Hi,
Since wcwidth() reportedly has become a bottleneck [1][2], and some of the time
the gnulib wcwidth() replacement spends is in localcharset(), let me optimize
localcharset().
Patch 1 removes support for Linux libc5 (obsolete since ca. 2001), glibc 2.0.x
(last used in Red Hat Linux 5.2, obsolete since ca. 2003 [3][4]) and
Mac OS X 10.2 (obsolete since 2003-2005 [5]).
Patch 2 adds a simple manual test, so that I can verify the results are as
expected when doing changes to the code.
Patch 3 removes the ability to specify the platform-dependent mapping in an
external file. This ability was useful up until ca. 2007. config.charset
has not changed for Unix platforms since 2010, therefore it is safe to assume
that the current mappings are nearly correct, i.e. not many people will need
to adjust them, and those that do can report it here or change the source code
locally.
At the same time, introduce a binary search for the mapping lookup.
The ultimate optimization of the tables would be through gperf, but this
comes with the cost of several extra files in the source code tree, and is
mostly relevant for old platforms only.
Patch 4 adds missing mappings. Found while testing on various platforms.
Patch 5 is a micro-optimization.
I also attempted to replace the binary search that works with strcmp()
with one that progresses one character at a time. This has the same overall
asymptotic complexity, and uses ca. 20% less memory accesses, but
is slower by a factor of 2. 'perf annotate' told me that this is because
apparently strcmp() has a fast implementation in glibc, whereas the
"one character at a time" algorithm uses plain x86_64 instructions
throughout. And apparently the cost of calling the function strcmp()
is negligible.
Bruno
[1] https://lists.gnu.org/archive/html/bug-gnulib/2018-04/msg00059.html
[2] https://lists.gnu.org/archive/html/coreutils/2018-05/msg00013.html
[3] https://distrowatch.com/table.php?distribution=redhat
[4] https://en.wikipedia.org/wiki/Red_Hat_Linux
[5] https://en.wikipedia.org/wiki/Darwin_(operating_system)#Release_history
>From c3edcfb79a446dc8dba5d7ce2c55b3efc97cf4ea Mon Sep 17 00:00:00 2001
From: Bruno Haible <br...@clisp.org>
Date: Sun, 20 May 2018 01:19:30 +0200
Subject: [PATCH 1/5] localcharset: Remove support for obsolete platforms.
* lib/config.charset: Remove support for Linux/libc5, glibc-2.0.x, and
Mac OS X 10.2. Comment out dubious entry for Solaris.
---
ChangeLog | 6 +++
lib/config.charset | 144 +----------------------------------------------------
2 files changed, 7 insertions(+), 143 deletions(-)
diff --git a/ChangeLog b/ChangeLog
index 7414575..251d106 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,9 @@
+2018-05-19 Bruno Haible <br...@clisp.org>
+
+ localcharset: Remove support for obsolete platforms.
+ * lib/config.charset: Remove support for Linux/libc5, glibc-2.0.x, and
+ Mac OS X 10.2. Comment out dubious entry for Solaris.
+
2018-05-19 Jim Meyering <meyer...@fb.com>
gnupload: adjust comment
diff --git a/lib/config.charset b/lib/config.charset
index acb99ba..a13065f 100644
--- a/lib/config.charset
+++ b/lib/config.charset
@@ -124,112 +124,11 @@ echo "# It was automatically generated from config.charset."
# List of references, updated during installation:
echo "# Packages using this file: "
case "$os" in
- linux-gnulibc1*)
- # Linux libc5 doesn't have nl_langinfo(CODESET); therefore
- # localcharset.c falls back to using the full locale name
- # from the environment variables.
- echo "C ASCII"
- echo "POSIX ASCII"
- for l in af af_ZA ca ca_ES da da_DK de de_AT de_BE de_CH de_DE de_LU \
- en en_AU en_BW en_CA en_DK en_GB en_IE en_NZ en_US en_ZA \
- en_ZW es es_AR es_BO es_CL es_CO es_DO es_EC es_ES es_GT \
- es_HN es_MX es_PA es_PE es_PY es_SV es_US es_UY es_VE et \
- et_EE eu eu_ES fi fi_FI fo fo_FO fr fr_BE fr_CA fr_CH fr_FR \
- fr_LU ga ga_IE gl gl_ES id id_ID in in_ID is is_IS it it_CH \
- it_IT kl kl_GL nl nl_BE nl_NL no no_NO pt pt_BR pt_PT sv \
- sv_FI sv_SE; do
- echo "$l ISO-8859-1"
- echo "$l.iso-8859-1 ISO-8859-1"
- echo "$l.iso-8859-15 ISO-8859-15"
- echo "$l.iso-8859-15@euro ISO-8859-15"
- echo "$l@euro ISO-8859-15"
- echo "$l.cp-437 CP437"
- echo "$l.cp-850 CP850"
- echo "$l.cp-1252 CP1252"
- echo "$l.cp-1252@euro CP1252"
- #echo "$l.atari-st ATARI-ST" # not a commonly used encoding
- echo "$l.utf-8 UTF-8"
- echo "$l.utf-8@euro UTF-8"
- done
- for l in cs cs_CZ hr hr_HR hu hu_HU pl pl_PL ro ro_RO sk sk_SK sl \
- sl_SI sr sr_CS sr_YU; do
- echo "$l ISO-8859-2"
- echo "$l.iso-8859-2 ISO-8859-2"
- echo "$l.cp-852 CP852"
- echo "$l.cp-1250 CP1250"
- echo "$l.utf-8 UTF-8"
- done
- for l in mk mk_MK ru ru_RU; do
- echo "$l ISO-8859-5"
- echo "$l.iso-8859-5 ISO-8859-5"
- echo "$l.koi8-r KOI8-R"
- echo "$l.cp-866 CP866"
- echo "$l.cp-1251 CP1251"
- echo "$l.utf-8 UTF-8"
- done
- for l in ar ar_SA; do
- echo "$l ISO-8859-6"
- echo "$l.iso-8859-6 ISO-8859-6"
- echo "$l.cp-864 CP864"
- #echo "$l.cp-868 CP868" # not a commonly used encoding
- echo "$l.cp-1256 CP1256"
- echo "$l.utf-8 UTF-8"
- done
- for l in el el_GR gr gr_GR; do
- echo "$l ISO-8859-7"
- echo "$l.iso-8859-7 ISO-8859-7"
- echo "$l.cp-869 CP869"
- echo "$l.cp-1253 CP1253"
- echo "$l.cp-1253@euro CP1253"
- echo "$l.utf-8 UTF-8"
- echo "$l.utf-8@euro UTF-8"
- done
- for l in he he_IL iw iw_IL; do
- echo "$l ISO-8859-8"
- echo "$l.iso-8859-8 ISO-8859-8"
- echo "$l.cp-862 CP862"
- echo "$l.cp-1255 CP1255"
- echo "$l.utf-8 UTF-8"
- done
- for l in tr tr_TR; do
- echo "$l ISO-8859-9"
- echo "$l.iso-8859-9 ISO-8859-9"
- echo "$l.cp-857 CP857"
- echo "$l.cp-1254 CP1254"
- echo "$l.utf-8 UTF-8"
- done
- for l in lt lt_LT lv lv_LV; do
- #echo "$l BALTIC" # not a commonly used encoding, wrong encoding name
- echo "$l ISO-8859-13"
- done
- for l in ru_UA uk uk_UA; do
- echo "$l KOI8-U"
- done
- for l in zh zh_CN; do
- #echo "$l GB_2312-80" # not a commonly used encoding, wrong encoding name
- echo "$l GB2312"
- done
- for l in ja ja_JP ja_JP.EUC; do
- echo "$l EUC-JP"
- done
- for l in ko ko_KR; do
- echo "$l EUC-KR"
- done
- for l in th th_TH; do
- echo "$l TIS-620"
- done
- for l in fa fa_IR; do
- #echo "$l ISIRI-3342" # a broken encoding
- echo "$l.utf-8 UTF-8"
- done
- ;;
linux* | *-gnu*)
# With glibc-2.1 or newer, we don't need any canonicalization,
# because glibc has iconv and both glibc and libiconv support all
# GNU canonical names directly. Therefore, the Makefile does not
# need to install the alias file at all.
- # The following applies only to glibc-2.0.x and older libcs.
- echo "ISO_646.IRV:1983 ASCII"
;;
aix*)
echo "ISO8859-1 ISO-8859-1"
@@ -341,7 +240,7 @@ case "$os" in
echo "GB18030 GB18030"
echo "cns11643 EUC-TW"
echo "5601 EUC-KR"
- echo "ko_KR.johap92 JOHAB"
+ #echo "ko_KR.johap92 JOHAB"
echo "eucJP EUC-JP"
echo "PCK SHIFT_JIS"
echo "TIS620.2533 TIS-620"
@@ -409,47 +308,6 @@ case "$os" in
echo "ISO8859-13 ISO-8859-13"
echo "ISO8859-15 ISO-8859-15"
;;
- darwin[56]*)
- # Darwin 6.8 doesn't have nl_langinfo(CODESET); therefore
- # localcharset.c falls back to using the full locale name
- # from the environment variables.
- echo "C ASCII"
- for l in en_AU en_CA en_GB en_US la_LN; do
- echo "$l.US-ASCII ASCII"
- done
- for l in da_DK de_AT de_CH de_DE en_AU en_CA en_GB en_US es_ES \
- fi_FI fr_BE fr_CA fr_CH fr_FR is_IS it_CH it_IT nl_BE \
- nl_NL no_NO pt_PT sv_SE; do
- echo "$l ISO-8859-1"
- echo "$l.ISO8859-1 ISO-8859-1"
- echo "$l.ISO8859-15 ISO-8859-15"
- done
- for l in la_LN; do
- echo "$l.ISO8859-1 ISO-8859-1"
- echo "$l.ISO8859-15 ISO-8859-15"
- done
- for l in cs_CZ hr_HR hu_HU la_LN pl_PL sl_SI; do
- echo "$l.ISO8859-2 ISO-8859-2"
- done
- for l in la_LN lt_LT; do
- echo "$l.ISO8859-4 ISO-8859-4"
- done
- for l in ru_RU; do
- echo "$l.KOI8-R KOI8-R"
- echo "$l.ISO8859-5 ISO-8859-5"
- echo "$l.CP866 CP866"
- done
- for l in bg_BG; do
- echo "$l.CP1251 CP1251"
- done
- echo "uk_UA.KOI8-U KOI8-U"
- echo "zh_TW.BIG5 BIG5"
- echo "zh_TW.Big5 BIG5"
- echo "zh_CN.EUC GB2312"
- echo "ja_JP.EUC EUC-JP"
- echo "ja_JP.SJIS SHIFT_JIS"
- echo "ko_KR.EUC EUC-KR"
- ;;
darwin*)
# Darwin 7.5 has nl_langinfo(CODESET), but sometimes its value is
# useless:
--
2.7.4
>From 9d70137bfec6f93a90eb77f63536dda25672152d Mon Sep 17 00:00:00 2001
From: Bruno Haible <br...@clisp.org>
Date: Sat, 19 May 2018 17:07:34 +0200
Subject: [PATCH 2/5] localcharset: Add a manual test.
* tests/test-localcharset.c: New file.
* modules/localcharset-tests: New file.
---
ChangeLog | 6 ++++++
modules/localcharset-tests | 9 +++++++++
tests/test-localcharset.c | 39 +++++++++++++++++++++++++++++++++++++++
3 files changed, 54 insertions(+)
create mode 100644 modules/localcharset-tests
create mode 100644 tests/test-localcharset.c
diff --git a/ChangeLog b/ChangeLog
index 251d106..1ae4351 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,5 +1,11 @@
2018-05-19 Bruno Haible <br...@clisp.org>
+ localcharset: Add a manual test.
+ * tests/test-localcharset.c: New file.
+ * modules/localcharset-tests: New file.
+
+2018-05-19 Bruno Haible <br...@clisp.org>
+
localcharset: Remove support for obsolete platforms.
* lib/config.charset: Remove support for Linux/libc5, glibc-2.0.x, and
Mac OS X 10.2. Comment out dubious entry for Solaris.
diff --git a/modules/localcharset-tests b/modules/localcharset-tests
new file mode 100644
index 0000000..52e4c60
--- /dev/null
+++ b/modules/localcharset-tests
@@ -0,0 +1,9 @@
+Files:
+tests/test-localcharset.c
+
+Depends-on:
+
+configure.ac:
+
+Makefile.am:
+noinst_PROGRAMS += test-localcharset
diff --git a/tests/test-localcharset.c b/tests/test-localcharset.c
new file mode 100644
index 0000000..69bca0b
--- /dev/null
+++ b/tests/test-localcharset.c
@@ -0,0 +1,39 @@
+/* Manual test of localcharset() function.
+ Copyright (C) 2018 Free Software Foundation, Inc.
+
+ 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 3 of the License, or
+ (at your option) any later version.
+
+ This program 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.
+
+ You should have received a copy of the GNU General Public License
+ along with this program. If not, see <https://www.gnu.org/licenses/>. */
+
+/* This program prints the result of locale_charset in the current locale.
+ One way to use it is:
+ $ for l in `locale -a`; do
+ echo -n "$l "; LANG=$l ./test-localcharset;
+ done \
+ | sort -k 2
+ */
+
+#include <config.h>
+
+#include "localcharset.h"
+
+#include <locale.h>
+#include <stdio.h>
+
+int
+main (void)
+{
+ setlocale (LC_ALL, "");
+ printf ("%s\n", locale_charset ());
+
+ return 0;
+}
--
2.7.4
>From bb5cc1c59c039836699b096a27a82b0d1abee038 Mon Sep 17 00:00:00 2001
From: Bruno Haible <br...@clisp.org>
Date: Sat, 19 May 2018 17:17:32 +0200
Subject: [PATCH 3/5] localcharset: Move mapping tables into the code.
* lib/localcharset.h: Document the GNU canonical names for character
encodings here.
* lib/localcharset.c: Don't include <fcntl.h>, <unistd.h>,
relocatable.h, configmake.h.
(O_NOFOLLOW, ISSLASH, DIRECTORY_SEPARATOR, getc, volatile): Remove
macros.
(charset_aliases): Remove variable.
(get_charset_aliases): Remove function.
(struct table_entry): New type.
(alias_table, locale_table): New constants.
(locale_charset): Use the alias_table or locale_table to get the
canonicalized encoding name.
* lib/config.charset: Remove file.
* lib/ref-add.sin: Remove file.
* lib/ref-del.sin: Remove file.
* m4/localcharset.m4 (gl_LOCALCHARSET): Don't require gl_FCNTL_O_FLAGS,
AC_CANONICAL_HOST, gl_GLIBC21. Don't check for getc_unlocked.
* modules/localcharset (Notice): Remove.
(Files): Remove config.charset, ref-add.sin, ref-del.sin, fcntl-o.m4,
glibc21.m4.
(Depends-on): Remove configmake.
(configure.ac): Define LOCALCHARSET_TESTS_ENVIRONMENT to empty.
(Makefile.am): Simplify.
