glibc/fnmatch-collating-elements.patch
Andreas Schwab 9627e11e8b Accepting request 454919 from home:Andreas_Schwab:Factory
- Update to glibc 2.25
  * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
    24731-2:2010, is supported to enable declarations of functions from that
    TR.
  * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
    18661-1:2014, is supported to enable declarations of functions and macros
    from that TS.
  * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
    18661-4:2015, is supported to enable declarations of functions and macros
    from that TS.
  * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
    now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
  * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated.
  * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
    fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
    femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
  * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
    CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
    UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
    <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
    INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
    UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
    INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
    UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
    UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
    UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
    UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
    WINT_WIDTH.
  * New <math.h> features are added from TS 18661-1:2014:
    - Signaling NaN macros: SNANF, SNAN, SNANL.

OBS-URL: https://build.opensuse.org/request/show/454919
OBS-URL: https://build.opensuse.org/package/show/Base:System/glibc?expand=0&rev=451
2017-02-06 10:25:17 +00:00

468 lines
13 KiB
Diff

Fix fnmatch handling of collating elements (BZ #17396, BZ #16976)
[BZ #16976]
[BZ #17396]
* posix/fnmatch_loop.c (internal_fnmatch, internal_fnwmatch): When
looking up collating elements match against (wide) character
sequence instead of name. Correct alignment adjustment.
* posix/fnmatch.c: Don't include "../locale/elem-hash.h".
* posix/Makefile (tests): Add tst-fnmatch4 and tst-fnmatch5.
* posix/tst-fnmatch4.c: New file.
* posix/tst-fnmatch5.c: New file.
Index: glibc-2.22/posix/Makefile
===================================================================
--- glibc-2.22.orig/posix/Makefile
+++ glibc-2.22/posix/Makefile
@@ -87,6 +87,7 @@ tests := tstgetopt testfnm runtests run
bug-getopt5 tst-getopt_long1 bug-regex34 bug-regex35 \
tst-pathconf tst-getaddrinfo4 tst-rxspencer-no-utf8 \
tst-fnmatch3 bug-regex36 tst-getaddrinfo5 \
+ tst-fnmatch4 tst-fnmatch5 \
tst-posix_spawn-fd \
tst-posix_fadvise tst-posix_fadvise64
xtests := bug-ga2
Index: glibc-2.22/posix/fnmatch.c
===================================================================
--- glibc-2.22.orig/posix/fnmatch.c
+++ glibc-2.22/posix/fnmatch.c
@@ -53,7 +53,6 @@
we support a correct implementation only in glibc. */
#ifdef _LIBC
# include "../locale/localeinfo.h"
-# include "../locale/elem-hash.h"
# include "../locale/coll-lookup.h"
# include <shlib-compat.h>
Index: glibc-2.22/posix/fnmatch_loop.c
===================================================================
--- glibc-2.22.orig/posix/fnmatch_loop.c
+++ glibc-2.22/posix/fnmatch_loop.c
@@ -503,26 +503,12 @@ FCT (pattern, string, string_end, no_lea
{
int32_t table_size;
const int32_t *symb_table;
-# if WIDE_CHAR_VERSION
- char str[c1];
- unsigned int strcnt;
-# else
-# define str (startp + 1)
-# endif
const unsigned char *extra;
int32_t idx;
int32_t elem;
- int32_t second;
- int32_t hash;
-
# if WIDE_CHAR_VERSION
- /* We have to convert the name to a single-byte
- string. This is possible since the names
- consist of ASCII characters and the internal
- representation is UCS4. */
- for (strcnt = 0; strcnt < c1; ++strcnt)
- str[strcnt] = startp[1 + strcnt];
-#endif
+ int32_t *wextra;
+# endif
table_size =
_NL_CURRENT_WORD (LC_COLLATE,
@@ -534,71 +520,55 @@ FCT (pattern, string, string_end, no_lea
