Redefined the bit WebTouchEvent.causesScrollingIfUncanceled.

Redefined the bit WebTouchEvent.causesScrollingIfUncanceled.

The "causesScrollingIfUncanceled" name is confusing because this bit
essentially means that a touchmove has exceeded the touch slop region.
A touchmove with this bit set doesn't necessarily cause scrolling,
even when uncanceled, e.g. when the touch sequence started on a node
with "touch-action:none".

This CL redefines the bit: this bit remains unset as long as the
touch-point is within the slop region, and is set after the first
touchmove beyond the slop region.

The bit is set by gesture-provider, and currently used in two places:
for touchmove slop suppression and for suppressing PointerEvent firing
for events that cause panning/zooming. This new definition was chosen
to be useful for both the current slop suppression and to enable us to
detect tap-like touchend events from blink (eg. for crbug.com/582140).
We are fixing the PointerEvent use case through crbug.com/567740.

BUG=582140, 567740
Committed: https://crrev.com/620be1f0cde37e56b8f8eb9e238e3d28b6471fc7
Cr-Commit-Position: refs/heads/master@{#374204}

[mustaq, 2016-01-29 17:06:10]

The CQ bit was checked by mustaq@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-01-29 17:06:30]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1645613007/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1645613007/1

[mustaq, 2016-01-29 17:06:53]

Description was changed from

==========
Clarified the bit name WebTouchEvent.causesScrollingIfUncanceled.

The "causesScrollingIfUncanceled" name is confusing because this bit
essentially means that a touchmove has exceeded the touch slop region.
A touchmove with this bit set may not cause scrolling on
prevent-default, e.g. when the touch sequence started on a node with
"touch-action:none".

The bit is set by gesture-provider, and currently used in two places:
for touchmove slop suppression and for suppressing PointerEvent firing
for events that cause panning/zooming. The renaming here conforms to the
first use, and we are addressing the second use through
crbug.com(567740).

BUG=582140,567740
==========

to

==========
Clarified the bit name WebTouchEvent.causesScrollingIfUncanceled.

The "causesScrollingIfUncanceled" name is confusing because this bit
essentially means that a touchmove has exceeded the touch slop region.
A touchmove with this bit set may not cause scrolling on
prevent-default, e.g. when the touch sequence started on a node with
"touch-action:none".

The bit is set by gesture-provider, and currently used in two places:
for touchmove slop suppression and for suppressing PointerEvent firing
for events that cause panning/zooming. The renaming here conforms to the
first use, and we are fixing the second use through crbug.com(567740).

BUG=582140,567740
==========

[commit-bot: I haz the power, 2016-01-29 17:09:17]

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

[commit-bot: I haz the power, 2016-01-29 17:09:18]

Dry run: Try jobs failed on following builders:
  mac_chromium_gn_rel on tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/mac_chromium_gn_r...)

[mustaq, 2016-01-29 19:20:17]

The CQ bit was checked by mustaq@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-01-29 19:21:19]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1645613007/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1645613007/20001

[mustaq, 2016-01-29 19:22:45]

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

[mustaq, 2016-01-29 20:11:16]

On 2016/01/29 19:22:45, mustaq wrote:

ptal

[commit-bot: I haz the power, 2016-01-29 21:22:26]

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

[commit-bot: I haz the power, 2016-01-29 21:22:26]

Dry run: This issue passed the CQ dry run.

[Rick Byers, 2016-02-02 04:24:16]

Thanks for doing this.  At a minimum let's cripen up the definition a bit to
make it completely accurate.  If you want to avoid changing the actual
implementation of the bit beyond what you need (eg. to let us have a simpler /
more useful definition) that's OK - I can do it as part of my patch that will
depend on it.

https://codereview.chromium.org/1645613007/diff/20001/third_party/WebKit/publ...
File third_party/WebKit/public/web/WebInputEvent.h (right):

https://codereview.chromium.org/1645613007/diff/20001/third_party/WebKit/publ...
third_party/WebKit/public/web/WebInputEvent.h:586: // fling-generating touchend
events.
Two questions that I think should be covered by this comment:
1) How does this behave for multi-finger scenarios?  Is it "true if ANY finger
has moved beyond the slop region"?
2) What about a touchend that occurs during a scroll but doesn't cause a fling
(due to 0 velocity)?

