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 153927 - Make Thai module prefer OpenType if exists
Make Thai module prefer OpenType if exists
Status: RESOLVED FIXED
Product: pango
Classification: Platform
Component: general
1.5.x
Other Linux
: Normal enhancement
: ---
Assigned To: pango-maint
pango-maint
Depends on:
Blocks:
 
 
Reported: 2004-09-28 08:06 UTC by Theppitak Karoonboonyanan
Modified: 2004-12-22 21:47 UTC
See Also:
GNOME target: ---
GNOME version: 2.7/2.8

Attachments
Proposed patch to make Thai module prefer OT shaping (6.92 KB, patch)
2004-09-28 08:09 UTC, Theppitak Karoonboonyanan 		none Details | Review
Revised patch using preprocessing in shaper (6.91 KB, patch)
2004-09-30 12:43 UTC, Theppitak Karoonboonyanan 		none Details | Review

Description Theppitak Karoonboonyanan 2004-09-28 08:06:56 UTC 
I have prepared a patch to make Thai module prefer OpenType shaping to the old
glyph substitution method if the corresponding GPOS/GSUB tables are available.
Comment 1 Theppitak Karoonboonyanan 2004-09-28 08:09:47 UTC 
Created attachment 32006 [details] [review]
Proposed patch to make Thai module prefer OT shaping
Comment 2 Theppitak Karoonboonyanan 2004-09-29 05:33:47 UTC 
With the proposed patch, some GSUB rules seem to be missed. They are sequences
of rules that require either moving backward or multi-pass substitution.
Comment 3 Theppitak Karoonboonyanan 2004-09-30 12:43:39 UTC 
Created attachment 32106 [details] [review]
Revised patch using preprocessing in shaper

As Owen remarked in 
http://mail.gnome.org/archives/gtk-i18n-list/2004-September/msg00017.html
I have tweaked the patch a little to take advantage of glyph decomposition &
reordering in the clusterization process, rather than relying on the GSUB
rules.
So, just use TIS fontset instead of ISO10646.

In fact, the ISO10646 fontset is now somewhat useless. I think it should be 
removed from the code soon.
Comment 4 Theppitak Karoonboonyanan 2004-10-02 03:36:08 UTC 
Patch committed to CVS HEAD.