Update from gnulib.
diff --git a/libcharset/lib/ChangeLog b/libcharset/lib/ChangeLog
index ab7196b..6dd0455 100644
--- a/libcharset/lib/ChangeLog
+++ b/libcharset/lib/ChangeLog
@@ -1,3 +1,41 @@
+2016-12-02 Bruno Haible <bruno@clisp.org>
+
+ * config.charset: Update from gnulib.
+ 2015-01-07 KO Myung-Hun <komh@chollian.net>
+ * config.charset: Don't output aliases if "$os" is os2*.
+ 2012-06-22 Bruno Haible <bruno@clisp.org>
+ * connfig.charset: Write "Mac OS X" instead of "MacOS X".
+
+ * localcharset.c: Update from gnulib.
+ 2016-12-02 Bruno Haible <bruno@clisp.org>
+ * localcharset.c (locale_charset) [WINDOWS_NATIVE]: Don't use
+ the return value from setlocale if it would lead to a buffer
+ overrun.
+ 2015-01-07 KO Myung-Hun <komh@chollian.net>
+ * localcharset.c (get_charset_aliases) [OS2]: Hardcode the
+ result for OS/2.
+ (locale_charset) [OS2]: Use system codepage if codeset is
+ omitted from the locale name which is neither "C" nor "POSIX".
+ 2014-07-15 Eli Zaretskii <eliz@gnu.org>
+ * localcharset.c (locale_charset) [WINDOWS_NATIVE]: Before
+ falling back on the default system codepage, try extracting
+ the codepage from what 'setlocale' returns. This allows to
+ take into account changes of the codeset due to non-default
+ locale set by a previous call to 'setlocale'.
+ 2013-07-20 Daiki Ueno <ueno@gnu.org>
+ * localcharset.c (locale_charset) [DARWIN7]: Use MB_CUR_MAX_L
+ instead of MB_CUR_MAX.
+ 2012-09-16 Paul Eggert <eggert@cs.ucla.edu>
+ * localcharset.c (locale_charset) [DARWIN7]:
+ Return "ASCII" if the system reports "UTF-8" and
+ MB_CUR_MAX <= 1, as these two values are incompatible. Problem
+ reported by Max Horn. For more discussion, please see
+ <http://lists.gnu.org/archive/html/bug-gnulib/2012-09/msg00061.html>.
+ 2012-06-22 Bruno Haible <bruno@clisp.org>
+ * localcharset.c: Write "Mac OS X" instead of "MacOS X".
+
+ * relocatable.c: Update from gnulib.
+
2016-11-22 Bruno Haible <bruno@clisp.org>
Update support for building with MSVC.
diff --git a/libcharset/lib/config.charset b/libcharset/lib/config.charset
index fae71f3..594e8ed 100644
--- a/libcharset/lib/config.charset
+++ b/libcharset/lib/config.charset
@@ -29,7 +29,7 @@
# The current list of GNU canonical charset names is as follows.
#
# name MIME? used by which systems
-# (darwin = MacOS X, woe32 = native Windows)
+# (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
@@ -348,12 +348,10 @@
#echo "sun_eu_greek ?" # what is this?
echo "UTF-8 UTF-8"
;;
- freebsd* | os2*)
+ 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.
- # Likewise for OS/2. OS/2 has XFree86 just like FreeBSD. Just
- # reuse FreeBSD's locale data for OS/2.
