[Android] Always insert a TouchCancel if window or tab focus is lost

[Android] Always insert a TouchCancel if window or tab focus is lost

This fixes a recent regression where loss of tab focus would cancel a gesture
sequence, but not the generating touch sequence.  Resolve this by inserting a
synthetic touch cancellation event into the pipeline, which will cause
cancellation of both touch and gesture sequences.

BUG=348528
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=255030

[jdduke (slow), 2014-03-03 22:18:08]

PTAL.  This looks like a big change but it's fairly straightforward.

Note that this patch allows us to synthesize WebTouchEvent's from any
ui::MotionEvent, which could come in handy for other purposes.

[tdresser, 2014-03-04 13:53:03]

On 2014/03/03 22:18:08, jdduke wrote:
> PTAL.  This looks like a big change but it's fairly straightforward.
> 
> Note that this patch allows us to synthesize WebTouchEvent's from any
> ui::MotionEvent, which could come in handy for other purposes.

Most recent patch suffers from ye olde chunk mismatch. Can you try another
upload?

[jdduke (slow), 2014-03-04 15:47:44]

On 2014/03/04 13:53:03, tdresser wrote:
> On 2014/03/03 22:18:08, jdduke wrote:
> > PTAL.  This looks like a big change but it's fairly straightforward.
> > 
> > Note that this patch allows us to synthesize WebTouchEvent's from any
> > ui::MotionEvent, which could come in handy for other purposes.
> 
> Most recent patch suffers from ye olde chunk mismatch. Can you try another
> upload?

Ugg, sorry, uploaded a fresh set.

[tdresser, 2014-03-04 16:19:33]

LGTM with a few small nits.

https://codereview.chromium.org/178193022/diff/80001/content/browser/android/...
File content/browser/android/content_view_core_impl.cc (right):

https://codereview.chromium.org/178193022/diff/80001/content/browser/android/...
content/browser/android/content_view_core_impl.cc:1000: if (!rwhv)
Is this check just for testing? Add a comment indicating when this would be the
case.

https://codereview.chromium.org/178193022/diff/80001/content/browser/android/...
content/browser/android/content_view_core_impl.cc:1003: const ui::MotionEvent*
current_down_event =
Maybe add a comment indicating why you can't just use
GestureProvider::CancelActiveTouchSequence here.

https://codereview.chromium.org/178193022/diff/80001/content/browser/android/...
File content/browser/android/content_view_core_impl.h (right):

https://codereview.chromium.org/178193022/diff/80001/content/browser/android/...
content/browser/android/content_view_core_impl.h:348: // Utility method for
synthesizing a cancellation touch event and dispatching
Probably more consistent if "cancellation touch event" -> "touch cancel event".

[jdduke (slow), 2014-03-04 19:07:02]

https://codereview.chromium.org/178193022/diff/80001/content/browser/android/...
File content/browser/android/content_view_core_impl.cc (right):

https://codereview.chromium.org/178193022/diff/80001/content/browser/android/...
content/browser/android/content_view_core_impl.cc:1000: if (!rwhv)
On 2014/03/04 16:19:34, tdresser wrote:
> Is this check just for testing? Add a comment indicating when this would be
the
> case.

Done.

https://codereview.chromium.org/178193022/diff/80001/content/browser/android/...
content/browser/android/content_view_core_impl.cc:1003: const ui::MotionEvent*
current_down_event =
On 2014/03/04 16:19:34, tdresser wrote:
> Maybe add a comment indicating why you can't just use
> GestureProvider::CancelActiveTouchSequence here.

I went ahead and removed that method, as it really shouldn't be used.

https://codereview.chromium.org/178193022/diff/80001/content/browser/android/...
File content/browser/android/content_view_core_impl.h (right):

https://codereview.chromium.org/178193022/diff/80001/content/browser/android/...
content/browser/android/content_view_core_impl.h:348: // Utility method for
synthesizing a cancellation touch event and dispatching
On 2014/03/04 16:19:34, tdresser wrote:
> Probably more consistent if "cancellation touch event" -> "touch cancel
event".

Done.

[jdduke (slow), 2014-03-04 19:09:55]

The CQ bit was checked by jdduke@chromium.org

[commit-bot: I haz the power, 2014-03-04 19:12:16]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/jdduke@chromium.org/178193022/100001

[commit-bot: I haz the power, 2014-03-04 22:41:32]

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

[commit-bot: I haz the power, 2014-03-04 22:41:32]

Failed to apply patch for
ui/events/gesture_detection/gesture_provider_unittest.cc:
While running patch -p1 --forward --force --no-backup-if-mismatch;
  patching file ui/events/gesture_detection/gesture_provider_unittest.cc
  Hunk #1 FAILED at 71.
  Hunk #2 FAILED at 301.
  Hunk #3 FAILED at 970.
  Hunk #4 succeeded at 980 with fuzz 1 (offset 1 line).
  Hunk #5 FAILED at 992.
  4 out of 5 hunks FAILED -- saving rejects to file
ui/events/gesture_detection/gesture_provider_unittest.cc.rej