Also for a long drag where all the touch events are consumed, it seems odd to
have this be false for the touchend, even though it was true for the touchmoves.
 Or perhaps the "fling-generating" means prior to touch disposition gesture
filtering?

https://codereview.chromium.org/1645613007/diff/20001/ui/events/gesture_detec...
File ui/events/gesture_detection/filtered_gesture_provider_unittest.cc (right):

https://codereview.chromium.org/1645613007/diff/20001/ui/events/gesture_detec...
ui/events/gesture_detection/filtered_gesture_provider_unittest.cc:25:
TEST_F(FilteredGestureProviderTest, TouchMayGenerateScroll) {
nit: call it TouchMovedBeyondSlopRegion just to clarify that we're testing this
bit.

https://codereview.chromium.org/1645613007/diff/20001/ui/events/gesture_detec...
ui/events/gesture_detection/filtered_gesture_provider_unittest.cc:43: //
Exceeding the slop should triggering scrolling and be reflected in the API.
nit: the comments probably need updating (helps to clarify the exact semantics
of this bit).

https://codereview.chromium.org/1645613007/diff/20001/ui/events/gesture_detec...
ui/events/gesture_detection/filtered_gesture_provider_unittest.cc:52:
EXPECT_TRUE(result.succeeded);
Oh, this doesn't match the definition!  Need to either tweak the definition or
change the implementation.

https://codereview.chromium.org/1645613007/diff/20001/ui/events/gesture_detec...
ui/events/gesture_detection/filtered_gesture_provider_unittest.cc:65:
EXPECT_FALSE(result.moved_beyond_slop_region);
IIRC there's nothing actually depending on the value of this bit for touchend
events today, right?  If so I think it would be simpler (and help my use case)
if we updated the definition to make this case true (i.e. one the bit became
true it was always true).

[mustaq, 2016-02-04 21:16:21]

Description was changed from

==========
Clarified the bit name WebTouchEvent.causesScrollingIfUncanceled.

The "causesScrollingIfUncanceled" name is confusing because this bit
essentially means that a touchmove has exceeded the touch slop region.
A touchmove with this bit set may not cause scrolling on
prevent-default, e.g. when the touch sequence started on a node with
"touch-action:none".

The bit is set by gesture-provider, and currently used in two places:
for touchmove slop suppression and for suppressing PointerEvent firing
for events that cause panning/zooming. The renaming here conforms to the
first use, and we are fixing the second use through crbug.com(567740).

BUG=582140,567740
==========

to

==========
Redefined the bit WebTouchEvent.causesScrollingIfUncanceled.

The "causesScrollingIfUncanceled" name is confusing because this bit
essentially means that a touchmove has exceeded the touch slop region.
A touchmove with this bit set may not cause scrolling on
prevent-default, e.g. when the touch sequence started on a node with
"touch-action:none".

This CL redefines the bit: this bit remains unset as long as the
touch-point is within the slop region, and is set after the first
touchmove beyond the slop region.

The bit is set by gesture-provider, and currently used in two places:
for touchmove slop suppression and for suppressing PointerEvent firing
for events that cause panning/zooming. The change here conforms to the
first use, and we are fixing the second use through crbug.com(567740).

BUG=582140,567740
==========

[mustaq, 2016-02-04 21:17:05]

Description was changed from

==========
Redefined the bit WebTouchEvent.causesScrollingIfUncanceled.

The "causesScrollingIfUncanceled" name is confusing because this bit
essentially means that a touchmove has exceeded the touch slop region.
A touchmove with this bit set may not cause scrolling on
prevent-default, e.g. when the touch sequence started on a node with
"touch-action:none".

