[Webkit-unassigned] [Bug 51450] Glyphs in vertical text tests are rotated 90 degrees clockwise on Chromium Windows

Thu Jan 20 00:39:42 PST 2011

https://bugs.webkit.org/show_bug.cgi?id=51450

--- Comment #15 from Hironori Bono <hbono at chromium.org>  2011-01-20 00:39:42 PST ---
(In reply to comment #14)
> Bono-san, Jungshik, do you have any comments on this?

For his solution to prevent the crash, I do not have any ideas about it since I do not have any backgrounds about why this function compares the returned font. Nevertheless, to read the comments in the function, we seem to compare the returned family name also for filtering out your case: preventing returning '@Malgun Gothic' when we need a '@Times New Roman' font. So, it is better to ask Jungshik or Darin, who probably know the backgrounds about this code. (I'm adding more fall-back fonts to FontCache::getLastResortFallbackFont() to prevent crashes or assertions at Bug 52422, though.)

By the way, I'm a little wondering why this change ALWAYS uses '@'-fonts for rendering vertical text. (I would note that I may be wrong since I do not have any context about '-webkit-writing-mode'.) It is pretty reasonable to use '@'-fonts so we can use alternative fonts that have vertical glyphs when we render JAPANESE text with the '-webkit-writing-mode: vertical-rl; font-family: Times New Roman;' styles. On the other hand, it may be better to just rotate the 'Times New Roman' font than to use alternative fonts when we render ASCII text with the same styles? (Even though I do not have clear ideas now, it may be better to add '@' only when we need CJK fonts?)
Also, it may be a good idea to check if the 'family' is started with '@' in Line 583 of FontCacheChromiumWin.cpp to prevent adding '@' twice when users explicitly use '@'-fonts in the 'font-family' property?

Regards,

Hironori Bono

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.