RESOLVED INVALID 74701
[Chromium] Add UVS support on Windows
https://bugs.webkit.org/show_bug.cgi?id=74701
Summary [Chromium] Add UVS support on Windows
Kenichi Ishibashi
Reported 2011-12-16 02:30:56 PST
Add UVS support on Windows.
Attachments
Patch (6.61 KB, patch)
2011-12-16 02:39 PST, Kenichi Ishibashi
no flags
Kenichi Ishibashi
Comment 1 2011-12-16 02:39:07 PST
Tony Chang
Comment 2 2011-12-16 10:16:57 PST
This looks OK, but I don't know anything about uniscribe. Adding Brett for an informal review.
Kenichi Ishibashi
Comment 3 2011-12-19 15:44:10 PST
Could you take a look, Brett?
Kenichi Ishibashi
Comment 4 2011-12-20 18:39:33 PST
(In reply to comment #3) > Could you take a look, Brett? Should I ask an informal review to other developers?
Eric Seidel (no email)
Comment 5 2011-12-21 13:21:59 PST
A "LGTM" from brett or Jungshik would be helpful, yes.
Brett Wilson (Google)
Comment 6 2011-12-27 14:15:36 PST
My question is what the uniscribe helper is doing for you in this patch. Normally, it manages all this complex state for doing things like measuring and laying out a run. In your case, this one function assumes that the object was created with a two character string, and ignores everything else. As far as I can tell, the only thing you need from the helper is the cached DC, the rest of the code can just be called directly. Maybe we can add a static function to create & return the cached DC so that your code in SimpleFontData can use it.
Kenichi Ishibashi
Comment 7 2011-12-27 19:19:00 PST
(In reply to comment #6) > My question is what the uniscribe helper is doing for you in this patch. > > Normally, it manages all this complex state for doing things like measuring and laying out a run. In your case, this one function assumes that the object was created with a two character string, and ignores everything else. > > As far as I can tell, the only thing you need from the helper is the cached DC, the rest of the code can just be called directly. Maybe we can add a static function to create & return the cached DC so that your code in SimpleFontData can use it. Thank you for review. You are right. I can implement the function following your suggestion. In any case, I begin to think the current implementation of supporting UVS isn't good. I should have implemented it by using complex path because UVS requires glyph substitution depending on the context. This way, UVS should be supported without any changes on SimpleFontData and UniscribeHelper. I'll change the implementation soon. I'm sorry to bothering you.
Kenichi Ishibashi
Comment 8 2011-12-27 20:29:40 PST
Filed a bug entry for using complex path for UVS. https://bugs.webkit.org/show_bug.cgi?id=75289 Closing this bug as INVALID.
Note You need to log in before you can comment on or make changes to this bug.