4af0c5bfdd
- strcoll-remove-strdiff-opt.patch: Remove incorrect STRDIFF-based optimization (BZ #18589) - iconv-reset-input-buffer.patch: Fix iconv buffer handling with IGNORE error handler (BZ #18830) (forwarded request 338135 from Andreas_Schwab) OBS-URL: https://build.opensuse.org/request/show/338136 OBS-URL: https://build.opensuse.org/package/show/openSUSE:Factory/glibc?expand=0&rev=196
367 lines
11 KiB
Diff
367 lines
11 KiB
Diff
2015-10-09 Carlos O'Donell <carlos@redhat.com>
|
|
|
|
[BZ #18589]
|
|
* string/bug-strcoll2.c: Adjust copyright, and remove contributed by.
|
|
* string/Makefile ($(objpfx)bug-strcoll2.out): Depend on
|
|
$(gen-locales).
|
|
|
|
2015-10-08 Carlos O'Donell <carlos@redhat.com>
|
|
|
|
[BZ #18589]
|
|
* string/Makefile (tests): Add bug-strcoll2.
|
|
(LOCALES): Add cs_CZ.UTF-8.
|
|
|
|
2015-09-28 Martin Sebor <msebor@redhat.com>
|
|
|
|
[BZ #18969]
|
|
* string/Makefile (LOCALES): Define.
|
|
(gen-locales.mk): Include.
|
|
(test-strcasecmp.out, test-strncasecmp.out, tst-strxfrm.out)
|
|
(tst-strxfrm2.out): Add deppendency on $(gen-locales).
|
|
* string/tst-strxfrm2.c (do_test): Print the name of the locale
|
|
on setlocale failure.
|
|
|
|
2015-10-08 Carlos O'Donell <carlos@redhat.com>
|
|
|
|
[BZ #18589]
|
|
* string/bug-strcoll2.c: New file.
|
|
* locale/categories.def: Revert commit
|
|
f13c2a8dff2329c6692a80176262ceaaf8a6f74e.
|
|
* locale/langinfo.h: Likewise.
|
|
* locale/localeinfo.h: Likewise.
|
|
* locale/C-collate.c: Likewise.
|
|
* programs/ld-collate.c (collate_output): Likewise.
|
|
* string/strcoll_l.c (STRDIFF): Likewise.
|
|
(STRCOLL): Likewise.
|
|
* wcsmbs/wcscoll_l.c: Likewise.
|
|
|
|
Index: glibc-2.22/locale/C-collate.c
|
|
===================================================================
|
|
--- glibc-2.22.orig/locale/C-collate.c
|
|
+++ glibc-2.22/locale/C-collate.c
|
|
@@ -144,8 +144,6 @@ const struct __locale_data _nl_C_LC_COLL
|
|
/* _NL_COLLATE_COLLSEQWC */
|
|
{ .string = (const char *) collseqwc },
|
|
/* _NL_COLLATE_CODESET */
|
|
- { .string = _nl_C_codeset },
|
|
- /* _NL_COLLATE_ENCODING_TYPE */
|
|
- { .word = __cet_8bit }
|
|
+ { .string = _nl_C_codeset }
|
|
}
|
|
};
|
|
Index: glibc-2.22/locale/categories.def
|
|
===================================================================
|
|
--- glibc-2.22.orig/locale/categories.def
|
|
+++ glibc-2.22/locale/categories.def
|
|
@@ -58,7 +58,6 @@ DEFINE_CATEGORY
|
|
DEFINE_ELEMENT (_NL_COLLATE_COLLSEQMB, "collate-collseqmb", std, wstring)
|
|
DEFINE_ELEMENT (_NL_COLLATE_COLLSEQWC, "collate-collseqwc", std, wstring)
|
|
DEFINE_ELEMENT (_NL_COLLATE_CODESET, "collate-codeset", std, string)
|
|
- DEFINE_ELEMENT (_NL_COLLATE_ENCODING_TYPE, "collate-encoding-type", std, word)
|
|
), NO_POSTLOAD)
|
|
|
|
|
|
Index: glibc-2.22/locale/langinfo.h
|
|
===================================================================
|
|
--- glibc-2.22.orig/locale/langinfo.h
|
|
+++ glibc-2.22/locale/langinfo.h
|
|
@@ -255,7 +255,6 @@ enum
|
|
_NL_COLLATE_COLLSEQMB,
|
|
_NL_COLLATE_COLLSEQWC,
|
|
_NL_COLLATE_CODESET,
|
|
- _NL_COLLATE_ENCODING_TYPE,
|
|
_NL_NUM_LC_COLLATE,
|
|
|
|
/* LC_CTYPE category: character classification.
