This is the mail archive of the
libc-alpha@cygnus.com
mailing list for the glibc project.
localedef can not treat multi byte characters
- To: libc-alpha@cygnus.com
- Subject: localedef can not treat multi byte characters
- From: Shinya Hanataka <hanataka@abyss.rim.or.jp>
- Date: Sun, 24 Jan 1999 16:44:44 +0900
>Submitter-Id: net
>Originator: Shinya Hanataka
>Organization: no
>Confidential: no
>Synopsis: localedef can not treat multi byte characters
>Severity: non-critical
>Priority: medium
>Category: libc
>Class: sw-bug
>Release: libc-2.0.111
>Environment:
PC-clone, pentium CPU, Linux 2.2.0pre7
Host type: i586-pc-linux-gnu
System: Linux chaos 2.2.0-pre7 #13 SMP Sun Jan 17 00:07:47 JST 1999 i586 unknown
Architecture: i586
Addons: crypt linuxthreads
Build CFLAGS: -O3 -mpentium
Build CC: gcc
Compiler version: pgcc-2.91.60 19981201 (egcs-1.1.1 release)
Kernel headers: 2.2.0-pre7
Symbol versioning: yes
Build static: yes
Build shared: yes
Build pic-default: no
Build profile: yes
Build omitfp: no
Build bounded: no
Build static-nss: no
Stdio: libio
>Description:
(1) localedef program can not treat multibyte character on LC_XXXX .
(for all east asian character code);
(2) localedef program can not treat charmaps with MB_CUR_MAX > 4.
(for some ISO-2022 based character code);
>How-To-Repeat:
(1) Define multi bytes character at charmap file.
<j0101>...<j0194> /xa1/xa1 JIS X 0208 KANJI
Use that code on locale source file.
currency_symbol "<j0179>"
(2) Define <mb_cur_max> 5 on charmap file.
>Fix:
(1) Simple, but not essential patch, below.
(2) I have no idea, maybe need char* instead of long for MB characters.
--- glibc-2.0.111.orig/locale/programs/stringtrans.c Sun Jan 24 16:11:55 1999
+++ glibc-2.0.111/locale/programs/stringtrans.c Sun Jan 24 11:57:23 1999
@@ -115,9 +115,17 @@
switch (encoding_method)
{
case ENC_UCS1:
- if (value > 255)
- return -1;
- *(*cpp)++ = (char) value;
+ /* We NEED multi byte character, here */
+ if (value >= 0x1000000) {
+ *(*cpp)++ = (char) ((value >> 24) & 0xff);
+ }
+ if (value >= 0x10000) {
+ *(*cpp)++ = (char) ((value >> 16) & 0xff);
+ }
+ if (value >= 0x100) {
+ *(*cpp)++ = (char) ((value >> 8) & 0xff);
+ }
+ *(*cpp)++ = (char) (value & 0xff);
break;
case ENC_UCS4:
--
HANATAKA Shinya <hanataka@abyss.rim.or.jp>