GitLab

displays a bubble pointing at "Most Visited"; this is dismissed after 10 displays (try opening the ntp 11 times), or after the user clicks a navigation dot. See bug for more background.

BUG=92434
TEST=manual


Review URL: http://codereview.chromium.org/7461160

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96680 0039d316-1c4b-4281-b951-d872f2087c98

Prevents crashes when destroying a decoder while messages are pending.

BUG=None
TEST=Pepper Flash


Review URL: http://codereview.chromium.org/7629022

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96679 0039d316-1c4b-4281-b951-d872f2087c98

This is needed by my new unified resoruce tracker, which will use this file from the shared_impl in the new resource base object.

I fixed up the namespaces for the callers. Longer term, I want to put the
proxy in the ppapi namespace which will eliminate some of this mess.

Review URL: http://codereview.chromium.org/7623018

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96678 0039d316-1c4b-4281-b951-d872f2087c98

An ill-timed lost context could destroy the compositor context, while a
composite has been scheduled. This makes sure we reset
is_accelerated_compositing_active_ so that we don't try to use the 3D path in
this case.

BUG=None
TEST=Fullscreen with Pepper Flash on ARM devices


Review URL: http://codereview.chromium.org/7628023

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96677 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=none


Review URL: http://codereview.chromium.org/7628026

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96676 0039d316-1c4b-4281-b951-d872f2087c98

I removed the _p.c files from the associated iaccessible2 previously to fix windows incremental link, but didn't add the required _p.c one to the proxy dll.

BUG=Red tree
TEST=IAccessible2Proxy links

Review URL: http://codereview.chromium.org/7649003
Patch from scottmg@chromium.org <scottmg@chromium.org>.

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96675 0039d316-1c4b-4281-b951-d872f2087c98

BUG=87856
TEST=See bug


Review URL: http://codereview.chromium.org/7640004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96674 0039d316-1c4b-4281-b951-d872f2087c98

R=jhawkins@chromium.org
BUG=none
TEST=WebIntentsTableTest.GetAllIntents,WebDataServiceTest.WebIntents.GetAll,WebIntentsRegistryTest.GetAllIntents,


Review URL: http://codereview.chromium.org/7633011

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96673 0039d316-1c4b-4281-b951-d872f2087c98

BUG=84992
TEST=printing still works.

Review URL: http://codereview.chromium.org/7639014

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96672 0039d316-1c4b-4281-b951-d872f2087c98

Allows the re-enabling of BrowserAccelerators test.


BUG=92563
TEST=*BrowserAccelerators


Review URL: http://codereview.chromium.org/7634028

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96671 0039d316-1c4b-4281-b951-d872f2087c98

"number of wheel handlers" to browser process.

We want to implement two-finger-swipe for history like safari does on
os x lion. The way this works is:
* If a page has no horizontal scrollbars, or the page is all the way
  to the left, scrolling to the left will go back in history instead.
* Same for right/forward.

Hence, this data needs to be available in the browser process.

This is modelled after the WebKit2 functions
WKPageIsPinnedToLeftSide(), WKPageIsPinnedToRightSide(),
WKPageHasHorizontalScrollbar(), and
webPageProxy::willHandleHorizontalScrollEvents().

Depends on https://bugs.webkit.org/show_bug.cgi?id=66094

BUG=90228
TEST=No observable effect yet


Review URL: http://codereview.chromium.org/7635011

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96670 0039d316-1c4b-4281-b951-d872f2087c98

As far as I can tell, the _p.c files should only be included into a proxy when
calling across domains. When linking incrementally it appears those files
aren't dead-stripped which causes unresolved external references to
CStdStubBuffer_Release.

BUG=92175
TEST=no new functionality


Review URL: http://codereview.chromium.org/7599005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96669 0039d316-1c4b-4281-b951-d872f2087c98

BUG=92582
TEST=doesn't crash when you paste a non-32bpp image that's not square.


Review URL: http://codereview.chromium.org/7628005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96668 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96664 0039d316-1c4b-4281-b951-d872f2087c98

It was found to be not the culprit of redness

Whitespace change: http://codereview.chromium.org/7633020

BUG=none
TEST=none
Review URL: http://codereview.chromium.org/7633030

TBR=vandebo@chromium.org
Review URL: http://codereview.chromium.org/7618026

TBR=ctguil@chromium.org
Review URL: http://codereview.chromium.org/7633047

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96662 0039d316-1c4b-4281-b951-d872f2087c98

Flash forwards some window messages when it has no handler set. The only ones I've observed are WM_MOUSEWHEEL and WM_GESTURE (gestures are supported only on Win7). So, we need to allow these messages through the UIPI boundary. 

One important note is that I use per-process message filters on Vista, but per-window filters on Win7 or later (because they're supported). 