_NL_CURRENT (LC_COLLATE,
_NL_COLLATE_SYMB_EXTRAMB);
- /* Locate the character in the hashing table. */
- hash = elem_hash (str, c1);
-
- idx = 0;
- elem = hash % table_size;
- if (symb_table[2 * elem] != 0)
- {
- second = hash % (table_size - 2) + 1;
-
- do
- {
- /* First compare the hashing value. */
- if (symb_table[2 * elem] == hash
- && (c1
- == extra[symb_table[2 * elem + 1]])
- && memcmp (str,
- &extra[symb_table[2 * elem
- + 1]
- + 1], c1) == 0)
- {
- /* Yep, this is the entry. */
- idx = symb_table[2 * elem + 1];
- idx += 1 + extra[idx];
- break;
- }
-
- /* Next entry. */
- elem += second;
- }
- while (symb_table[2 * elem] != 0);
- }
+ for (elem = 0; elem < table_size; elem++)
+ if (symb_table[2 * elem] != 0)
+ {
+ idx = symb_table[2 * elem + 1];
+ /* Skip the name of collating element. */
+ idx += 1 + extra[idx];
+# if WIDE_CHAR_VERSION
+ /* Skip the byte sequence of the
+ collating element. */
+ idx += 1 + extra[idx];
+ /* Adjust for the alignment. */
+ idx = (idx + 3) & ~3;
+
+ wextra = (int32_t *) &extra[idx + 4];
+
+ if (/* Compare the length of the sequence. */
+ c1 == wextra[0]
+ /* Compare the wide char sequence. */
+ && memcmp (startp + 1, &wextra[1],
+ c1 * sizeof (UCHAR)) == 0)
+ /* Yep, this is the entry. */
+ break;
+# else
+ if (/* Compare the length of the sequence. */
+ c1 == extra[idx]
+ /* Compare the byte sequence. */
+ && memcmp (startp + 1,
+ &extra[idx + 1], c1) == 0)
+ /* Yep, this is the entry. */
+ break;
+# endif
+ }
- if (symb_table[2 * elem] != 0)
+ if (elem < table_size)
{
/* Compare the byte sequence but only if
this is not part of a range. */
-# if WIDE_CHAR_VERSION
- int32_t *wextra;
+ if (! is_range
- idx += 1 + extra[idx];
- /* Adjust for the alignment. */
- idx = (idx + 3) & ~3;
-
- wextra = (int32_t *) &extra[idx + 4];
-# endif
-
- if (! is_range)
- {
# if WIDE_CHAR_VERSION
- for (c1 = 0;
- (int32_t) c1 < wextra[idx];
- ++c1)
- if (n[c1] != wextra[1 + c1])
- break;
-
- if ((int32_t) c1 == wextra[idx])
- goto matched;
+ && memcmp (n, &wextra[1],
+ c1 * sizeof (UCHAR)) == 0
# else
- for (c1 = 0; c1 < extra[idx]; ++c1)
- if (n[c1] != extra[1 + c1])
- break;
-
- if (c1 == extra[idx])
- goto matched;
+ && memcmp (n, &extra[idx + 1], c1) == 0
# endif
+ )
+ {
+ n += c1 - 1;
+ goto matched;
}
/* Get the collation sequence value. */
@@ -606,9 +576,9 @@ FCT (pattern, string, string_end, no_lea
# if WIDE_CHAR_VERSION
cold = wextra[1 + wextra[idx]];
# else
- /* Adjust for the alignment. */
idx += 1 + extra[idx];
- idx = (idx + 3) & ~4;
+ /* Adjust for the alignment. */
+ idx = (idx + 3) & ~3;
cold = *((int32_t *) &extra[idx]);
# endif
@@ -618,10 +588,10 @@ FCT (pattern, string, string_end, no_lea
{
/* No valid character. Match it as a
single byte. */
- if (!is_range && *n == str[0])
+ if (!is_range && *n == startp[1])
goto matched;
- cold = str[0];
+ cold = startp[1];
c = *p++;
}
else
@@ -629,7 +599,6 @@ FCT (pattern, string, string_end, no_lea
}
}
else
-# undef str
#endif
{
c = FOLD (c);
@@ -721,25 +690,11 @@ FCT (pattern, string, string_end, no_lea
{
int32_t table_size;
const int32_t *symb_table;
-# if WIDE_CHAR_VERSION
- char str[c1];
- unsigned int strcnt;
-# else
-# define str (startp + 1)
-# endif
const unsigned char *extra;
int32_t idx;
int32_t elem;
- int32_t second;
- int32_t hash;
-
# if WIDE_CHAR_VERSION
- /* We have to convert the name to a single-byte
- string. This is possible since the names
- consist of ASCII characters and the internal
- representation is UCS4. */
- for (strcnt = 0; strcnt < c1; ++strcnt)
- str[strcnt] = startp[1 + strcnt];
+ int32_t *wextra;
# endif
table_size =
@@ -752,51 +707,44 @@ FCT (pattern, string, string_end, no_lea
_NL_CURRENT (LC_COLLATE,
_NL_COLLATE_SYMB_EXTRAMB);
- /* Locate the character in the hashing
- table. */
- hash = elem_hash (str, c1);
-
- idx = 0;
- elem = hash % table_size;
- if (symb_table[2 * elem] != 0)
- {
- second = hash % (table_size - 2) + 1;
-
- do
- {
- /* First compare the hashing value. */
- if (symb_table[2 * elem] == hash
- && (c1
- == extra[symb_table[2 * elem + 1]])
- && memcmp (str,
- &extra[symb_table[2 * elem + 1]