This CL redefines the bit: this bit remains unset as long as the
touch-point is within the slop region, and is set after the first
touchmove beyond the slop region.

The bit is set by gesture-provider, and currently used in two places:
for touchmove slop suppression and for suppressing PointerEvent firing
for events that cause panning/zooming. The change here conforms to the
first use, and we are fixing the second use through crbug.com(567740).

BUG=582140,567740
==========

to

==========
Redefined the bit WebTouchEvent.causesScrollingIfUncanceled.

The "causesScrollingIfUncanceled" name is confusing because this bit
essentially means that a touchmove has exceeded the touch slop region.
A touchmove with this bit set may not cause scrolling on
prevent-default, e.g. when the touch sequence started on a node with
"touch-action:none".

This CL redefines the bit: this bit remains unset as long as the
touch-point is within the slop region, and is set after the first
touchmove beyond the slop region.

The bit is set by gesture-provider, and currently used in two places:
for touchmove slop suppression and for suppressing PointerEvent firing
for events that cause panning/zooming. The change here conforms to the
first use, and we are fixing the second use through crbug.com/567740.

BUG=582140,567740
==========

[mustaq, 2016-02-04 21:17:22]

ptal

https://codereview.chromium.org/1645613007/diff/20001/third_party/WebKit/publ...
File third_party/WebKit/public/web/WebInputEvent.h (right):

https://codereview.chromium.org/1645613007/diff/20001/third_party/WebKit/publ...
third_party/WebKit/public/web/WebInputEvent.h:586: // fling-generating touchend
events.
On 2016/02/02 04:24:16, Rick Byers wrote:
> Two questions that I think should be covered by this comment:
> 1) How does this behave for multi-finger scenarios?  Is it "true if ANY finger
> has moved beyond the slop region"?
> 2) What about a touchend that occurs during a scroll but doesn't cause a fling
> (due to 0 velocity)?
> 
> Also for a long drag where all the touch events are consumed, it seems odd to
> have this be false for the touchend, even though it was true for the
touchmoves.
>  Or perhaps the "fling-generating" means prior to touch disposition gesture
> filtering?

Done.

https://codereview.chromium.org/1645613007/diff/20001/ui/events/gesture_detec...
File ui/events/gesture_detection/filtered_gesture_provider_unittest.cc (right):

https://codereview.chromium.org/1645613007/diff/20001/ui/events/gesture_detec...
ui/events/gesture_detection/filtered_gesture_provider_unittest.cc:25:
TEST_F(FilteredGestureProviderTest, TouchMayGenerateScroll) {
On 2016/02/02 04:24:16, Rick Byers wrote:
> nit: call it TouchMovedBeyondSlopRegion just to clarify that we're testing
this
> bit.

Done. Also added a multitouch test in a separate method.

https://codereview.chromium.org/1645613007/diff/20001/ui/events/gesture_detec...
ui/events/gesture_detection/filtered_gesture_provider_unittest.cc:43: //
Exceeding the slop should triggering scrolling and be reflected in the API.
On 2016/02/02 04:24:16, Rick Byers wrote:
> nit: the comments probably need updating (helps to clarify the exact semantics
> of this bit).

Done.

https://codereview.chromium.org/1645613007/diff/20001/ui/events/gesture_detec...
ui/events/gesture_detection/filtered_gesture_provider_unittest.cc:52:
EXPECT_TRUE(result.succeeded);
On 2016/02/02 04:24:16, Rick Byers wrote:
> Oh, this doesn't match the definition!  Need to either tweak the definition or
> change the implementation.

Done.

https://codereview.chromium.org/1645613007/diff/20001/ui/events/gesture_detec...
ui/events/gesture_detection/filtered_gesture_provider_unittest.cc:65:
EXPECT_FALSE(result.moved_beyond_slop_region);
On 2016/02/02 04:24:16, Rick Byers wrote:
> IIRC there's nothing actually depending on the value of this bit for touchend
> events today, right?  If so I think it would be simpler (and help my use case)
> if we updated the definition to make this case true (i.e. one the bit became
> true it was always true).

