Re: [PATCH 02/14] Fix character encoding aliases for OS/2

[Daiki Ueno]

Sorry for the delay.

KO Myung-Hun <address@hidden> writes:

>> I'm attaching two patches: the first one is an update of the patch we
>> are currently working on (I added mappings from the "unusable" codepages
>> to their equivalents), and the second one is an alternative
>> implementation following the kLIBC implementation.
>
> I prefer second one. It was my first idea and more simple. ^^

Okay, if nobody objects, I'll push the attached (I did a bit more
research on OS/2 codepages and added mappings from them to the GNU
canonical charset names as possible).

Regards,
--
Daiki Ueno

>From 0d7bfa1a8177175adb04f12d163cd606602d4285 Mon Sep 17 00:00:00 2001
From: KO Myung-Hun <address@hidden>
Date: Thu, 23 Feb 2012 22:37:21 +0900
Subject: [PATCH] localcharset: improve charset detection on OS/2

Use system codepage if appropriate.  Map the OS/2 codepages to the
GNU canonical charset names.
* lib/config.charset: Remove os2* from case "$os" in
* lib/localcharset.c (get_charset_aliases) [OS2]: Embed the
mapping from the OS/2 codepages to the GNU canonical charset
names.
(locale_charset) [OS2]: Use system codepage if codeset is omitted
from the locale name and the locale name is not "C" nor "POSIX".
---
 ChangeLog          | 12 ++++++++++++
 lib/config.charset |  4 +---
 lib/localcharset.c | 42 ++++++++++++++++++++++++++++++++++++++----
 3 files changed, 51 insertions(+), 7 deletions(-)

diff --git a/ChangeLog b/ChangeLog
index 9f5611d..ec8a19f 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,15 @@
+2015-01-05  KO Myung-Hun  <address@hidden>
+
+       localcharset: improve charset detection on OS/2
+       Use system codepage if appropriate.  Map the OS/2 codepages to the
+       GNU canonical charset names.
+       * lib/config.charset: Remove os2* from case "$os" in
+       * lib/localcharset.c (get_charset_aliases) [OS2]: Embed the
+       mapping from the OS/2 codepages to the GNU canonical charset
+       names.
+       (locale_charset) [OS2]: Use system codepage if codeset is omitted
+       from the locale name and the locale name is not "C" nor "POSIX".
+
 2015-01-04  Paul Eggert  <address@hidden>
 
        lib-symbol-versions: cache script check
diff --git a/lib/config.charset b/lib/config.charset
index 1b8a3f7..ac10e4b 100644
--- a/lib/config.charset
+++ b/lib/config.charset
@@ -348,12 +348,10 @@ case "$os" in
     #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/lib/localcharset.c b/lib/localcharset.c
index d54dbfb..fa25d4d 100644
--- a/lib/localcharset.c
+++ b/lib/localcharset.c
@@ -128,7 +128,7 @@ get_charset_aliases (void)
   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;
@@ -342,6 +342,36 @@ get_charset_aliases (void)
            "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 based on "List of OS/2 Codepages"
+         assembled by Alex Taylor:
+         <http://altsan.org/os2/toolkits/uls/index.html#codepages>
+         and "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;
@@ -499,6 +529,8 @@ locale_charset (void)
   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");
@@ -530,10 +562,12 @@ locale_charset (void)
             }
         }
 
-      /* 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))
-- 
2.1.0