Rebaseline layout test text-underline-position-cjk.html

Rebaseline layout test text-underline-position-cjk.html

BUG=690486
Review-Url: https://codereview.chromium.org/2712443003
Cr-Commit-Position: refs/heads/master@{#456289}
Committed: https://chromium.googlesource.com/chromium/src/+/18ffa64cff913d0b56d09276fd72257b74278f61

[qyearsley, 2017-02-21 17:33:57]

qyearsley@chromium.org changed reviewers:
+ kojii@chromium.org

[qyearsley, 2017-02-21 17:37:27]

The CQ bit was checked by qyearsley@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2017-02-21 17:38:15]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2017-02-21 18:50:01]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2017-02-21 18:50:02]

Dry run: This issue passed the CQ dry run.

[kojii, 2017-02-22 04:54:52]

huh, strange, win7 still fails?

[qyearsley, 2017-02-22 16:06:12]

On 2017/02/22 at 04:54:52, kojii wrote:
> huh, strange, win7 still fails?

Aye -- and looking at the results from the try job[1], it appears that the
difference between the two images is the size and appearance of the text with
font-family set to "'Yu Gothic', YuGothic"[2].

If this is because some of the win7_blink_rel slaves have different fonts
installed, then we should expect the results to be flaky. If it's an issue
related to configuration of the bots, do you know what we could do about this?

[1]
https://storage.googleapis.com/chromium-layout-test-archives/win7_blink_rel/2...
[2]
https://chromium.googlesource.com/chromium/src/+/6a23306766d02a480c7992f91b51...

[kojii, 2017-02-23 09:16:02]

Yu Gothic is rather a new font, bundled since Mac 10.9 and Win 8.1, so Win7
should simply fallback to default font.

So it sounds like some slaves fallback to different font when this font-family
is set?

Maybe we can add another, such as "'Yu Gothic', YuGothic', Meiryo, MSGothic" to
ensure consistent fallback?

[kojii, 2017-02-23 09:24:59]

Or only some slaves have Yu Gothic manually installed? Is this a possibility?

[qyearsley, 2017-02-23 19:56:41]

On 2017/02/23 at 09:24:59, kojii wrote:
> Or only some slaves have Yu Gothic manually installed? Is this a possibility?

I'm not sure -- I think that it appears that either some win7 machines have
different fallbacks than others, or some win7 machines have Yu Gothic installed.

If it's the former, then explicitly listing fallbacks will hopefully fix it.

Looking at the actual and expected images for
https://storage.googleapis.com/chromium-layout-test-archives/win7_blink_rel/2...,
it looks like neither of them look like
https://www.fonts.com/font/jiyu-kobo/yu-gothic, but the font size is small so
I'm not sure. Maybe you could confirm whether it's using a fall-back in both
actual and expected?

[qyearsley, 2017-02-23 21:54:07]

On 2017/02/23 at 19:56:41, qyearsley wrote:
> On 2017/02/23 at 09:24:59, kojii wrote:
> > Or only some slaves have Yu Gothic manually installed? Is this a
possibility?
> 
> I'm not sure -- I think that it appears that either some win7 machines have
different fallbacks than others, or some win7 machines have Yu Gothic installed.
> 
> If it's the former, then explicitly listing fallbacks will hopefully fix it.
> 
> Looking at the actual and expected images for
https://storage.googleapis.com/chromium-layout-test-archives/win7_blink_rel/2...,
it looks like neither of them look like
https://www.fonts.com/font/jiyu-kobo/yu-gothic, but the font size is small so
I'm not sure. Maybe you could confirm whether it's using a fall-back in both
actual and expected?

Note, I just uploaded a patch for this CL that changes the test to explicitly
list more fallbacks; if we can get win7_blink_rel to consistently pass after
this, then that should solve the issue.

[qyearsley, 2017-02-26 19:23:06]

The CQ bit was checked by qyearsley@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2017-02-26 19:23:10]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2017-02-26 21:34:29]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2017-02-26 21:34:30]

Dry run: Try jobs failed on following builders:
  win_chromium_rel_ng on master.tryserver.chromium.win (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.win/builders/win_chromium_rel_...)

[qyearsley, 2017-03-02 20:05:36]

Description was changed from

==========
Rebaseline win7/mac10.9 fast/.../text-underline-position-cjk.html

BUG=690486
==========

to

==========
Rebaseline layout test text-underline-position-cjk.html

BUG=690486
==========

[qyearsley, 2017-03-07 18:35:29]

The CQ bit was checked by qyearsley@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2017-03-07 18:36:19]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2017-03-07 19:56:25]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2017-03-07 19:56:27]

Dry run: This issue passed the CQ dry run.