|
|
Index: glibc-2.22/locale/localeinfo.h
|
|
===================================================================
|
|
--- glibc-2.22.orig/locale/localeinfo.h
|
|
+++ glibc-2.22/locale/localeinfo.h
|
|
@@ -110,14 +110,6 @@ enum coll_sort_rule
|
|
sort_mask
|
|
};
|
|
|
|
-/* Collation encoding type. */
|
|
-enum collation_encoding_type
|
|
-{
|
|
- __cet_other,
|
|
- __cet_8bit,
|
|
- __cet_utf8
|
|
-};
|
|
-
|
|
/* We can map the types of the entries into a few categories. */
|
|
enum value_type
|
|
{
|
|
Index: glibc-2.22/locale/programs/ld-collate.c
|
|
===================================================================
|
|
--- glibc-2.22.orig/locale/programs/ld-collate.c
|
|
+++ glibc-2.22/locale/programs/ld-collate.c
|
|
@@ -32,7 +32,6 @@
|
|
#include "linereader.h"
|
|
#include "locfile.h"
|
|
#include "elem-hash.h"
|
|
-#include "../localeinfo.h"
|
|
|
|
/* Uncomment the following line in the production version. */
|
|
/* #define NDEBUG 1 */
|
|
@@ -2131,8 +2130,6 @@ collate_output (struct localedef_t *loca
|
|
/* The words have to be handled specially. */
|
|
if (idx == _NL_ITEM_INDEX (_NL_COLLATE_SYMB_HASH_SIZEMB))
|
|
add_locale_uint32 (&file, 0);
|
|
- else if (idx == _NL_ITEM_INDEX (_NL_COLLATE_ENCODING_TYPE))
|
|
- add_locale_uint32 (&file, __cet_other);
|
|
else
|
|
add_locale_empty (&file);
|
|
}
|
|
@@ -2496,12 +2493,6 @@ collate_output (struct localedef_t *loca
|
|
add_locale_raw_data (&file, collate->mbseqorder, 256);
|
|
add_locale_collseq_table (&file, &collate->wcseqorder);
|
|
add_locale_string (&file, charmap->code_set_name);
|
|
- if (strcmp (charmap->code_set_name, "UTF-8") == 0)
|
|
- add_locale_uint32 (&file, __cet_utf8);
|
|
- else if (charmap->mb_cur_max == 1)
|
|
- add_locale_uint32 (&file, __cet_8bit);
|
|
- else
|
|
- add_locale_uint32 (&file, __cet_other);
|
|
write_locale_data (output_path, LC_COLLATE, "LC_COLLATE", &file);
|
|
|
|
obstack_free (&weightpool, NULL);
|
|
Index: glibc-2.22/string/Makefile
|
|
===================================================================
|
|
--- glibc-2.22.orig/string/Makefile
|
|
+++ glibc-2.22/string/Makefile
|
|
@@ -54,7 +54,7 @@ tests := tester inl-tester noinl-tester
|
|
tst-strtok tst-strxfrm bug-strcoll1 tst-strfry \
|
|
bug-strtok1 $(addprefix test-,$(strop-tests)) \
|
|
bug-envz1 tst-strxfrm2 tst-endian tst-svc2 \
|
|
- tst-strtok_r
|
|
+ tst-strtok_r bug-strcoll2
|
|
|
|
xtests = tst-strcoll-overflow
|
|
|
|
@@ -75,4 +75,17 @@ ifeq ($(run-built-tests),yes)
|
|
$(objpfx)tst-svc-cmp.out: tst-svc.expect $(objpfx)tst-svc.out
|
|
cmp $^ > $@; \
|
|
$(evaluate-test)
|
|
+
|
|
+LOCALES := de_DE.UTF-8 en_US.ISO-8859-1 en_US.UTF-8 \
|
|
+ tr_TR.ISO-8859-9 tr_TR.UTF-8 cs_CZ.UTF-8 \
|
|
+ da_DK.ISO-8859-1
|
|
+include ../gen-locales.mk
|
|
+
|
|
+$(objpfx)test-strcasecmp.out: $(gen-locales)
|
|
+$(objpfx)test-strncasecmp.out: $(gen-locales)
|
|
+$(objpfx)tst-strxfrm.out: $(gen-locales)
|
|
+$(objpfx)tst-strxfrm2.out: $(gen-locales)
|
|
+# bug-strcoll2 needs cs_CZ.UTF-8 and da_DK.ISO-8859-1.