BUG=86810
TEST=None.
Review URL: http://codereview.chromium.org/7617019

TBR=jschuh@chromium.org
Review URL: http://codereview.chromium.org/7648011

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96661 0039d316-1c4b-4281-b951-d872f2087c98

TEST=none
BUG=none

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96660 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=Launches on XP/Vista
TBR=georgey
Review URL: http://codereview.chromium.org/7633044

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96657 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96656 0039d316-1c4b-4281-b951-d872f2087c98

Whitespace change: http://codereview.chromium.org/7633020

BUG=none
TEST=none
Review URL: http://codereview.chromium.org/7633030

TBR=vandebo@chromium.org
Review URL: http://codereview.chromium.org/7618026

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96655 0039d316-1c4b-4281-b951-d872f2087c98

BUG=92498
TEST=With remoting, Connect and keep the session there for > 10m, hit continue and do it again. Expect dialog to come up again.

Review URL: http://codereview.chromium.org/7650003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96652 0039d316-1c4b-4281-b951-d872f2087c98

When we switch to P2P Pepper API we will need to run networking code on the
main plugin thread. Switching to MessageLoopProxy for network thread, so that
it's easier to switch network thread in the future.

BUG=None
TEST=None

Review URL: http://codereview.chromium.org/7633009

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96651 0039d316-1c4b-4281-b951-d872f2087c98

BUG=
TEST=


Review URL: http://codereview.chromium.org/7602021

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96650 0039d316-1c4b-4281-b951-d872f2087c98

The DCHECKs don't trigger in release, and in any case it's a bit rude to have
the renderer crash on a plugin that did the wrong thing, so send an error
instead.

BUG=None
TEST=Pepper Flash


Review URL: http://codereview.chromium.org/7628025

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96649 0039d316-1c4b-4281-b951-d872f2087c98

BUG=92231
TEST=Second remoting connection from the same browser works.

Review URL: http://codereview.chromium.org/7633036

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96647 0039d316-1c4b-4281-b951-d872f2087c98

BUG=19775,32624,89304,92571
TEST=none
TBR=jhawkins
Review URL: http://codereview.chromium.org/7648003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96646 0039d316-1c4b-4281-b951-d872f2087c98

BUG=91620
TEST=Host doesn't crash when shutting down.

Review URL: http://codereview.chromium.org/7635005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96645 0039d316-1c4b-4281-b951-d872f2087c98

BUG=chromium-os:17469
TEST=Click battery icon, confirm it meets spec

Review URL: http://codereview.chromium.org/7491083

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96643 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96641 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=tree green
Review URL: http://codereview.chromium.org/7637025

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96640 0039d316-1c4b-4281-b951-d872f2087c98

Whitespace change: http://codereview.chromium.org/7633020

BUG=none
TEST=none
Review URL: http://codereview.chromium.org/7633030

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96639 0039d316-1c4b-4281-b951-d872f2087c98

For a detailed description see part #1, http://codereview.chromium.org/6478005/

This CL fixes the bug fully on Linux. More platforms will follow.

BUG=61062
TEST=unit_tests, possibly more

Review URL: http://codereview.chromium.org/7628006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96638 0039d316-1c4b-4281-b951-d872f2087c98

This makes WebUILoginView not a drag-n-drop target as TabContents::ShouldAcceptDragAndDrop would return false now. Besides, WebUILoginView is popup in some sense.

BUG=chromium-os:18781
TEST=Verify crash no longer happens for chromium-os:18781.


Review URL: http://codereview.chromium.org/7640023

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96636 0039d316-1c4b-4281-b951-d872f2087c98

On both Lucid and Natty, cups-config --libs includes cflags and other
unneeded output.  Parse the cups-config output to emit only the relevant
subsets of flags.

Review URL: http://codereview.chromium.org/7633022

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96635 0039d316-1c4b-4281-b951-d872f2087c98

Now that pending preview requests are being canceled when a new request is issued, the event listeners of the printing controls should not check for |hasPendingPreviewRequest| before issuing new requests, instead they should immediately request a new preview so that the previous one is canceled.

Also all functions that are associated with a specific print request should get the requestID as a parameter, in order to ignore obsolete responses.

I found out that when the print preview tab is reloaded, responses from requests before the reload are reaching the ui code. If reloading the preview tab repeatedly all these request have id=0 and are not being ignored as they should. Therefore I changed the id of the initial request to have a random variable.

BUG=91556, 91769
TEST=See bug description

Review URL: http://codereview.chromium.org/7550022

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96634 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96633 0039d316-1c4b-4281-b951-d872f2087c98

BUG=91619
TEST=Set up a remoting sesssion and make sure it works.

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=96327

Review URL: http://codereview.chromium.org/7491070

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96632 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: http://codereview.chromium.org/7638002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96631 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: http://codereview.chromium.org/7637026

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96630 0039d316-1c4b-4281-b951-d872f2087c98

BUG=92521
TEST=as in bug

Review URL: http://codereview.chromium.org/7639022

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@96629 0039d316-1c4b-4281-b951-d872f2087c98