linux: Fix grabs for popups belonging to inactive tabs.

linux: Fix grabs for popups belonging to inactive tabs.

When displaying a popup for an inactive tab, grab input on
behalf of the popup's window rather than its parent window.

Also removes window-group code that doesn't make sense to me
here (the popup wants all input rather than just input for a
subset of browser windows, as shown by the way that it's
doing X grabs).

BUG=126296
TEST=manual

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=135966

[Daniel Erat, 2012-05-07 23:57:51]

Evan for review, Scott for approval.

[Evan Stade, 2012-05-08 00:01:11]

http://codereview.chromium.org/10380044/diff/2001/content/browser/renderer_ho...
File content/browser/renderer_host/render_widget_host_view_gtk.cc (right):

http://codereview.chromium.org/10380044/diff/2001/content/browser/renderer_ho...
content/browser/renderer_host/render_widget_host_view_gtk.cc:595: // tab), skip
the grab.
how does the popup window know to close if there's no grab? will clicking on the
page still cause it to dismiss?

[Daniel Erat, 2012-05-08 00:11:55]

http://codereview.chromium.org/10380044/diff/2001/content/browser/renderer_ho...
File content/browser/renderer_host/render_widget_host_view_gtk.cc (right):

http://codereview.chromium.org/10380044/diff/2001/content/browser/renderer_ho...
content/browser/renderer_host/render_widget_host_view_gtk.cc:595: // tab), skip
the grab.
On 2012/05/08 00:01:13, Evan Stade wrote:
> how does the popup window know to close if there's no grab? will clicking on
the
> page still cause it to dismiss?

Hmm, it doesn't get closed until I click on an item. :-/  I initially tried just
closing the popup like you recommended, but doing so at this early point (at
least via the host_->Shutdown() route that we use for closing via the Escape
key) was causing a segfault.  I can look into it a bit more.

[Daniel Erat, 2012-05-08 17:42:12]

Another look?  I don't think that the window group code was necessary.

[Evan Stade, 2012-05-08 18:04:12]

https://chromiumcodereview.appspot.com/10380044/diff/9002/content/browser/ren...
File content/browser/renderer_host/render_widget_host_view_gtk.cc (right):

https://chromiumcodereview.appspot.com/10380044/diff/9002/content/browser/ren...
content/browser/renderer_host/render_widget_host_view_gtk.cc:613: // shown on
behalf of an inactive tab).
so you might show a popup for inactive tab? wouldn't it look rather odd to have
a select popup seemingly in the middle of nowhere

[Daniel Erat, 2012-05-08 18:10:12]

https://chromiumcodereview.appspot.com/10380044/diff/9002/content/browser/ren...
File content/browser/renderer_host/render_widget_host_view_gtk.cc (right):

https://chromiumcodereview.appspot.com/10380044/diff/9002/content/browser/ren...
content/browser/renderer_host/render_widget_host_view_gtk.cc:613: // shown on
behalf of an inactive tab).
On 2012/05/08 18:04:12, Evan Stade wrote:
> so you might show a popup for inactive tab? wouldn't it look rather odd to
have
> a select popup seemingly in the middle of nowhere

Yeah, it's weird, but I'm just trying to fix the linked bug in a general way. 
I'll file a separate bug about menus being openable for inactive tabs (which
works on Windows Chrome as well).

[Evan Stade, 2012-05-08 18:15:33]

ok lgtm

[sky, 2012-05-08 19:43:19]

Rubber stamp LGTM

[commit-bot: I haz the power, 2012-05-08 21:00:44]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/derat@chromium.org/10380044/9002

[commit-bot: I haz the power, 2012-05-08 21:21:58]

Try job failure for 10380044-9002 on win_rel for step "update".
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=win_rel&nu...

Step "update" is always a major failure.
Look at the try server FAQ for more details.

[commit-bot: I haz the power, 2012-05-08 21:23:56]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/derat@chromium.org/10380044/9002

[commit-bot: I haz the power, 2012-05-09 00:58:58]

Try job failure for 10380044-9002 (retry) on win_rel for step "browser_tests".
It's a second try, previously, step "browser_tests" failed.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=win_rel&nu...