After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 572230 - text mode program rendering is strange in cjk locale.
text mode program rendering is strange in cjk locale.
Status: RESOLVED FIXED
Product: vte
Classification: Core
Component: general
0.19.x
Other All
: Normal normal
: ---
Assigned To: VTE Maintainers
VTE Maintainers
: 566670 (view as bug list)
Depends on:
Blocks:
 
 
Reported: 2009-02-18 02:43 UTC by sangu
Modified: 2010-03-02 19:31 UTC
See Also:
GNOME target: ---
GNOME version: 2.25/2.26

Attachments
setup in vte (locale ja_JP.UTF-8) (28.54 KB, image/png)
2009-02-18 02:44 UTC, sangu 		Details
setup in $LANG=en_US.UTF-8 vte. (23.85 KB, image/png)
2009-08-05 15:33 UTC, sangu 		Details

Description sangu 2009-02-18 02:43:52 UTC 
Please describe the problem:
text mode program rendering is strange in cjk  locale.

Steps to reproduce:
1. $LANG=ja_JP.UTF-8 vte
2. $ mc or setup in vte
3. 


Actual results:


Expected results:


Does this happen every time?
yes, always.

Other information:
vte-0.19.4-3.fc11
Comment 1 sangu 2009-02-18 02:44:49 UTC 
Created attachment 128939 [details]
setup in vte (locale ja_JP.UTF-8)
Comment 2 sangu 2009-08-05 15:31:41 UTC 
This issue still happens in vte-0.20.5.
---
"$ LANG=en_US.UTF-8 vte" has good rendering.

slang-2.2.0-1.fc12.x86_64
ncurses-5.7-3.20090207.fc12.x86_64
Comment 3 sangu 2009-08-05 15:33:38 UTC 
Created attachment 139955 [details]
setup in $LANG=en_US.UTF-8 vte.
Comment 4 Behdad Esfahbod 2009-08-05 17:22:09 UTC 
Humm, I'm beng more convinved that we shouldn't wide-char the ambiguous chars under any UTF-8 locale by default, and let VTE_CJK_WIDTH choose it.  I'll give it a shot.
Comment 5 Behdad Esfahbod 2009-08-06 01:21:18 UTC 
Humm, weird, setup under ja_JP.UTF-8 works fine here.  What version of vte is that?  Can you test with master?
Comment 6 Behdad Esfahbod 2009-08-06 01:22:35 UTC 
Humm, my bad.
Comment 7 Behdad Esfahbod 2009-08-06 01:29:37 UTC 
Ok, turned of ambiguous-width guessing based on locale if charset is UTF-8.  The Unicode specs is clear about that case, so we follow it.

commit 3eaca3e36fb806608c961ef594baf99cecf9fdfa
Author: Behdad Esfahbod <behdad@behdad.org>
Date:   Wed Aug 5 21:26:05 2009 -0400

    Bug 572230 – text mode program rendering is strange in cjk locale.
    
    Under UTF-8 encoding, always use width=1 for ambiguous-width chars,
    only allow override using VTE_CJK_WIDTH=1.  Preivously we were using
    width=2 for CJKS locales.  That's against Unicode recommendations,
    and breaks a bunch of programs.
Comment 8 Behdad Esfahbod 2010-03-02 19:31:14 UTC 
*** Bug 566670 has been marked as a duplicate of this bug. ***