Done.

[Rick Byers, 2016-02-04 22:29:20]

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

[Rick Byers, 2016-02-04 22:29:21]

LGTM - thanks!
+tdresser for OWNERS

For the record, this tweaked definition was chosen to be useful for both the
current slop suppression and to enable us to detect tap-like touchend events
from blink (eg. for http://crbug.com/582140)

nit: In the CL description "A touchmove with this bit set may not cause
scrolling on prevent-default, e.g. when the touch sequence started on a node
with "touch-action:none"."  A touchmove with preventDefault never causes
scrolling.

Perhaps you meant something like "A touchmove with this bit set doesn't
necessarily cause scrolling, even when uncanceled, e.g. ..."?

[tdresser, 2016-02-05 14:19:00]

https://codereview.chromium.org/1645613007/diff/40001/ui/events/blink/blink_e...
File ui/events/blink/blink_event_util.cc (right):

https://codereview.chromium.org/1645613007/diff/40001/ui/events/blink/blink_e...
ui/events/blink/blink_event_util.cc:208: bool IsTouchSequenceStart(const
MotionEvent& event) {
Isn't this equivalent to event.GetAction() == ACTION_DOWN?

http://developer.android.com/reference/android/view/MotionEvent.html#ACTION_DOWN

https://codereview.chromium.org/1645613007/diff/40001/ui/events/gesture_detec...
File ui/events/gesture_detection/filtered_gesture_provider.cc (right):

https://codereview.chromium.org/1645613007/diff/40001/ui/events/gesture_detec...
ui/events/gesture_detection/filtered_gesture_provider.cc:36:
last_touch_event_moved_beyond_slop_region_ = false;
The exact semantics of this bit aren't clear to me.

It looks like this bit is true if _any_ touch event moved beyond the slop
region, not just if the most recent one did.

[mustaq, 2016-02-05 15:29:18]

@tdresser: ptal

https://codereview.chromium.org/1645613007/diff/40001/ui/events/blink/blink_e...
File ui/events/blink/blink_event_util.cc (right):

https://codereview.chromium.org/1645613007/diff/40001/ui/events/blink/blink_e...
ui/events/blink/blink_event_util.cc:208: bool IsTouchSequenceStart(const
MotionEvent& event) {
On 2016/02/05 14:19:00, tdresser wrote:
> Isn't this equivalent to event.GetAction() == ACTION_DOWN?
> 
>
http://developer.android.com/reference/android/view/MotionEvent.html#ACTION_DOWN

Thanks for the catch: I was trying to return true on multi-tap, but my
filtered_gesture_provider logic works as-is with your suggestion.

https://codereview.chromium.org/1645613007/diff/40001/ui/events/gesture_detec...
File ui/events/gesture_detection/filtered_gesture_provider.cc (right):

https://codereview.chromium.org/1645613007/diff/40001/ui/events/gesture_detec...
ui/events/gesture_detection/filtered_gesture_provider.cc:36:
last_touch_event_moved_beyond_slop_region_ = false;
On 2016/02/05 14:19:00, tdresser wrote:
> The exact semantics of this bit aren't clear to me.
> 
> It looks like this bit is true if _any_ touch event moved beyond the slop
> region, not just if the most recent one did.

Done.

[mustaq, 2016-02-05 15:32:20]

Description was changed from

==========
Redefined the bit WebTouchEvent.causesScrollingIfUncanceled.

The "causesScrollingIfUncanceled" name is confusing because this bit
essentially means that a touchmove has exceeded the touch slop region.
A touchmove with this bit set may not cause scrolling on
prevent-default, e.g. when the touch sequence started on a node with
"touch-action:none".

This CL redefines the bit: this bit remains unset as long as the
touch-point is within the slop region, and is set after the first
touchmove beyond the slop region.

The bit is set by gesture-provider, and currently used in two places:
for touchmove slop suppression and for suppressing PointerEvent firing
for events that cause panning/zooming. The change here conforms to the
first use, and we are fixing the second use through crbug.com/567740.

