New flingCancel flag to prevent fling boosting.

New flingCancel flag to prevent fling boosting.

This flag is used by Chromium in order to prevent flingCancel
from being used as a part of the fling boost events sequence.
Previously (without this flag), when Chromium called native
cancelFling method, flingCancel event was deferred,
but the proper behaviour should be to cancel fling instantly.

See: https://codereview.chromium.org/792833002/

BUG=440886
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=187163

[wjarosik, 2014-12-10 15:13:15]

wjarosik@opera.com changed reviewers:
+ jdduke@chromium.org

[jdduke (slow), 2014-12-10 18:53:40]

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

[jdduke (slow), 2014-12-10 18:53:41]

Could you also add a link to the Chromium patch?

+rbyers@ for owner review. This approach seems pretty reasonable (particularly
as we may eventually support boosting on the main thread as well). 

The alternative we discussed was exposing an additional "synthetic" gesture
source type, implicitly avoiding accumulation between synthetic and touchscreen
gestures, but that will probably involve a good number of changes to update all
the places we branch on the source type.

[Rick Byers, 2014-12-10 19:01:02]

Please also add a BUG that describes the problem being solved.

LGTM with nit.

It seems a little awkward that flingCancel now means "maybe cancel" and we're
adding a "yes, really I want you to cancel" bit. At some point we should
consider overhauling the cancellation system.  But this seems like an OK
incremental step.

https://codereview.chromium.org/789223002/diff/1/public/web/WebInputEvent.h
File public/web/WebInputEvent.h (right):

https://codereview.chromium.org/789223002/diff/1/public/web/WebInputEvent.h#n...
public/web/WebInputEvent.h:471: // Don't treat flingCancel as a part of fling
boost
Nit: Add "if true" since we don't want to imply that most flingCancel events
will have this set.

[wjarosik, 2014-12-11 16:33:45]

On 2014/12/10 18:53:41, jdduke wrote:
> Could you also add a link to the Chromium patch?
> 
> +rbyers@ for owner review. This approach seems pretty reasonable (particularly
> as we may eventually support boosting on the main thread as well). 
> 
> The alternative we discussed was exposing an additional "synthetic" gesture
> source type, implicitly avoiding accumulation between synthetic and
touchscreen
> gestures, but that will probably involve a good number of changes to update
all
> the places we branch on the source type.

Link added.

[wjarosik, 2014-12-11 16:34:07]

On 2014/12/10 19:01:02, Rick Byers wrote:
> Please also add a BUG that describes the problem being solved.
> 
> LGTM with nit.
> 
> It seems a little awkward that flingCancel now means "maybe cancel" and we're
> adding a "yes, really I want you to cancel" bit. At some point we should
> consider overhauling the cancellation system.  But this seems like an OK
> incremental step.
> 
> https://codereview.chromium.org/789223002/diff/1/public/web/WebInputEvent.h
> File public/web/WebInputEvent.h (right):
> 
>
https://codereview.chromium.org/789223002/diff/1/public/web/WebInputEvent.h#n...
> public/web/WebInputEvent.h:471: // Don't treat flingCancel as a part of fling
> boost
> Nit: Add "if true" since we don't want to imply that most flingCancel events
> will have this set.

OK, thanks. It's done.

[wjarosik, 2014-12-11 16:45:14]

Changes made (I haven't clicked Publish+Mail earlier so I'm doing it now :)).

[Rick Byers, 2014-12-12 21:23:58]

On 2014/12/11 16:45:14, wjarosik wrote:
> Changes made (I haven't clicked Publish+Mail earlier so I'm doing it now :)).

Still LGTM - you can CQ it (didn't sound like Jared had any outstanding
concerns).

[jdduke (slow), 2014-12-15 16:51:44]

The CQ bit was checked by jdduke@chromium.org

[commit-bot: I haz the power, 2014-12-15 16:52:32]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/789223002/20001

[commit-bot: I haz the power, 2014-12-15 18:40:38]

Committed patchset #2 (id:20001) as
https://src.chromium.org/viewvc/blink?view=rev&revision=187163