GitLab

Details:
- Changing the interface of SuggestionsService, so results are returned via callbacks. Also updating tests.
- To support calls to SuggestionsService::FetchSuggestionsData() while data fetch takes place, we queue callbacks, and serve all data when data fetch completes.
- Adding SuggestionsService::Shutdown() to handle mid-flight shutdown.
- Making chrome://suggestions unavailable for incognito mode.

BUG=None
Test=SuggestionsServiceTest

Review URL: https://codereview.chromium.org/163953005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251935 0039d316-1c4b-4281-b951-d872f2087c98

Also updated the browser test to test the case where media pipeline starts to
ask the CDM to do decryption before the session is fully loaded.

BUG=344651
TEST=Updated test.
R=ddorwin@chromium.org

Review URL: https://codereview.chromium.org/166273009

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251934 0039d316-1c4b-4281-b951-d872f2087c98

BUG=344703
TEST=manual
NOTRY=True

Review URL: https://codereview.chromium.org/169053006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251933 0039d316-1c4b-4281-b951-d872f2087c98

BUG=344079

Review URL: https://codereview.chromium.org/166903006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251932 0039d316-1c4b-4281-b951-d872f2087c98

TBR=scottmg@chromium.org

Review URL: https://codereview.chromium.org/162613005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251931 0039d316-1c4b-4281-b951-d872f2087c98

TBR=inferno@chromium.org

Review URL: https://codereview.chromium.org/171503004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251928 0039d316-1c4b-4281-b951-d872f2087c98

R=estade@chromium.org
BUG=281069
TEST=bug repro steps

Review URL: https://codereview.chromium.org/168183002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251927 0039d316-1c4b-4281-b951-d872f2087c98

Consolidate the various implementations of SyncChangeProcessorDelegate and
fake into sync/api.  Replace many of the fake SyncChangeProcessor
implementations with a slightly more advanced FakeSyncChangeProcessor.

TBR=brettw@chromium.org
BUG=

Review URL: https://codereview.chromium.org/151963002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251926 0039d316-1c4b-4281-b951-d872f2087c98

BUG=341820

Review URL: https://codereview.chromium.org/169303002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251925 0039d316-1c4b-4281-b951-d872f2087c98

Previously during EndSession, a nested message loop and a task posted to
the FILE thread that posted a MessageLoop::Quit were used to communicate
that the FILE thread had completed outstanding writes. However, during
shutdown the GPU process might be terminated before the browser has run
through this code. In that case, the GPU process host might process the
"lost context" and try to re-establish (synchronously) a connection with
the GPU process. In this case, Windows can deny the launch of the gpu
process because the window station is terminating. This results in a
browser hang waiting for the GPU, which turns into a crash when the
watchdog timer kills the browser.

To avoid all this, use the same mechanism as Linux -- create a waitable
event and simply have the FILE thread signal it (meaning it's completed
previous writes). By not re-entering a UI-thread message loop, we avoid
having the GPU process host messages get processed.

As a side-benefit, this also means that GPU paint messages are not
processed, which (I think) will avoid the ugliness of sometimes having
sad tabs rendered during shutdown. This could happen because the
renderers are also asynchronously killed during shutdown and so it's a
race between painting the next frame and the browser window being
closed.

R=piman@chromium.org, cpu@chromium.org, darin@chromium.org
BUG=318527,47908,142501

Review URL: https://codereview.chromium.org/170023002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251922 0039d316-1c4b-4281-b951-d872f2087c98

BUG=338952,325567

Review URL: https://codereview.chromium.org/171163003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251921 0039d316-1c4b-4281-b951-d872f2087c98

BUG=297672

Review URL: https://codereview.chromium.org/101463005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251920 0039d316-1c4b-4281-b951-d872f2087c98

R=leng@chromium.org, markusheintz@chromium.org
BUG=343234

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=251392

Review URL: https://codereview.chromium.org/162713003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251919 0039d316-1c4b-4281-b951-d872f2087c98

Code to do this was recently removed because it was no longer being
used in r241225 (https://codereview.chromium.org/116103002). This is
now necessary again to throttle renderers when using CoreAnimation.

