Implement direction-specific touch-action values (blink side)

Implement direction-specific touch-action values (blink side)

Add PanLeft, PanRight, PanUp and PanDown to the touch-action
CSS property value set.

These values are only activated in experimental mode.

Depends on chromium side change: https://codereview.chromium.org/1131093002/

BUG=476556
TEST=Tools/Scripts/run-webkit-tests --debug fast/css/touch-action-parsing.html

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=195179

[dtapuska, 2015-05-07 19:01:18]

dtapuska@chromium.org changed reviewers:
+ rbyers@chromium.org

[Rick Byers, 2015-05-08 13:35:47]

rbyers@chromium.org changed reviewers:
+ vollick@chromium.org

[Rick Byers, 2015-05-08 13:35:48]

LGTM with nit.

I updated your CL description to point to the CL this depends on. Obviously
you'll need to wait for that CL to land (I _think_ the blink bots pick up ToT
chromium but it might lag slightly).

https://codereview.chromium.org/1137483003/diff/1/public/web/WebTouchAction.h
File public/web/WebTouchAction.h (right):

https://codereview.chromium.org/1137483003/diff/1/public/web/WebTouchAction.h...
public/web/WebTouchAction.h:36: // Flags for permitted touch actions, specified
in http://www.w3.org/TR/pointerevents/#the-touch-action-css-property.
Nit, please update to the current editors draft
(w3c.github.io/pointerevents/#the-touch-action-css-property) since the new
values are only defined there.

[Rick Byers, 2015-05-08 13:37:00]

+vollick for Source/platform OWNERS

[Rick Byers, 2015-05-08 13:50:21]

By the way, normally we have some test coverage verifying that new APIs aren't
shipped to the web as 'stable' unintentionally.  Mostly we rely on
LayoutTests/webexposed for this and the virtual/stable virtual test suite.  But
in this case we're just adding new property values, not something we detect
automatically.

I gave some thought to how we might add some test coverage here.  Some options:
- make fast/css/touch-action-parsing run in virtual/stable and expect some
failures for now
- add a new webexposed/css-syntax test which has a list of 'style' declarations
and for each outputs whether it parses or not.  Then we'd have different
expectations in stable/

The latter seems somewhat appealing to me, but it may be overkill.  This isn't a
new problem and if we have to rely on people to explicitly add entries for the
new css syntax they're protecting against, it's unlikely that this pattern would
catch and prevent any accidental shipping.  I suppose we could have some
additional automation that verified that each token in CSSValueKeywords.in is
represented by this test somewhere.

Anyway, I'm not convinced the value of adding additional test coverage here is
worth the cost.  So I think we should just proceed with this CL as-is.  Ian, any
opinion?

[dtapuska, 2015-05-08 13:52:46]

https://codereview.chromium.org/1137483003/diff/1/public/web/WebTouchAction.h
File public/web/WebTouchAction.h (right):

https://codereview.chromium.org/1137483003/diff/1/public/web/WebTouchAction.h...
public/web/WebTouchAction.h:36: // Flags for permitted touch actions, specified
in http://www.w3.org/TR/pointerevents/#the-touch-action-css-property.
On 2015/05/08 13:35:48, Rick Byers wrote:
> Nit, please update to the current editors draft
> (w3c.github.io/pointerevents/#the-touch-action-css-property) since the new
> values are only defined there.

Done.

[Ian Vollick, 2015-05-08 14:00:26]

On 2015/05/08 at 13:52:46, dtapuska wrote:
> https://codereview.chromium.org/1137483003/diff/1/public/web/WebTouchAction.h
> File public/web/WebTouchAction.h (right):
> 
>
https://codereview.chromium.org/1137483003/diff/1/public/web/WebTouchAction.h...
> public/web/WebTouchAction.h:36: // Flags for permitted touch actions,
specified in http://www.w3.org/TR/pointerevents/#the-touch-action-css-property.
> On 2015/05/08 13:35:48, Rick Byers wrote:
> > Nit, please update to the current editors draft
> > (w3c.github.io/pointerevents/#the-touch-action-css-property) since the new
> > values are only defined there.
> 
> Done.

Source/platform lgtm

[dtapuska, 2015-05-11 13:30:59]

The CQ bit was checked by dtapuska@chromium.org

[dtapuska, 2015-05-11 13:31:00]

The patchset sent to the CQ was uploaded after l-g-t-m from rbyers@chromium.org
Link to the patchset: https://codereview.chromium.org/1137483003/#ps20001
(title: "Fix minor nit from patch set 1")

[commit-bot: I haz the power, 2015-05-11 13:31:15]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1137483003/20001

[commit-bot: I haz the power, 2015-05-11 14:51:48]

Committed patchset #2 (id:20001) as
https://src.chromium.org/viewvc/blink?view=rev&revision=195179