echo "C ASCII"
echo "US-ASCII ASCII"
for l in la_LN lt_LN; do
diff --git a/libcharset/lib/localcharset.c b/libcharset/lib/localcharset.c
index 1756d5d..f3d7f27 100644
--- a/libcharset/lib/localcharset.c
+++ b/libcharset/lib/localcharset.c
@@ -29,11 +29,12 @@
#include <stdlib.h>
#if defined __APPLE__ && defined __MACH__ && HAVE_LANGINFO_CODESET
-# define DARWIN7 /* Darwin 7 or newer, i.e. MacOS X 10.3 or newer */
+# define DARWIN7 /* Darwin 7 or newer, i.e. Mac OS X 10.3 or newer */
#endif
#if defined _WIN32 || defined __WIN32__
# define WINDOWS_NATIVE
+# include <locale.h>
#endif
#if defined __EMX__
@@ -65,6 +66,11 @@
# include <os2.h>
#endif
+/* For MB_CUR_MAX_L */
+#if defined DARWIN7
+# include <xlocale.h>
+#endif
+
#if ENABLE_RELOCATABLE
# include "relocatable.h"
#else
@@ -122,7 +128,7 @@
cp = charset_aliases;
if (cp == NULL)
{
-#if !(defined DARWIN7 || defined VMS || defined WINDOWS_NATIVE || defined __CYGWIN__)
+#if !(defined DARWIN7 || defined VMS || defined WINDOWS_NATIVE || defined __CYGWIN__ || defined OS2)
const char *dir;
const char *base = "charset.alias";
char *file_name;
@@ -336,6 +342,36 @@
"CP54936" "\0" "GB18030" "\0"
"CP65001" "\0" "UTF-8" "\0";
# 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":
+ <http://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";
+# endif
#endif
charset_aliases = cp;
@@ -456,14 +492,34 @@
static char buf[2 + 10 + 1];
- /* The Windows API has a function returning the locale's codepage as a
- number: GetACP().
- When the output goes to a console window, it needs to be provided in
- GetOEMCP() encoding if the console is using a raster font, or in
- GetConsoleOutputCP() encoding if it is using a TrueType font.
- But in GUI programs and for output sent to files and pipes, GetACP()
- encoding is the best bet. */
- sprintf (buf, "CP%u", GetACP ());
+ /* The Windows API has a function returning the locale's codepage as
+ a number, but the value doesn't change according to what the
+ 'setlocale' call specified. So we use it as a last resort, in
+ case the string returned by 'setlocale' doesn't specify the
+ codepage. */
+ char *current_locale = setlocale (LC_ALL, NULL);
+ char *pdot;
+
+ /* If they set different locales for different categories,
+ 'setlocale' will return a semi-colon separated list of locale
+ values. To make sure we use the correct one, we choose LC_CTYPE. */
+ if (strchr (current_locale, ';'))
+ current_locale = setlocale (LC_CTYPE, NULL);
+
+ pdot = strrchr (current_locale, '.');
+ if (pdot && 2 + strlen (pdot + 1) + 1 <= sizeof (buf))
+ sprintf (buf, "CP%s", pdot + 1);
+ else
+ {
+ /* The Windows API has a function returning the locale's codepage as a
+ number: GetACP().
+ When the output goes to a console window, it needs to be provided in
+ GetOEMCP() encoding if the console is using a raster font, or in
+ GetConsoleOutputCP() encoding if it is using a TrueType font.