* build-aux/prefix-gnulib-mk: Remove special code for the removed files.
---
ChangeLog | 28 ++
build-aux/prefix-gnulib-mk | 9 -
lib/.cvsignore | 1 -
lib/config.charset | 540 -----------------------
lib/localcharset.c | 1039 +++++++++++++++++++++++++++++---------------
lib/localcharset.h | 96 +++-
lib/ref-add.sin | 29 --
lib/ref-del.sin | 24 -
m4/localcharset.m4 | 8 +-
modules/localcharset | 83 +---
10 files changed, 813 insertions(+), 1044 deletions(-)
delete mode 100644 lib/config.charset
delete mode 100644 lib/ref-add.sin
delete mode 100644 lib/ref-del.sin
diff --git a/ChangeLog b/ChangeLog
index 1ae4351..ad2e064 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,5 +1,33 @@
2018-05-19 Bruno Haible <br...@clisp.org>
+ localcharset: Move mapping tables into the code. Use a binary search.
+ * lib/localcharset.h: Document the GNU canonical names for character
+ encodings here.
+ * lib/localcharset.c: Don't include <fcntl.h>, <unistd.h>,
+ relocatable.h, configmake.h.
+ (O_NOFOLLOW, ISSLASH, DIRECTORY_SEPARATOR, getc, volatile): Remove
+ macros.
+ (charset_aliases): Remove variable.
+ (get_charset_aliases): Remove function.
+ (struct table_entry): New type.
+ (alias_table, locale_table): New constants.
+ (locale_charset): Use the alias_table or locale_table to get the
+ canonicalized encoding name.
+ * lib/config.charset: Remove file.
+ * lib/ref-add.sin: Remove file.
+ * lib/ref-del.sin: Remove file.
+ * m4/localcharset.m4 (gl_LOCALCHARSET): Don't require gl_FCNTL_O_FLAGS,
+ AC_CANONICAL_HOST, gl_GLIBC21. Don't check for getc_unlocked.
+ * modules/localcharset (Notice): Remove.
+ (Files): Remove config.charset, ref-add.sin, ref-del.sin, fcntl-o.m4,
+ glibc21.m4.
+ (Depends-on): Remove configmake.
+ (configure.ac): Define LOCALCHARSET_TESTS_ENVIRONMENT to empty.
+ (Makefile.am): Simplify.
+ * build-aux/prefix-gnulib-mk: Remove special code for the removed files.
+
+2018-05-19 Bruno Haible <br...@clisp.org>
+
localcharset: Add a manual test.
* tests/test-localcharset.c: New file.
* modules/localcharset-tests: New file.
diff --git a/build-aux/prefix-gnulib-mk b/build-aux/prefix-gnulib-mk
index 67b723f..544654f 100755
--- a/build-aux/prefix-gnulib-mk
+++ b/build-aux/prefix-gnulib-mk
@@ -146,15 +146,6 @@ sub prefix ($)
s{^([\w.]+\s*\+?=)(.*)$}
{prefix_assignment($1, $2)}gem;
- # These three guys escape all the other regular rules.
- # Require the leading white space to avoid inserting the prefix
- # on a line like this:
- # charset_alias = $(DESTDIR)$(libdir)/charset.alias
- # With $(libdir), it would be erroneous.
- s{(\s)(charset\.alias|ref-add\.sed|ref-del\.sed)}{$1$prefix$2}g;
- # Unfortunately, as a result we sometimes have lib/lib.
- s{($prefix){2}}{$1}g;
-
# $(srcdir)/ is actually $(top_srcdir)/$prefix/.
# The trailing slash is required to avoid matching this rule:
# test '$(srcdir)' = . || rm -f $(top_builddir)/GNUmakefile
diff --git a/lib/.cvsignore b/lib/.cvsignore
index f2e6007..dc5a42e 100644
--- a/lib/.cvsignore
+++ b/lib/.cvsignore
@@ -1,7 +1,6 @@
.deps
Makefile
alloca.h
-charset.alias
fnmatch.h
fts.h
getdate.c
diff --git a/lib/config.charset b/lib/config.charset
deleted file mode 100644
index a13065f..0000000
--- a/lib/config.charset
+++ /dev/null
@@ -1,540 +0,0 @@
-#! /bin/sh
-# Output a system dependent table of character encoding aliases.
-#
-# Copyright (C) 2000-2004, 2006-2018 Free Software Foundation, Inc.
-#
-# 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.
-#
-# This program 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.
-#
-# You should have received a copy of the GNU General Public License along
-# with this program; if not, see <https://www.gnu.org/licenses/>.
-#
-# The table consists of lines of the form
-# ALIAS CANONICAL
-#
-# ALIAS is the (system dependent) result of "nl_langinfo (CODESET)".
-# ALIAS is compared in a case sensitive way.
-#
-# CANONICAL is the GNU canonical name for this character encoding.
-# It must be an encoding supported by libiconv. Support by GNU libc is
-# also desirable. CANONICAL is case insensitive. Usually an upper case
-# MIME charset name is preferred.
-# The current list of GNU canonical charset names is as follows.
-#
-# name MIME? used by which systems
-# (darwin = Mac OS X, woe32 = native Windows)
-#
-# ASCII, ANSI_X3.4-1968 glibc solaris freebsd netbsd darwin cygwin
-# ISO-8859-1 Y glibc aix hpux irix osf solaris freebsd netbsd openbsd darwin cygwin
-# ISO-8859-2 Y glibc aix hpux irix osf solaris freebsd netbsd openbsd darwin cygwin
-# ISO-8859-3 Y glibc solaris cygwin
-# ISO-8859-4 Y osf solaris freebsd netbsd openbsd darwin
-# ISO-8859-5 Y glibc aix hpux irix osf solaris freebsd netbsd openbsd darwin cygwin
-# ISO-8859-6 Y glibc aix hpux solaris cygwin
-# ISO-8859-7 Y glibc aix hpux irix osf solaris netbsd openbsd darwin cygwin
-# ISO-8859-8 Y glibc aix hpux osf solaris cygwin
-# ISO-8859-9 Y glibc aix hpux irix osf solaris darwin cygwin
-# ISO-8859-13 glibc netbsd openbsd darwin cygwin
-# ISO-8859-14 glibc cygwin
-# ISO-8859-15 glibc aix osf solaris freebsd netbsd openbsd darwin cygwin
-# KOI8-R Y glibc solaris freebsd netbsd openbsd darwin
-# KOI8-U Y glibc freebsd netbsd openbsd darwin cygwin
-# KOI8-T glibc
-# CP437 dos
-# CP775 dos
-# CP850 aix osf dos
-# CP852 dos
-# CP855 dos
-# CP856 aix
-# CP857 dos
-# CP861 dos
-# CP862 dos
-# CP864 dos
-# CP865 dos
-# CP866 freebsd netbsd openbsd darwin dos
-# CP869 dos
-# CP874 woe32 dos
-# CP922 aix
-# CP932 aix cygwin woe32 dos
-# CP943 aix
-# CP949 osf darwin woe32 dos
-# CP950 woe32 dos
-# CP1046 aix
-# CP1124 aix
-# CP1125 dos
-# CP1129 aix
-# CP1131 darwin
-# CP1250 woe32
-# CP1251 glibc solaris netbsd openbsd darwin cygwin woe32
-# CP1252 aix woe32
-# CP1253 woe32
-# CP1254 woe32
-# CP1255 glibc woe32
-# CP1256 woe32
-# CP1257 woe32
-# GB2312 Y glibc aix hpux irix solaris freebsd netbsd darwin
-# EUC-JP Y glibc aix hpux irix osf solaris freebsd netbsd darwin
-# EUC-KR Y glibc aix hpux irix osf solaris freebsd netbsd darwin cygwin
-# EUC-TW glibc aix hpux irix osf solaris netbsd
-# BIG5 Y glibc aix hpux osf solaris freebsd netbsd darwin cygwin
-# BIG5-HKSCS glibc solaris darwin
-# GBK glibc aix osf solaris darwin cygwin woe32 dos
-# GB18030 glibc solaris netbsd darwin
-# SHIFT_JIS Y hpux osf solaris freebsd netbsd darwin
-# JOHAB glibc solaris woe32
-# TIS-620 glibc aix hpux osf solaris cygwin
-# VISCII Y glibc
-# TCVN5712-1 glibc
-# ARMSCII-8 glibc darwin
-# GEORGIAN-PS glibc cygwin
-# PT154 glibc
-# HP-ROMAN8 hpux
-# HP-ARABIC8 hpux
-# HP-GREEK8 hpux
-# HP-HEBREW8 hpux
-# HP-TURKISH8 hpux
-# HP-KANA8 hpux
-# DEC-KANJI osf
-# DEC-HANYU osf
-# UTF-8 Y glibc aix hpux osf solaris netbsd darwin cygwin
-#
-# Note: Names which are not marked as being a MIME name should not be used in
-# Internet protocols for information interchange (mail, news, etc.).
-#
-# Note: ASCII and ANSI_X3.4-1968 are synonymous canonical names. Applications
-# must understand both names and treat them as equivalent.
-#
-# The first argument passed to this file is the canonical host specification,
-# CPU_TYPE-MANUFACTURER-OPERATING_SYSTEM
-# or
-# CPU_TYPE-MANUFACTURER-KERNEL-OPERATING_SYSTEM
-
-host="$1"
-os=`echo "$host" | sed -e 's/^[^-]*-[^-]*-\(.*\)$/\1/'`
-echo "# This file contains a table of character encoding aliases,"
-echo "# suitable for operating system '${os}'."
-echo "# It was automatically generated from config.charset."
-# List of references, updated during installation:
-echo "# Packages using this file: "
-case "$os" in
- linux* | *-gnu*)
- # With glibc-2.1 or newer, we don't need any canonicalization,
- # because glibc has iconv and both glibc and libiconv support all
- # GNU canonical names directly. Therefore, the Makefile does not
- # need to install the alias file at all.
- ;;
- aix*)
- echo "ISO8859-1 ISO-8859-1"
- echo "ISO8859-2 ISO-8859-2"
- echo "ISO8859-5 ISO-8859-5"
- echo "ISO8859-6 ISO-8859-6"
- echo "ISO8859-7 ISO-8859-7"
- echo "ISO8859-8 ISO-8859-8"
- echo "ISO8859-9 ISO-8859-9"
- echo "ISO8859-15 ISO-8859-15"
- echo "IBM-850 CP850"
- echo "IBM-856 CP856"
- echo "IBM-921 ISO-8859-13"
- echo "IBM-922 CP922"
- echo "IBM-932 CP932"
- echo "IBM-943 CP943"
- echo "IBM-1046 CP1046"
- echo "IBM-1124 CP1124"
- echo "IBM-1129 CP1129"
- echo "IBM-1252 CP1252"
- echo "IBM-eucCN GB2312"
- echo "IBM-eucJP EUC-JP"
- echo "IBM-eucKR EUC-KR"
- echo "IBM-eucTW EUC-TW"
- echo "big5 BIG5"
- echo "GBK GBK"
- echo "TIS-620 TIS-620"
- echo "UTF-8 UTF-8"
- ;;
- hpux*)
- echo "iso88591 ISO-8859-1"
- echo "iso88592 ISO-8859-2"
- echo "iso88595 ISO-8859-5"
- echo "iso88596 ISO-8859-6"
- echo "iso88597 ISO-8859-7"
- echo "iso88598 ISO-8859-8"
- echo "iso88599 ISO-8859-9"
- echo "iso885915 ISO-8859-15"
- echo "roman8 HP-ROMAN8"
- echo "arabic8 HP-ARABIC8"
- echo "greek8 HP-GREEK8"
- echo "hebrew8 HP-HEBREW8"
- echo "turkish8 HP-TURKISH8"
- echo "kana8 HP-KANA8"
- echo "tis620 TIS-620"
- echo "big5 BIG5"
- echo "eucJP EUC-JP"
- echo "eucKR EUC-KR"
- echo "eucTW EUC-TW"
- echo "hp15CN GB2312"
- #echo "ccdc ?" # what is this?
- echo "SJIS SHIFT_JIS"
- echo "utf8 UTF-8"
- ;;
- irix*)
- echo "ISO8859-1 ISO-8859-1"
- echo "ISO8859-2 ISO-8859-2"
- echo "ISO8859-5 ISO-8859-5"
- echo "ISO8859-7 ISO-8859-7"
- echo "ISO8859-9 ISO-8859-9"
- echo "eucCN GB2312"
- echo "eucJP EUC-JP"
- echo "eucKR EUC-KR"
- echo "eucTW EUC-TW"
- ;;
- osf*)
- echo "ISO8859-1 ISO-8859-1"
- echo "ISO8859-2 ISO-8859-2"
- echo "ISO8859-4 ISO-8859-4"
- echo "ISO8859-5 ISO-8859-5"
- echo "ISO8859-7 ISO-8859-7"
- echo "ISO8859-8 ISO-8859-8"
- echo "ISO8859-9 ISO-8859-9"
- echo "ISO8859-15 ISO-8859-15"
- echo "cp850 CP850"
- echo "big5 BIG5"
- echo "dechanyu DEC-HANYU"
- echo "dechanzi GB2312"
- echo "deckanji DEC-KANJI"
- echo "deckorean EUC-KR"
- echo "eucJP EUC-JP"
- echo "eucKR EUC-KR"
- echo "eucTW EUC-TW"
- echo "GBK GBK"
- echo "KSC5601 CP949"
- echo "sdeckanji EUC-JP"
- echo "SJIS SHIFT_JIS"
- echo "TACTIS TIS-620"
- echo "UTF-8 UTF-8"
- ;;
- solaris*)
- echo "646 ASCII"
- echo "ISO8859-1 ISO-8859-1"
- echo "ISO8859-2 ISO-8859-2"
- echo "ISO8859-3 ISO-8859-3"
- echo "ISO8859-4 ISO-8859-4"
- echo "ISO8859-5 ISO-8859-5"
- echo "ISO8859-6 ISO-8859-6"
- echo "ISO8859-7 ISO-8859-7"
- echo "ISO8859-8 ISO-8859-8"
- echo "ISO8859-9 ISO-8859-9"
- echo "ISO8859-15 ISO-8859-15"
- echo "koi8-r KOI8-R"
- echo "ansi-1251 CP1251"
- echo "BIG5 BIG5"
- echo "Big5-HKSCS BIG5-HKSCS"
- echo "gb2312 GB2312"
- echo "GBK GBK"
- echo "GB18030 GB18030"
- echo "cns11643 EUC-TW"
- echo "5601 EUC-KR"
- #echo "ko_KR.johap92 JOHAB"
- echo "eucJP EUC-JP"
- echo "PCK SHIFT_JIS"
- echo "TIS620.2533 TIS-620"
- #echo "sun_eu_greek ?" # what is this?
- echo "UTF-8 UTF-8"
- ;;
- freebsd*)
- # FreeBSD 4.2 doesn't have nl_langinfo(CODESET); therefore
- # localcharset.c falls back to using the full locale name
- # from the environment variables.