BUG=582140,567740
==========

to

==========
Redefined the bit WebTouchEvent.causesScrollingIfUncanceled.

The "causesScrollingIfUncanceled" name is confusing because this bit
essentially means that a touchmove has exceeded the touch slop region.
A touchmove with this bit set doesn't necessarily cause scrolling,
even when uncanceled, e.g. when the touch sequence started on a node
with "touch-action:none".

This CL redefines the bit: this bit remains unset as long as the
touch-point is within the slop region, and is set after the first
touchmove beyond the slop region.

The bit is set by gesture-provider, and currently used in two places:
for touchmove slop suppression and for suppressing PointerEvent firing
for events that cause panning/zooming. The change here conforms to the
first use, and we are fixing the second use through crbug.com/567740.

BUG=582140,567740
==========

[tdresser, 2016-02-05 15:42:44]

LGTM, thanks.

[mustaq, 2016-02-05 16:17:29]

Description was changed from

==========
Redefined the bit WebTouchEvent.causesScrollingIfUncanceled.

The "causesScrollingIfUncanceled" name is confusing because this bit
essentially means that a touchmove has exceeded the touch slop region.
A touchmove with this bit set doesn't necessarily cause scrolling,
even when uncanceled, e.g. when the touch sequence started on a node
with "touch-action:none".

This CL redefines the bit: this bit remains unset as long as the
touch-point is within the slop region, and is set after the first
touchmove beyond the slop region.

The bit is set by gesture-provider, and currently used in two places:
for touchmove slop suppression and for suppressing PointerEvent firing
for events that cause panning/zooming. The change here conforms to the
first use, and we are fixing the second use through crbug.com/567740.

BUG=582140,567740
==========

to

==========
Redefined the bit WebTouchEvent.causesScrollingIfUncanceled.

The "causesScrollingIfUncanceled" name is confusing because this bit
essentially means that a touchmove has exceeded the touch slop region.
A touchmove with this bit set doesn't necessarily cause scrolling,
even when uncanceled, e.g. when the touch sequence started on a node
with "touch-action:none".

This CL redefines the bit: this bit remains unset as long as the
touch-point is within the slop region, and is set after the first
touchmove beyond the slop region.

The bit is set by gesture-provider, and currently used in two places:
for touchmove slop suppression and for suppressing PointerEvent firing
for events that cause panning/zooming. This new definition was chosen
to be useful for both the current slop suppression and to enable us to
detect tap-like touchend events from blink (eg. for crbug.com/582140).
We are fixing the PointerEvent use case through crbug.com/567740.

BUG=582140,567740
==========

[mustaq, 2016-02-05 16:26:16]

mustaq@chromium.org changed reviewers:
+ dtrainor@chromium.org, sadrul@chromium.org

[mustaq, 2016-02-05 16:26:17]

dtrainor@chromium.org: Please review changes in
    blimp/client/feature/compositor/blimp_input_manager.cc

sadrul@chromium.org: Please review changes in
    components/html_viewer/touch_handler.cc

[sadrul, 2016-02-08 15:47:32]

lgtm

[David Trainor- moved to gerrit, 2016-02-08 15:48:16]

blimp/ lgtm

[mustaq, 2016-02-08 15:51:50]

The CQ bit was checked by mustaq@chromium.org

[mustaq, 2016-02-08 15:51:51]

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/1645613007/#ps60001
(title: " ")

[commit-bot: I haz the power, 2016-02-08 15:52:36]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1645613007/60001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1645613007/60001

[commit-bot: I haz the power, 2016-02-08 16:34:10]

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

[commit-bot: I haz the power, 2016-02-08 16:34:11]

Try jobs failed on following builders:
  linux_chromium_chromeos_ozone_rel_ng on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)

[mustaq, 2016-02-08 18:33:51]

The CQ bit was checked by mustaq@chromium.org

[mustaq, 2016-02-08 18:33:52]

The patchset sent to the CQ was uploaded after l-g-t-m from
dtrainor@chromium.org, sadrul@chromium.org, tdresser@chromium.org,
rbyers@chromium.org
Link to the patchset: https://codereview.chromium.org/1645613007/#ps80001
(title: "Fixed an aura unittest.")

