diff options
| author | thakis@chromium.org <thakis@chromium.org@0039d316-1c4b-4281-b951-d872f2087c98> | 2009-11-06 02:43:51 +0000 |
|---|---|---|
| committer | thakis@chromium.org <thakis@chromium.org@0039d316-1c4b-4281-b951-d872f2087c98> | 2009-11-06 02:43:51 +0000 |
| commit | 5c87690ac9aed7e975132b301aed90abbe4e0550 (patch) | |
| tree | c882124c8abfa9f4e06f77c4f5a8ebb06cb301a9 /chrome/browser/tab_contents | |
| parent | 34437af8727dac421df5ff694eea7c8392ed1038 (diff) | |
| download | chromium_src-5c87690ac9aed7e975132b301aed90abbe4e0550.zip chromium_src-5c87690ac9aed7e975132b301aed90abbe4e0550.tar.gz chromium_src-5c87690ac9aed7e975132b301aed90abbe4e0550.tar.bz2 | |
Do not send input menu toggle events to the web.
This changes us back from the sendEvent:-based approach to the performKeyEquivalent:-based one. The main problem with the pKE: approach was that ctrl-tab toggled the view loop. However, I found an SPI in WebKit's source that enables us to work around that.
I tested lots of stuff, but there are of course chances that this regresses something. The only sane way to write unit tests for all that keyhandling crap however is to sendEvent: keys to NSApp and have a complete browser with renderers running, and browser tests don't work on OS X yet :-/
BUG=26115
TEST=Open SysPrefs->Keyboard and configure cmd-space to switch input languages. Go to a page that blocks js events (e.g. http://unixpapa.com/js/testkey.html , check the two left checkboxes), cmd-space should still switch input languages. ) All the cases listed at the end of http://sites.google.com/a/chromium.org/dev/developers/os-x-keyboard-handling .
Review URL: http://codereview.chromium.org/363024
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@31197 0039d316-1c4b-4281-b951-d872f2087c98
Diffstat (limited to 'chrome/browser/tab_contents')
0 files changed, 0 insertions, 0 deletions