GitLab

This patch is a small clean-up to the schedule tasks.

BUG=none
R=reveman@chromium.org

Review URL: https://chromiumcodereview.appspot.com/23619048

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223420 0039d316-1c4b-4281-b951-d872f2087c98

BUG=54748
TEST=on Windows open a password-protected PDF; a password dialog should appear

Review URL: https://chromiumcodereview.appspot.com/23498038

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223419 0039d316-1c4b-4281-b951-d872f2087c98

BUG=232657
TEST=none
R=beaudoin@chromium.org

Review URL: https://codereview.chromium.org/23536053

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223418 0039d316-1c4b-4281-b951-d872f2087c98

BUG=291251

Review URL: https://chromiumcodereview.appspot.com/24052004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223417 0039d316-1c4b-4281-b951-d872f2087c98

See bug for context. This change makes calling
an IAccessible2 API on the web content's root
accessible object enable full web accessibility
support. This supports popular Windows screen readers
but won't enable accessibility under other
circumstances when we don't need it.

However, this doesn't always detect the screen
reader right away when the app opens, so we also
modify Views accessibility to detect IAccessible2
and query the IAccessible2 API on each web view.
That catches use of IAccessible2 immediately.

BUG=292719

Review URL: https://chromiumcodereview.appspot.com/23531055

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223416 0039d316-1c4b-4281-b951-d872f2087c98

BUG=285083

Review URL: https://chromiumcodereview.appspot.com/23503056

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223415 0039d316-1c4b-4281-b951-d872f2087c98

For a list of major sites with dynamic password forms and whether they are fixed with this CL, please refer to the 'JavaScript Forms' column in https://docs.google.com/a/google.com/spreadsheet/ccc?key=0AkbIXgaqCC1tdGQ4NHlwd0phbWhBOEs1eGFTa1BsZFE#gid=0.

Each remaining issue is tracked in separate bugs, crbug/26186,  crbug/43219, crbug/234292, crbug/282487, crbug/282488, crbug/282522


BUG=123955

Review URL: https://chromiumcodereview.appspot.com/22926033

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223413 0039d316-1c4b-4281-b951-d872f2087c98

R=thakis@chromium.org
NOTRY=true

Review URL: https://chromiumcodereview.appspot.com/23717053

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223411 0039d316-1c4b-4281-b951-d872f2087c98

This test is already running in components_perftests.

BUG=None
TEST=components_perftests, perf_tests
R=thestig@chromium.org,joi@chromium.org,boliu@chromium.org

Review URL: https://chromiumcodereview.appspot.com/23724030

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223408 0039d316-1c4b-4281-b951-d872f2087c98

BUG=286074
R=isherman@chromium.org
TBR=ananta

Review URL: https://codereview.chromium.org/23876026

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223407 0039d316-1c4b-4281-b951-d872f2087c98

BUG=

Review URL: https://chromiumcodereview.appspot.com/24179002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223406 0039d316-1c4b-4281-b951-d872f2087c98

InfoBarService doesn't delete InfoBarDelegates directly (on shutdown or at any
other time), but rather relies on a listening InfoBarContainer to delete the
corresponding InfoBars, which in turn would delete the InfoBarDelegates.

If InfoBarContainer was destroyed before InfoBarService, it could still be
holding infobars, which would then leak, because there would be no container to
do the deletions at InfoBarService destruction time.  To prevent this leak,
InfoBarContainer was calling CloseSoon() and then Hide() on all InfoBars on
destruction, which would force them to be deleted and delete their delegates.

However, this (always!) caused a write-to-freed-memory when InfoBarService shut
down, because in this case it would still be holding pointers to all the
corresponding (deleted) InfoBarDelegates, and would attempt to access them
(specifically, to call clear_owner()) when removing them.  (The clear_owner()
call is sort of a hack; really, InfoBar and InfoBarDelegate should not have
separate members pointing to the owning InfoBarService.  Unfortunately this is
necessary to avoid some other bugs in the current system.)

The core problem was that CloseSoon() tells the InfoBar it's unowned, but in
this case the InfoBar isn't actually unowned yet.  Removing the CloseSoon() call
from InfoBarContainer fixes this, at the cost of re-introducing the leak
described above.

The correct fix is for InfoBarService to talk to InfoBars instead of
InfoBarDelegates, so that regardless of the destruction order of InfoBarService
versus InfoBarContainer, InfoBars will stay alive until they're both un-owned
and hidden, then delete themselves.  In this world we also don't need
InfoBarDelegates to have a pointer back to an "owning" InfoBarService; they're
owned by long-lived InfoBars.

If this sounds familiar, that's because it's a precise description of the new
ownership model that I've written (years ago) but still not yet landed.  (Though
I'm getting close; I'm now only blocked by some Android work.)  Until this model
is landed to fix the problem correctly, we have little choice but to accept the
potential leak in this scenario.

