Re-add extra hit-tests on GestureTap

Re-add extra hit-tests on GestureTap

In r178098 I removed 3 hit tests on GestureTap for the 3 derived mouse
events.  This broke a few websites in subtle ways.  This CL adds
explicit hit tests back while keeping the cleanup that gives us more
direct control over how mouse events are sent during a tap.  I'll be
taking a more sophisticated approach to eliminating these hit tests in
the common scenarios.

This also adds a number of new tests and updates the expectations for
a few tests.

BUG=396652, 397983, 398205
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=179413

[Rick Byers, 2014-07-31 15:55:58]

David,
Can you please take a look at this CL?  Zeeshan is out of the office.

Thanks,
   Rick

[bokan, 2014-07-31 17:42:58]

On 2014/07/31 15:55:58, Rick Byers wrote:
> David,
> Can you please take a look at this CL?  Zeeshan is out of the office.
> 
> Thanks,
>    Rick

Everything looks reasonable to me, so LGTM with the caveat that I have no
expertise here.

[Rick Byers, 2014-07-31 17:50:23]

On 2014/07/31 17:42:58, bokan wrote:
> On 2014/07/31 15:55:58, Rick Byers wrote:
> > David,
> > Can you please take a look at this CL?  Zeeshan is out of the office.
> > 
> > Thanks,
> >    Rick
> 
> Everything looks reasonable to me, so LGTM with the caveat that I have no
> expertise here.

Thanks.  I've CC'd Zeeshan and Elliott who have some more context here - let me
know if you guys have any feedback after the fact, but I won't block on it. 
This should be pretty non-controversial

Rick

[Rick Byers, 2014-07-31 17:50:27]

The CQ bit was checked by rbyers@chromium.org

[commit-bot: I haz the power, 2014-07-31 17:51:02]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/rbyers@chromium.org/429793004/40001

[Zeeshan Qureshi, 2014-07-31 17:57:32]

I don't quite understand how the paragraph hit testing case works but looks good
to me overall.

[commit-bot: I haz the power, 2014-07-31 18:28:58]

FYI, CQ is re-trying this CL (attempt #1).
The failing builders are:
  mac_blink_rel on tryserver.blink
(http://build.chromium.org/p/tryserver.blink/builders/mac_blink_rel/builds/17921)
  win_blink_rel on tryserver.blink
(http://build.chromium.org/p/tryserver.blink/builders/win_blink_rel/builds/19928)

[Rick Byers, 2014-07-31 18:29:36]

On 2014/07/31 17:57:32, Zeeshan Qureshi wrote:
> I don't quite understand how the paragraph hit testing case works but looks
good
> to me overall.

Thanks.  Yeah I don't fully understand what's going on there either, except that
the general problem is that the touch adjustment algorithm is quite
sophisticated and doesn't necessarily return (point, node) pairs that are
possible to get via normal hit testing.  This was previously masked by the extra
hit test, so all sorts of bugs with the node touch adjustment returns could go
unnoticed.  I spent a couple hours trying to figure out why we were getting such
unpredictable/inconsistent results out of adjustment here but it's wasn't easy. 
I'll follow up separately in the new bug I filed and create some more discipline
around this (eg. have a mode where we can ASSERT that doing a redundant hit-test
after adjustment is a no-op).

[commit-bot: I haz the power, 2014-07-31 19:08:21]

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

[commit-bot: I haz the power, 2014-07-31 19:08:22]

Try jobs failed on following builders:
  mac_blink_rel on tryserver.blink
(http://build.chromium.org/p/tryserver.blink/builders/mac_blink_rel/builds/17934)

[donnd, 2014-07-31 23:07:17]

On 2014/07/31 19:08:22, I haz the power (commit-bot) wrote:
> Try jobs failed on following builders:
>   mac_blink_rel on tryserver.blink
>
(http://build.chromium.org/p/tryserver.blink/builders/mac_blink_rel/builds/17934)

Looks like this works for my feature.  - Donn

[Rick Byers, 2014-08-01 12:52:33]

On 2014/07/31 23:07:17, Donn wrote:
> On 2014/07/31 19:08:22, I haz the power (commit-bot) wrote:
> > Try jobs failed on following builders:
> >   mac_blink_rel on tryserver.blink
> >
>
(http://build.chromium.org/p/tryserver.blink/builders/mac_blink_rel/builds/17934)
> 
> Looks like this works for my feature.  - Donn

Thanks Donn!

David, can you please review the change?  You can just review the diff from PS
#3 (except it shows the addition of the RefPtr guard which was actually added in
the other CL you reviewed).

