Remove dependency on element in applying color properties

Remove dependency on element in applying color properties

This patch removes the state.element().isLink() calls in applying color
properties for visited links. This has a small behavioural change where
the -webkit-link color keyword will now be the visited link color for a
visited link descendent (bug 82498). I don't think we can currently test
visited link colors in a layout test.

This also fixes a crash when trying to set color in a pseudo-element
animation on a link element (or descendent), since we go through the
StyleBuilder with a null element. Note that the other places in the
StyleBuilder where we access the element already check for null.

BUG=82498, 343048
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=168435

[dstockwell, 2014-02-28 02:02:39]

I don't really understand 82498. It was marked as WontFix as the current
behavior was believed to be correct?

Why was state.element().isLink() being checked before?

[Timothy Loh, 2014-02-28 02:57:26]

On 2014/02/28 02:02:39, dstockwell wrote:
> I don't really understand 82498. It was marked as WontFix as the current
> behavior was believed to be correct?

The bug was incorrectly closed, current versions of chrome still have the
problem described in the bug. Honestly we probably don't care too much about
fixing the behaviour of a prefixed keyword, but for the sake of removing the
element dependency I would rather fix the crash this way instead of
null-checking the element.

> Why was state.element().isLink() being checked before?

Unclear; it looks like it was originally added in r57292
(https://bugs.webkit.org/attachment.cgi?id=52778&action=prettypatch) but I can't
tell why.

[dstockwell, 2014-03-04 01:51:59]

lgtm

Looks like it was just trying to avoid accidental leakage. A follow up change
might be to replace the forVisitedLink param with a separate function:
visitedLinkColorFromPrimitiveValue.

[Timothy Loh, 2014-03-04 02:38:49]

The CQ bit was checked by timloh@chromium.org

[commit-bot: I haz the power, 2014-03-04 02:39:00]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/timloh@chromium.org/179363005/40002

[commit-bot: I haz the power, 2014-03-04 20:41:00]

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

[commit-bot: I haz the power, 2014-03-04 20:41:00]

Retried try job too often on win_layout for step(s) webkit_lint
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=win_layout...

[Timothy Loh, 2014-03-04 23:07:30]

The CQ bit was checked by timloh@chromium.org

[commit-bot: I haz the power, 2014-03-04 23:08:06]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/timloh@chromium.org/179363005/40002

[commit-bot: I haz the power, 2014-03-05 06:06:38]

Change committed as 168435