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 124175 - Thai combining characters shifted in framebuffer
Thai combining characters shifted in framebuffer
Status: RESOLVED FIXED
Product: pango
Classification: Platform
Component: general
1.2.x
Other Linux
: Normal normal
: ---
Assigned To: pango-maint
pango-maint
Depends on: 121488
Blocks:
 
 
Reported: 2003-10-09 04:50 UTC by Theppitak Karoonboonyanan
Modified: 2004-12-22 21:47 UTC
See Also:
GNOME target: ---
GNOME version: 2.3/2.4


Attachments
Possible Pango patch (1.33 KB, patch)
2003-10-09 04:55 UTC, Theppitak Karoonboonyanan
none Details | Review
Example Thai TTF (38.80 KB, application/octet-stream)
2004-01-28 05:01 UTC, Theppitak Karoonboonyanan
  Details
A Thai test message (686 bytes, text/plain)
2004-01-28 05:11 UTC, Theppitak Karoonboonyanan
  Details
A shorter test message (40 bytes, text/plain)
2004-01-28 05:23 UTC, Theppitak Karoonboonyanan
  Details

Description Theppitak Karoonboonyanan 2003-10-09 04:50:28 UTC
GTK+ on framebuffer mode renders Thai combining characters
shifted to the left of its base consonants. I have verified that
pango-thai-ft2 was loaded to do the job, both by checking
/proc/<pid>/maps and by printing some debug messages.

It's weird that the same logic works on X, but fails on FB backend.
So, I'm not sure if this is the bug in Pango or GTK+.
But the attached patch just makes it work.

(P.S. This requires Bug #121488 to be fixed first.
I have applied Owen's patch before checking this bug.)
Comment 1 Theppitak Karoonboonyanan 2003-10-09 04:55:09 UTC
Created attachment 20582 [details] [review]
Possible Pango patch
Comment 2 Owen Taylor 2004-01-26 17:53:32 UTC
I've applied this patch in CVS, and it looks reasonable to me.
But I don't undderstand why it would make a difference to 
the displayed output. So there might be a bug somewhere else
we should track down. Can you provide an example of
the text and font that was giving problems?

Mon Jan 26 10:30:06 2004  Owen Taylor  <otaylor@redhat.com>
 
        * pango/modules/thai/thai-c: Fix up glyph addition
        code so that it works properly with the typical
        TrueType fonts that have 0 width for combining
        characters. (#124175, Theppitak Karoonboonyanan)
Comment 3 Theppitak Karoonboonyanan 2004-01-28 05:01:13 UTC
Created attachment 23829 [details]
Example Thai TTF
Comment 4 Theppitak Karoonboonyanan 2004-01-28 05:11:24 UTC
Created attachment 23830 [details]
A Thai test message
Comment 5 Theppitak Karoonboonyanan 2004-01-28 05:23:37 UTC
Created attachment 23831 [details]
A shorter test message
Comment 6 Theppitak Karoonboonyanan 2004-01-28 05:32:12 UTC
Please find test font and texts above. Note, however, that I have
already lost my previous configuration for testing framebuffer.
So, I haven't verified yet if the problem remains in recent CVS head.
Comment 7 Owen Taylor 2004-02-13 16:29:01 UTC
Hmm, with that font and that text I get identical results
for pangoft2topgm with and without the patch. So I don't
know what was going on.  Looks like I accidentally committed
a revert of the patch earlier, so fixed that up now:

Mon Jan 26 10:30:06 2004  Owen Taylor  <otaylor@redhat.com>
 
        * pango/modules/thai/thai-shaper.c: Reapply the patch
        from #124175 that somehow got reverted.