- echo "C ASCII"
- echo "US-ASCII ASCII"
- for l in la_LN lt_LN; do
- echo "$l.ASCII ASCII"
- done
- for l in da_DK de_AT de_CH de_DE en_AU en_CA en_GB en_US es_ES \
- fi_FI fr_BE fr_CA fr_CH fr_FR is_IS it_CH it_IT la_LN \
- lt_LN nl_BE nl_NL no_NO pt_PT sv_SE; do
- echo "$l.ISO_8859-1 ISO-8859-1"
- echo "$l.DIS_8859-15 ISO-8859-15"
- done
- for l in cs_CZ hr_HR hu_HU la_LN lt_LN pl_PL sl_SI; do
- echo "$l.ISO_8859-2 ISO-8859-2"
- done
- for l in la_LN lt_LT; do
- echo "$l.ISO_8859-4 ISO-8859-4"
- done
- for l in ru_RU ru_SU; do
- echo "$l.KOI8-R KOI8-R"
- echo "$l.ISO_8859-5 ISO-8859-5"
- echo "$l.CP866 CP866"
- done
- echo "uk_UA.KOI8-U KOI8-U"
- echo "zh_TW.BIG5 BIG5"
- echo "zh_TW.Big5 BIG5"
- echo "zh_CN.EUC GB2312"
- echo "ja_JP.EUC EUC-JP"
- echo "ja_JP.SJIS SHIFT_JIS"
- echo "ja_JP.Shift_JIS SHIFT_JIS"
- echo "ko_KR.EUC EUC-KR"
- ;;
- netbsd*)
- echo "646 ASCII"
- echo "ISO8859-1 ISO-8859-1"
- echo "ISO8859-2 ISO-8859-2"
- echo "ISO8859-4 ISO-8859-4"
- echo "ISO8859-5 ISO-8859-5"
- echo "ISO8859-7 ISO-8859-7"
- echo "ISO8859-13 ISO-8859-13"
- echo "ISO8859-15 ISO-8859-15"
- echo "eucCN GB2312"
- echo "eucJP EUC-JP"
- echo "eucKR EUC-KR"
- echo "eucTW EUC-TW"
- echo "BIG5 BIG5"
- echo "SJIS SHIFT_JIS"
- ;;
- openbsd*)
- echo "646 ASCII"
- echo "ISO8859-1 ISO-8859-1"
- echo "ISO8859-2 ISO-8859-2"
- echo "ISO8859-4 ISO-8859-4"
- echo "ISO8859-5 ISO-8859-5"
- echo "ISO8859-7 ISO-8859-7"
- echo "ISO8859-13 ISO-8859-13"
- echo "ISO8859-15 ISO-8859-15"
- ;;
- darwin*)
- # Darwin 7.5 has nl_langinfo(CODESET), but sometimes its value is
- # useless:
- # - It returns the empty string when LANG is set to a locale of the
- # form ll_CC, although ll_CC/LC_CTYPE is a symlink to an UTF-8
- # LC_CTYPE file.
- # - The environment variables LANG, LC_CTYPE, LC_ALL are not set by
- # the system; nl_langinfo(CODESET) returns "US-ASCII" in this case.
- # - The documentation says:
- # "... all code that calls BSD system routines should ensure
- # that the const *char parameters of these routines are in UTF-8
- # encoding. All BSD system functions expect their string
- # parameters to be in UTF-8 encoding and nothing else."
- # It also says
- # "An additional caveat is that string parameters for files,
- # paths, and other file-system entities must be in canonical
- # UTF-8. In a canonical UTF-8 Unicode string, all decomposable
- # characters are decomposed ..."
- # but this is not true: You can pass non-decomposed UTF-8 strings
- # to file system functions, and it is the OS which will convert
- # them to decomposed UTF-8 before accessing the file system.
- # - The Apple Terminal application displays UTF-8 by default.
- # - However, other applications are free to use different encodings:
- # - xterm uses ISO-8859-1 by default.
- # - TextEdit uses MacRoman by default.
- # We prefer UTF-8 over decomposed UTF-8-MAC because one should
- # minimize the use of decomposed Unicode. Unfortunately, through the
- # Darwin file system, decomposed UTF-8 strings are leaked into user
- # space nevertheless.
- # Then there are also the locales with encodings other than US-ASCII
- # and UTF-8. These locales can be occasionally useful to users (e.g.
- # when grepping through ISO-8859-1 encoded text files), when all their
- # file names are in US-ASCII.
- echo "ISO8859-1 ISO-8859-1"
- echo "ISO8859-2 ISO-8859-2"
- echo "ISO8859-4 ISO-8859-4"
- echo "ISO8859-5 ISO-8859-5"
- echo "ISO8859-7 ISO-8859-7"
- echo "ISO8859-9 ISO-8859-9"
- echo "ISO8859-13 ISO-8859-13"
- echo "ISO8859-15 ISO-8859-15"
- echo "KOI8-R KOI8-R"
- echo "KOI8-U KOI8-U"
- echo "CP866 CP866"
- echo "CP949 CP949"
- echo "CP1131 CP1131"
- echo "CP1251 CP1251"
- echo "eucCN GB2312"
- echo "GB2312 GB2312"
- echo "eucJP EUC-JP"
- echo "eucKR EUC-KR"
- echo "Big5 BIG5"
- echo "Big5HKSCS BIG5-HKSCS"
- echo "GBK GBK"
- echo "GB18030 GB18030"
- echo "SJIS SHIFT_JIS"
- echo "ARMSCII-8 ARMSCII-8"
- echo "PT154 PT154"
- #echo "ISCII-DEV ?"
- echo "* UTF-8"
- ;;
- beos* | haiku*)
- # BeOS and Haiku have a single locale, and it has UTF-8 encoding.
- echo "* UTF-8"
- ;;
- msdosdjgpp*)
- # DJGPP 2.03 doesn't have nl_langinfo(CODESET); therefore
- # localcharset.c falls back to using the full locale name
- # from the environment variables.
- echo "#"
- echo "# The encodings given here may not all be correct."
- echo "# If you find that the encoding given for your language and"
- echo "# country is not the one your DOS machine actually uses, just"
- echo "# correct it in this file, and send a mail to"
- echo "# Juan Manuel Guerrero <juan.guerr...@gmx.de>"
- echo "# and Bruno Haible <br...@clisp.org>."
- echo "#"
- echo "C ASCII"
- # ISO-8859-1 languages
- echo "ca CP850"
- echo "ca_ES CP850"
- echo "da CP865" # not CP850 ??
- echo "da_DK CP865" # not CP850 ??
- echo "de CP850"
- echo "de_AT CP850"
- echo "de_CH CP850"
- echo "de_DE CP850"
- echo "en CP850"
- echo "en_AU CP850" # not CP437 ??
- echo "en_CA CP850"
- echo "en_GB CP850"
- echo "en_NZ CP437"
- echo "en_US CP437"
- echo "en_ZA CP850" # not CP437 ??
- echo "es CP850"
- echo "es_AR CP850"
- echo "es_BO CP850"
- echo "es_CL CP850"
- echo "es_CO CP850"
- echo "es_CR CP850"
- echo "es_CU CP850"
- echo "es_DO CP850"
- echo "es_EC CP850"
- echo "es_ES CP850"
- echo "es_GT CP850"
- echo "es_HN CP850"
- echo "es_MX CP850"
- echo "es_NI CP850"
- echo "es_PA CP850"
- echo "es_PY CP850"
- echo "es_PE CP850"
- echo "es_SV CP850"
- echo "es_UY CP850"
- echo "es_VE CP850"
- echo "et CP850"
- echo "et_EE CP850"
- echo "eu CP850"
- echo "eu_ES CP850"
- echo "fi CP850"
- echo "fi_FI CP850"
- echo "fr CP850"
- echo "fr_BE CP850"
- echo "fr_CA CP850"
- echo "fr_CH CP850"
- echo "fr_FR CP850"
- echo "ga CP850"
- echo "ga_IE CP850"
- echo "gd CP850"
- echo "gd_GB CP850"
- echo "gl CP850"
- echo "gl_ES CP850"
- echo "id CP850" # not CP437 ??
- echo "id_ID CP850" # not CP437 ??
- echo "is CP861" # not CP850 ??
- echo "is_IS CP861" # not CP850 ??
- echo "it CP850"
- echo "it_CH CP850"
- echo "it_IT CP850"
- echo "lt CP775"
- echo "lt_LT CP775"
- echo "lv CP775"
- echo "lv_LV CP775"
- echo "nb CP865" # not CP850 ??
- echo "nb_NO CP865" # not CP850 ??
- echo "nl CP850"
- echo "nl_BE CP850"
- echo "nl_NL CP850"
- echo "nn CP865" # not CP850 ??
- echo "nn_NO CP865" # not CP850 ??
- echo "no CP865" # not CP850 ??
- echo "no_NO CP865" # not CP850 ??
- echo "pt CP850"
- echo "pt_BR CP850"
- echo "pt_PT CP850"
- echo "sv CP850"
- echo "sv_SE CP850"
- # ISO-8859-2 languages
- echo "cs CP852"
- echo "cs_CZ CP852"
- echo "hr CP852"
- echo "hr_HR CP852"
- echo "hu CP852"
- echo "hu_HU CP852"
- echo "pl CP852"
- echo "pl_PL CP852"
- echo "ro CP852"
- echo "ro_RO CP852"
- echo "sk CP852"
- echo "sk_SK CP852"
- echo "sl CP852"
- echo "sl_SI CP852"
- echo "sq CP852"
- echo "sq_AL CP852"
- echo "sr CP852" # CP852 or CP866 or CP855 ??
- echo "sr_CS CP852" # CP852 or CP866 or CP855 ??
- echo "sr_YU CP852" # CP852 or CP866 or CP855 ??
- # ISO-8859-3 languages
- echo "mt CP850"
- echo "mt_MT CP850"
- # ISO-8859-5 languages
- echo "be CP866"
- echo "be_BE CP866"
- echo "bg CP866" # not CP855 ??
- echo "bg_BG CP866" # not CP855 ??
- echo "mk CP866" # not CP855 ??
- echo "mk_MK CP866" # not CP855 ??
- echo "ru CP866"
- echo "ru_RU CP866"
- echo "uk CP1125"
- echo "uk_UA CP1125"
- # ISO-8859-6 languages
- echo "ar CP864"
- echo "ar_AE CP864"
- echo "ar_DZ CP864"
- echo "ar_EG CP864"
- echo "ar_IQ CP864"
- echo "ar_IR CP864"
- echo "ar_JO CP864"
- echo "ar_KW CP864"
- echo "ar_MA CP864"
- echo "ar_OM CP864"
- echo "ar_QA CP864"
- echo "ar_SA CP864"
- echo "ar_SY CP864"
- # ISO-8859-7 languages
- echo "el CP869"
- echo "el_GR CP869"
- # ISO-8859-8 languages
- echo "he CP862"
- echo "he_IL CP862"
- # ISO-8859-9 languages
- echo "tr CP857"
- echo "tr_TR CP857"
- # Japanese
- echo "ja CP932"
- echo "ja_JP CP932"
- # Chinese
- echo "zh_CN GBK"
- echo "zh_TW CP950" # not CP938 ??
- # Korean
- echo "kr CP949" # not CP934 ??
- echo "kr_KR CP949" # not CP934 ??
