FontConfig character/lang match requires weak family.

FontConfig character/lang match requires weak family.

In FontConfig the normal priority order of matches is character, family,
language; however family can be marked as weak which changes the order to
character, language, family. This is what is desired in
matchFamilyStyleCharacter.

Committed: https://skia.googlesource.com/skia/+/1d4ec0a1ad4998c32c6a28ff4b85000553ff686d

[bungeman-chromium, 2015-04-29 18:13:58]

bungeman@chromium.org changed reviewers:
+ bungeman@chromium.org, mtklein@google.com, reed@google.com

[bungeman-chromium, 2015-04-29 18:13:59]

Fixing the test in a previous CL revealed this bug. The issue is that the family
name is being prioritized over the language, which isn't what we want here.

[reed1, 2015-04-29 18:26:01]

lgtm

[bungeman-chromium, 2015-04-29 18:27:12]

The CQ bit was checked by bungeman@chromium.org

[commit-bot: I haz the power, 2015-04-29 18:27:21]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1118553002/1

[commit-bot: I haz the power, 2015-04-29 18:33:06]

Committed patchset #1 (id:1) as
https://skia.googlesource.com/skia/+/1d4ec0a1ad4998c32c6a28ff4b85000553ff686d

[bungeman-chromium, 2015-04-29 19:21:06]

On 2015/04/29 18:33:06, I haz the power (commit-bot) wrote:
> Committed patchset #1 (id:1) as
> https://skia.googlesource.com/skia/+/1d4ec0a1ad4998c32c6a28ff4b85000553ff686d

Note that this fixed 'Droid Sans Fallback' and 'Droid Sans Japanese' since they
actually exhibited this issue. On the other hand, there appear to be a number of
fonts which seem to just be set up with the wrong language settings. I'll have
to take a look into that separately.

[bungeman-chromium, 2015-04-29 19:30:09]

On 2015/04/29 19:21:06, bungeman2 wrote:
> On 2015/04/29 18:33:06, I haz the power (commit-bot) wrote:
> > Committed patchset #1 (id:1) as
> >
https://skia.googlesource.com/skia/+/1d4ec0a1ad4998c32c6a28ff4b85000553ff686d
> 
> Note that this fixed 'Droid Sans Fallback' and 'Droid Sans Japanese' since
they
> actually exhibited this issue. On the other hand, there appear to be a number
of
> fonts which seem to just be set up with the wrong language settings. I'll have
> to take a look into that separately.

This also fixed ChromeOS completely, since it appears they have a more
controlled and explicit configuration for FontConfig.