GNOME Bugzilla – Bug 335799
Nepali texts rendered incorrectly
Last modified: 2007-01-28 22:42:34 UTC
Please describe the problem: Some vowels in nepali texts are put on incorrect consonants. Half-letters aren't combined with next full letter. Steps to reproduce: 1. Visit http://kantipuronline.com/kantipur.php 2. Copy some text to Gedit 3. Notice the difference in the text rendered by Epiphany and Gedit Actual results: Gedit renders the text correctly, epiphany does not. Expected results: I would have expected epiphany to render the text like gedit Does this happen every time? Yes. Other information: Epiphany is started with MOZ_PANGO_ENABLED=1.
Created attachment 61889 [details] Screenshot comparing nepali text rendering by epiphany and gedit The screenshot shows how nepali text rendering by epiphany and gedit differ--gedit's way is correct. We can see the inconsistency in the heading of the news. Compare the second word of the heading. Gedit put the vowel ि on the second letter whereas epiphany puts it on the third letter. In the same word, in epiphany, a half-letter is denoted by a dash below the letter ( ् ), gedit combines the half-letter with the next letter.
Thanks for the bug report! Can you try the following: Make sure all epiphany windows are closed and no instance is running (check with "ps aux | grep epiphany" shell command). Then start epiphany from a shell with export MOZ_ENABLE_PANGO=1 epiphany Does that fix the rendering? Also, can you please paste the output of "locale" command here? Thanks!
That fixes the problem, but introduces another one: the space between letters is too much in justify mode. Output of locale command is: LANG=en_US.UTF-8 LANGUAGE=en_US.UTF-8 LC_CTYPE="en_US.UTF-8" LC_NUMERIC="en_US.UTF-8" LC_TIME="en_US.UTF-8" LC_COLLATE="en_US.UTF-8" LC_MONETARY="en_US.UTF-8" LC_MESSAGES="en_US.UTF-8" LC_PAPER="en_US.UTF-8" LC_NAME="en_US.UTF-8" LC_ADDRESS="en_US.UTF-8" LC_TELEPHONE="en_US.UTF-8" LC_MEASUREMENT="en_US.UTF-8" LC_IDENTIFICATION="en_US.UTF-8" LC_ALL=
Created attachment 61932 [details] Screenshot showing too much space between letters.
The text spacing in letters in justify mode when using pango is a mozilla bug. I'm going to introduce a way to enable pango when the locale is one of those that require pango rendering (ne and a few others).
Fixed in CVS. The fix will go into the next patch release. Thank you for your bug report. Fixed in the development version. The fix will be available in the next major release. Thank you for your bug report.
Hi, I am now using version 2.16.1 (the one in Ubuntu Edgy) and the problem of too much text spacing in justify mode still exists. For example compare the pages: http://kantipuronline.com/kantipur.php (this is fine) http://kantipuronline.com/kolnepalinews.php?&nid=89788 (this is messed up) I have enabled pango using: export MOZ_ENABLE_PANGO=1 Output of locale command is: LANG=en_US.UTF-8 LANGUAGE=en_US.UTF-8 LC_CTYPE="en_US.UTF-8" LC_NUMERIC="en_US.UTF-8" LC_TIME="en_US.UTF-8" LC_COLLATE="en_US.UTF-8" LC_MONETARY="en_US.UTF-8" LC_MESSAGES="en_US.UTF-8" LC_PAPER="en_US.UTF-8" LC_NAME="en_US.UTF-8" LC_ADDRESS="en_US.UTF-8" LC_TELEPHONE="en_US.UTF-8" LC_MEASUREMENT="en_US.UTF-8" LC_IDENTIFICATION="en_US.UTF-8" LC_ALL=
Does firefox show the page correctly using MOZ_ENABLE_PANGO ? Btw, you shouldn't need MOZ_ENABLE_PANGO in epiphany, since it automatically uses it if you have "ne" locale. (Or you can use gconf to enable it, set "/apps/epiphany/web/enable_pango" key to true.)
Firefox shows the page incorrectly using either MOZ_ENABLE_PANGO = 1 or 0. Thanks for the info about gconf key. I have enabled it there now.
So that means it is a bug in the mozilla rendering engine, not in epiphany.
Definitely a gecko bug (and probably fixed in xulrunner 1.9).