|
|
+$(objpfx)bug-strcoll2.out: $(gen-locales)
|
|
+
|
|
endif
|
|
Index: glibc-2.22/string/bug-strcoll2.c
|
|
===================================================================
|
|
--- /dev/null
|
|
+++ glibc-2.22/string/bug-strcoll2.c
|
|
@@ -0,0 +1,92 @@
|
|
+/* Bug 18589: sort-test.sh fails at random.
|
|
+ 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 <string.h>
|
|
+#include <locale.h>
|
|
+
|
|
+/* An incorrect strcoll optimization resulted in incorrect
|
|
+ results from strcoll for cs_CZ and da_DK. */
|
|
+
|
|
+int
|
|
+test_cs_CZ (void)
|
|
+{
|
|
+ const char t1[] = "config";
|
|
+ const char t2[] = "choose";
|
|
+ if (setlocale (LC_ALL, "cs_CZ.UTF-8") == NULL)
|
|
+ {
|
|
+ perror ("setlocale");
|
|
+ return 1;
|
|
+ }
|
|
+ /* In Czech the digraph ch sorts after c, therefore we expect
|
|
+ config to sort before choose. */
|
|
+ int a = strcoll (t1, t2);
|
|
+ int b = strcoll (t2, t1);
|
|
+ printf ("strcoll (\"%s\", \"%s\") = %d\n", t1, t2, a);
|
|
+ printf ("strcoll (\"%s\", \"%s\") = %d\n", t2, t1, b);
|
|
+ if (a < 0 && b > 0)
|
|
+ {
|
|
+ puts ("PASS: config < choose");
|
|
+ return 0;
|
|
+ }
|
|
+ else
|
|
+ {
|
|
+ puts ("FAIL: Wrong sorting in cs_CZ.UTF-8.");
|
|
+ return 1;
|
|
+ }
|
|
+}
|
|
+
|
|
+int
|
|
+test_da_DK (void)
|
|
+{
|
|
+ const char t1[] = "AS";
|
|
+ const char t2[] = "AA";
|
|
+ if (setlocale (LC_ALL, "da_DK.ISO-8859-1") == NULL)
|
|
+ {
|
|
+ perror ("setlocale");
|
|
+ return 1;
|
|
+ }
|
|
+ /* AA should be treated as the last letter of the Danish alphabet,
|
|
+ hence sorting after AS. */
|
|
+ int a = strcoll (t1, t2);
|
|
+ int b = strcoll (t2, t1);
|
|
+ printf ("strcoll (\"%s\", \"%s\") = %d\n", t1, t2, a);
|
|
+ printf ("strcoll (\"%s\", \"%s\") = %d\n", t2, t1, b);
|
|
+ if (a < 0 && b > 0)
|
|
+ {
|
|
+ puts ("PASS: AS < AA");
|
|
+ return 0;
|
|
+ }
|
|
+ else
|
|
+ {
|
|
+ puts ("FAIL: Wrong sorting in da_DK.ISO-8859-1");
|
|
+ return 1;
|
|
+ }
|
|
+}
|
|
+
|
|
+static int
|
|
+do_test (void)
|
|
+{
|
|
+ int err = 0;
|
|
+ err |= test_cs_CZ ();
|
|
+ err |= test_da_DK ();
|
|
+ return err;
|
|
+}
|
|
+
|
|
+#define TEST_FUNCTION do_test ()
|
|
+#include "../test-skeleton.c"
|
|
Index: glibc-2.22/string/strcoll_l.c
|
|
===================================================================
|
|
--- glibc-2.22.orig/string/strcoll_l.c
|
|
+++ glibc-2.22/string/strcoll_l.c
|
|
@@ -29,7 +29,6 @@
|
|
# define STRING_TYPE char
|
|
# define USTRING_TYPE unsigned char
|
|
# define STRCOLL __strcoll_l
|
|
-# define STRDIFF __strdiff
|
|
# define STRCMP strcmp
|
|
# define WEIGHT_H "../locale/weight.h"
|
|
# define SUFFIX MB
|
|
@@ -42,20 +41,6 @@
|
|
#include "../locale/localeinfo.h"
|
|
#include WEIGHT_H
|
|
|
|
-#define MASK_UTF8_7BIT (1 << 7)
|
|
-#define MASK_UTF8_START (3 << 6)
|
|
-
|
|
-size_t
|
|