- # Thai
- echo "th CP874"
- echo "th_TH CP874"
- # Other
- echo "eo CP850"
- echo "eo_EO CP850"
- ;;
-esac
diff --git a/lib/localcharset.c b/lib/localcharset.c
index 50a5d68..94e0f8a 100644
--- a/lib/localcharset.c
+++ b/lib/localcharset.c
@@ -22,7 +22,6 @@
/* Specification. */
#include "localcharset.h"
-#include <fcntl.h>
#include <stddef.h>
#include <stdio.h>
#include <string.h>
@@ -45,11 +44,10 @@
#endif
#if !defined WINDOWS_NATIVE
-# include <unistd.h>
# if HAVE_LANGINFO_CODESET
# include <langinfo.h>
# else
-# if 0 /* see comment below */
+# if 0 /* see comment regarding use of setlocale(), below */
# include <locale.h>
# endif
# endif
@@ -71,323 +69,569 @@
# include <xlocale.h>
#endif
-#if ENABLE_RELOCATABLE
-# include "relocatable.h"
-#else
-# define relocate(pathname) (pathname)
-# define relocate2(pathname,allocatedp) (*(allocatedp) = NULL, (pathname))
-#endif
-
-/* Get LIBDIR. */
-#ifndef LIBDIR
-# include "configmake.h"
-#endif
-
-/* Define O_NOFOLLOW to 0 on platforms where it does not exist. */
-#ifndef O_NOFOLLOW
-# define O_NOFOLLOW 0
-#endif
-
-#if defined _WIN32 || defined __CYGWIN__ || defined __EMX__ || defined __DJGPP__
- /* Native Windows, Cygwin, OS/2, DOS */
-# define ISSLASH(C) ((C) == '/' || (C) == '\\')
-#endif
-#ifndef DIRECTORY_SEPARATOR
-# define DIRECTORY_SEPARATOR '/'
-#endif
+#if HAVE_LANGINFO_CODESET || defined WINDOWS_NATIVE || defined OS2
-#ifndef ISSLASH
-# define ISSLASH(C) ((C) == DIRECTORY_SEPARATOR)
-#endif
+/* On these platforms, we use a mapping from non-canonical encoding name
+ to GNU canonical encoding name. */
-#if HAVE_DECL_GETC_UNLOCKED
-# undef getc
-# define getc getc_unlocked
-#endif
+/* With glibc-2.1 or newer, we don't need any canonicalization,
+ because glibc has iconv and both glibc and libiconv support all
+ GNU canonical names directly. */
+# if !((defined __GNU_LIBRARY__ && __GLIBC__ >= 2) || defined __UCLIBC__)
-/* The following static variable is declared 'volatile' to avoid a
- possible multithread problem in the function get_charset_aliases. If we
- are running in a threaded environment, and if two threads initialize
- 'charset_aliases' simultaneously, both will produce the same value,
- and everything will be ok if the two assignments to 'charset_aliases'
- are atomic. But I don't know what will happen if the two assignments mix. */
-#if __STDC__ != 1
-# define volatile /* empty */
-#endif
-/* Pointer to the contents of the charset.alias file, if it has already been
- read, else NULL. Its format is:
- ALIAS_1 '\0' CANONICAL_1 '\0' ... ALIAS_n '\0' CANONICAL_n '\0' '\0' */
-static const char * volatile charset_aliases;
-
-/* Return a pointer to the contents of the charset.alias file. */
-static const char *
-get_charset_aliases (void)
+struct table_entry
{
- const char *cp;
-
- cp = charset_aliases;
- if (cp == NULL)
- {
-#if !(defined DARWIN7 || defined VMS || defined WINDOWS_NATIVE || defined __CYGWIN__ || defined OS2)
- char *malloc_dir = NULL;
- const char *dir;
- const char *base = "charset.alias";
- char *file_name;
-
- /* Make it possible to override the charset.alias location. This is
- necessary for running the testsuite before "make install". */
- dir = getenv ("CHARSETALIASDIR");
- if (dir == NULL || dir[0] == '\0')
- dir = relocate2 (LIBDIR, &malloc_dir);
-
- /* Concatenate dir and base into freshly allocated file_name. */
- {
- size_t dir_len = strlen (dir);
- size_t base_len = strlen (base);
- int add_slash = (dir_len > 0 && !ISSLASH (dir[dir_len - 1]));
- file_name = (char *) malloc (dir_len + add_slash + base_len + 1);
- if (file_name != NULL)
- {
- memcpy (file_name, dir, dir_len);
- if (add_slash)
- file_name[dir_len] = DIRECTORY_SEPARATOR;
- memcpy (file_name + dir_len + add_slash, base, base_len + 1);
- }
- }
-
- free (malloc_dir);
-
- if (file_name == NULL)
- /* Out of memory. Treat the file as empty. */
- cp = "";
- else
- {
- int fd;
-
- /* Open the file. Reject symbolic links on platforms that support
- O_NOFOLLOW. This is a security feature. Without it, an attacker
- could retrieve parts of the contents (namely, the tail of the
- first line that starts with "* ") of an arbitrary file by placing
- a symbolic link to that file under the name "charset.alias" in
- some writable directory and defining the environment variable
- CHARSETALIASDIR to point to that directory. */
- fd = open (file_name,
- O_RDONLY | (HAVE_WORKING_O_NOFOLLOW ? O_NOFOLLOW : 0));
- if (fd < 0)
- /* File not found. Treat it as empty. */
- cp = "";
- else
- {
- FILE *fp;
-
- fp = fdopen (fd, "r");
- if (fp == NULL)
- {
- /* Out of memory. Treat the file as empty. */
- close (fd);
- cp = "";
- }
- else
- {
- /* Parse the file's contents. */
- char *res_ptr = NULL;
- size_t res_size = 0;
-
- for (;;)
- {
- int c;
- char buf1[50+1];
- char buf2[50+1];
- size_t l1, l2;
- char *old_res_ptr;
-
- c = getc (fp);
- if (c == EOF)
- break;
- if (c == '\n' || c == ' ' || c == '\t')
- continue;
- if (c == '#')
- {
- /* Skip comment, to end of line. */
- do
- c = getc (fp);
- while (!(c == EOF || c == '\n'));
- if (c == EOF)
- break;
- continue;
- }
- ungetc (c, fp);
- if (fscanf (fp, "%50s %50s", buf1, buf2) < 2)
- break;
- l1 = strlen (buf1);
- l2 = strlen (buf2);
- old_res_ptr = res_ptr;
- if (res_size == 0)
- {
- res_size = l1 + 1 + l2 + 1;
- res_ptr = (char *) malloc (res_size + 1);
- }
- else
- {
- res_size += l1 + 1 + l2 + 1;
- res_ptr = (char *) realloc (res_ptr, res_size + 1);
- }
- if (res_ptr == NULL)
- {
- /* Out of memory. */
- res_size = 0;
- free (old_res_ptr);
- break;
- }
- strcpy (res_ptr + res_size - (l2 + 1) - (l1 + 1), buf1);
- strcpy (res_ptr + res_size - (l2 + 1), buf2);
- }
- fclose (fp);
- if (res_size == 0)
- cp = "";
- else
- {
- *(res_ptr + res_size) = '\0';
- cp = res_ptr;
- }
- }
- }
+ const char alias[11+1];
+ const char canonical[11+1];
+};
+
+/* Table of platform-dependent mappings, sorted in ascending order. */
+static const struct table_entry alias_table[] =
+ {
+# if defined __FreeBSD__ /* FreeBSD */
+ { "C", "ASCII" },
+ { "US-ASCII", "ASCII" }
+# define alias_table_defined
+# endif
+# if defined __NetBSD__ /* NetBSD */
+ { "646", "ASCII" },
+ { "BIG5", "BIG5" },
+ { "ISO8859-1", "ISO-8859-1" },
+ { "ISO8859-13", "ISO-8859-13" },
+ { "ISO8859-15", "ISO-8859-15" },
+ { "ISO8859-2", "ISO-8859-2" },
+ { "ISO8859-4", "ISO-8859-4" },
+ { "ISO8859-5", "ISO-8859-5" },
+ { "ISO8859-7", "ISO-8859-7" },
+ { "SJIS", "SHIFT_JIS" },
+ { "eucCN", "GB2312" },
+ { "eucJP", "EUC-JP" },
+ { "eucKR", "EUC-KR" },
+ { "eucTW", "EUC-TW" }
+# define alias_table_defined
+# endif
+# if defined __OpenBSD__ /* OpenBSD */
+ { "646", "ASCII" },
+ { "ISO8859-1", "ISO-8859-1" },
+ { "ISO8859-13", "ISO-8859-13" },
+ { "ISO8859-15", "ISO-8859-15" },
+ { "ISO8859-2", "ISO-8859-2" },
+ { "ISO8859-4", "ISO-8859-4" },
+ { "ISO8859-5", "ISO-8859-5" },
+ { "ISO8859-7", "ISO-8859-7" }
+# define alias_table_defined
+# endif
+# if defined __APPLE__ && defined __MACH__ /* Mac OS X */
+ /* Darwin 7.5 has nl_langinfo(CODESET), but sometimes its value is
+ useless:
+ - It returns the empty string when LANG is set to a locale of the
+ form ll_CC, although ll_CC/LC_CTYPE is a symlink to an UTF-8
+ LC_CTYPE file.
+ - The environment variables LANG, LC_CTYPE, LC_ALL are not set by
+ the system; nl_langinfo(CODESET) returns "US-ASCII" in this case.
+ - The documentation says:
+ "... all code that calls BSD system routines should ensure
+ that the const *char parameters of these routines are in UTF-8
+ encoding. All BSD system functions expect their string
+ parameters to be in UTF-8 encoding and nothing else."
+ It also says
+ "An additional caveat is that string parameters for files,
+ paths, and other file-system entities must be in canonical
+ UTF-8. In a canonical UTF-8 Unicode string, all decomposable
+ characters are decomposed ..."
+ but this is not true: You can pass non-decomposed UTF-8 strings
+ to file system functions, and it is the OS which will convert
+ them to decomposed UTF-8 before accessing the file system.
+ - The Apple Terminal application displays UTF-8 by default.
+ - However, other applications are free to use different encodings:
+ - xterm uses ISO-8859-1 by default.
+ - TextEdit uses MacRoman by default.
+ We prefer UTF-8 over decomposed UTF-8-MAC because one should
+ minimize the use of decomposed Unicode. Unfortunately, through the
+ Darwin file system, decomposed UTF-8 strings are leaked into user
+ space nevertheless.
+ Then there are also the locales with encodings other than US-ASCII
+ and UTF-8. These locales can be occasionally useful to users (e.g.
+ when grepping through ISO-8859-1 encoded text files), when all their
+ file names are in US-ASCII.
+ */
+ { "ARMSCII-8", "ARMSCII-8" },
+ { "Big5", "BIG5" },
+ { "Big5HKSCS", "BIG5-HKSCS" },
+ { "CP1131", "CP1131" },
+ { "CP1251", "CP1251" },
+ { "CP866", "CP866" },
+ { "CP949", "CP949" },
+ { "GB18030", "GB18030" },
+ { "GB2312", "GB2312" },
+ { "GBK", "GBK" },
+ /*{ "ISCII-DEV", "?" },*/
+ { "ISO8859-1", "ISO-8859-1" },
+ { "ISO8859-13", "ISO-8859-13" },
+ { "ISO8859-15", "ISO-8859-15" },
+ { "ISO8859-2", "ISO-8859-2" },
+ { "ISO8859-4", "ISO-8859-4" },
+ { "ISO8859-5", "ISO-8859-5" },
+ { "ISO8859-7", "ISO-8859-7" },
+ { "ISO8859-9", "ISO-8859-9" },
+ { "KOI8-R", "KOI8-R" },
+ { "KOI8-U", "KOI8-U" },
+ { "PT154", "PT154" },
+ { "SJIS", "SHIFT_JIS" },
+ { "eucCN", "GB2312" },
+ { "eucJP", "EUC-JP" },
+ { "eucKR", "EUC-KR" }
+# define alias_table_defined
+# endif
+# if defined _AIX /* AIX */
+ { "GBK", "GBK" },
+ { "IBM-1046", "CP1046" },
+ { "IBM-1124", "CP1124" },
+ { "IBM-1129", "CP1129" },
+ { "IBM-1252", "CP1252" },
+ { "IBM-850", "CP850" },
+ { "IBM-856", "CP856" },
+ { "IBM-921", "ISO-8859-13" },
+ { "IBM-922", "CP922" },
+ { "IBM-932", "CP932" },
+ { "IBM-943", "CP943" },
+ { "IBM-eucCN", "GB2312" },
+ { "IBM-eucJP", "EUC-JP" },
+ { "IBM-eucKR", "EUC-KR" },
+ { "IBM-eucTW", "EUC-TW" },
+ { "ISO8859-1", "ISO-8859-1" },
+ { "ISO8859-15", "ISO-8859-15" },
+ { "ISO8859-2", "ISO-8859-2" },
+ { "ISO8859-5", "ISO-8859-5" },
+ { "ISO8859-6", "ISO-8859-6" },
+ { "ISO8859-7", "ISO-8859-7" },
+ { "ISO8859-8", "ISO-8859-8" },
+ { "ISO8859-9", "ISO-8859-9" },
+ { "TIS-620", "TIS-620" },
+ { "UTF-8", "UTF-8" },
+ { "big5", "BIG5" }
+# define alias_table_defined
+# endif
+# if defined __hpux /* HP-UX */
+ { "SJIS", "SHIFT_JIS" },
+ { "arabic8", "HP-ARABIC8" },
+ { "big5", "BIG5" },
+ { "eucJP", "EUC-JP" },
+ { "eucKR", "EUC-KR" },
+ { "eucTW", "EUC-TW" },
+ { "greek8", "HP-GREEK8" },
+ { "hebrew8", "HP-HEBREW8" },
+ { "hp15CN", "GB2312" },
+ { "iso88591", "ISO-8859-1" },
+ { "iso885915", "ISO-8859-15" },
+ { "iso88592", "ISO-8859-2" },
+ { "iso88595", "ISO-8859-5" },
+ { "iso88596", "ISO-8859-6" },
+ { "iso88597", "ISO-8859-7" },
+ { "iso88598", "ISO-8859-8" },
+ { "iso88599", "ISO-8859-9" },
+ { "kana8", "HP-KANA8" },
+ { "roman8", "HP-ROMAN8" },
+ { "tis620", "TIS-620" },
+ { "turkish8", "HP-TURKISH8" },
+ { "utf8", "UTF-8" }
+# define alias_table_defined
+# endif
+# if defined __sgi /* IRIX */
+ { "ISO8859-1", "ISO-8859-1" },
+ { "ISO8859-2", "ISO-8859-2" },
+ { "ISO8859-5", "ISO-8859-5" },
+ { "ISO8859-7", "ISO-8859-7" },
+ { "ISO8859-9", "ISO-8859-9" },
+ { "eucCN", "GB2312" },
+ { "eucJP", "EUC-JP" },
+ { "eucKR", "EUC-KR" },
+ { "eucTW", "EUC-TW" }
+# define alias_table_defined
+# endif
+# if defined __osf__ /* OSF/1 */
+ { "GBK", "GBK" },
+ { "ISO8859-1", "ISO-8859-1" },
+ { "ISO8859-15", "ISO-8859-15" },
+ { "ISO8859-2", "ISO-8859-2" },
+ { "ISO8859-4", "ISO-8859-4" },
+ { "ISO8859-5", "ISO-8859-5" },
+ { "ISO8859-7", "ISO-8859-7" },
+ { "ISO8859-8", "ISO-8859-8" },
+ { "ISO8859-9", "ISO-8859-9" },
+ { "KSC5601", "CP949" },
+ { "SJIS", "SHIFT_JIS" },
+ { "TACTIS", "TIS-620" },
+ { "UTF-8", "UTF-8" },
+ { "big5", "BIG5" },
+ { "cp850", "CP850" },
+ { "dechanyu", "DEC-HANYU" },
+ { "dechanzi", "GB2312" },
+ { "deckanji", "DEC-KANJI" },
+ { "deckorean", "EUC-KR" },
+ { "eucJP", "EUC-JP" },
+ { "eucKR", "EUC-KR" },
+ { "eucTW", "EUC-TW" },
+ { "sdeckanji", "EUC-JP" }
+# define alias_table_defined
+# endif
+# if defined __sun /* Solaris */
+ { "5601", "EUC-KR" },
+ { "646", "ASCII" },
+ { "BIG5", "BIG5" },
+ { "Big5-HKSCS", "BIG5-HKSCS" },
+ { "GB18030", "GB18030" },
+ { "GBK", "GBK" },
+ { "ISO8859-1", "ISO-8859-1" },
+ { "ISO8859-15", "ISO-8859-15" },
+ { "ISO8859-2", "ISO-8859-2" },
+ { "ISO8859-3", "ISO-8859-3" },
+ { "ISO8859-4", "ISO-8859-4" },
+ { "ISO8859-5", "ISO-8859-5" },
+ { "ISO8859-6", "ISO-8859-6" },
+ { "ISO8859-7", "ISO-8859-7" },
+ { "ISO8859-8", "ISO-8859-8" },
+ { "ISO8859-9", "ISO-8859-9" },
+ { "PCK", "SHIFT_JIS" },
+ { "TIS620.2533", "TIS-620" },
+ { "UTF-8", "UTF-8" },
+ { "ansi-1251", "CP1251" },
+ { "cns11643", "EUC-TW" },
+ { "eucJP", "EUC-JP" },
+ { "gb2312", "GB2312" },
+ { "koi8-r", "KOI8-R" }
+# define alias_table_defined
+# endif
+# if defined WINDOWS_NATIVE || defined __CYGWIN__ /* Windows */
+ { "CP1361", "JOHAB" },
+ { "CP20127", "ASCII" },
+ { "CP20866", "KOI8-R" },
+ { "CP20936", "GB2312" },
+ { "CP21866", "KOI8-RU" },
+ { "CP28591", "ISO-8859-1" },
+ { "CP28592", "ISO-8859-2" },
+ { "CP28593", "ISO-8859-3" },
+ { "CP28594", "ISO-8859-4" },
+ { "CP28595", "ISO-8859-5" },
+ { "CP28596", "ISO-8859-6" },
+ { "CP28597", "ISO-8859-7" },
+ { "CP28598", "ISO-8859-8" },
+ { "CP28599", "ISO-8859-9" },
+ { "CP28605", "ISO-8859-15" },
+ { "CP38598", "ISO-8859-8" },
+ { "CP51932", "EUC-JP" },
+ { "CP51936", "GB2312" },
+ { "CP51949", "EUC-KR" },
+ { "CP51950", "EUC-TW" },
+ { "CP54936", "GB18030" },
+ { "CP65001", "UTF-8" },
+ { "CP936", "GBK" }
+# define alias_table_defined
+# endif
+# if defined OS2 /* OS/2 */
+ /* The list of encodings is taken from "List of OS/2 Codepages"
+ by Alex Taylor:
+ <http://altsan.org/os2/toolkits/uls/index.html#codepages>.
