Exit mouse lock permision prompt on tab switch or close.

Exit mouse lock permision prompt on tab switch or close.

BUG=131702
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=143941

[scheib, 2012-06-22 00:16:31]

I've selected several notification types that seem relevant for situations we
would want to exit mouse lock (or pending mouse lock). I'm not confident that
these aren't overkill, but there's not much harm in that. Any recommendations of
people to ask about what notifications would be the safest minimum set?

[yzshen1, 2012-06-22 22:18:04]

http://codereview.chromium.org/10642008/diff/5001/chrome/browser/ui/fullscree...
File chrome/browser/ui/fullscreen_controller.cc (right):

http://codereview.chromium.org/10642008/diff/5001/chrome/browser/ui/fullscree...
chrome/browser/ui/fullscreen_controller.cc:263: if (contents ==
fullscreened_tab_ ||
Do we need to check NULL here?

http://codereview.chromium.org/10642008/diff/5001/chrome/browser/ui/fullscree...
File chrome/browser/ui/fullscreen_controller_browsertest.cc (right):

http://codereview.chromium.org/10642008/diff/5001/chrome/browser/ui/fullscree...
chrome/browser/ui/fullscreen_controller_browsertest.cc:380:
IN_PROC_BROWSER_TEST_F(FullscreenControllerTest,
nit: it might be useful to also test another case: the tab with mouse lock
bubble is not the last tab in the window to close.

[scheib, 2012-06-22 22:27:16]

http://codereview.chromium.org/10642008/diff/5001/chrome/browser/ui/fullscree...
File chrome/browser/ui/fullscreen_controller.cc (right):

http://codereview.chromium.org/10642008/diff/5001/chrome/browser/ui/fullscree...
chrome/browser/ui/fullscreen_controller.cc:263: if (contents ==
fullscreened_tab_ ||
On 2012/06/22 22:18:04, yzshen1 wrote:
> Do we need to check NULL here?

Done.

http://codereview.chromium.org/10642008/diff/5001/chrome/browser/ui/fullscree...
File chrome/browser/ui/fullscreen_controller_browsertest.cc (right):

http://codereview.chromium.org/10642008/diff/5001/chrome/browser/ui/fullscree...
chrome/browser/ui/fullscreen_controller_browsertest.cc:380:
IN_PROC_BROWSER_TEST_F(FullscreenControllerTest,
On 2012/06/22 22:18:04, yzshen1 wrote:
> nit: it might be useful to also test another case: the tab with mouse lock
> bubble is not the last tab in the window to close.
Done.

Acutlaly, there were already two tabs in this test. The new tab exists, and then
this test explicitly adds a BlankURL tab. However, adding a 3rd won't hurt, and
will ease readability in case anyone has the same concern.

[yzshen1, 2012-06-22 22:28:10]

LGTM

Thanks!

[scheib, 2012-06-22 22:34:08]

sky, owners please.

[sky, 2012-06-25 14:20:48]

LGTM