+ But in GUI programs and for output sent to files and pipes, GetACP()
+ encoding is the best bet. */
+ sprintf (buf, "CP%u", GetACP ());
+ }
codeset = buf;
#elif defined OS2
@@ -473,6 +529,8 @@
ULONG cp[3];
ULONG cplen;
+ codeset = NULL;
+
/* Allow user to override the codeset, as set in the operating system,
with standard language environment variables. */
locale = getenv ("LC_ALL");
@@ -504,10 +562,12 @@
}
}
- /* Resolve through the charset.alias file. */
- codeset = locale;
+ /* For the POSIX locale, don't use the system's codepage. */
+ if (strcmp (locale, "C") == 0 || strcmp (locale, "POSIX") == 0)
+ codeset = "";
}
- else
+
+ if (codeset == NULL)
{
/* OS/2 has a function returning the locale's codepage as a number. */
if (DosQueryCp (sizeof (cp), cp, &cplen))
@@ -542,5 +602,12 @@
if (codeset[0] == '\0')
codeset = "ASCII";
+#ifdef DARWIN7
+ /* Mac OS X sets MB_CUR_MAX to 1 when LC_ALL=C, and "UTF-8"
+ (the default codeset) does not work when MB_CUR_MAX is 1. */
+ if (strcmp (codeset, "UTF-8") == 0 && MB_CUR_MAX_L (uselocale (NULL)) <= 1)
+ codeset = "ASCII";
+#endif
+
return codeset;
}
diff --git a/libcharset/lib/relocatable.c b/libcharset/lib/relocatable.c
index 1b48ab2..6d3658f 100644
--- a/libcharset/lib/relocatable.c
+++ b/libcharset/lib/relocatable.c
@@ -47,6 +47,14 @@
# include <windows.h>
#endif
+#ifdef __EMX__
+# define INCL_DOS
+# include <os2.h>
+
+# define strcmp stricmp
+# define strncmp strnicmp
+#endif
+
#if DEPENDS_ON_LIBCHARSET
# include <libcharset.h>
#endif
@@ -335,6 +343,45 @@
return TRUE;
}
+#elif defined __EMX__
+
+extern int _CRT_init (void);
+extern void _CRT_term (void);
+extern void __ctordtorInit (void);
+extern void __ctordtorTerm (void);
+
+unsigned long _System
+_DLL_InitTerm (unsigned long hModule, unsigned long ulFlag)
+{
+ static char location[CCHMAXPATH];
+
+ switch (ulFlag)
+ {
+ case 0:
+ if (_CRT_init () == -1)
+ return 0;
+
+ __ctordtorInit();
+
+ /* See http://cyberkinetica.homeunix.net/os2tk45/cp1/1247_L2H_DosQueryModuleNameSy.html
+ for specification of DosQueryModuleName(). */
+ if (DosQueryModuleName (hModule, sizeof (location), location))
+ return 0;
+
+ _fnslashify (location);
+ shared_library_fullname = strdup (location);
+ break;
+
+ case 1:
+ __ctordtorTerm();
+
+ _CRT_term ();
+ break;
+ }
+
+ return 1;
+}
+
#else /* Unix */
static void
@@ -390,15 +437,16 @@
#endif
}
-#endif /* Native Windows / Unix */
+#endif /* Native Windows / EMX / Unix */
/* Return the full pathname of the current shared library.
Return NULL if unknown.
- Guaranteed to work only on Linux, Cygwin, and native Windows. */
+ Guaranteed to work only on Linux, EMX, Cygwin, and native Windows. */
static char *
get_shared_library_fullname ()
{
-#if !((defined _WIN32 || defined __WIN32__) && !defined __CYGWIN__)
+#if (!((defined _WIN32 || defined __WIN32__) && !defined __CYGWIN__) \
+ && !defined __EMX__)
static bool tried_find_shared_library_fullname;
if (!tried_find_shared_library_fullname)
{
@@ -489,6 +537,39 @@
}
}
}
+
+#ifdef __EMX__
+# ifdef __KLIBC__
+# undef strncmp
+
+ if (pathname && strncmp (pathname, "/@unixroot", 10) == 0
+ && (pathname[10] == '\0' || pathname[10] == '/' || pathname[10] == '\\'))
+ {
+ /* kLIBC itself processes /@unixroot prefix */
+
+ return pathname;
+ }
+ else
+# endif
+ if (pathname && ISSLASH (pathname[0]))
+ {
+ const char *unixroot = getenv ("UNIXROOT");
+
+ if (unixroot && HAS_DEVICE (unixroot) && !unixroot[2])
+ {
+ char *result = (char *) xmalloc (2 + strlen (pathname) + 1);
+#ifdef NO_XMALLOC
+ if (result != NULL)
+#endif
+ {
+ strcpy (result, unixroot);
+ strcpy (result + 2, pathname);
+ return result;
+ }
+ }
+ }
+#endif
+
/* Nothing to relocate. */
return pathname;
}
