<?xml version="1.0" encoding="UTF-8" standalone="yes" ?>
<!DOCTYPE bugzilla SYSTEM "http://bugs.gentoo.org/bugzilla.dtd">

<bugzilla version="2.22.7"
          urlbase="http://bugs.gentoo.org/"
          maintainer="bugzilla@gentoo.org"
>

    <bug>
          <bug_id>142056</bug_id>
          
          <creation_ts>2006-07-28 21:32 0000</creation_ts>
          <short_desc>app-i18n/scim-1.4.4 - kanji input to terminals stoped working</short_desc>
          <delta_ts>2007-07-07 03:42:08 0000</delta_ts>
          <reporter_accessible>1</reporter_accessible>
          <cclist_accessible>1</cclist_accessible>
          <classification_id>1</classification_id>
          <classification>Unclassified</classification>
          <product>Gentoo Linux</product>
          <component>Applications</component>
          <version>unspecified</version>
          <rep_platform>All</rep_platform>
          <op_sys>Linux</op_sys>
          <bug_status>RESOLVED</bug_status>
          <resolution>FIXED</resolution>
          
          
          
          <priority>P1</priority>
          <bug_severity>major</bug_severity>
          <target_milestone>---</target_milestone>
          
          
          
          <everconfirmed>1</everconfirmed>
          <reporter>galtgendo@o2.pl</reporter>
          <assigned_to>cjk@gentoo.org</assigned_to>
          <cc>Andreas.Klauer@metamorpher.de</cc>
    
    <cc>hiyuh.root@gmail.com</cc>

      

      
          <long_desc isprivate="0">
            <who>galtgendo@o2.pl</who>
            <bug_when>2006-07-28 21:32:52 0000</bug_when>
            <thetext>First, list of packages:
app-i18n/anthy-7900
app-i18n/scim-1.4.4
app-i18n/scim-anthy-1.0.0
gcc-4.1.1
glibc-2.4-r3
2.6.17-gentoo-r4
sys-devel/binutils-2.16.1-r3

Now, the bug. A rather stange one. You see, after upgrade in the first three packages, kanji input to terminals stoped working. Everything works fine for firefox and other gtk apps, however the same is not true for both xterm and urxvt, for them things work only up to the point, when you press Enter to commit
selected kanji (kanji conversion seems to work), then instead of a kanji a char sequence, often a one containing escape sequences, is input to the terminal. Both of those terminals handle kanji correctly, I can paste them without any problem.</thetext>
          </long_desc>
          <long_desc isprivate="0">
            <who>Andreas.Klauer@metamorpher.de</who>
            <bug_when>2006-08-02 00:34:15 0000</bug_when>
            <thetext>Same here. xterm/urxvt/OpenOffice (all XIM applications as far as I know) are affected the same way by this problem. SCIM input window comes up and displays kana/kanji correctly, but when I hit enter, only unknown control sequences get inserted into the application. Input works great in GTK apps and I can copy&amp;paste Japanese text into xterm/urxvt/OpenOffice without problems.</thetext>
          </long_desc>
          <long_desc isprivate="0">
            <who>galtgendo@o2.pl</who>
            <bug_when>2006-08-05 22:56:54 0000</bug_when>
            <thetext>Ok, more info.
It&apos;s a bug in scim 1.4.4, 1.4.3  works, cvs of scim still doesn&apos;t.
The bug lies in rawcode imengine, disabing it makes 1.4.4 work, what&apos;s more
reverting the changes made in modules/IMEngine/  scim_rawcode_imengine.cpp and
scim_rawcode_imengine.h between 1.4.3 and 1.4.4 makes it work with rawcode enabled.