Frames are acked only when they are drawn. Both the CoreAnimation and
the non-CoreAnimation paths are changed to behave this way. By virtue
of the fact that the non-CoreAnimation path draws immediately, this
should have no functional effect on that path.

This differs from the mechanism deleted in the aforementioned patch in
two ways.

First, it uses a scoped_ptr of a struct instead of a vector of pairs to
store the information about the swap being returned. This should improve
readability in that the ack has struct names instead of just first and
second in the pair, and in that the scoped_ptr does not suggest support
for multiple pending swaps (which does not exist).

Second, it does makes RWHVMac ack frames more aggressively when
inside a draw call, rather than adding a call from RWHImpl to the view,
to effect the ack (which is what the old way does).

BUG=340133

Review URL: https://codereview.chromium.org/165703002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251918 0039d316-1c4b-4281-b951-d872f2087c98

BUG=342316
TEST=MenuControllerMnemonicTest*.*

Review URL: https://codereview.chromium.org/168033003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251917 0039d316-1c4b-4281-b951-d872f2087c98

This will get rid of the 1 second delay from waiting for setInternal to fire.

BUG=

Review URL: https://codereview.chromium.org/171263002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251916 0039d316-1c4b-4281-b951-d872f2087c98

Long ago, we used to disable compositing on chrome: and about: URLs by default
since that caused a startup performance hit. We then had an --allow-webui-compositing
switch to override this behavior when desired.  The usefulness of this behavior
has long gone and we completely ignore the behavior and the flag on aura builds. This
nukes it from the code

Review URL: https://codereview.chromium.org/167713003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251915 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: https://codereview.chromium.org/170743008

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251911 0039d316-1c4b-4281-b951-d872f2087c98

If ANDROID_SDK_ROOT is set, it's still added to the path for now, but it's
no longer passed to gyp. If you want to set a custom sdk root, call gyp like
"build/gyp_chromium -Dandroid_sdk_root=path".

If ANDROID_SDK_ROOT is not set, it's still getting exported with a default
value that's identical to the default value of gyp's android_sdk_root variable.

ANDROID_SDK_TOOLS and ANDROID_SDK_BUILD_TOOLS_VERSION are now
completely ignored. Set the android_sdk_tools or android_sdk_build_tools_version
gyp defines instead if you need to change either. If they are not set,
envsetup no longer sets them.

(Dependent on an internal that changes a bot to set these gyp defines in
addition to the env vars.)

BUG=330631
NOTRY=true

Review URL: https://codereview.chromium.org/164193010

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251904 0039d316-1c4b-4281-b951-d872f2087c98

that --v=1 is functional.

Review URL: https://codereview.chromium.org/170743003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251903 0039d316-1c4b-4281-b951-d872f2087c98

Resubmission of https://codereview.chromium.org/130263007/ with the binary resources added with dcommit.

BUG=334309, 161119
TBR=oshima@chromium.org

Review URL: https://codereview.chromium.org/170843002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251901 0039d316-1c4b-4281-b951-d872f2087c98

BUG=None
TEST=None

Review URL: https://codereview.chromium.org/162943004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251900 0039d316-1c4b-4281-b951-d872f2087c98

The test seems to pass fine when I run it manually on Windows Ash (including with high values of --gtest_repeat)
This CL:
- Makes ToplevelWindowEventHandlerTest.RunMoveLoopFailsDuringInProgressDrag look more like ToplevelWindowEventHandlerTest.Caption which I don't think is flaky on Windows Ash.
- Adds an extra check which will help better understand the source of the flakiness should it occur again

BUG=343167
TEST=None

Review URL: https://codereview.chromium.org/168793002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251899 0039d316-1c4b-4281-b951-d872f2087c98

R=piman
BUG=

Review URL: https://codereview.chromium.org/167953003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251898 0039d316-1c4b-4281-b951-d872f2087c98

BUG=326023

Review URL: https://codereview.chromium.org/167703002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251897 0039d316-1c4b-4281-b951-d872f2087c98

We're not planning to enable this on any new platforms, so we can just
hardcode it off when not on ChromeOS. Leaving the disable flag for
ChromeOS for now, as we're working on disabling that code path there.