-STRDIFF (const STRING_TYPE *s, const STRING_TYPE *t)
|
|
-{
|
|
- size_t n;
|
|
-
|
|
- for (n = 0; *s != '\0' && *s++ == *t++; ++n)
|
|
- continue;
|
|
-
|
|
- return n;
|
|
-}
|
|
-
|
|
/* Track status while looking for sequences in a string. */
|
|
typedef struct
|
|
{
|
|
@@ -269,29 +254,9 @@ STRCOLL (const STRING_TYPE *s1, const ST
|
|
const USTRING_TYPE *extra;
|
|
const int32_t *indirect;
|
|
|
|
- /* In case there is no locale specific sort order (C / POSIX). */
|
|
if (nrules == 0)
|
|
return STRCMP (s1, s2);
|
|
|
|
- /* Fast forward to the position of the first difference. Needs to be
|
|
- encoding aware as the byte-by-byte comparison can stop in the middle
|
|
- of a char sequence for multibyte encodings like UTF-8. */
|
|
- uint_fast32_t encoding =
|
|
- current->values[_NL_ITEM_INDEX (_NL_COLLATE_ENCODING_TYPE)].word;
|
|
- if (encoding != __cet_other)
|
|
- {
|
|
- size_t diff = STRDIFF (s1, s2);
|
|
- if (diff > 0)
|
|
- {
|
|
- if (encoding == __cet_utf8 && (*(s1 + diff) & MASK_UTF8_7BIT) != 0)
|
|
- do
|
|
- diff--;
|
|
- while (diff > 0 && (*(s1 + diff) & MASK_UTF8_START) != MASK_UTF8_START);
|
|
- s1 += diff;
|
|
- s2 += diff;
|
|
- }
|
|
- }
|
|
-
|
|
/* Catch empty strings. */
|
|
if (__glibc_unlikely (*s1 == '\0') || __glibc_unlikely (*s2 == '\0'))
|
|
return (*s1 != '\0') - (*s2 != '\0');
|
|
@@ -358,8 +323,7 @@ STRCOLL (const STRING_TYPE *s1, const ST
|
|
byte-level comparison to ensure that we don't waste time
|
|
going through multiple passes for totally equal strings
|
|
before proceeding to subsequent passes. */
|
|
- if (pass == 0 && encoding == __cet_other &&
|
|
- STRCMP (s1, s2) == 0)
|
|
+ if (pass == 0 && STRCMP (s1, s2) == 0)
|
|
return result;
|
|
else
|
|
break;
|
|
Index: glibc-2.22/string/tst-strxfrm2.c
|
|
===================================================================
|
|
--- glibc-2.22.orig/string/tst-strxfrm2.c
|
|
+++ glibc-2.22/string/tst-strxfrm2.c
|
|
@@ -5,6 +5,8 @@
|
|
static int
|
|
do_test (void)
|
|
{
|
|
+ static const char test_locale[] = "de_DE.UTF-8";
|
|
+
|
|
int res = 0;
|
|
|
|
char buf[20];
|
|
@@ -38,9 +40,9 @@ do_test (void)
|
|
res = 1;
|
|
}
|
|
|
|
- if (setlocale (LC_ALL, "de_DE.UTF-8") == NULL)
|
|
+ if (setlocale (LC_ALL, test_locale) == NULL)
|
|
{
|
|
- puts ("setlocale failed");
|
|
+ printf ("cannot set locale \"%s\"\n", test_locale);
|
|
res = 1;
|
|
}
|
|
else
|
|
Index: glibc-2.22/wcsmbs/wcscoll_l.c
|
|
===================================================================
|
|
--- glibc-2.22.orig/wcsmbs/wcscoll_l.c
|
|
+++ glibc-2.22/wcsmbs/wcscoll_l.c
|
|
@@ -23,7 +23,6 @@
|
|
#define STRING_TYPE wchar_t
|
|
#define USTRING_TYPE wint_t
|
|
#define STRCOLL __wcscoll_l
|
|
-#define STRDIFF __wcsdiff
|
|
#define STRCMP __wcscmp
|
|
#define WEIGHT_H "../locale/weightwc.h"
|
|
#define SUFFIX WC
|