return REG_NOERROR;
}
else
@ -2938,9 +2934,8 @@ parse_bracket_exp (regexp, dfa, token, syntax, err)
re_charset_t *mbcset;
int coll_sym_alloc = 0, range_alloc = 0, mbchar_alloc = 0;
int equiv_class_alloc = 0, char_class_alloc = 0;
#else /* not RE_ENABLE_I18N */
int non_match = 0;
#endif /* not RE_ENABLE_I18N */
int non_match = 0;
bin_tree_t *work_tree;
int token_len;
int first_round = 1;
@ -2985,9 +2980,8 @@ parse_bracket_exp (regexp, dfa, token, syntax, err)
{
#ifdef RE_ENABLE_I18N
mbcset->non_match = 1;
#else /* not RE_ENABLE_I18N */
non_match = 1;
#endif /* not RE_ENABLE_I18N */
non_match = 1;
if (syntax & RE_HAT_LISTS_NOT_NEWLINE)
bitset_set (sbcset, '\0');
re_string_skip_bytes (regexp, token_len); /* Skip a token. */
@ -3066,11 +3060,18 @@ parse_bracket_exp (regexp, dfa, token, syntax, err)
token_len = peek_token_bracket (token, regexp, syntax);
*err = build_range_exp (sbcset,
#ifdef RE_ENABLE_I18N
mbcset, &range_alloc,
#endif /* RE_ENABLE_I18N */
#ifdef _LIBC
*err = build_range_exp (sbcset, mbcset, &range_alloc,
&start_elem, &end_elem);
#else
# ifdef RE_ENABLE_I18N
*err = build_range_exp (sbcset,
dfa->mb_cur_max > 1 ? mbcset : NULL,
&range_alloc, &start_elem, &end_elem);
# else
*err = build_range_exp (sbcset, &start_elem, &end_elem);
# endif
#endif /* RE_ENABLE_I18N */
if (BE (*err != REG_NOERROR, 0))
goto parse_bracket_exp_free_return;
}
@ -3144,12 +3145,9 @@ parse_bracket_exp (regexp, dfa, token, syntax, err)
re_string_skip_bytes (regexp, token_len); /* Skip a token. */
/* If it is non-matching list. */
#ifdef RE_ENABLE_I18N
if (mbcset->non_match)
#else /* not RE_ENABLE_I18N */
if (non_match)
#endif /* not RE_ENABLE_I18N */
bitset_not (sbcset);
#ifdef RE_ENABLE_I18N
/* Ensure only single byte characters are set. */
if (dfa->mb_cur_max > 1)
@ -3324,7 +3322,7 @@ build_equiv_class (sbcset, name)
re_bitset_ptr_t sbcset;
const unsigned char *name;
{
#if defined _LIBC && defined RE_ENABLE_I18N
#if defined _LIBC
uint32_t nrules = _NL_CURRENT_WORD (LC_COLLATE, _NL_COLLATE_NRULES);
if (nrules != 0)
{
@ -3393,7 +3391,7 @@ build_equiv_class (sbcset, name)
mbcset->equiv_classes[mbcset->nequiv_classes++] = idx1;
}
else
#endif /* _LIBC && RE_ENABLE_I18N */
#endif /* _LIBC */
{
if (BE (strlen ((const char *) name) != 1, 0))
return REG_ECOLLATE;
@ -3489,20 +3487,18 @@ build_charclass (trans, sbcset, class_name, syntax)
}
static bin_tree_t *
build_charclass_op (dfa, trans, class_name, extra, not, err)
build_charclass_op (dfa, trans, class_name, extra, non_match, err)
re_dfa_t *dfa;
unsigned RE_TRANSLATE_TYPE trans;
const unsigned char *class_name;
const unsigned char *extra;
int not;
int non_match;
reg_errcode_t *err;
{
re_bitset_ptr_t sbcset;
#ifdef RE_ENABLE_I18N
re_charset_t *mbcset;
int alloc = 0;
#else /* not RE_ENABLE_I18N */
int non_match = 0;
#endif /* not RE_ENABLE_I18N */
reg_errcode_t ret;
re_token_t br_token;
@ -3523,7 +3519,7 @@ build_charclass_op (dfa, trans, class_name, extra, not, err)
return NULL;
}
if (not)
if (non_match)
{
#ifdef RE_ENABLE_I18N
/*
@ -3531,8 +3527,6 @@ build_charclass_op (dfa, trans, class_name, extra, not, err)
bitset_set(cset->sbcset, '\0');
*/
mbcset->non_match = 1;
#else /* not RE_ENABLE_I18N */
non_match = 1;
#endif /* not RE_ENABLE_I18N */
}
@ -3557,11 +3551,7 @@ build_charclass_op (dfa, trans, class_name, extra, not, err)
bitset_set (sbcset, *extra);
/* If it is non-matching list. */
#ifdef RE_ENABLE_I18N
if (mbcset->non_match)
#else /* not RE_ENABLE_I18N */
if (non_match)
#endif /* not RE_ENABLE_I18N */
bitset_not (sbcset);
#ifdef RE_ENABLE_I18N