+ See also "IBM Globalization - Code page identifiers":
+ <https://www-01.ibm.com/software/globalization/cp/cp_cpgid.html>. */
+ { "CP1089", "ISO-8859-6" },
+ { "CP1208", "UTF-8" },
+ { "CP1381", "GB2312" },
+ { "CP1386", "GBK" },
+ { "CP3372", "EUC-JP" },
+ { "CP813", "ISO-8859-7" },
+ { "CP819", "ISO-8859-1" },
+ { "CP878", "KOI8-R" },
+ { "CP912", "ISO-8859-2" },
+ { "CP913", "ISO-8859-3" },
+ { "CP914", "ISO-8859-4" },
+ { "CP915", "ISO-8859-5" },
+ { "CP916", "ISO-8859-8" },
+ { "CP920", "ISO-8859-9" },
+ { "CP921", "ISO-8859-13" },
+ { "CP923", "ISO-8859-15" },
+ { "CP954", "EUC-JP" },
+ { "CP964", "EUC-TW" },
+ { "CP970", "EUC-KR" }
+# define alias_table_defined
+# endif
+# if defined VMS /* OpenVMS */
+ /* The list of encodings is taken from the OpenVMS 7.3-1 documentation
+ "Compaq C Run-Time Library Reference Manual for OpenVMS systems"
+ section 10.7 "Handling Different Character Sets". */
+ { "DECHANYU", "DEC-HANYU" },
+ { "DECHANZI", "GB2312" },
+ { "DECKANJI", "DEC-KANJI" },
+ { "DECKOREAN", "EUC-KR" },
+ { "ISO8859-1", "ISO-8859-1" },
+ { "ISO8859-2", "ISO-8859-2" },
+ { "ISO8859-5", "ISO-8859-5" },
+ { "ISO8859-7", "ISO-8859-7" },
+ { "ISO8859-8", "ISO-8859-8" },
+ { "ISO8859-9", "ISO-8859-9" },
+ { "SDECKANJI", "EUC-JP" },
+ { "SJIS", "SHIFT_JIS" },
+ { "eucJP", "EUC-JP" },
+ { "eucTW", "EUC-TW" }
+# define alias_table_defined
+# endif
+# ifndef alias_table_defined
+ /* Just a dummy entry, to avoid a C syntax error. */
+ { "", "" }
+# endif
+ };
- free (file_name);
- }
+# endif
#else
-# if defined DARWIN7
- /* To avoid the trouble of installing a file that is shared by many
- GNU packages -- many packaging systems have problems with this --,
- simply inline the aliases here.
- For speed, map the most frequent case first. */
- cp = "UTF-8" "\0" "UTF-8" "\0"
- "ISO8859-1" "\0" "ISO-8859-1" "\0"
- "ISO8859-2" "\0" "ISO-8859-2" "\0"
- "ISO8859-4" "\0" "ISO-8859-4" "\0"
- "ISO8859-5" "\0" "ISO-8859-5" "\0"
- "ISO8859-7" "\0" "ISO-8859-7" "\0"
- "ISO8859-9" "\0" "ISO-8859-9" "\0"
- "ISO8859-13" "\0" "ISO-8859-13" "\0"
- "ISO8859-15" "\0" "ISO-8859-15" "\0"
- "KOI8-R" "\0" "KOI8-R" "\0"
- "KOI8-U" "\0" "KOI8-U" "\0"
- "CP866" "\0" "CP866" "\0"
- "CP949" "\0" "CP949" "\0"
- "CP1131" "\0" "CP1131" "\0"
- "CP1251" "\0" "CP1251" "\0"
- "eucCN" "\0" "GB2312" "\0"
- "GB2312" "\0" "GB2312" "\0"
- "eucJP" "\0" "EUC-JP" "\0"
- "eucKR" "\0" "EUC-KR" "\0"
- "Big5" "\0" "BIG5" "\0"
- "Big5HKSCS" "\0" "BIG5-HKSCS" "\0"
- "GBK" "\0" "GBK" "\0"
- "GB18030" "\0" "GB18030" "\0"
- "SJIS" "\0" "SHIFT_JIS" "\0"
- "ARMSCII-8" "\0" "ARMSCII-8" "\0"
- "PT154" "\0" "PT154" "\0"
- /*"ISCII-DEV" "\0" "?" "\0"*/
- "*" "\0" "UTF-8" "\0";
-# endif
+/* On these platforms, we use a mapping from locale name to GNU canonical
+ encoding name. */
-# if defined VMS
- /* To avoid the troubles of an extra file charset.alias_vms in the
- sources of many GNU packages, simply inline the aliases here. */
- /* The list of encodings is taken from the OpenVMS 7.3-1 documentation
- "Compaq C Run-Time Library Reference Manual for OpenVMS systems"
- section 10.7 "Handling Different Character Sets". */
- cp = "ISO8859-1" "\0" "ISO-8859-1" "\0"
- "ISO8859-2" "\0" "ISO-8859-2" "\0"
- "ISO8859-5" "\0" "ISO-8859-5" "\0"
- "ISO8859-7" "\0" "ISO-8859-7" "\0"
- "ISO8859-8" "\0" "ISO-8859-8" "\0"
- "ISO8859-9" "\0" "ISO-8859-9" "\0"
- /* Japanese */
- "eucJP" "\0" "EUC-JP" "\0"
- "SJIS" "\0" "SHIFT_JIS" "\0"
- "DECKANJI" "\0" "DEC-KANJI" "\0"
- "SDECKANJI" "\0" "EUC-JP" "\0"
- /* Chinese */
- "eucTW" "\0" "EUC-TW" "\0"
- "DECHANYU" "\0" "DEC-HANYU" "\0"
- "DECHANZI" "\0" "GB2312" "\0"
- /* Korean */
- "DECKOREAN" "\0" "EUC-KR" "\0";
+struct table_entry
+{
+ const char locale[17+1];
+ const char canonical[11+1];
+};
+
+/* Table of platform-dependent mappings, sorted in ascending order. */
+static const struct table_entry locale_table[] =
+ {
+# if defined __FreeBSD__ /* FreeBSD 4.2 */
+ { "cs_CZ.ISO_8859-2", "ISO-8859-2" },
+ { "da_DK.DIS_8859-15", "ISO-8859-15" },
+ { "da_DK.ISO_8859-1", "ISO-8859-1" },
+ { "de_AT.DIS_8859-15", "ISO-8859-15" },
+ { "de_AT.ISO_8859-1", "ISO-8859-1" },
+ { "de_CH.DIS_8859-15", "ISO-8859-15" },
+ { "de_CH.ISO_8859-1", "ISO-8859-1" },
+ { "de_DE.DIS_8859-15", "ISO-8859-15" },
+ { "de_DE.ISO_8859-1", "ISO-8859-1" },
+ { "en_AU.DIS_8859-15", "ISO-8859-15" },
+ { "en_AU.ISO_8859-1", "ISO-8859-1" },
+ { "en_CA.DIS_8859-15", "ISO-8859-15" },
+ { "en_CA.ISO_8859-1", "ISO-8859-1" },
+ { "en_GB.DIS_8859-15", "ISO-8859-15" },
+ { "en_GB.ISO_8859-1", "ISO-8859-1" },
+ { "en_US.DIS_8859-15", "ISO-8859-15" },
+ { "en_US.ISO_8859-1", "ISO-8859-1" },
+ { "es_ES.DIS_8859-15", "ISO-8859-15" },
+ { "es_ES.ISO_8859-1", "ISO-8859-1" },
+ { "fi_FI.DIS_8859-15", "ISO-8859-15" },
+ { "fi_FI.ISO_8859-1", "ISO-8859-1" },
+ { "fr_BE.DIS_8859-15", "ISO-8859-15" },
+ { "fr_BE.ISO_8859-1", "ISO-8859-1" },
+ { "fr_CA.DIS_8859-15", "ISO-8859-15" },
+ { "fr_CA.ISO_8859-1", "ISO-8859-1" },
+ { "fr_CH.DIS_8859-15", "ISO-8859-15" },
+ { "fr_CH.ISO_8859-1", "ISO-8859-1" },
+ { "fr_FR.DIS_8859-15", "ISO-8859-15" },
+ { "fr_FR.ISO_8859-1", "ISO-8859-1" },
+ { "hr_HR.ISO_8859-2", "ISO-8859-2" },
+ { "hu_HU.ISO_8859-2", "ISO-8859-2" },
+ { "is_IS.DIS_8859-15", "ISO-8859-15" },
+ { "is_IS.ISO_8859-1", "ISO-8859-1" },
+ { "it_CH.DIS_8859-15", "ISO-8859-15" },
+ { "it_CH.ISO_8859-1", "ISO-8859-1" },
+ { "it_IT.DIS_8859-15", "ISO-8859-15" },
+ { "it_IT.ISO_8859-1", "ISO-8859-1" },
+ { "ja_JP.EUC", "EUC-JP" },
+ { "ja_JP.SJIS", "SHIFT_JIS" },
+ { "ja_JP.Shift_JIS", "SHIFT_JIS" },
+ { "ko_KR.EUC", "EUC-KR" },
+ { "la_LN.ASCII", "ASCII" },
+ { "la_LN.DIS_8859-15", "ISO-8859-15" },
+ { "la_LN.ISO_8859-1", "ISO-8859-1" },
+ { "la_LN.ISO_8859-2", "ISO-8859-2" },
+ { "la_LN.ISO_8859-4", "ISO-8859-4" },
+ { "lt_LN.ASCII", "ASCII" },
+ { "lt_LN.DIS_8859-15", "ISO-8859-15" },
+ { "lt_LN.ISO_8859-1", "ISO-8859-1" },
+ { "lt_LN.ISO_8859-2", "ISO-8859-2" },
+ { "lt_LT.ISO_8859-4", "ISO-8859-4" },
+ { "nl_BE.DIS_8859-15", "ISO-8859-15" },
+ { "nl_BE.ISO_8859-1", "ISO-8859-1" },
+ { "nl_NL.DIS_8859-15", "ISO-8859-15" },
+ { "nl_NL.ISO_8859-1", "ISO-8859-1" },
+ { "no_NO.DIS_8859-15", "ISO-8859-15" },
+ { "no_NO.ISO_8859-1", "ISO-8859-1" },
+ { "pl_PL.ISO_8859-2", "ISO-8859-2" },
+ { "pt_PT.DIS_8859-15", "ISO-8859-15" },
+ { "pt_PT.ISO_8859-1", "ISO-8859-1" },
+ { "ru_RU.CP866", "CP866" },
+ { "ru_RU.ISO_8859-5", "ISO-8859-5" },
+ { "ru_RU.KOI8-R", "KOI8-R" },
+ { "ru_SU.CP866", "CP866" },
+ { "ru_SU.ISO_8859-5", "ISO-8859-5" },
+ { "ru_SU.KOI8-R", "KOI8-R" },
+ { "sl_SI.ISO_8859-2", "ISO-8859-2" },
+ { "sv_SE.DIS_8859-15", "ISO-8859-15" },
+ { "sv_SE.ISO_8859-1", "ISO-8859-1" },
+ { "uk_UA.KOI8-U", "KOI8-U" },
+ { "zh_CN.EUC", "GB2312" },
+ { "zh_TW.BIG5", "BIG5" },
+ { "zh_TW.Big5", "BIG5" }
+# define locale_table_defined
# endif
-
-# if defined WINDOWS_NATIVE || defined __CYGWIN__
- /* To avoid the troubles of installing a separate file in the same
- directory as the DLL and of retrieving the DLL's directory at
- runtime, simply inline the aliases here. */
-
- cp = "CP936" "\0" "GBK" "\0"
- "CP1361" "\0" "JOHAB" "\0"
- "CP20127" "\0" "ASCII" "\0"
- "CP20866" "\0" "KOI8-R" "\0"
- "CP20936" "\0" "GB2312" "\0"
- "CP21866" "\0" "KOI8-RU" "\0"
- "CP28591" "\0" "ISO-8859-1" "\0"
- "CP28592" "\0" "ISO-8859-2" "\0"
- "CP28593" "\0" "ISO-8859-3" "\0"
- "CP28594" "\0" "ISO-8859-4" "\0"
- "CP28595" "\0" "ISO-8859-5" "\0"
- "CP28596" "\0" "ISO-8859-6" "\0"
- "CP28597" "\0" "ISO-8859-7" "\0"
- "CP28598" "\0" "ISO-8859-8" "\0"
- "CP28599" "\0" "ISO-8859-9" "\0"
- "CP28605" "\0" "ISO-8859-15" "\0"
- "CP38598" "\0" "ISO-8859-8" "\0"
- "CP51932" "\0" "EUC-JP" "\0"
- "CP51936" "\0" "GB2312" "\0"
- "CP51949" "\0" "EUC-KR" "\0"
- "CP51950" "\0" "EUC-TW" "\0"
- "CP54936" "\0" "GB18030" "\0"
- "CP65001" "\0" "UTF-8" "\0";
+# if defined __DJGPP__ /* DOS / DJGPP 2.03 */
+ /* The encodings given here may not all be correct.