[commit-bot: I haz the power, 2016-02-08 18:39:27]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1645613007/80001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1645613007/80001

[commit-bot: I haz the power, 2016-02-08 22:47:56]

Description was changed from

==========
Redefined the bit WebTouchEvent.causesScrollingIfUncanceled.

The "causesScrollingIfUncanceled" name is confusing because this bit
essentially means that a touchmove has exceeded the touch slop region.
A touchmove with this bit set doesn't necessarily cause scrolling,
even when uncanceled, e.g. when the touch sequence started on a node
with "touch-action:none".

This CL redefines the bit: this bit remains unset as long as the
touch-point is within the slop region, and is set after the first
touchmove beyond the slop region.

The bit is set by gesture-provider, and currently used in two places:
for touchmove slop suppression and for suppressing PointerEvent firing
for events that cause panning/zooming. This new definition was chosen
to be useful for both the current slop suppression and to enable us to
detect tap-like touchend events from blink (eg. for crbug.com/582140).
We are fixing the PointerEvent use case through crbug.com/567740.

BUG=582140,567740
==========

to

==========
Redefined the bit WebTouchEvent.causesScrollingIfUncanceled.

The "causesScrollingIfUncanceled" name is confusing because this bit
essentially means that a touchmove has exceeded the touch slop region.
A touchmove with this bit set doesn't necessarily cause scrolling,
even when uncanceled, e.g. when the touch sequence started on a node
with "touch-action:none".

This CL redefines the bit: this bit remains unset as long as the
touch-point is within the slop region, and is set after the first
touchmove beyond the slop region.

The bit is set by gesture-provider, and currently used in two places:
for touchmove slop suppression and for suppressing PointerEvent firing
for events that cause panning/zooming. This new definition was chosen
to be useful for both the current slop suppression and to enable us to
detect tap-like touchend events from blink (eg. for crbug.com/582140).
We are fixing the PointerEvent use case through crbug.com/567740.

BUG=582140,567740
==========

[commit-bot: I haz the power, 2016-02-08 22:47:57]

Committed patchset #5 (id:80001)

[commit-bot: I haz the power, 2016-02-08 22:48:57]

Description was changed from

==========
Redefined the bit WebTouchEvent.causesScrollingIfUncanceled.

The "causesScrollingIfUncanceled" name is confusing because this bit
essentially means that a touchmove has exceeded the touch slop region.
A touchmove with this bit set doesn't necessarily cause scrolling,
even when uncanceled, e.g. when the touch sequence started on a node
with "touch-action:none".

This CL redefines the bit: this bit remains unset as long as the
touch-point is within the slop region, and is set after the first
touchmove beyond the slop region.

The bit is set by gesture-provider, and currently used in two places:
for touchmove slop suppression and for suppressing PointerEvent firing
for events that cause panning/zooming. This new definition was chosen
to be useful for both the current slop suppression and to enable us to
detect tap-like touchend events from blink (eg. for crbug.com/582140).
We are fixing the PointerEvent use case through crbug.com/567740.

BUG=582140,567740
==========

to

==========
Redefined the bit WebTouchEvent.causesScrollingIfUncanceled.

The "causesScrollingIfUncanceled" name is confusing because this bit
essentially means that a touchmove has exceeded the touch slop region.
A touchmove with this bit set doesn't necessarily cause scrolling,
even when uncanceled, e.g. when the touch sequence started on a node
with "touch-action:none".

This CL redefines the bit: this bit remains unset as long as the
touch-point is within the slop region, and is set after the first
touchmove beyond the slop region.