BUG=290976
TEST=none
R=erg@chromium.org

Review URL: https://codereview.chromium.org/24151002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223405 0039d316-1c4b-4281-b951-d872f2087c98

BUG=292418

Review URL: https://chromiumcodereview.appspot.com/23533047

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223404 0039d316-1c4b-4281-b951-d872f2087c98

BUG=291114
TEST=manual
R=alemate@chromium.org, nkostylev@chromium.org, skuhne@chromium.org

Review URL: https://codereview.chromium.org/23531054

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223403 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223402 0039d316-1c4b-4281-b951-d872f2087c98

Sizes increase on linux. Investigating if it is related to a broken -fPIC setting.

BUG=279335
TBR=tomfinegan

Review URL: https://codereview.chromium.org/23514060

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223401 0039d316-1c4b-4281-b951-d872f2087c98

webview's cookies file, for the first instance of the app, so
cookies are not lost.

BUG=b/9962984

TBRing safebrowsing_service and profile_impl_io_data changes since these are mechanical in nature.

TBR=mattm@chromium.org,mmenke@chromiumorg

adding no try since we have comments that are longer than 80 chars, and since this is allowed under style guidance.
NOTRY=true

Review URL: https://chromiumcodereview.appspot.com/23964011

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223400 0039d316-1c4b-4281-b951-d872f2087c98

This CL: 
 * Closes the P1 security hole described in http://crbug.com/284792 by changing the message contents to contain Pepper file open flags instead of base::PlatformFileFlags and checking those in FileAPIMessageFilter.
 * Ports the rest of FileAPIMessageFilter to use new CPSP calls.
 * Ports one call in ResourceDispatcherHostImpl.
 * Makes base::PlatformFileFlags-based methods private in CPSP.

Refactoring document / plans here:
https://docs.google.com/a/google.com/document/d/1QGkGWuwgSuaRqovz4wyb0upqPKDVsgYOFKt44E7gmOE/edit?usp=sharing 

BUG=262142,284792

Review URL: https://chromiumcodereview.appspot.com/23760004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223399 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none

Review URL: https://chromiumcodereview.appspot.com/23542045

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223398 0039d316-1c4b-4281-b951-d872f2087c98

Partial re-land of https://codereview.chromium.org/22198004/) -- Always enable FCM on Windows -- Part 1/3.

On top of fixing some of the issues there; only lands the part that enables the GPU blacklist in tests as the former CL is too hard to land all at once.

Also keeping --skip-gpu-data-loading around for now to be able to deal with failures caused in layout_tests by this on their own later (this needs to land ASAP and layout_tests don't need this ASAP).

Another CL will follow to always enable FCM on non-blacklisted Windows machines.

This part re-enables loading the blacklist in tests (and adds a content_browsertest to make sure that the configuration we expect to be testing is indeed the one we are testing -- this uncertainty is basically the only reason the blacklist was explicitly disabled before).

This CL also cleans up compositor_util.cc which was enforcing the blacklist twice.

The original plan was to do this only for Windows as Mac/Linux was causing trouble, but it turns out to be harder to do it only on Windows; so taking care of http://crbug.com/277242 in this CL too after all... 

BUG=233830, 267038, 190942, 277242
TBR=jcivelli, piman 

Originally Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=219132 
Then Reverted: https://src.chromium.org/viewvc/chrome?view=rev&revision=219159

Then re-committed (part 1/3): https://src.chromium.org/viewvc/chrome?view=rev&revision=221114
Reverted in: https://src.chromium.org/viewvc/chrome?view=rev&revision=221145

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=222105

Review URL: https://chromiumcodereview.appspot.com/23534006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223397 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: https://chromiumcodereview.appspot.com/24170002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223396 0039d316-1c4b-4281-b951-d872f2087c98

Otherwise, parent window can be interacted with which causes
redraw problems.

R=estade@chromium.org
BUG=284368

Review URL: https://chromiumcodereview.appspot.com/23889021

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223395 0039d316-1c4b-4281-b951-d872f2087c98

Multi-line text rendering in the Chromium UI is currently done by slicing the string into multiple lines by using text metrics from cross-platform libraries and rendering text by using different RenderText instances - one for each line. This approach has poor performance and is very bug-prone.

This CL adds a cross-platform interface and data structures to RenderText to support multi-line text rendering. This CL also implements the required platform-specific bits for Windows. Support for other platforms will be implemented in subsequent CLs.

Multi-line rendering as implemented in this CL is limited. Newline characters are ignored. RTL and complex scripts are supported, while there are be issues with selection highlights. Text-space <-> view-space mappings do not support RTL/complex scripts.