+ If you find that the encoding given for your language and
+ country is not the one your DOS machine actually uses, just
+ correct it in this file, and send a mail to
+ Juan Manuel Guerrero <juan.guerr...@gmx.de>
+ and <bug-gnulib@gnu.org>. */
+ { "C", "ASCII" },
+ { "ar", "CP864" },
+ { "ar_AE", "CP864" },
+ { "ar_DZ", "CP864" },
+ { "ar_EG", "CP864" },
+ { "ar_IQ", "CP864" },
+ { "ar_IR", "CP864" },
+ { "ar_JO", "CP864" },
+ { "ar_KW", "CP864" },
+ { "ar_MA", "CP864" },
+ { "ar_OM", "CP864" },
+ { "ar_QA", "CP864" },
+ { "ar_SA", "CP864" },
+ { "ar_SY", "CP864" },
+ { "be", "CP866" },
+ { "be_BE", "CP866" },
+ { "bg", "CP866" }, /* not CP855 ?? */
+ { "bg_BG", "CP866" }, /* not CP855 ?? */
+ { "ca", "CP850" },
+ { "ca_ES", "CP850" },
+ { "cs", "CP852" },
+ { "cs_CZ", "CP852" },
+ { "da", "CP865" }, /* not CP850 ?? */
+ { "da_DK", "CP865" }, /* not CP850 ?? */
+ { "de", "CP850" },
+ { "de_AT", "CP850" },
+ { "de_CH", "CP850" },
+ { "de_DE", "CP850" },
+ { "el", "CP869" },
+ { "el_GR", "CP869" },
+ { "en", "CP850" },
+ { "en_AU", "CP850" }, /* not CP437 ?? */
+ { "en_CA", "CP850" },
+ { "en_GB", "CP850" },
+ { "en_NZ", "CP437" },
+ { "en_US", "CP437" },
+ { "en_ZA", "CP850" }, /* not CP437 ?? */
+ { "eo", "CP850" },
+ { "eo_EO", "CP850" },
+ { "es", "CP850" },
+ { "es_AR", "CP850" },
+ { "es_BO", "CP850" },
+ { "es_CL", "CP850" },
+ { "es_CO", "CP850" },
+ { "es_CR", "CP850" },
+ { "es_CU", "CP850" },
+ { "es_DO", "CP850" },
+ { "es_EC", "CP850" },
+ { "es_ES", "CP850" },
+ { "es_GT", "CP850" },
+ { "es_HN", "CP850" },
+ { "es_MX", "CP850" },
+ { "es_NI", "CP850" },
+ { "es_PA", "CP850" },
+ { "es_PE", "CP850" },
+ { "es_PY", "CP850" },
+ { "es_SV", "CP850" },
+ { "es_UY", "CP850" },
+ { "es_VE", "CP850" },
+ { "et", "CP850" },
+ { "et_EE", "CP850" },
+ { "eu", "CP850" },
+ { "eu_ES", "CP850" },
+ { "fi", "CP850" },
+ { "fi_FI", "CP850" },
+ { "fr", "CP850" },
+ { "fr_BE", "CP850" },
+ { "fr_CA", "CP850" },
+ { "fr_CH", "CP850" },
+ { "fr_FR", "CP850" },
+ { "ga", "CP850" },
+ { "ga_IE", "CP850" },
+ { "gd", "CP850" },
+ { "gd_GB", "CP850" },
+ { "gl", "CP850" },
+ { "gl_ES", "CP850" },
+ { "he", "CP862" },
+ { "he_IL", "CP862" },
+ { "hr", "CP852" },
+ { "hr_HR", "CP852" },
+ { "hu", "CP852" },
+ { "hu_HU", "CP852" },
+ { "id", "CP850" }, /* not CP437 ?? */
+ { "id_ID", "CP850" }, /* not CP437 ?? */
+ { "is", "CP861" }, /* not CP850 ?? */
+ { "is_IS", "CP861" }, /* not CP850 ?? */
+ { "it", "CP850" },
+ { "it_CH", "CP850" },
+ { "it_IT", "CP850" },
+ { "ja", "CP932" },
+ { "ja_JP", "CP932" },
+ { "kr", "CP949" }, /* not CP934 ?? */
+ { "kr_KR", "CP949" }, /* not CP934 ?? */
+ { "lt", "CP775" },
+ { "lt_LT", "CP775" },
+ { "lv", "CP775" },
+ { "lv_LV", "CP775" },
+ { "mk", "CP866" }, /* not CP855 ?? */
+ { "mk_MK", "CP866" }, /* not CP855 ?? */
+ { "mt", "CP850" },
+ { "mt_MT", "CP850" },
+ { "nb", "CP865" }, /* not CP850 ?? */
+ { "nb_NO", "CP865" }, /* not CP850 ?? */
+ { "nl", "CP850" },
+ { "nl_BE", "CP850" },
+ { "nl_NL", "CP850" },
+ { "nn", "CP865" }, /* not CP850 ?? */
+ { "nn_NO", "CP865" }, /* not CP850 ?? */
+ { "no", "CP865" }, /* not CP850 ?? */
+ { "no_NO", "CP865" }, /* not CP850 ?? */
+ { "pl", "CP852" },
+ { "pl_PL", "CP852" },
+ { "pt", "CP850" },
+ { "pt_BR", "CP850" },
+ { "pt_PT", "CP850" },
+ { "ro", "CP852" },
+ { "ro_RO", "CP852" },
+ { "ru", "CP866" },
+ { "ru_RU", "CP866" },
+ { "sk", "CP852" },
+ { "sk_SK", "CP852" },
+ { "sl", "CP852" },
+ { "sl_SI", "CP852" },
+ { "sq", "CP852" },
+ { "sq_AL", "CP852" },
+ { "sr", "CP852" }, /* CP852 or CP866 or CP855 ?? */
+ { "sr_CS", "CP852" }, /* CP852 or CP866 or CP855 ?? */
+ { "sr_YU", "CP852" }, /* CP852 or CP866 or CP855 ?? */
+ { "sv", "CP850" },
+ { "sv_SE", "CP850" },
+ { "th", "CP874" },
+ { "th_TH", "CP874" },
+ { "tr", "CP857" },
+ { "tr_TR", "CP857" },
+ { "uk", "CP1125" },
+ { "uk_UA", "CP1125" },
+ { "zh_CN", "GBK" },
+ { "zh_TW", "CP950" } /* not CP938 ?? */
+# define locale_table_defined
# endif
-# if defined OS2
- /* To avoid the troubles of installing a separate file in the same
- directory as the DLL and of retrieving the DLL's directory at
- runtime, simply inline the aliases here. */
-
- /* The list of encodings is taken from "List of OS/2 Codepages"
- by Alex Taylor:
- <http://altsan.org/os2/toolkits/uls/index.html#codepages>.
- See also "IBM Globalization - Code page identifiers":
- <https://www-01.ibm.com/software/globalization/cp/cp_cpgid.html>. */
- cp = "CP813" "\0" "ISO-8859-7" "\0"
- "CP878" "\0" "KOI8-R" "\0"
- "CP819" "\0" "ISO-8859-1" "\0"
- "CP912" "\0" "ISO-8859-2" "\0"
- "CP913" "\0" "ISO-8859-3" "\0"
- "CP914" "\0" "ISO-8859-4" "\0"
- "CP915" "\0" "ISO-8859-5" "\0"
- "CP916" "\0" "ISO-8859-8" "\0"
- "CP920" "\0" "ISO-8859-9" "\0"
- "CP921" "\0" "ISO-8859-13" "\0"
- "CP923" "\0" "ISO-8859-15" "\0"
- "CP954" "\0" "EUC-JP" "\0"
- "CP964" "\0" "EUC-TW" "\0"
- "CP970" "\0" "EUC-KR" "\0"
- "CP1089" "\0" "ISO-8859-6" "\0"
- "CP1208" "\0" "UTF-8" "\0"
- "CP1381" "\0" "GB2312" "\0"
- "CP1386" "\0" "GBK" "\0"
- "CP3372" "\0" "EUC-JP" "\0";
+# ifndef locale_table_defined
+ /* Just a dummy entry, to avoid a C syntax error. */
+ { "", "" }
# endif
-#endif
+ };
- charset_aliases = cp;
- }
+#endif
- return cp;
-}
/* Determine the current locale's character encoding, and canonicalize it
- into one of the canonical names listed in config.charset.
+ into one of the canonical names listed in localcharset.h.
The result must not be freed; it is statically allocated.
If the canonical name cannot be determined, the result is a non-canonical
name. */
@@ -399,9 +643,8 @@ const char *
locale_charset (void)
{
const char *codeset;
- const char *aliases;
-#if !(defined WINDOWS_NATIVE || defined OS2)
+#if HAVE_LANGINFO_CODESET || defined WINDOWS_NATIVE || defined OS2
# if HAVE_LANGINFO_CODESET
@@ -464,37 +707,11 @@ locale_charset (void)
}
# endif
-# else
-
- /* On old systems which lack it, use setlocale or getenv. */
- const char *locale = NULL;
-
- /* But most old systems don't have a complete set of locales. Some
- (like SunOS 4 or DJGPP) have only the C locale. Therefore we don't
- use setlocale here; it would return "C" when it doesn't support the
- locale name the user has set. */
-# if 0
- locale = setlocale (LC_CTYPE, NULL);
-# endif
- if (locale == NULL || locale[0] == '\0')
- {
- locale = getenv ("LC_ALL");
- if (locale == NULL || locale[0] == '\0')
- {
- locale = getenv ("LC_CTYPE");
- if (locale == NULL || locale[0] == '\0')
- locale = getenv ("LANG");
- }
- }
-
- /* On some old systems, one used to set locale = "iso8859_1". On others,
- you set it to "language_COUNTRY.charset". In any case, we resolve it
- through the charset.alias file. */
- codeset = locale;
-
-# endif
+ if (codeset == NULL)
+ /* The canonical name cannot be determined. */
+ codeset = "";
-#elif defined WINDOWS_NATIVE
+# elif defined WINDOWS_NATIVE
static char buf[2 + 10 + 1];
@@ -528,7 +745,7 @@ locale_charset (void)
}
codeset = buf;
-#elif defined OS2
+# elif defined OS2
const char *locale;
static char buf[2 + 10 + 1];
@@ -585,28 +802,144 @@ locale_charset (void)
}
}
-#endif
+# else
- if (codeset == NULL)
- /* The canonical name cannot be determined. */
- codeset = "";
+# error "Add code for other platforms here."
- /* Resolve alias. */
- for (aliases = get_charset_aliases ();
- *aliases != '\0';
- aliases += strlen (aliases) + 1, aliases += strlen (aliases) + 1)
- if (strcmp (codeset, aliases) == 0
- || (aliases[0] == '*' && aliases[1] == '\0'))
+# endif
+
+ /* Resolve alias. */
+ {
+# ifdef alias_table_defined
+ /* On some platforms, UTF-8 locales are the most frequently used ones.
+ Speed up the common case and slow down the less common cases by
+ testing for this case first. */
+# if defined __OpenBSD__ || (defined __APPLE__ && defined __MACH__) || defined __sun || defined __CYGWIN__
+ if (strcmp (codeset, "UTF-8") == 0)
+ goto done_table_lookup;
+ else
+# endif
{
- codeset = aliases + strlen (aliases) + 1;
- break;
+ const struct table_entry * const table = alias_table;
+ size_t const table_size =
+ sizeof (alias_table) / sizeof (struct table_entry);
+ /* The table is sorted. Perform a binary search. */
+ size_t hi = table_size;
+ size_t lo = 0;
+ while (lo < hi)
+ {
+ /* Invariant:
+ for i < lo, strcmp (table[i].alias, codeset) < 0,
+ for i >= hi, strcmp (table[i].alias, codeset) > 0. */
+ size_t mid = (hi + lo) >> 1; /* >= lo, < hi */
+ int cmp = strcmp (table[mid].alias, codeset);
+ if (cmp < 0)
+ lo = mid + 1;
+ else if (cmp > 0)
+ hi = mid;
+ else
+ {
+ /* Found an i with
+ strcmp (table[i].alias, codeset) == 0. */
+ codeset = table[mid].canonical;
+ goto done_table_lookup;
+ }
+ }
}
+ if (0)
+ done_table_lookup: ;
+ else
+# endif
+ {
+ /* Did not find it in the table. */
+ /* On Mac OS X, all modern locales use the UTF-8 encoding.
+ BeOS and Haiku have a single locale, and it has UTF-8 encoding. */
+# if (defined __APPLE__ && defined __MACH__) || defined __BEOS__ || defined __HAIKU__
+ codeset = "UTF-8";
+# else
+ /* Don't return an empty string. GNU libc and GNU libiconv interpret
+ the empty string as denoting "the locale's character encoding",
+ thus GNU libiconv would call this function a second time. */
+ if (codeset[0] == '\0')
+ codeset = "ASCII";
+# endif
+ }
+ }
- /* Don't return an empty string. GNU libc and GNU libiconv interpret
- the empty string as denoting "the locale's character encoding",
- thus GNU libiconv would call this function a second time. */
- if (codeset[0] == '\0')
- codeset = "ASCII";
+#else
+
+ /* On old systems which lack it, use setlocale or getenv. */
+ const char *locale = NULL;
+
+ /* But most old systems don't have a complete set of locales. Some
+ (like DJGPP) have only the C locale. Therefore we don't use setlocale
+ here; it would return "C" when it doesn't support the locale name the
+ user has set. */
+# if 0
+ locale = setlocale (LC_CTYPE, NULL);
+# endif
+ if (locale == NULL || locale[0] == '\0')
+ {
+ locale = getenv ("LC_ALL");
+ if (locale == NULL || locale[0] == '\0')
+ {
+ locale = getenv ("LC_CTYPE");
+ if (locale == NULL || locale[0] == '\0')
+ locale = getenv ("LANG");
+ if (locale == NULL)
+ locale = "";
+ }
+ }
+
+ /* Map locale name to canonical encoding name. */
+ {
+# ifdef locale_table_defined
+ const struct table_entry * const table = locale_table;
+ size_t const table_size =
+ sizeof (locale_table) / sizeof (struct table_entry);
+ /* The table is sorted. Perform a binary search. */
+ size_t hi = table_size;
+ size_t lo = 0;
+ while (lo < hi)
+ {
+ /* Invariant:
+ for i < lo, strcmp (table[i].locale, locale) < 0,
+ for i >= hi, strcmp (table[i].locale, locale) > 0. */
+ size_t mid = (hi + lo) >> 1; /* >= lo, < hi */
+ int cmp = strcmp (table[mid].locale, locale);
+ if (cmp < 0)
+ lo = mid + 1;
+ else if (cmp > 0)
+ hi = mid;
+ else
+ {
+ /* Found an i with
+ strcmp (table[i].locale, locale) == 0. */
+ codeset = table[mid].canonical;
+ goto done_table_lookup;
+ }
+ }
+ if (0)
+ done_table_lookup: ;
+ else
+# endif
+ {
+ /* Did not find it in the table. */
+ /* On Mac OS X, all modern locales use the UTF-8 encoding.
+ BeOS and Haiku have a single locale, and it has UTF-8 encoding. */
+# if (defined __APPLE__ && defined __MACH__) || defined __BEOS__ || defined __HAIKU__
+ codeset = "UTF-8";
+# else
+ /* The canonical name cannot be determined. */
+ /* Don't return an empty string. GNU libc and GNU libiconv interpret
+ the empty string as denoting "the locale's character encoding",
+ thus GNU libiconv would call this function a second time. */
+ codeset = "ASCII";
+# endif
+ }
+ }
+
+#endif
#ifdef DARWIN7
/* Mac OS X sets MB_CUR_MAX to 1 when LC_ALL=C, and "UTF-8"
diff --git a/lib/localcharset.h b/lib/localcharset.h
index 917e473..56d5aaa 100644
--- a/lib/localcharset.h
+++ b/lib/localcharset.h
@@ -25,12 +25,106 @@ extern "C" {
/* Determine the current locale's character encoding, and canonicalize it
- into one of the canonical names listed in config.charset.
+ into one of the canonical names listed below.
The result must not be freed; it is statically allocated.
If the canonical name cannot be determined, the result is a non-canonical
name. */
extern const char * locale_charset (void);
+/* About GNU canonical names for character encodings:
+
+ Every canonical name must be supported by GNU libiconv. Support by GNU libc
+ is also desirable.
+
+ The name is case insensitive. Usually an upper case MIME charset name is
+ preferred.