The bit is set by gesture-provider, and currently used in two places:
for touchmove slop suppression and for suppressing PointerEvent firing
for events that cause panning/zooming. This new definition was chosen
to be useful for both the current slop suppression and to enable us to
detect tap-like touchend events from blink (eg. for crbug.com/582140).
We are fixing the PointerEvent use case through crbug.com/567740.

BUG=582140,567740

Committed: https://crrev.com/620be1f0cde37e56b8f8eb9e238e3d28b6471fc7
Cr-Commit-Position: refs/heads/master@{#374204}
==========

[commit-bot: I haz the power, 2016-02-08 22:48:59]

Patchset 5 (id:??) landed as
https://crrev.com/620be1f0cde37e56b8f8eb9e238e3d28b6471fc7
Cr-Commit-Position: refs/heads/master@{#374204}

[please use gerrit instead, 2016-02-09 00:11:56]

This patch may have broken WebViewFocusInteractiveTest.FocusAndVisibility test
on "Win 7 Tests x64 (1)"

https://build.chromium.org/p/chromium.win/builders/Win%207%20Tests%20x64%20%2...

[ RUN      ] WebViewFocusInteractiveTest.FocusAndVisibility
[4020:4512:0208/154721:WARNING:suggestions_service.cc(169)] Token error: Not
authorized.
[4020:4512:0208/154721:INFO:CONSOLE(15)] "Setting URL to
"http://localhost:54274/extensions/platform_apps/web_view/focus_visibility/guest.html".",
source: chrome-extension://cgbednfbmhloindaacgdnimfalljikph/window.js (15)
c:uild\slave\win_x64_builderuild\src\chromerowserpps\guest_view\web_view_interactive_browsertest.cc(1312):
error: Value of: webview_button_focused_listener.WaitUntilSatisfied()
  Actual: false
Expected: true
c:uild\slave\win_x64_builderuild\src\chromerowserpps\guest_view\web_view_interactive_browsertest.cc(1331):
error: Value of: webview_button_not_focused_listener.WaitUntilSatisfied()
  Actual: false
Expected: true
[  FAILED  ] WebViewFocusInteractiveTest.FocusAndVisibility, where TypeParam = 
and GetParam() =  (1321 ms)

If the failure persists or the test starts to flake, I will be reverting this
change.

[mustaq, 2016-02-09 17:03:05]

On 2016/02/09 00:11:56, Rouslan wrote:
> This patch may have broken WebViewFocusInteractiveTest.FocusAndVisibility test
> on "Win 7 Tests x64 (1)"
> 
>
https://build.chromium.org/p/chromium.win/builders/Win%207%20Tests%20x64%20%2...
> 
> [ RUN      ] WebViewFocusInteractiveTest.FocusAndVisibility
> [4020:4512:0208/154721:WARNING:suggestions_service.cc(169)] Token error: Not
> authorized.
> [4020:4512:0208/154721:INFO:CONSOLE(15)] "Setting URL to
>
"http://localhost:54274/extensions/platform_apps/web_view/focus_visibility/guest.html".",
> source: chrome-extension://cgbednfbmhloindaacgdnimfalljikph/window.js (15)
>
c:uild\slave\win_x64_builderuild\src\chromerowserpps\guest_view\web_view_interactive_browsertest.cc(1312):
> error: Value of: webview_button_focused_listener.WaitUntilSatisfied()
>   Actual: false
> Expected: true
>
c:uild\slave\win_x64_builderuild\src\chromerowserpps\guest_view\web_view_interactive_browsertest.cc(1331):
> error: Value of: webview_button_not_focused_listener.WaitUntilSatisfied()
>   Actual: false
> Expected: true
> [  FAILED  ] WebViewFocusInteractiveTest.FocusAndVisibility, where TypeParam =

> and GetParam() =  (1321 ms)
> 
> If the failure persists or the test starts to flake, I will be reverting this
> change.

I think the failing test involves only keyboard events (tabbing through elements
to be precise), so should be unaffected by this CL.

https://code.google.com/p/chromium/codesearch#chromium/src/chrome/browser/app...