- + 1], c1) == 0)
- {
- /* Yep, this is the entry. */
- idx = symb_table[2 * elem + 1];
- idx += 1 + extra[idx];
- break;
- }
-
- /* Next entry. */
- elem += second;
- }
- while (symb_table[2 * elem] != 0);
- }
-
- if (symb_table[2 * elem] != 0)
- {
- /* Compare the byte sequence but only if
- this is not part of a range. */
+ for (elem = 0; elem < table_size; elem++)
+ if (symb_table[2 * elem] != 0)
+ {
+ idx = symb_table[2 * elem + 1];
+ /* Skip the name of collating
+ element. */
+ idx += 1 + extra[idx];
# if WIDE_CHAR_VERSION
- int32_t *wextra;
-
- idx += 1 + extra[idx];
- /* Adjust for the alignment. */
- idx = (idx + 3) & ~4;
-
- wextra = (int32_t *) &extra[idx + 4];
+ /* Skip the byte sequence of the
+ collating element. */
+ idx += 1 + extra[idx];
+ /* Adjust for the alignment. */
+ idx = (idx + 3) & ~3;
+
+ wextra = (int32_t *) &extra[idx + 4];
+
+ if (/* Compare the length of the
+ sequence. */
+ c1 == wextra[0]
+ /* Compare the wide char sequence. */
+ && memcmp (startp + 1, &wextra[1],
+ c1 * sizeof (int32_t)) == 0)
+ /* Yep, this is the entry. */
+ break;
+# else
+ if (/* Compare the length of the
+ sequence. */
+ c1 == extra[idx]
+ /* Compare the byte sequence. */
+ && memcmp (startp + 1,
+ &extra[idx + 1], c1) == 0)
+ /* Yep, this is the entry. */
+ break;
# endif
+ }
+
+ if (elem < table_size)
+ {
/* Get the collation sequence value. */
is_seqval = 1;
# if WIDE_CHAR_VERSION
@@ -804,19 +752,18 @@ FCT (pattern, string, string_end, no_lea
# else
/* Adjust for the alignment. */
idx += 1 + extra[idx];
- idx = (idx + 3) & ~4;
+ idx = (idx + 3) & ~3;
cend = *((int32_t *) &extra[idx]);
# endif
}
- else if (symb_table[2 * elem] != 0 && c1 == 1)
+ else if (c1 == 1)
{
- cend = str[0];
+ cend = startp[1];
c = *p++;
}
else
return FNM_NOMATCH;
}
-# undef str
}
else
{
Index: glibc-2.22/posix/tst-fnmatch4.c
===================================================================
--- /dev/null
+++ glibc-2.22/posix/tst-fnmatch4.c
@@ -0,0 +1,51 @@
+/* Test for fnmatch handling of collating elements
+ Copyright (C) 2015 Free Software Foundation, Inc.
+ This file is part of the GNU C Library.
+
+ 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.
+
+ 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
+ Lesser General Public License for more details.
+
+ 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 <stdio.h>
+#include <locale.h>
+#include <fnmatch.h>
+
+static int
+do_test_locale (const char *locale)
+{
+ const char *pattern = "[[.ll.]]";
+
+ if (setlocale (LC_ALL, locale) == NULL)
+ {
+ printf ("could not set locale %s\n", locale);
+ return 1;
+ }
+
+ if (fnmatch (pattern, "ll", 0) != 0)
+ {
+ printf ("%s didn't match in locale %s\n", pattern, locale);
+ return 1;
+ }
+
+ return 0;
+}
+
+static int
+do_test (void)
+{
+ return (do_test_locale ("es_US.ISO-8859-1")
+ || do_test_locale ("es_US.UTF-8"));
+}
+
+#define TEST_FUNCTION do_test ()
+#include "../test-skeleton.c"
Index: glibc-2.22/posix/tst-fnmatch5.c
===================================================================
--- /dev/null
+++ glibc-2.22/posix/tst-fnmatch5.c
@@ -0,0 +1,53 @@
+/* Test for fnmatch handling of collating elements
+ Copyright (C) 2015 Free Software Foundation, Inc.
+ This file is part of the GNU C Library.
+
+ 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.
+
+ 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
+ Lesser General Public License for more details.
+
+ 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 <fnmatch.h>
+#include <locale.h>
+#include <stdio.h>
+#include <string.h>
+
+#define LENGTH 20000000
+
+char pattern[LENGTH + 7];
+
+static int
+do_test (void)
+{
+ if (setlocale (LC_ALL, "en_US.UTF-8") == NULL)
+ {
+ puts ("could not set locale");
+ return 1;
+ }
+ pattern[0] = '[';
+ pattern[1] = '[';
+ pattern[2] = '.';
+ memset (pattern + 3, 'a', LENGTH);
+ pattern[LENGTH + 3] = '.';
+ pattern[LENGTH + 4] = ']';
+ pattern[LENGTH + 5] = ']';
+ int ret = fnmatch (pattern, "a", 0);
+ if (ret == 0)
+ {
+ puts ("fnmatch returned 0 for invalid pattern");
+ return 1;
+ }
+ return 0;
+}
+
+#define TEST_FUNCTION do_test ()
+#include "../test-skeleton.c"