+
+ The current list of these GNU canonical names is:
+
+ name MIME? used by which systems
+ (darwin = Mac OS X, windows = native Windows)
+
+ ASCII, ANSI_X3.4-1968 glibc solaris freebsd netbsd darwin cygwin
+ ISO-8859-1 Y glibc aix hpux irix osf solaris freebsd netbsd openbsd darwin cygwin
+ ISO-8859-2 Y glibc aix hpux irix osf solaris freebsd netbsd openbsd darwin cygwin
+ ISO-8859-3 Y glibc solaris cygwin
+ ISO-8859-4 Y osf solaris freebsd netbsd openbsd darwin
+ ISO-8859-5 Y glibc aix hpux irix osf solaris freebsd netbsd openbsd darwin cygwin
+ ISO-8859-6 Y glibc aix hpux solaris cygwin
+ ISO-8859-7 Y glibc aix hpux irix osf solaris netbsd openbsd darwin cygwin
+ ISO-8859-8 Y glibc aix hpux osf solaris cygwin
+ ISO-8859-9 Y glibc aix hpux irix osf solaris darwin cygwin
+ ISO-8859-13 glibc netbsd openbsd darwin cygwin
+ ISO-8859-14 glibc cygwin
+ ISO-8859-15 glibc aix osf solaris freebsd netbsd openbsd darwin cygwin
+ KOI8-R Y glibc solaris freebsd netbsd openbsd darwin
+ KOI8-U Y glibc freebsd netbsd openbsd darwin cygwin
+ KOI8-T glibc
+ CP437 dos
+ CP775 dos
+ CP850 aix osf dos
+ CP852 dos
+ CP855 dos
+ CP856 aix
+ CP857 dos
+ CP861 dos
+ CP862 dos
+ CP864 dos
+ CP865 dos
+ CP866 freebsd netbsd openbsd darwin dos
+ CP869 dos
+ CP874 windows dos
+ CP922 aix
+ CP932 aix cygwin windows dos
+ CP943 aix
+ CP949 osf darwin windows dos
+ CP950 windows dos
+ CP1046 aix
+ CP1124 aix
+ CP1125 dos
+ CP1129 aix
+ CP1131 darwin
+ CP1250 windows
+ CP1251 glibc solaris netbsd openbsd darwin cygwin windows
+ CP1252 aix windows
+ CP1253 windows
+ CP1254 windows
+ CP1255 glibc windows
+ CP1256 windows
+ CP1257 windows
+ GB2312 Y glibc aix hpux irix solaris freebsd netbsd darwin
+ EUC-JP Y glibc aix hpux irix osf solaris freebsd netbsd darwin
+ EUC-KR Y glibc aix hpux irix osf solaris freebsd netbsd darwin cygwin
+ EUC-TW glibc aix hpux irix osf solaris netbsd
+ BIG5 Y glibc aix hpux osf solaris freebsd netbsd darwin cygwin
+ BIG5-HKSCS glibc solaris darwin
+ GBK glibc aix osf solaris darwin cygwin windows dos
+ GB18030 glibc solaris netbsd darwin
+ SHIFT_JIS Y hpux osf solaris freebsd netbsd darwin
+ JOHAB glibc solaris windows
+ TIS-620 glibc aix hpux osf solaris cygwin
+ VISCII Y glibc
+ TCVN5712-1 glibc
+ ARMSCII-8 glibc darwin
+ GEORGIAN-PS glibc cygwin
+ PT154 glibc
+ HP-ROMAN8 hpux
+ HP-ARABIC8 hpux
+ HP-GREEK8 hpux
+ HP-HEBREW8 hpux
+ HP-TURKISH8 hpux
+ HP-KANA8 hpux
+ DEC-KANJI osf
+ DEC-HANYU osf
+ UTF-8 Y glibc aix hpux osf solaris netbsd darwin cygwin
+
+ Note: Names which are not marked as being a MIME name should not be used in
+ Internet protocols for information interchange (mail, news, etc.).
+
+ Note: ASCII and ANSI_X3.4-1968 are synonymous canonical names. Applications
+ must understand both names and treat them as equivalent.
+ */
+
#ifdef __cplusplus
}
diff --git a/lib/ref-add.sin b/lib/ref-add.sin
deleted file mode 100644
index 139494f..0000000
--- a/lib/ref-add.sin
+++ /dev/null
@@ -1,29 +0,0 @@
-# Add this package to a list of references stored in a text file.
-#
-# Copyright (C) 2000, 2009-2018 Free Software Foundation, Inc.
-#
-# 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.
-#
-# This program 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.
-#
-# You should have received a copy of the GNU General Public License along
-# with this program; if not, see <https://www.gnu.org/licenses/>.
-#
-# Written by Bruno Haible <hai...@clisp.cons.org>.
-#
-/^# Packages using this file: / {
- s/# Packages using this file://
- ta
- :a
- s/ @PACKAGE@ / @PACKAGE@ /
- tb
- s/ $/ @PACKAGE@ /
- :b
- s/^/# Packages using this file:/
-}
diff --git a/lib/ref-del.sin b/lib/ref-del.sin
deleted file mode 100644
index da00348..0000000
--- a/lib/ref-del.sin
+++ /dev/null
@@ -1,24 +0,0 @@
-# Remove this package from a list of references stored in a text file.
-#
-# Copyright (C) 2000, 2009-2018 Free Software Foundation, Inc.
-#
-# 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.
-#
-# This program 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.
-#
-# You should have received a copy of the GNU General Public License along
-# with this program; if not, see <https://www.gnu.org/licenses/>.
-#
-# Written by Bruno Haible <hai...@clisp.cons.org>.
-#
-/^# Packages using this file: / {
- s/# Packages using this file://
- s/ @PACKAGE@ / /
- s/^/# Packages using this file:/
-}
diff --git a/m4/localcharset.m4 b/m4/localcharset.m4
index 2d7ca64..d38a2c9 100644
--- a/m4/localcharset.m4
+++ b/m4/localcharset.m4
@@ -1,4 +1,4 @@
-# localcharset.m4 serial 7
+# localcharset.m4 serial 8
dnl Copyright (C) 2002, 2004, 2006, 2009-2018 Free Software Foundation, Inc.
dnl This file is free software; the Free Software Foundation
dnl gives unlimited permission to copy and/or distribute it,
@@ -8,10 +8,4 @@ AC_DEFUN([gl_LOCALCHARSET],
[
dnl Prerequisites of lib/localcharset.c.
AC_REQUIRE([AM_LANGINFO_CODESET])
- AC_REQUIRE([gl_FCNTL_O_FLAGS])
- AC_CHECK_DECLS_ONCE([getc_unlocked])
-
- dnl Prerequisites of the lib/Makefile.am snippet.
- AC_REQUIRE([AC_CANONICAL_HOST])
- AC_REQUIRE([gl_GLIBC21])
])
diff --git a/modules/localcharset b/modules/localcharset
index 1c2dd19..65ab529 100644
--- a/modules/localcharset
+++ b/modules/localcharset
@@ -1,100 +1,23 @@
Description:
Return current locale's character encoding.
-Notice:
-If your package's tests make use of the locale_charset() function directly or
-indirectly, you may need to define the CHARSETALIASDIR environment variable,
-so that "make check" works before "make install". In Makefile.am syntax:
-TESTS_ENVIRONMENT += @LOCALCHARSET_TESTS_ENVIRONMENT@
-
Files:
lib/localcharset.h
lib/localcharset.c
-lib/config.charset
-lib/ref-add.sin
-lib/ref-del.sin
m4/codeset.m4
-m4/fcntl-o.m4
-m4/glibc21.m4
m4/localcharset.m4
Depends-on:
-configmake
extensions
configure.ac:
gl_LOCALCHARSET
-LOCALCHARSET_TESTS_ENVIRONMENT="CHARSETALIASDIR=\"\$(abs_top_builddir)/$gl_source_base\""
+dnl For backward compatibility. Some packages still use this.
+LOCALCHARSET_TESTS_ENVIRONMENT=
AC_SUBST([LOCALCHARSET_TESTS_ENVIRONMENT])
Makefile.am:
-lib_SOURCES += localcharset.h localcharset.c
-
-# We need the following in order to install a simple file in $(libdir)
-# which is shared with other installed packages. We use a list of referencing
-# packages so that "make uninstall" will remove the file if and only if it
-# is not used by another installed package.
-# On systems with glibc-2.1 or newer, the file is redundant, therefore we
-# avoid installing it.
-
-all-local: charset.alias ref-add.sed ref-del.sed
-
-charset_alias = $(DESTDIR)$(libdir)/charset.alias
-charset_tmp = $(DESTDIR)$(libdir)/charset.tmp
-install-exec-local: install-exec-localcharset
-install-exec-localcharset: all-local
- if test $(GLIBC21) = no; then \
- case '$(host_os)' in \
- darwin[56]*) \
- need_charset_alias=true ;; \
- darwin* | cygwin* | mingw* | pw32* | cegcc*) \
- need_charset_alias=false ;; \
- *) \
- need_charset_alias=true ;; \
- esac ; \
- else \
- need_charset_alias=false ; \
- fi ; \
- if $$need_charset_alias; then \
- $(mkinstalldirs) $(DESTDIR)$(libdir) ; \
- fi ; \
- if test -f $(charset_alias); then \
- sed -f ref-add.sed $(charset_alias) > $(charset_tmp) ; \
- $(INSTALL_DATA) $(charset_tmp) $(charset_alias) ; \
- rm -f $(charset_tmp) ; \
- else \
- if $$need_charset_alias; then \
- sed -f ref-add.sed charset.alias > $(charset_tmp) ; \
- $(INSTALL_DATA) $(charset_tmp) $(charset_alias) ; \
- rm -f $(charset_tmp) ; \
- fi ; \
- fi
-
-uninstall-local: uninstall-localcharset
-uninstall-localcharset: all-local
- if test -f $(charset_alias); then \
- sed -f ref-del.sed $(charset_alias) > $(charset_tmp); \
- if grep '^# Packages using this file: $$' $(charset_tmp) \
- > /dev/null; then \
- rm -f $(charset_alias); \
- else \
- $(INSTALL_DATA) $(charset_tmp) $(charset_alias); \
- fi; \
- rm -f $(charset_tmp); \
- fi
-
-charset.alias: config.charset
- $(AM_V_GEN)rm -f t-$@ $@ && \
- $(SHELL) $(srcdir)/config.charset '$(host)' > t-$@ && \
- mv t-$@ $@
-
-SUFFIXES += .sed .sin
-.sin.sed:
- $(AM_V_GEN)rm -f t-$@ $@ && \
- sed -e '/^#/d' -e 's/@''PACKAGE''@/$(PACKAGE)/g' $< > t-$@ && \
- mv t-$@ $@
-
-CLEANFILES += charset.alias ref-add.sed ref-del.sed
+lib_SOURCES += localcharset.c
Include:
"localcharset.h"
--
2.7.4
>From d10d9f0f571cdfc3a5590abc8ca9e94d08862591 Mon Sep 17 00:00:00 2001
From: Bruno Haible <br...@clisp.org>
Date: Sun, 20 May 2018 01:37:22 +0200
Subject: [PATCH 4/5] localcharset: Map the locale encodings found in newer
OSes.
* lib/localcharset.c (alias_table): Add mapping for locale encodings
found in FreeBSD 11, NetBSD 7, Solaris 10, Openindiana, HP-UX 11.31,
IRIX 6.5, Minix 3.3.
* lib/localcharset.h: Update comments accordingly. Also for Cygwin 2.9.
---
ChangeLog | 8 +++++++
lib/localcharset.c | 66 +++++++++++++++++++++++++++++++++++++++++++++---------
lib/localcharset.h | 32 +++++++++++++-------------
3 files changed, 79 insertions(+), 27 deletions(-)
diff --git a/ChangeLog b/ChangeLog
index ad2e064..8a02c1a 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,5 +1,13 @@
2018-05-19 Bruno Haible <br...@clisp.org>
+ localcharset: Map the locale encodings found in newer OSes.
+ * lib/localcharset.c (alias_table): Add mapping for locale encodings
+ found in FreeBSD 11, NetBSD 7, Solaris 10, Openindiana, HP-UX 11.31,
+ IRIX 6.5, Minix 3.3.
+ * lib/localcharset.h: Update comments accordingly. Also for Cygwin 2.9.
+
+2018-05-19 Bruno Haible <br...@clisp.org>
+
localcharset: Move mapping tables into the code. Use a binary search.
* lib/localcharset.h: Document the GNU canonical names for character
encodings here.