Thanks,
   Rick

[bokan, 2014-08-01 13:41:19]

On 2014/08/01 12:52:33, Rick Byers wrote:
> On 2014/07/31 23:07:17, Donn wrote:
> > On 2014/07/31 19:08:22, I haz the power (commit-bot) wrote:
> > > Try jobs failed on following builders:
> > >   mac_blink_rel on tryserver.blink
> > >
> >
>
(http://build.chromium.org/p/tryserver.blink/builders/mac_blink_rel/builds/17934)
> > 
> > Looks like this works for my feature.  - Donn
> 
> Thanks Donn!
> 
> David, can you please review the change?  You can just review the diff from PS
> #3 (except it shows the addition of the RefPtr guard which was actually added
in
> the other CL you reviewed).
> 
> Thanks,
>    Rick

lgtm

[Rick Byers, 2014-08-01 17:24:59]

The CQ bit was checked by rbyers@chromium.org

[commit-bot: I haz the power, 2014-08-01 17:25:46]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/rbyers@chromium.org/429793004/100001

[commit-bot: I haz the power, 2014-08-01 17:27:16]

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

[commit-bot: I haz the power, 2014-08-01 17:27:17]

Failed to apply patch for
LayoutTests/fast/events/touch/gesture/gesture-tap-frame-removed-expected.txt:
While running patch -p1 --forward --force --no-backup-if-mismatch;
  The next patch would create the file
LayoutTests/fast/events/touch/gesture/gesture-tap-frame-removed-expected.txt,
  which already exists!  Skipping patch.
  1 out of 1 hunk ignored

Patch:   N  
LayoutTests/fast/events/touch/gesture/gesture-tap-frame-removed-expected.txt
Index:
LayoutTests/fast/events/touch/gesture/gesture-tap-frame-removed-expected.txt
diff --git
a/LayoutTests/fast/events/touch/gesture/gesture-tap-frame-removed-expected.txt
b/LayoutTests/fast/events/touch/gesture/gesture-tap-frame-removed-expected.txt
new file mode 100644
index
0000000000000000000000000000000000000000..6e7049ae03ded181e24053ee9fd8b4d4b3600d0a
--- /dev/null
+++
b/LayoutTests/fast/events/touch/gesture/gesture-tap-frame-removed-expected.txt
@@ -0,0 +1,15 @@
+Verifies that a tap occuring on an iframe that gets removed during tap handling
doesn't cause a crash.
+
+On success, you will see a series of "PASS" messages, followed by "TEST
COMPLETE".
+
+
+Sending GestureTapDown
+Sending GestureShowPress
+Sending GestureTap
+Received mousemove in child frame
+Removing iframe
+PASS document.getElementById('target') is null
+PASS successfullyParsed is true
+
+TEST COMPLETE
+

[Rick Byers, 2014-08-01 17:27:30]

The CQ bit was checked by rbyers@chromium.org

[Rick Byers, 2014-08-01 17:27:36]

The CQ bit was unchecked by rbyers@chromium.org

[commit-bot: I haz the power, 2014-08-01 17:27:38]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/rbyers@chromium.org/429793004/100001

[commit-bot: I haz the power, 2014-08-01 17:27:51]

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

[commit-bot: I haz the power, 2014-08-01 17:27:51]

Failed to apply patch for
LayoutTests/fast/events/touch/gesture/gesture-tap-frame-removed-expected.txt:
While running patch -p1 --forward --force --no-backup-if-mismatch;
  The next patch would create the file
LayoutTests/fast/events/touch/gesture/gesture-tap-frame-removed-expected.txt,
  which already exists!  Skipping patch.
  1 out of 1 hunk ignored

Patch:   N  
LayoutTests/fast/events/touch/gesture/gesture-tap-frame-removed-expected.txt
Index:
LayoutTests/fast/events/touch/gesture/gesture-tap-frame-removed-expected.txt
diff --git
a/LayoutTests/fast/events/touch/gesture/gesture-tap-frame-removed-expected.txt
b/LayoutTests/fast/events/touch/gesture/gesture-tap-frame-removed-expected.txt
new file mode 100644
index
0000000000000000000000000000000000000000..6e7049ae03ded181e24053ee9fd8b4d4b3600d0a
--- /dev/null
+++
b/LayoutTests/fast/events/touch/gesture/gesture-tap-frame-removed-expected.txt
@@ -0,0 +1,15 @@
+Verifies that a tap occuring on an iframe that gets removed during tap handling
doesn't cause a crash.
+
+On success, you will see a series of "PASS" messages, followed by "TEST
COMPLETE".
+
+
+Sending GestureTapDown
+Sending GestureShowPress
+Sending GestureTap
+Received mousemove in child frame
+Removing iframe
+PASS document.getElementById('target') is null
+PASS successfullyParsed is true
+
+TEST COMPLETE
+

[Rick Byers, 2014-08-01 17:39:39]

The CQ bit was checked by rbyers@chromium.org

[commit-bot: I haz the power, 2014-08-01 17:39:51]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/rbyers@chromium.org/429793004/120001

[commit-bot: I haz the power, 2014-08-01 18:42:46]

FYI, CQ is re-trying this CL (attempt #1).
The failing builders are:
  mac_blink_rel on tryserver.blink
(http://build.chromium.org/p/tryserver.blink/builders/mac_blink_rel/builds/18103)

[commit-bot: I haz the power, 2014-08-01 19:20:27]

Change committed as 179413

[qinmin, 2014-09-30 23:38:41]

qinmin@chromium.org changed reviewers:
+ qinmin@chromium.org

[qinmin, 2014-09-30 23:38:42]

This CL is causing embedded youtube video fail to play: crbug.com/418800