Patch:       ui/events/gesture_detection/gesture_provider_unittest.cc
Index: ui/events/gesture_detection/gesture_provider_unittest.cc
diff --git a/ui/events/gesture_detection/gesture_provider_unittest.cc
b/ui/events/gesture_detection/gesture_provider_unittest.cc
index
9cede09e157420878bc47184c4b38ebb38727aba..f3987df978a87d34ba91afe246fe7aef0326bd64
100644
--- a/ui/events/gesture_detection/gesture_provider_unittest.cc
+++ b/ui/events/gesture_detection/gesture_provider_unittest.cc
@@ -71,6 +71,13 @@ class GestureProviderTest : public testing::Test, public
GestureProviderClient {
     gestures_.push_back(gesture);
   }
 
+  bool CancelActiveTouchSequence() {
+    if (!gesture_provider_->current_down_event())
+      return false;
+    return gesture_provider_->OnTouchEvent(
+        *gesture_provider_->current_down_event()->Cancel());
+  }
+
   bool HasReceivedGesture(GestureEventType type) const {
     for (size_t i = 0; i < gestures_.size(); ++i) {
       if (gestures_[i].type == type)
@@ -301,7 +308,7 @@ TEST_F(GestureProviderTest, TapCancelledWhenWindowFocusLost)
{
   EXPECT_EQ(GESTURE_LONG_PRESS, GetMostRecentGestureEventType());
 
   // The long press triggers window focus loss by opening a context menu
-  gesture_provider_->CancelActiveTouchSequence();
+  EXPECT_TRUE(CancelActiveTouchSequence());
   EXPECT_EQ(GESTURE_TAP_CANCEL, GetMostRecentGestureEventType());
 }
 
@@ -970,7 +977,7 @@ TEST_F(GestureProviderTest,
GesturesCancelledAfterLongPressCausesLostFocus) {
   Wait(long_press_timeout);
   EXPECT_EQ(GESTURE_LONG_PRESS, GetMostRecentGestureEventType());
 
-  gesture_provider_->CancelActiveTouchSequence();
+  EXPECT_TRUE(CancelActiveTouchSequence());
   EXPECT_EQ(GESTURE_TAP_CANCEL, GetMostRecentGestureEventType());
 
   event = ObtainMotionEvent(event_time + long_press_timeout,
@@ -979,12 +986,12 @@ TEST_F(GestureProviderTest,
GesturesCancelledAfterLongPressCausesLostFocus) {
   EXPECT_FALSE(HasReceivedGesture(GESTURE_LONG_TAP));
 }
 
-// Verify that ignoring the remaining touch sequence triggers proper touch and
-// gesture cancellation.
+// Verify that inserting a touch cancel event will trigger proper touch and
+// gesture sequence cancellation.
 TEST_F(GestureProviderTest, CancelActiveTouchSequence) {
   base::TimeTicks event_time = base::TimeTicks::Now();
 
-  gesture_provider_->CancelActiveTouchSequence();
+  EXPECT_FALSE(CancelActiveTouchSequence());
   EXPECT_EQ(0U, GetReceivedGestureCount());
 
   MockMotionEvent event =
@@ -992,7 +999,7 @@ TEST_F(GestureProviderTest, CancelActiveTouchSequence) {
   EXPECT_TRUE(gesture_provider_->OnTouchEvent(event));
   EXPECT_EQ(GESTURE_TAP_DOWN, GetMostRecentGestureEventType());
 
-  gesture_provider_->CancelActiveTouchSequence();
+  ASSERT_TRUE(CancelActiveTouchSequence());
   EXPECT_EQ(GESTURE_TAP_CANCEL, GetMostRecentGestureEventType());
 
   // Subsequent MotionEvent's are dropped until ACTION_DOWN.

[jdduke (slow), 2014-03-04 22:50:21]

The CQ bit was checked by jdduke@chromium.org

[commit-bot: I haz the power, 2014-03-04 22:52:00]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/jdduke@chromium.org/178193022/110001

[commit-bot: I haz the power, 2014-03-05 00:35:16]

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

[commit-bot: I haz the power, 2014-03-05 00:35:16]

Retried try job too often on win_rel for step(s) base_unittests, browser_tests,
interactive_ui_tests, net_unittests, unit_tests
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=win_rel&nu...

[jdduke (slow), 2014-03-05 01:30:59]

The CQ bit was checked by jdduke@chromium.org

[commit-bot: I haz the power, 2014-03-05 01:38:25]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/jdduke@chromium.org/178193022/110001

[commit-bot: I haz the power, 2014-03-05 06:59:41]

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

[commit-bot: I haz the power, 2014-03-05 06:59:42]

Retried try job too often on win_rel for step(s) browser_tests, unit_tests
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=win_rel&nu...

[jdduke (slow), 2014-03-05 13:29:54]

The CQ bit was checked by jdduke@chromium.org

[commit-bot: I haz the power, 2014-03-05 13:30:03]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/jdduke@chromium.org/178193022/110001

[commit-bot: I haz the power, 2014-03-05 14:12:47]

Change committed as 255030