BUG=248597

Review URL: https://chromiumcodereview.appspot.com/16867016

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223394 0039d316-1c4b-4281-b951-d872f2087c98

Ensure tooltips are closed when their anchor views go away. Also, close the
tooltip when shelf alignment changes.

BUG=288838
TEST=added to ash_unittests LauncherViewTest

Review URL: https://chromiumcodereview.appspot.com/24141003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223393 0039d316-1c4b-4281-b951-d872f2087c98

This CL adds code which parses the JS heap snapshot retrieved from the
Dev tools.

BUG=274456

Review URL: https://chromiumcodereview.appspot.com/24076016

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223392 0039d316-1c4b-4281-b951-d872f2087c98

This patch centralizes renderer binding management in ChildProcessLauncher,
simplifying ChildProcessConnection and encapsulating the logic in a nested
class.

BindingManager becomes responsible for all tasks related to adding, removing
(possibly in a delayed way) and inspecting the state of renderer out-of-memory
bindings.

BUG=290351

Review URL: https://chromiumcodereview.appspot.com/23532058

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223391 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223390 0039d316-1c4b-4281-b951-d872f2087c98

http://build.chromium.org/f/chromium/perf/dashboard/ui/changelog_blink.html?url=/trunk&range=157793:157830&mode=html
TBR=
BUG=

Review URL: https://codereview.chromium.org/23503061

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223389 0039d316-1c4b-4281-b951-d872f2087c98

The code was sending it directly to the RenderWidgetHostImpl, which doesn't work if the surface is connected to the browser compositor. It needs to be sent to the GPU process, which can route it correctly.

BUG=

Review URL: https://chromiumcodereview.appspot.com/23843005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223388 0039d316-1c4b-4281-b951-d872f2087c98

The script optimize-png-files.sh requires optipng, advdef, and pngout when
running at optimization levels 1 and 2, but checks for these tools only at
level 2. This CL changes optimize-png-files.sh to also check for the
required tools at level 1 optimization.

TEST=Uninstall optipng and run the script with -o1 parameter.
R=oshima@chromium.org
BUG=None

Review URL: https://chromiumcodereview.appspot.com/23757043

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223387 0039d316-1c4b-4281-b951-d872f2087c98

This changes AwContents to work correctly when in a WRAP_CONTENT
layout mode.

TBR=danakj@chromium.org
BUG=246621

Review URL: https://chromiumcodereview.appspot.com/23478022

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223386 0039d316-1c4b-4281-b951-d872f2087c98

BUG=289178

Review URL: https://chromiumcodereview.appspot.com/23441060

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223385 0039d316-1c4b-4281-b951-d872f2087c98

Enable end to end tests for Mac OS. The timeouts should be fixed now the number of window.open calls have been reduced (https://chromiumcodereview.appspot.com/19014003/). 

Two tests are showing other errors on Mac - they have separate bugs and will be looked into separately. 

This change also:
   - adds ability to customize expected logging to a particular OS
   - disable a test for particular OS.
   - some minor cleanup to make things more readable

BUG=245594

Review URL: https://chromiumcodereview.appspot.com/23465025

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223379 0039d316-1c4b-4281-b951-d872f2087c98

I'll try to re-enable the linux webrtc content browser tests and see if they've magically fixed themselves. If not I'll try switching from the embedded test server to the spawned one. I need to do this on the real bots since it doesn't reproduce locally.

R=tommi@chromium.org
BUG=281492

Review URL: https://chromiumcodereview.appspot.com/23766035

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223378 0039d316-1c4b-4281-b951-d872f2087c98

Previously we didn't register dooms in the backend until after
asynchronous operations on the entry; this meant that multiple
registrations were possible if different entries were being doomed, and null dereferences abounded after doom operations. Now fixed.

This reenables DiskCacheBackendTest.SimpleCacheDoomAll.

R=pasko,clamy,ttuttle
BUG=290866,237450,289542

Review URL: https://chromiumcodereview.appspot.com/23591048

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223376 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
R=jar@chromium.org, mmenke@chromium.org

Review URL: https://codereview.chromium.org/15809005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223375 0039d316-1c4b-4281-b951-d872f2087c98

BUG=292745
R=jered@chromium.org, jfweitz@chromium.org

Review URL: https://codereview.chromium.org/23578034

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223372 0039d316-1c4b-4281-b951-d872f2087c98

R=thakis@chromium.org

Review URL: https://codereview.chromium.org/23526052

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223370 0039d316-1c4b-4281-b951-d872f2087c98

it fails with EINTR. Specifically, in this case, the FD will already have been
closed.

BUG=269623
R=thakis@chromium.org

Review URL: https://codereview.chromium.org/23455051

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223369 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223368 0039d316-1c4b-4281-b951-d872f2087c98