PS. Somebody should change bug description to reflect this.</thetext>
          </long_desc>
          <long_desc isprivate="0">
            <who>galtgendo@o2.pl</who>
            <bug_when>2006-08-07 22:00:06 0000</bug_when>
            <thetext>Created an attachment (id=93723)
fixes rawcode imengine problem affecting scim-anthy

Final update, I think.
Attached is minimal reverting patch that makes things work.
It&apos;s a bit strange, cause set_languages calls set_locales.
It&apos;s probably a typo, but I can&apos;t figure it out - 
maybe there&apos;s some difference between String and unsigned char* 
I&apos;m missing, maybe there&apos;s a bug in set_locales that&apos;s doesn&apos;t happen
when it&apos;s called by set_languages. As it&apos;s an upstream bug, somebody should report it there.</thetext>
          </long_desc>
          <long_desc isprivate="0">
            <who>Andreas.Klauer@metamorpher.de</who>
            <bug_when>2006-08-18 08:48:17 0000</bug_when>
            <thetext>That patch indeed seems to fix the problem. Thank you!</thetext>
          </long_desc>
          <long_desc isprivate="0">
            <who>liquidx@gentoo.org</who>
            <bug_when>2006-08-20 03:07:26 0000</bug_when>
            <thetext>submitted this upstream

http://sourceforge.net/tracker/index.php?func=detail&amp;aid=1543409&amp;group_id=108454&amp;atid=650539</thetext>
          </long_desc>
          <long_desc isprivate="0">
            <who>galtgendo@o2.pl</who>
            <bug_when>2006-10-10 09:01:57 0000</bug_when>
            <thetext>Update for scim-1.4.5:
Bug is still there, fix still works.
Please consider adding to ebuild.</thetext>
          </long_desc>
          <long_desc isprivate="0">
            <who>matsuu@gentoo.org</who>
            <bug_when>2006-10-10 09:31:47 0000</bug_when>
            <thetext>Fixed in cvs(1.4.5). thanks.
</thetext>
          </long_desc>
          <long_desc isprivate="0">
            <who>Andreas.Klauer@metamorpher.de</who>
            <bug_when>2007-01-09 21:54:29 0000</bug_when>
            <thetext>I&apos;m still suffering from this problem. I cannot input into xterm anymore with scim-1.4.5(-r1), but strangely enough, downgrading to 1.4.4 fixes the problem. This is on ~x86 Gentoo. I have no idea what is going on.</thetext>
          </long_desc>
          <long_desc isprivate="0">
            <who>Andreas.Klauer@metamorpher.de</who>
            <bug_when>2007-01-09 22:14:34 0000</bug_when>
            <thetext>Removing the scim-1.4.5-imengine.patch from scim-1.4.5-r1.ebuild fixed the problem for me. *shrug*</thetext>
          </long_desc>
          <long_desc isprivate="0">
            <who>galtgendo@o2.pl</who>
            <bug_when>2007-01-26 19:00:51 0000</bug_when>
            <thetext>I&apos;m reopening this bug for somebody to close it for good.
As per comment nr 5, it was submited upstream on 2006-08-20 03:06.
As of 2006-11-21 18:51, a proposed patch was given by upstream.
As of today, I checked that patch and it seemms to work.
Somebody please double check it and if sucessful, take the upstream patch and close it for good.</thetext>
          </long_desc>
          <long_desc isprivate="0">
            <who>galtgendo@o2.pl</who>
            <bug_when>2007-04-14 16:16:14 0000</bug_when>
            <thetext>OK, sorry. I was wrong.
That upstream patch actually does NOT work. The actual problem is that for some reason calling set_locales(string) doesn&apos;t work, while set_languages(string) does.</thetext>
          </long_desc>
          <long_desc isprivate="0">
            <who>galtgendo@o2.pl</who>
            <bug_when>2007-04-28 17:18:59 0000</bug_when>
            <thetext>Of all the silly things... Well, anyway...
As far as I can tell scim 1.4.6 is no longer affected by this bug.</thetext>
          </long_desc>
          <long_desc isprivate="0">
            <who>matsuu@gentoo.org</who>
            <bug_when>2007-07-07 03:42:08 0000</bug_when>
            <thetext>fixed in 1.4.6.
</thetext>
          </long_desc>
      
          <attachment
              isobsolete="0"
              ispatch="1"
              isprivate="0"
          >
            <attachid>93723</attachid>
            <date>2006-08-07 22:00 0000</date>
            <desc>fixes rawcode imengine problem affecting scim-anthy</desc>
            <filename>xim-fix.diff</filename>
            <type>text/plain</type>
            <data encoding="base64">ZGlmZiAtdXJOIHNjaW0tMS40LjQvbW9kdWxlcy9JTUVuZ2luZS9zY2ltX3Jhd2NvZGVfaW1lbmdp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</data>        

          </attachment>
    </bug>

</bugzilla>