Added missing pen tilt when creating PointerEvent from ME.

Added missing pen tilt when creating PointerEvent from ME.

Added back the missing plumbing. Also added EventSender support to pass stylus
info for testing.

BUG=593850
Committed: https://crrev.com/abca6ff6e9ffb9791242701d4499e0144ed94283
Cr-Commit-Position: refs/heads/master@{#398933}

[mustaq, 2016-06-07 15:16:33]

mustaq@chromium.org changed reviewers:
+ dtapuska@chromium.org, nzolghadr@chromium.org

[mustaq, 2016-06-07 15:16:34]

ptal

[dtapuska, 2016-06-07 15:29:50]

lgtm % formatting nit

https://codereview.chromium.org/2043053002/diff/1/components/test_runner/even...
File components/test_runner/event_sender.cc (right):

https://codereview.chromium.org/2043053002/diff/1/components/test_runner/even...
components/test_runner/event_sender.cc:2775: e.button_type,
Is this with "git cl format" 

Seems a lot of formatting changes; and we should standardize on making sure git
cl format is the standardized formatting behavior.

[Navid Zolghadr, 2016-06-07 15:30:56]

https://codereview.chromium.org/2043053002/diff/1/components/test_runner/even...
File components/test_runner/event_sender.cc (right):

https://codereview.chromium.org/2043053002/diff/1/components/test_runner/even...
components/test_runner/event_sender.cc:145: }
Looking at this now I remembered why I chose mouse id to be different from pen
ids. Note that I was also enforcing that in this block. Basically since
EventSender keep track of the state of the pointers in one array based on the
rawPointerId only then they should be unique across mouse and pen.  Although
that wouldn't be any problem for the actual code as PointerEventFactory keeps
the pair of (pointerType, rawPointerId) as the key.

Maybe not important that much as that is only about how we use APIs in our tests
but now EventSender can send a pen pointerdown and set the mouse event
properties in EventSender. Right?

[mustaq, 2016-06-07 15:51:49]

https://codereview.chromium.org/2043053002/diff/1/components/test_runner/even...
File components/test_runner/event_sender.cc (right):

https://codereview.chromium.org/2043053002/diff/1/components/test_runner/even...
components/test_runner/event_sender.cc:145: }
On 2016/06/07 15:30:56, Navid Zolghadr wrote:
> Looking at this now I remembered why I chose mouse id to be different from pen
> ids. Note that I was also enforcing that in this block. Basically since
> EventSender keep track of the state of the pointers in one array based on the
> rawPointerId only then they should be unique across mouse and pen.  Although
> that wouldn't be any problem for the actual code as PointerEventFactory keeps
> the pair of (pointerType, rawPointerId) as the key.
> 
> Maybe not important that much as that is only about how we use APIs in our
tests
> but now EventSender can send a pen pointerdown and set the mouse event
> properties in EventSender. Right?

I think the test code should keep track of ids while eventSender does the
minimal bookkeeping to ease the testing. This lets testing behavior in
non-standard event stream. Enforcing event consistency perfectly would make
eventSender too bulky while restricting new test scenarios.

You are right that the test code can now send pen pointerdown with
mouse-properties, but eventSender traditionally supported such calls (touchmove
w/o touchstart, 2+ mousedowns consecutively) to possibly test weird input cases.
E.g. what would happen if a pointerdown and the corresponding pointerup have
different ids and/or isPrimary values?

I agree that both sides have benefits. When there is no clear winner, I prefer
simpler solutions to ease code maintenance overhead.

https://codereview.chromium.org/2043053002/diff/1/components/test_runner/even...
components/test_runner/event_sender.cc:2775: e.button_type,
On 2016/06/07 15:29:50, dtapuska wrote:
> Is this with "git cl format" 
> 
> Seems a lot of formatting changes; and we should standardize on making sure
git
> cl format is the standardized formatting behavior.

Done, even though I hate the auto-formatted "paragraphs" for long arg lists.