R=piman
BUG=

Review URL: https://codereview.chromium.org/167973003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251896 0039d316-1c4b-4281-b951-d872f2087c98

Generated using:
> cd src/ui
> git grep -l pp_ifdef | xargs sed -i "s:pp_ifdef('\([^']\+\)'):\1:g"

BUG=343235
R=arv@chromium.org, joi@chromium.org, oshima@chromium.org

Review URL: https://codereview.chromium.org/161313002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251895 0039d316-1c4b-4281-b951-d872f2087c98

One common coding patern is to define an enum as in:
  enum Color { RED, GREEN, BLUE, COLOR_LAST=BLUE };
but this is fragile when someone adds a new constant and forgets
to update the COLOR_LAST constant.

This change looks for enums that have a xxx_LAST or xxxLast member,
and warns if there are any higher-valued constants present.

Review URL: https://codereview.chromium.org/150943005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251894 0039d316-1c4b-4281-b951-d872f2087c98

BUG=344704
NOTRY=true
TBR=vitalybuka@chromium.org

Review URL: https://codereview.chromium.org/171673003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251893 0039d316-1c4b-4281-b951-d872f2087c98

This will allow the embedder to change the autofill authentication based on
whether the functionality is available but has been turned off.

BUG=341492
R=gcasto@chromium.org, isherman@chromium.org

Review URL: https://codereview.chromium.org/170783005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251892 0039d316-1c4b-4281-b951-d872f2087c98

These files are automatically generated for the PDF extension. The long term
solution is not to check in vulcanize-generated files. These files may not
be needed once HTMLImports are fully implemented in chrome.

BUG=303491
R=arv@chromium.org, maruel@chromium.org

Review URL: https://codereview.chromium.org/154223003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251890 0039d316-1c4b-4281-b951-d872f2087c98

BUG=254545
R=cpu@chromium.org
TBR=jam

Review URL: https://codereview.chromium.org/165353006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251889 0039d316-1c4b-4281-b951-d872f2087c98

Begin blacklisting datamngr.dl, lmrn.dll and hk.dll from the browser
process.

BUG=329023

Review URL: https://codereview.chromium.org/171153002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251886 0039d316-1c4b-4281-b951-d872f2087c98

When opening a bubble on linux aura, we send a message to our anchor
Widget telling it to not render as if it was disabled. We never reenable
this when we go away.

BUG=342008

Review URL: https://codereview.chromium.org/171253002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251883 0039d316-1c4b-4281-b951-d872f2087c98

R=ericde@chromium.org
TBR=engedy@chromium.org
BUG=344643

Review URL: https://codereview.chromium.org/171483002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251881 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none

Review URL: https://codereview.chromium.org/164983009

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251878 0039d316-1c4b-4281-b951-d872f2087c98

This is a follow up CL of r251277

Now we invalidate the weak pointers during Decrypting*::Stop(), so all callbacks
are canceled. We don't need to manually cancel the NewKeyCB anymore.

TEST=Media unittests updated.

Review URL: https://codereview.chromium.org/167333004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251876 0039d316-1c4b-4281-b951-d872f2087c98

r251624 mistakenly removed this GetSystemColor call.

BUG=344593
TEST=New Profile Management UI buttons have the expected blue background on hover; otherwise look as expected.
TBR=sky@chromium.org,noms@chromium.org

Review URL: https://codereview.chromium.org/166273005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251872 0039d316-1c4b-4281-b951-d872f2087c98

BUG=341969
R=nkostylev@chromium.org

Review URL: https://codereview.chromium.org/169993002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251871 0039d316-1c4b-4281-b951-d872f2087c98

Only affects SDK build, not Chrome.

NOTRY=true
BUG=344575
TEST=build_sdk.py --bionic

see also parent issue:
  https://code.google.com/p/chromium/issues/detail?id=344574

Review URL: https://codereview.chromium.org/170973003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@251870 0039d316-1c4b-4281-b951-d872f2087c98