Fixes in UserGestureIndicator

Fixes in UserGestureIndicator

This CL fixes problems discovered in https://codereview.chromium.org/235983016/ post landing.

BUG=163072
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=172128

[vabr (Chromium), 2014-04-16 16:25:15]

Thanks for pointing this out in https://codereview.chromium.org/235983016/.

PTAL. In particular, I'm not sure if I did the right thing in the
UserGestureIndicator(PassRefPtr<UserGestureToken> token) constructor.

Cheers,
Vaclav

[vabr (Chromium), 2014-04-16 16:26:10]

On 2014/04/16 16:25:15, vabr (Chromium) wrote:
> Thanks for pointing this out in https://codereview.chromium.org/235983016/.
This should have been:

Hi Jochen,
Thanks for pointing this out (...)

:)
> 
> PTAL. In particular, I'm not sure if I did the right thing in the
> UserGestureIndicator(PassRefPtr<UserGestureToken> token) constructor.
> 
> Cheers,
> Vaclav

[jochen (gone - plz use gerrit), 2014-04-17 08:25:37]

https://codereview.chromium.org/239963005/diff/60001/Source/platform/UserGest...
File Source/platform/UserGestureIndicator.cpp (right):

https://codereview.chromium.org/239963005/diff/60001/Source/platform/UserGest...
Source/platform/UserGestureIndicator.cpp:160: s_processedUserGestureInPast =
true;
here we're not really adding a gesture, we're just moving it from one token to
another

[jochen (gone - plz use gerrit), 2014-04-17 08:25:59]

also, your tests fail :)

[vabr (Chromium), 2014-04-17 13:55:55]

Hi Jochen, PTAL.

Also sorry about the failing tests: yesterday I only one the new test and that
ran OK. But running in parallel influenced the initial value of the new static
bool flag in UserGestureIndicator. The intended use of the new API is to always
clear the flag before use, so I corrected the test to reflect that.

The current failure on linux_blink_dbg looks unrelated, and does not reproduce
with my local debug build.

Cheers,
Vaclav

https://codereview.chromium.org/239963005/diff/60001/Source/platform/UserGest...
File Source/platform/UserGestureIndicator.cpp (right):

https://codereview.chromium.org/239963005/diff/60001/Source/platform/UserGest...
Source/platform/UserGestureIndicator.cpp:160: s_processedUserGestureInPast =
true;
On 2014/04/17 08:25:38, jochen wrote:
> here we're not really adding a gesture, we're just moving it from one token to
> another

Good point, I will remove the line.
The flag was flipped to true during the original creation of the token, which
only happens in the other UserGestureIndicator constructor (if I can trust Code
Search). That means the flag was flipped there already.

[vabr (Chromium), 2014-04-22 09:52:06]

Hi Jochen, pinging you about this review.
(linux_blink_dbg looks like a flake.)
Cheers,
Vaclav

[jochen (gone - plz use gerrit), 2014-04-22 09:56:42]

lgtm

[vabr (Chromium), 2014-04-22 09:57:35]

The CQ bit was checked by vabr@chromium.org

[commit-bot: I haz the power, 2014-04-22 09:57:48]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/vabr@chromium.org/239963005/80001

[commit-bot: I haz the power, 2014-04-22 10:56:41]

Change committed as 172128