Make parseColumnsShorthand not mark 'auto' initial value as implicit

Make parseColumnsShorthand not mark 'auto' initial value as implicit

There does not seem to be a reason to use implicit for the 'auto' initial
value. Also this way we can simplify the logic somewhat.

BUG=471917
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=201695

[rwlbuis, 2015-09-02 22:00:55]

rob.buis@samsung.com changed reviewers:
+ timloh@chromium.org

[rwlbuis, 2015-09-02 22:00:56]

PTAL.

[rwlbuis, 2015-09-02 22:32:43]

I updated the Description likely. BTW I assume with issue 471917 the object is
to completely remove implicit ? Or only 'initial' marked as implicit?

[Timothy Loh, 2015-09-03 01:02:45]

lgtm

On 2015/09/02 22:32:43, rwlbuis wrote:
> I updated the Description likely. BTW I assume with issue 471917 the object is
> to completely remove implicit ? Or only 'initial' marked as implicit?

The bug is about removing implicit completely, yep. I initially thought it would
be fairly straightforwards, but due to our handling of shorthands (bug 501673)
it will probably actually take a bit of work to resolve. At this stage I'm not
sure we want to fix it right now since there are behaviour changes and we'd need
to check it doesn't hurt performance. I think the CSSParserValue removal is the
main priority for the CSS parser at the moment :-)

[rwlbuis, 2015-09-03 01:42:32]

On 2015/09/03 01:02:45, Timothy Loh wrote:
> lgtm
> 
> On 2015/09/02 22:32:43, rwlbuis wrote:
> > I updated the Description likely. BTW I assume with issue 471917 the object
is
> > to completely remove implicit ? Or only 'initial' marked as implicit?
> 
> The bug is about removing implicit completely, yep. I initially thought it
would
> be fairly straightforwards, but due to our handling of shorthands (bug 501673)
> it will probably actually take a bit of work to resolve. At this stage I'm not
> sure we want to fix it right now since there are behaviour changes and we'd
need
> to check it doesn't hurt performance. I think the CSSParserValue removal is
the
> main priority for the CSS parser at the moment :-)

Yeah the implicit stuff sounds like a nice goal but not trivial. Sure
CSSParserValue removal is my highest prio for the CSS stuff I do :)

[rwlbuis, 2015-09-03 01:42:39]

The CQ bit was checked by rob.buis@samsung.com

[commit-bot: I haz the power, 2015-09-03 01:42:53]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1311503006/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1311503006/40001

[commit-bot: I haz the power, 2015-09-03 03:53:17]

Committed patchset #3 (id:40001) as
https://src.chromium.org/viewvc/blink?view=rev&revision=201695