[mustaq, 2016-06-07 15:52:30]

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

[mustaq, 2016-06-07 15:52:31]

rbyers@chromium.org: ptal.

[Navid Zolghadr, 2016-06-07 16:01:43]

LGTM

https://codereview.chromium.org/2043053002/diff/1/components/test_runner/even...
File components/test_runner/event_sender.cc (right):

https://codereview.chromium.org/2043053002/diff/1/components/test_runner/even...
components/test_runner/event_sender.cc:145: }
On 2016/06/07 15:51:49, mustaq wrote:
> On 2016/06/07 15:30:56, Navid Zolghadr wrote:
> > Looking at this now I remembered why I chose mouse id to be different from
pen
> > ids. Note that I was also enforcing that in this block. Basically since
> > EventSender keep track of the state of the pointers in one array based on
the
> > rawPointerId only then they should be unique across mouse and pen.  Although
> > that wouldn't be any problem for the actual code as PointerEventFactory
keeps
> > the pair of (pointerType, rawPointerId) as the key.
> > 
> > Maybe not important that much as that is only about how we use APIs in our
> tests
> > but now EventSender can send a pen pointerdown and set the mouse event
> > properties in EventSender. Right?
> 
> I think the test code should keep track of ids while eventSender does the
> minimal bookkeeping to ease the testing. This lets testing behavior in
> non-standard event stream. Enforcing event consistency perfectly would make
> eventSender too bulky while restricting new test scenarios.
> 
> You are right that the test code can now send pen pointerdown with
> mouse-properties, but eventSender traditionally supported such calls
(touchmove
> w/o touchstart, 2+ mousedowns consecutively) to possibly test weird input
cases.
> E.g. what would happen if a pointerdown and the corresponding pointerup have
> different ids and/or isPrimary values?


isPrimary is unrelated as it doesn't come with the input to PointerEventManager.
But having different ids for pointerdown and up would break the bookkeeping we
do in PointerEventManager.

> 
> I agree that both sides have benefits. When there is no clear winner, I prefer
> simpler solutions to ease code maintenance overhead.

[Rick Byers, 2016-06-08 17:50:19]

RS LGTM

[mustaq, 2016-06-08 17:55:05]

The CQ bit was checked by mustaq@chromium.org

[mustaq, 2016-06-08 17:55:06]

The patchset sent to the CQ was uploaded after l-g-t-m from
dtapuska@chromium.org
Link to the patchset: https://codereview.chromium.org/2043053002/#ps20001
(title: " ")

[commit-bot: I haz the power, 2016-06-08 17:55:25]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/2043053002/20001

[commit-bot: I haz the power, 2016-06-08 22:30:03]

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

[commit-bot: I haz the power, 2016-06-08 22:30:04]

Try jobs failed on following builders:
  win_chromium_rel_ng on tryserver.chromium.win (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.win/builders/win_chromium_rel_...)

[mustaq, 2016-06-09 13:43:23]

The CQ bit was checked by mustaq@chromium.org

[commit-bot: I haz the power, 2016-06-09 13:43:38]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/2043053002/20001

[commit-bot: I haz the power, 2016-06-09 17:30:07]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2016-06-09 17:35:24]

Description was changed from

==========
Added missing pen tilt when creating PointerEvent from ME.

Added back the missing plumbing. Also added EventSender support to pass stylus
info for testing.

BUG=593850
==========

to

==========
Added missing pen tilt when creating PointerEvent from ME.

Added back the missing plumbing. Also added EventSender support to pass stylus
info for testing.

BUG=593850

Committed: https://crrev.com/abca6ff6e9ffb9791242701d4499e0144ed94283
Cr-Commit-Position: refs/heads/master@{#398933}
==========

[commit-bot: I haz the power, 2016-06-09 17:35:25]

Patchset 2 (id:??) landed as
https://crrev.com/abca6ff6e9ffb9791242701d4499e0144ed94283
Cr-Commit-Position: refs/heads/master@{#398933}