Partially revert r181835 to fix link highlighting

Partially revert r181835 to fix link highlighting

The changes in LinkHighlight code in http://crrev.com/541913002 causes
the link-highlights to disappear in many cases (see the bug). Eight
changes have been committed to the same code afterwards, making it
impossible to fully revert the target CL. The partial revert here fixes
the problem.

BUG=430244
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=185245

[mustaq, 2014-11-12 16:43:58]

mustaq@chromium.org changed reviewers:
+ rbyers@chromium.org, schenney@chromium.org

[mustaq, 2014-11-12 16:45:10]

mustaq@chromium.org changed reviewers:
+ chrishtr@chromium.org

[mustaq, 2014-11-12 16:45:10]

chrishtr@chromium.org: Need approval

[Stephen Chennney, 2014-11-12 17:27:59]

LGTM. Meanwhile I'll try to figure out why it fails.

[Stephen Chennney, 2014-11-12 17:29:20]

On 2014/11/12 17:27:59, Stephen Chenney wrote:
> LGTM. Meanwhile I'll try to figure out why it fails.

Bet it's a problem with the clip rect.

[chrishtr, 2014-11-12 17:31:54]

The CQ bit was checked by chrishtr@chromium.org

[chrishtr, 2014-11-12 17:31:55]

lgtm

[commit-bot: I haz the power, 2014-11-12 17:32:12]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/719053004/1

[Rick Byers, 2014-11-12 18:02:38]

Thanks for tracking this down Mustaq.  As discussed we'll try to get this
low-risk merged into M39 (and M40) ASAP to avoid sending this regression out to
stable.

Can you please file a bug to track adding test coverage for this?  We've got a
lot of different tests for link highlighting, why didn't any of them catch this?

We can leave it to Stephen to decide whether he wants to file a bug to re-land
the code you're reverting here.

Also note I tweaked the title/summary slightly to make it a little more
descriptive (since people often review lists of changes looking only at the
summary).

[mustaq, 2014-11-12 18:16:22]

On 2014/11/12 18:02:38, Rick Byers wrote:
> Thanks for tracking this down Mustaq.  As discussed we'll try to get this
> low-risk merged into M39 (and M40) ASAP to avoid sending this regression out
to
> stable.
> 
> Can you please file a bug to track adding test coverage for this?  We've got a
> lot of different tests for link highlighting, why didn't any of them catch
this?
> 
> We can leave it to Stephen to decide whether he wants to file a bug to re-land
> the code you're reverting here.
> 
> Also note I tweaked the title/summary slightly to make it a little more
> descriptive (since people often review lists of changes looking only at the
> summary).

David has filed http://crbug.com/432560 to track test coverage.

[commit-bot: I haz the power, 2014-11-12 19:23:32]

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

[commit-bot: I haz the power, 2014-11-12 19:23:32]

Try jobs failed on following builders:
  win_blink_rel on tryserver.blink
(http://build.chromium.org/p/tryserver.blink/builders/win_blink_rel/builds/36126)

[mustaq, 2014-11-12 21:15:53]

The CQ bit was checked by mustaq@chromium.org

[commit-bot: I haz the power, 2014-11-12 21:16:10]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/719053004/1

[commit-bot: I haz the power, 2014-11-12 21:16:55]

Committed patchset #1 (id:1) as 185245