ba737b94fd
2002-03-14 Jakub Jelinek <jakub@redhat.com> * locale/broken_cur_max.c (__ctype_get_mb_cur_max): Use nl_langinfo. * locale/Versions (_nl_current_LC_COLLATE, _nl_current_LC_CTYPE): Remove. 2002-03-14 Jakub Jelinek <jakub@redhat.com> * sysdeps/generic/mp_clz_tab.c: New file. * sysdeps/i386/mp_clz_tab.c: New file. * sysdeps/hppa/mp_clz_tab.c: New file. * sysdeps/powerpc/mp_clz_tab.c: New file. * stdlib/Makefile (aux): Revert last patch. * math/Makefile (gmp-objs): Likewise. 2002-03-13 Paul Eggert <eggert@twinsun.com> * time/strftime.c: Comment fixes for references to obsolescent standards, In most cases the simplest fix is to remove the confusing comments. Cross-referencing all the standards properly is a bit of a pain, and it should be enough to put that info in the documentation as I did in my recent time.texi patch. 2002-03-13 Paul Eggert <eggert@twinsun.com> * manual/time.texi (Formatting Calendar Time): ISO C99 also specifies the E and O modifiers. %P is a GNU extension, and is not in ISO C99. Mention that %r is equivalent to %I:%M:%S %p in the POSIX locale. %T is also in ISO C99. The RFC 822 example is not valid in arbitrary locales. Reword the POSIX.2 wording slightly, to make it a bit clearer that POSIX.2 formats are also supported by later POSIX versions. If a format was introduced in ISO C99 it is also required by POSIX.1-2001.
52 lines
2.1 KiB
C
52 lines
2.1 KiB
C
/* Return number of characters in multibyte representation for current
|
|
character set.
|
|
Copyright (C) 1996, 1999 Free Software Foundation, Inc.
|
|
This file is part of the GNU C Library.
|
|
Contributed by Ulrich Drepper <drepper@cygnus.com>, 1996.
|
|
|
|
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, write to the Free
|
|
Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
|
|
02111-1307 USA. */
|
|
|
|
#include <langinfo.h>
|
|
#include <locale.h>
|
|
#include <stdlib.h>
|
|
#include "localeinfo.h"
|
|
|
|
|
|
/* This is a gross hack to get broken programs running.
|
|
|
|
ISO C provides no mean to find out how many bytes the wide
|
|
character representation really uses. But it defines MB_CUR_LEN to
|
|
return the information for the multi-byte character representation.
|
|
Many programmers don't know the difference between the two and
|
|
think this means the same. But assuming all characters have a size
|
|
of MB_CUR_LEN after they have been processed by `mbrtowc' is wrong.
|
|
Instead the maximum number of characters used for the conversion is
|
|
MB_CUR_LEN.
|
|
|
|
It is known that some Motif applications have this problem. To
|
|
cure this one has to make sure the glibc uses the function in this
|
|
file instead of the one in locale/mb_cur_max.c. This can either be
|
|
done by linking with this file or by using the LD_PRELOAD feature
|
|
of the dynamic linker. */
|
|
size_t
|
|
__ctype_get_mb_cur_max (void)
|
|
{
|
|
union locale_data_value u;
|
|
|
|
u.string = nl_langinfo (_NL_CTYPE_MB_CUR_MAX);
|
|
return ((size_t []) { 1, 1, 1, 2, 2, 3, 4 })[u.word];
|
|
}
|