x11/test: Use a custom-deleter for the stored XEvent.

x11/test: Use a custom-deleter for the XEvent in ScopedXI2Event.

BUG=none
R=sky@chromium.org

Previously committed in r248512, and reverted in r248526 because it introduced leaks.

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=248683

[sky, 2014-02-03 04:48:30]

LGTM

https://codereview.chromium.org/136153006/diff/50001/ui/events/test/events_te...
File ui/events/test/events_test_utils_x11.cc (right):

https://codereview.chromium.org/136153006/diff/50001/ui/events/test/events_te...
ui/events/test/events_test_utils_x11.cc:138: void
XEventDeleter::operator()(XEvent* event) {
Can you add some comments as to why this is necessary. Or maybe if it's already
documented why else where in the point add a reference to that.

[sadrul, 2014-02-03 15:58:17]

https://codereview.chromium.org/136153006/diff/50001/ui/events/test/events_te...
File ui/events/test/events_test_utils_x11.cc (right):

https://codereview.chromium.org/136153006/diff/50001/ui/events/test/events_te...
ui/events/test/events_test_utils_x11.cc:138: void
XEventDeleter::operator()(XEvent* event) {
On 2014/02/03 04:48:31, sky wrote:
> Can you add some comments as to why this is necessary. Or maybe if it's
already
> documented why else where in the point add a reference to that.

Done (added a comment explaining why this is necessary).

[sadrul, 2014-02-03 17:38:08]

Committed patchset #3 manually as r248512 (presubmit successful).

[msw, 2014-02-03 18:06:26]

A revert of this CL has been created in
https://codereview.chromium.org/135393011/ by msw@chromium.org.

The reason for reverting is: This seems to cause leaks:
http://build.chromium.org/p/chromium.memory/builders/Linux%20ASAN%20Tests%20%...
http://build.chromium.org/p/chromium.memory/builders/Linux%20ASAN%20Tests%20%....

[sadrul, 2014-02-04 05:48:37]

On 2014/02/03 18:06:26, msw wrote:
> A revert of this CL has been created in
> https://codereview.chromium.org/135393011/ by mailto:msw@chromium.org.
> 
> The reason for reverting is: This seems to cause leaks:
>
http://build.chromium.org/p/chromium.memory/builders/Linux%2520ASAN%2520Tests...
>
http://build.chromium.org/p/chromium.memory/builders/Linux%2520ASAN%2520Tests....

Unfortunately the asan trybots failed to catch the problem because the relevant
tests didn't run on these bots. I have uploaded a fix in the latest patchset
(patchset 5, patchset 4 is the reverted patch), and run it through the valgrind
trybots, which ran at least some of the tests that failed in the waterfall. So I
am going to land this change with that fix.

[sadrul, 2014-02-04 07:04:42]

Committed patchset #5 manually as r248683 (presubmit successful).