[qyearsley, 2017-03-09 18:20:56]

On 2017/02/23 at 21:54:07, qyearsley wrote:
> On 2017/02/23 at 19:56:41, qyearsley wrote:
> > On 2017/02/23 at 09:24:59, kojii wrote:
> > > Or only some slaves have Yu Gothic manually installed? Is this a
possibility?
> > 
> > I'm not sure -- I think that it appears that either some win7 machines have
different fallbacks than others, or some win7 machines have Yu Gothic installed.
> > 
> > If it's the former, then explicitly listing fallbacks will hopefully fix it.
> > 
> > Looking at the actual and expected images for
https://storage.googleapis.com/chromium-layout-test-archives/win7_blink_rel/2...,
it looks like neither of them look like
https://www.fonts.com/font/jiyu-kobo/yu-gothic, but the font size is small so
I'm not sure. Maybe you could confirm whether it's using a fall-back in both
actual and expected?
> 
> Note, I just uploaded a patch for this CL that changes the test to explicitly
list more fallbacks; if we can get win7_blink_rel to consistently pass after
this, then that should solve the issue.

Looks like win7_blink_rel and the other platforms are passing consistently now
-- win7_blink_dbg failed twice, but for reasons that look unrelated.

OK to commit?

[kojii, 2017-03-10 04:48:05]

Thanks for doing this, lgtm.

[kojii, 2017-03-10 04:48:19]

The CQ bit was checked by kojii@chromium.org

[commit-bot: I haz the power, 2017-03-10 04:48:31]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2017-03-10 06:46:07]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2017-03-10 06:46:08]

Try jobs failed on following builders:
  win_chromium_rel_ng on master.tryserver.chromium.win (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.win/builders/win_chromium_rel_...)

[kojii, 2017-03-10 07:55:38]

ah, "yugothic" is still flaky on...what's the version of "win_chromium_rel_ng"?
Maybe depends on bots?

It's a little surprising YuGothic is flaky, we could remove it if there's no
other good workaround.

[qyearsley, 2017-03-10 16:36:08]

On 2017/03/10 at 07:55:38, kojii wrote:
> ah, "yugothic" is still flaky on...what's the version of
"win_chromium_rel_ng"? Maybe depends on bots?
> 
> It's a little surprising YuGothic is flaky, we could remove it if there's no
other good workaround.

Hm, yeah, at this point it seems like the easiest thing to do is to remove it
:-/ OK to remove it in this CL?

[qyearsley, 2017-03-10 22:25:58]

The CQ bit was checked by qyearsley@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2017-03-10 22:26:19]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[qyearsley, 2017-03-10 23:19:12]

The CQ bit was unchecked by qyearsley@chromium.org

[qyearsley, 2017-03-11 03:02:14]

On 2017/03/10 at 16:36:08, qyearsley wrote:
> On 2017/03/10 at 07:55:38, kojii wrote:
> > ah, "yugothic" is still flaky on...what's the version of
"win_chromium_rel_ng"? Maybe depends on bots?
> > 
> > It's a little surprising YuGothic is flaky, we could remove it if there's no
other good workaround.
> 
> Hm, yeah, at this point it seems like the easiest thing to do is to remove it
:-/ OK to remove it in this CL?

Looking at the slaves for
https://build.chromium.org/p/tryserver.chromium.win/builders/win_chromium_rel_ng,
win_chromium_rel_ng is supposed to be Windows 7. I think that in theory they're
all VMs that are configured the same way.

In the latest patchset, I've removed the text with the YuGothic font family, and
the CQ passed (win7_blink_rel failed a separate unrelated test).

So, if this change is OK, it could be committed any time.

[kojii, 2017-03-11 19:37:27]

yes, lgtm, thank you for patiently working on this trouble.

[kojii, 2017-03-11 19:37:38]

The CQ bit was checked by kojii@chromium.org

[commit-bot: I haz the power, 2017-03-11 19:37:48]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2017-03-11 22:48:28]

CQ is committing da patch.

Bot data: {"patchset_id": 160001, "attempt_start_ts": 1489261058106760,
"parent_rev": "dfe1dfe7bc684f54d0242262fd98ecde0afb29e0", "commit_rev":
"18ffa64cff913d0b56d09276fd72257b74278f61"}

[commit-bot: I haz the power, 2017-03-11 22:49:11]

Description was changed from

==========
Rebaseline layout test text-underline-position-cjk.html

BUG=690486
==========

to

==========
Rebaseline layout test text-underline-position-cjk.html

BUG=690486

Review-Url: https://codereview.chromium.org/2712443003
Cr-Commit-Position: refs/heads/master@{#456289}
Committed:
https://chromium.googlesource.com/chromium/src/+/18ffa64cff913d0b56d09276fd72...
==========

[commit-bot: I haz the power, 2017-03-11 22:49:12]

Committed patchset #9 (id:160001) as
https://chromium.googlesource.com/chromium/src/+/18ffa64cff913d0b56d09276fd72...