diff --git a/lib/localcharset.c b/lib/localcharset.c
index 94e0f8a..60c320a 100644
--- a/lib/localcharset.c
+++ b/lib/localcharset.c
@@ -90,13 +90,41 @@ struct table_entry
static const struct table_entry alias_table[] =
{
# if defined __FreeBSD__ /* FreeBSD */
- { "C", "ASCII" },
- { "US-ASCII", "ASCII" }
+ { "ARMSCII-8", "ARMSCII-8" },
+ { "Big5", "BIG5" },
+ { "C", "ASCII" },
+ { "CP1131", "CP1131" },
+ { "CP1251", "CP1251" },
+ { "CP866", "CP866" },
+ { "GB18030", "GB18030" },
+ { "GB2312", "GB2312" },
+ { "GBK", "GBK" },
+ /*{ "ISCII-DEV", "?" },*/
+ { "ISO8859-1", "ISO-8859-1" },
+ { "ISO8859-13", "ISO-8859-13" },
+ { "ISO8859-15", "ISO-8859-15" },
+ { "ISO8859-2", "ISO-8859-2" },
+ { "ISO8859-5", "ISO-8859-5" },
+ { "ISO8859-7", "ISO-8859-7" },
+ { "ISO8859-9", "ISO-8859-9" },
+ { "KOI8-R", "KOI8-R" },
+ { "KOI8-U", "KOI8-U" },
+ { "SJIS", "SHIFT_JIS" },
+ { "US-ASCII", "ASCII" },
+ { "eucCN", "GB2312" },
+ { "eucJP", "EUC-JP" },
+ { "eucKR", "EUC-KR" }
# define alias_table_defined
# endif
# if defined __NetBSD__ /* NetBSD */
{ "646", "ASCII" },
+ { "ARMSCII-8", "ARMSCII-8" },
{ "BIG5", "BIG5" },
+ { "Big5-HKSCS", "BIG5-HKSCS" },
+ { "CP1251", "CP1251" },
+ { "CP866", "CP866" },
+ { "GB18030", "GB18030" },
+ { "GB2312", "GB2312" },
{ "ISO8859-1", "ISO-8859-1" },
{ "ISO8859-13", "ISO-8859-13" },
{ "ISO8859-15", "ISO-8859-15" },
@@ -104,6 +132,9 @@ static const struct table_entry alias_table[] =
{ "ISO8859-4", "ISO-8859-4" },
{ "ISO8859-5", "ISO-8859-5" },
{ "ISO8859-7", "ISO-8859-7" },
+ { "KOI8-R", "KOI8-R" },
+ { "KOI8-U", "KOI8-U" },
+ { "PT154", "PT154" },
{ "SJIS", "SHIFT_JIS" },
{ "eucCN", "GB2312" },
{ "eucJP", "EUC-JP" },
@@ -217,21 +248,27 @@ static const struct table_entry alias_table[] =
{ "SJIS", "SHIFT_JIS" },
{ "arabic8", "HP-ARABIC8" },
{ "big5", "BIG5" },
+ { "cp1251", "CP1251" },
{ "eucJP", "EUC-JP" },
{ "eucKR", "EUC-KR" },
{ "eucTW", "EUC-TW" },
+ { "gb18030", "GB18030" },
{ "greek8", "HP-GREEK8" },
{ "hebrew8", "HP-HEBREW8" },
+ { "hkbig5", "BIG5-HKSCS" },
{ "hp15CN", "GB2312" },
{ "iso88591", "ISO-8859-1" },
+ { "iso885913", "ISO-8859-13" },
{ "iso885915", "ISO-8859-15" },
{ "iso88592", "ISO-8859-2" },
+ { "iso88594", "ISO-8859-4" },
{ "iso88595", "ISO-8859-5" },
{ "iso88596", "ISO-8859-6" },
{ "iso88597", "ISO-8859-7" },
{ "iso88598", "ISO-8859-8" },
{ "iso88599", "ISO-8859-9" },
{ "kana8", "HP-KANA8" },
+ { "koi8r", "KOI8-R" },
{ "roman8", "HP-ROMAN8" },
{ "tis620", "TIS-620" },
{ "turkish8", "HP-TURKISH8" },
@@ -239,15 +276,16 @@ static const struct table_entry alias_table[] =
# define alias_table_defined
# endif
# if defined __sgi /* IRIX */
- { "ISO8859-1", "ISO-8859-1" },
- { "ISO8859-2", "ISO-8859-2" },
- { "ISO8859-5", "ISO-8859-5" },
- { "ISO8859-7", "ISO-8859-7" },
- { "ISO8859-9", "ISO-8859-9" },
- { "eucCN", "GB2312" },
- { "eucJP", "EUC-JP" },
- { "eucKR", "EUC-KR" },
- { "eucTW", "EUC-TW" }
+ { "ISO8859-1", "ISO-8859-1" },
+ { "ISO8859-15", "ISO-8859-15" },
+ { "ISO8859-2", "ISO-8859-2" },
+ { "ISO8859-5", "ISO-8859-5" },
+ { "ISO8859-7", "ISO-8859-7" },
+ { "ISO8859-9", "ISO-8859-9" },
+ { "eucCN", "GB2312" },
+ { "eucJP", "EUC-JP" },
+ { "eucKR", "EUC-KR" },
+ { "eucTW", "EUC-TW" }
# define alias_table_defined
# endif
# if defined __osf__ /* OSF/1 */
@@ -284,6 +322,8 @@ static const struct table_entry alias_table[] =
{ "GB18030", "GB18030" },
{ "GBK", "GBK" },
{ "ISO8859-1", "ISO-8859-1" },
+ { "ISO8859-11", "TIS-620" },
+ { "ISO8859-13", "ISO-8859-13" },
{ "ISO8859-15", "ISO-8859-15" },
{ "ISO8859-2", "ISO-8859-2" },
{ "ISO8859-3", "ISO-8859-3" },
@@ -303,6 +343,10 @@ static const struct table_entry alias_table[] =
{ "koi8-r", "KOI8-R" }
# define alias_table_defined
# endif
+# if defined __minix /* Minix */
+ { "646", "ASCII" }
+# define alias_table_defined
+# endif
# if defined WINDOWS_NATIVE || defined __CYGWIN__ /* Windows */
{ "CP1361", "JOHAB" },
{ "CP20127", "ASCII" },
diff --git a/lib/localcharset.h b/lib/localcharset.h
index 56d5aaa..3a10f69 100644
--- a/lib/localcharset.h
+++ b/lib/localcharset.h
@@ -44,20 +44,20 @@ extern const char * locale_charset (void);
name MIME? used by which systems
(darwin = Mac OS X, windows = native Windows)
- ASCII, ANSI_X3.4-1968 glibc solaris freebsd netbsd darwin cygwin
+ ASCII, ANSI_X3.4-1968 glibc solaris freebsd netbsd darwin minix cygwin
ISO-8859-1 Y glibc aix hpux irix osf solaris freebsd netbsd openbsd darwin cygwin
ISO-8859-2 Y glibc aix hpux irix osf solaris freebsd netbsd openbsd darwin cygwin
ISO-8859-3 Y glibc solaris cygwin
- ISO-8859-4 Y osf solaris freebsd netbsd openbsd darwin
+ ISO-8859-4 Y hpux osf solaris freebsd netbsd openbsd darwin
ISO-8859-5 Y glibc aix hpux irix osf solaris freebsd netbsd openbsd darwin cygwin
ISO-8859-6 Y glibc aix hpux solaris cygwin
- ISO-8859-7 Y glibc aix hpux irix osf solaris netbsd openbsd darwin cygwin
+ ISO-8859-7 Y glibc aix hpux irix osf solaris freebsd netbsd openbsd darwin cygwin
ISO-8859-8 Y glibc aix hpux osf solaris cygwin
- ISO-8859-9 Y glibc aix hpux irix osf solaris darwin cygwin
- ISO-8859-13 glibc netbsd openbsd darwin cygwin
+ ISO-8859-9 Y glibc aix hpux irix osf solaris freebsd darwin cygwin
+ ISO-8859-13 glibc hpux solaris freebsd netbsd openbsd darwin cygwin
ISO-8859-14 glibc cygwin
- ISO-8859-15 glibc aix osf solaris freebsd netbsd openbsd darwin cygwin
- KOI8-R Y glibc solaris freebsd netbsd openbsd darwin
+ ISO-8859-15 glibc aix irix osf solaris freebsd netbsd openbsd darwin cygwin
+ KOI8-R Y glibc hpux solaris freebsd netbsd openbsd darwin
KOI8-U Y glibc freebsd netbsd openbsd darwin cygwin
KOI8-T glibc
CP437 dos
@@ -83,31 +83,31 @@ extern const char * locale_charset (void);
CP1124 aix
CP1125 dos
CP1129 aix
- CP1131 darwin
+ CP1131 freebsd darwin
CP1250 windows
- CP1251 glibc solaris netbsd openbsd darwin cygwin windows
+ CP1251 glibc hpux solaris freebsd netbsd openbsd darwin cygwin windows
CP1252 aix windows
CP1253 windows
CP1254 windows
CP1255 glibc windows
CP1256 windows
CP1257 windows
- GB2312 Y glibc aix hpux irix solaris freebsd netbsd darwin
- EUC-JP Y glibc aix hpux irix osf solaris freebsd netbsd darwin
+ GB2312 Y glibc aix hpux irix solaris freebsd netbsd darwin cygwin
+ EUC-JP Y glibc aix hpux irix osf solaris freebsd netbsd darwin cygwin
EUC-KR Y glibc aix hpux irix osf solaris freebsd netbsd darwin cygwin
EUC-TW glibc aix hpux irix osf solaris netbsd
BIG5 Y glibc aix hpux osf solaris freebsd netbsd darwin cygwin
- BIG5-HKSCS glibc solaris darwin
- GBK glibc aix osf solaris darwin cygwin windows dos
- GB18030 glibc solaris netbsd darwin
+ BIG5-HKSCS glibc hpux solaris netbsd darwin
+ GBK glibc aix osf solaris freebsd darwin cygwin windows dos
+ GB18030 glibc hpux solaris freebsd netbsd darwin
SHIFT_JIS Y hpux osf solaris freebsd netbsd darwin
JOHAB glibc solaris windows
TIS-620 glibc aix hpux osf solaris cygwin
VISCII Y glibc
TCVN5712-1 glibc
- ARMSCII-8 glibc darwin
+ ARMSCII-8 glibc freebsd netbsd darwin
GEORGIAN-PS glibc cygwin
- PT154 glibc
+ PT154 glibc netbsd cygwin
HP-ROMAN8 hpux
HP-ARABIC8 hpux
HP-GREEK8 hpux
--
2.7.4
>From 9431faf6172b7e6557d9b7f16f4df12210d58d32 Mon Sep 17 00:00:00 2001
From: Bruno Haible <br...@clisp.org>
Date: Sun, 20 May 2018 12:59:01 +0200
Subject: [PATCH 5/5] localcharset: Optimize.
* lib/localcharset.c (alias_table): Comment out no-op mappings for
platforms where these don't matter. This reduces the table size,
which in turn reduces the lookup time.
---
ChangeLog | 7 +++++++
lib/localcharset.c | 50 +++++++++++++++++++++++++-------------------------
2 files changed, 32 insertions(+), 25 deletions(-)
diff --git a/ChangeLog b/ChangeLog
index 8a02c1a..86c2863 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,10 @@
+2018-05-20 Bruno Haible <br...@clisp.org>
+
+ localcharset: Optimize.
+ * lib/localcharset.c (alias_table): Comment out no-op mappings for
+ platforms where these don't matter. This reduces the table size,
+ which in turn reduces the lookup time.
+
2018-05-19 Bruno Haible <br...@clisp.org>
localcharset: Map the locale encodings found in newer OSes.
diff --git a/lib/localcharset.c b/lib/localcharset.c
index 60c320a..a51eb0b 100644
--- a/lib/localcharset.c
+++ b/lib/localcharset.c
@@ -90,15 +90,15 @@ struct table_entry
static const struct table_entry alias_table[] =
{
# if defined __FreeBSD__ /* FreeBSD */
- { "ARMSCII-8", "ARMSCII-8" },
+ /*{ "ARMSCII-8", "ARMSCII-8" },*/
{ "Big5", "BIG5" },
{ "C", "ASCII" },
- { "CP1131", "CP1131" },
- { "CP1251", "CP1251" },
- { "CP866", "CP866" },
- { "GB18030", "GB18030" },
- { "GB2312", "GB2312" },
- { "GBK", "GBK" },
+ /*{ "CP1131", "CP1131" },*/
+ /*{ "CP1251", "CP1251" },*/
+ /*{ "CP866", "CP866" },*/
+ /*{ "GB18030", "GB18030" },*/
+ /*{ "GB2312", "GB2312" },*/
+ /*{ "GBK", "GBK" },*/
/*{ "ISCII-DEV", "?" },*/
{ "ISO8859-1", "ISO-8859-1" },
{ "ISO8859-13", "ISO-8859-13" },
@@ -107,8 +107,8 @@ static const struct table_entry alias_table[] =
{ "ISO8859-5", "ISO-8859-5" },
{ "ISO8859-7", "ISO-8859-7" },
{ "ISO8859-9", "ISO-8859-9" },
- { "KOI8-R", "KOI8-R" },
- { "KOI8-U", "KOI8-U" },
+ /*{ "KOI8-R", "KOI8-R" },*/
+ /*{ "KOI8-U", "KOI8-U" },*/
{ "SJIS", "SHIFT_JIS" },
{ "US-ASCII", "ASCII" },
{ "eucCN", "GB2312" },
@@ -118,13 +118,13 @@ static const struct table_entry alias_table[] =
# endif
# if defined __NetBSD__ /* NetBSD */
{ "646", "ASCII" },
- { "ARMSCII-8", "ARMSCII-8" },
- { "BIG5", "BIG5" },
+ /*{ "ARMSCII-8", "ARMSCII-8" },*/
+ /*{ "BIG5", "BIG5" },*/
{ "Big5-HKSCS", "BIG5-HKSCS" },
- { "CP1251", "CP1251" },
- { "CP866", "CP866" },
- { "GB18030", "GB18030" },
- { "GB2312", "GB2312" },
+ /*{ "CP1251", "CP1251" },*/
+ /*{ "CP866", "CP866" },*/
+ /*{ "GB18030", "GB18030" },*/
+ /*{ "GB2312", "GB2312" },*/
{ "ISO8859-1", "ISO-8859-1" },
{ "ISO8859-13", "ISO-8859-13" },
{ "ISO8859-15", "ISO-8859-15" },
@@ -132,9 +132,9 @@ static const struct table_entry alias_table[] =
{ "ISO8859-4", "ISO-8859-4" },
{ "ISO8859-5", "ISO-8859-5" },
{ "ISO8859-7", "ISO-8859-7" },
- { "KOI8-R", "KOI8-R" },
- { "KOI8-U", "KOI8-U" },
- { "PT154", "PT154" },
+ /*{ "KOI8-R", "KOI8-R" },*/
+ /*{ "KOI8-U", "KOI8-U" },*/
+ /*{ "PT154", "PT154" },*/
{ "SJIS", "SHIFT_JIS" },
{ "eucCN", "GB2312" },
{ "eucJP", "EUC-JP" },
@@ -216,7 +216,7 @@ static const struct table_entry alias_table[] =
# define alias_table_defined
# endif
# if defined _AIX /* AIX */
- { "GBK", "GBK" },
+ /*{ "GBK", "GBK" },*/
{ "IBM-1046", "CP1046" },
{ "IBM-1124", "CP1124" },
{ "IBM-1129", "CP1129" },
@@ -240,7 +240,7 @@ static const struct table_entry alias_table[] =
{ "ISO8859-8", "ISO-8859-8" },
{ "ISO8859-9", "ISO-8859-9" },
{ "TIS-620", "TIS-620" },
- { "UTF-8", "UTF-8" },
+ /*{ "UTF-8", "UTF-8" },*/
{ "big5", "BIG5" }
# define alias_table_defined
# endif
@@ -289,7 +289,7 @@ static const struct table_entry alias_table[] =
# define alias_table_defined
# endif
# if defined __osf__ /* OSF/1 */
- { "GBK", "GBK" },
+ /*{ "GBK", "GBK" },*/
{ "ISO8859-1", "ISO-8859-1" },
{ "ISO8859-15", "ISO-8859-15" },
{ "ISO8859-2", "ISO-8859-2" },
@@ -301,7 +301,7 @@ static const struct table_entry alias_table[] =
{ "KSC5601", "CP949" },
{ "SJIS", "SHIFT_JIS" },
{ "TACTIS", "TIS-620" },
- { "UTF-8", "UTF-8" },
+ /*{ "UTF-8", "UTF-8" },*/
{ "big5", "BIG5" },
{ "cp850", "CP850" },
{ "dechanyu", "DEC-HANYU" },
@@ -317,10 +317,10 @@ static const struct table_entry alias_table[] =
# if defined __sun /* Solaris */
{ "5601", "EUC-KR" },
{ "646", "ASCII" },
- { "BIG5", "BIG5" },
+ /*{ "BIG5", "BIG5" },*/
{ "Big5-HKSCS", "BIG5-HKSCS" },
{ "GB18030", "GB18030" },
- { "GBK", "GBK" },
+ /*{ "GBK", "GBK" },*/
{ "ISO8859-1", "ISO-8859-1" },
{ "ISO8859-11", "TIS-620" },
{ "ISO8859-13", "ISO-8859-13" },
@@ -335,7 +335,7 @@ static const struct table_entry alias_table[] =
{ "ISO8859-9", "ISO-8859-9" },
{ "PCK", "SHIFT_JIS" },
{ "TIS620.2533", "TIS-620" },
- { "UTF-8", "UTF-8" },
+ /*{ "UTF-8", "UTF-8" },*/
{ "ansi-1251", "CP1251" },
{ "cns11643", "EUC-TW" },
{ "eucJP", "EUC-JP" },
--
2.7.4
