Mark mouse events created for Enter/Leave notify as synthesized as they're not real mouse events.

Mark mouse events created for Enter/Leave notify as synthesized as they're not real mouse events.

 This is to prevent showing mouse cursor when a mouse enters into RootWindow when host window is created/deleted.
 Updated the test to make sure that synthesized events will not change the cursor visibility.

 The first CL (https://codereview.chromium.org/22051002/)
didn't solve all problem because EnterNotify sends
chrome MOUSE_MOVE event (which will send ui::ET_MOUSE_ENTERED/ui::ET_MOUSE_EXITED).



BUG=None
TEST=covered by test.

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=216105

[oshima, 2013-08-05 20:34:36]

r215531 wasn't enough, so this is 2nd attempt to fix the issue.

[sadrul, 2013-08-06 17:52:23]

This LGTM.

But can you explain the CL description why the previous change didn't work?

[oshima, 2013-08-06 18:21:34]

On 2013/08/06 17:52:23, sadrul wrote:
> This LGTM.
> 
> But can you explain the CL description why the previous change didn't work?
Done.

"The first CL (https://codereview.chromium.org/22051002/) 
didn't solve all problem because Enter/Leave Notify sends 
chrome MOUSE_MOVE event (which will send
ui::ET_MOUSE_ENTERED/ui::ET_MOUSE_EXITED)."

My local test didn't reveal all issues because of the timing issue.

[oshima, 2013-08-06 18:21:53]

sky -> ui/aura

[sky, 2013-08-06 21:11:05]

https://codereview.chromium.org/22264002/diff/3001/ui/aura/root_window_host_x...
File ui/aura/root_window_host_x11.cc (right):

https://codereview.chromium.org/22264002/diff/3001/ui/aura/root_window_host_x...
ui/aura/root_window_host_x11.cc:465: mouse_event.set_flags(mouse_event.flags() |
ui::EF_IS_SYNTHESIZED);
This seems like a real event to me. Why the synthesized? Is the bug that on
showing we should ignore and not here?

[sadrul, 2013-08-06 21:18:51]

https://codereview.chromium.org/22264002/diff/3001/ui/aura/root_window_host_x...
File ui/aura/root_window_host_x11.cc (right):

https://codereview.chromium.org/22264002/diff/3001/ui/aura/root_window_host_x...
ui/aura/root_window_host_x11.cc:465: mouse_event.set_flags(mouse_event.flags() |
ui::EF_IS_SYNTHESIZED);
On 2013/08/06 21:11:05, sky wrote:
> This seems like a real event to me. Why the synthesized? Is the bug that on
> showing we should ignore and not here?

My understanding is the EF_IS_SYNTHESIZED is set here since the Enter/Leave
events are system-generated events (by x11) as opposed to user-generated Motion
etc. events

[oshima, 2013-08-06 21:20:32]

https://codereview.chromium.org/22264002/diff/3001/ui/aura/root_window_host_x...
File ui/aura/root_window_host_x11.cc (right):

https://codereview.chromium.org/22264002/diff/3001/ui/aura/root_window_host_x...
ui/aura/root_window_host_x11.cc:465: mouse_event.set_flags(mouse_event.flags() |
ui::EF_IS_SYNTHESIZED);
On 2013/08/06 21:11:05, sky wrote:
> This seems like a real event to me.

This creates mouse MOVE event, even though mouse doesn't move,
hence synthesized. (MOVE event is generated separately when a mouse
enters by move)

> Why the synthesized? Is the bug that on
> showing we should ignore and not here?

Right now, I've seen this only in test bots so far, although
this can happen in real device. I can file if you want. let me know.

[sky, 2013-08-06 21:37:46]

https://codereview.chromium.org/22264002/diff/3001/ui/aura/root_window_host_x...
File ui/aura/root_window_host_x11.cc (right):

https://codereview.chromium.org/22264002/diff/3001/ui/aura/root_window_host_x...
ui/aura/root_window_host_x11.cc:465: mouse_event.set_flags(mouse_event.flags() |
ui::EF_IS_SYNTHESIZED);
On 2013/08/06 21:20:32, oshima wrote:
> On 2013/08/06 21:11:05, sky wrote:
> > This seems like a real event to me.
> 
> This creates mouse MOVE event, even though mouse doesn't move,
> hence synthesized. (MOVE event is generated separately when a mouse
> enters by move)

Then should the synthesized be on the move and not this?

> 
> > Why the synthesized? Is the bug that on
> > showing we should ignore and not here?
> 
> Right now, I've seen this only in test bots so far, although
> this can happen in real device. I can file if you want. let me know.
>

[oshima, 2013-08-06 22:07:16]

ah, you mean like this new patch?

[sky, 2013-08-06 22:32:43]

LGTM

[commit-bot: I haz the power, 2013-08-06 22:37:00]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/oshima@chromium.org/22264002/21001

[commit-bot: I haz the power, 2013-08-07 05:54:00]

Change committed as 216105