Created attachment 344553 [details] 0BB8-0BC2.jpg Description of problem: When inputing "U+0BB8 U+0BC2" or "U+0BB9 U+0BC2" in gedit, the results were wrong. Version-Release number of selected component (if applicable): pangomm-2.14.1-1.fc10.x86_64 pango-1.22.3-1.fc10.i386 pango-1.22.3-1.fc10.x86_64 pango-devel-1.22.3-1.fc10.x86_64 How reproducible: Always Steps to Reproduce: 1.Select "RAW CODE" in scim-bridge 2.Input "U+0BB8 U+0BC2" or "U+0BB9 U+0BC2" in gedit. Actual results: The result are U+0BB8 U+0BC2 ஸூ U+0BB9 U+0BC2 ஹூ and they are different from the correct images. Expected results: Please see the attachment: 0BB8-0BC2.jpg and 0BB9-0BC2.jpg
Created attachment 344554 [details] 0BB9-0BC2.jpg
"U+0B9C U+0BC2" has the same problem. Actural results: "U+0B9C U+0BC2" ஜூ Expected results: Please see the attachment 0B9C-0BC2.jpg
Created attachment 344555 [details] 0B9C-0BC2.jpg
felix, Can you give correct image for following combination? 1) U+0BB8 U+0BC2 2) U+0BB9 U+0BC2 3) U+0B9C U+0BC2 Or you can just tell us if above(comment#1 and comment#3) attached images are the correct rendering for these combinations?
The character classes for the aforementioned characters (forming respective syllable combination) are defined correctly in pango-1.22.3-1.fc10.i386. Additionally, I noticed "serafettin-cartoon-fonts" yields correct/desired results for following combination: 1) U+0BB8 U+0BC2 2) U+0BB9 U+0BC2 3) U+0B9C U+0BC2 Whereas, lohit-tamil-fonts fails to give correct results for above combination. I have filed a bug against lohit-tamil-fonts (see bug 524457). I have attached corrected font file in the same bug, which yields the expected results.
From Comment #5, Correct behaviour of following combination: 1) U+0BB8 U+0BC2 2) U+0BB9 U+0BC2 3) U+0B9C U+0BC2 can now be obtained from build lohit-tamil-fonts-2_4_4-1_fc12 (available in Rawhide).
*** This bug has been marked as a duplicate of bug 524457 ***