GitLab

Currently, an app can exceed the number of decoders that vendor backends
support (for example, by creating multiple video elements), leading to
unpredictable behaviour including crashes.  While most apps don't do
this, we have seen some problems crop up e.g. when ads are injected into
an existing page.

For now, restrict to 1 video decoder and 2 audio decoders.

BUG=internal b/26910920

Review-Url: https://codereview.chromium.org/2173593002
Cr-Commit-Position: refs/heads/master@{#407215}

For iframes, we don't want to use the source url for determining the
associated extension because it starts out with an about:blank context
that is scriptable by its parent.

BUG=573131

Review-Url: https://codereview.chromium.org/2151693002
Cr-Commit-Position: refs/heads/master@{#407214}

Also adds a unit test using the function that is used downstream to
prevent this from happening again.

BUG=630260

Review-Url: https://codereview.chromium.org/2173873002
Cr-Commit-Position: refs/heads/master@{#407213}

BUG=590369
R=yosin,esprehn

Review-Url: https://codereview.chromium.org/2144673003
Cr-Commit-Position: refs/heads/master@{#407212}

Before this patch, when root layer scrolling was enabled, both FV and PLSA would
perform scroll anchoring. FV doesn't need perform scroll anchoring for the main
frame because PLSA will do the right thing.

This issue was found because the following tests failed when ScrollAnchoring was
turned on by default:
virtual/rootlayerscrolls/fast/scrolling/scroll-into-view-collapsed-div.html
virtual/rootlayerscrolls/fast/scrolling/fractional-scroll-offset-document.html

BUG=597479,558575

Review-Url: https://codereview.chromium.org/2176523003
Cr-Commit-Position: refs/heads/master@{#407211}

This is sufficiently different from regular accelerated video decode
that we want to be able to blacklist it separately.

BUG=616318
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel

Review-Url: https://codereview.chromium.org/2167423002
Cr-Commit-Position: refs/heads/master@{#407210}

DevTools: keep widgets in widget hierarchy upon hide, split attach/detach cycle from show/hide (reland)

TBR=dgozman

Review-Url: https://codereview.chromium.org/2168223002
Cr-Commit-Position: refs/heads/master@{#407209}

Database data loss is a non-standard extension to IDBVersionChangeEvent
(https://www.w3.org/Bugs/Public/show_bug.cgi?id=22370). This state only
needs to be persisted as part of the database open. So moving this value
from IndexedDBCallbacks (used for any IDB call) to
IndexedDBPendingConnection (only used during database open).

BUG=627484

Review-Url: https://codereview.chromium.org/2140193002
Cr-Commit-Position: refs/heads/master@{#407208}

BUG=None

Review-Url: https://codereview.chromium.org/2179523002
Cr-Commit-Position: refs/heads/master@{#407207}

We used to return null on bind and avoid connecting in custom tabs
service if FRE is not completed. This change allows bind and connection
but returns false on any code that require network or browser UI
interaction. This way the client can still register a new session and
get callbacks.

BUG=630339

Review-Url: https://codereview.chromium.org/2174453002
Cr-Commit-Position: refs/heads/master@{#407206}

BUG=553459

Review-Url: https://codereview.chromium.org/2116163002
Cr-Commit-Position: refs/heads/master@{#407205}

Revert of Remove platform/text/ParserUtilities.h (patchset #2 id:20001 of https://codereview.chromium.org/2176623003/ )

Reason for revert:
Likely cause of failures:

---
https://build.chromium.org/p/chromium.webkit/builders/WebKit%20Linux/builds/67858

unexpected_failures:
fast/forms/color/color-suggestion-picker-appearance-zoom125.html
fast/forms/color/color-suggestion-picker-with-scrollbar-appearance.html
fast/forms/color/color-suggestion-picker-appearance-zoom200.html

all with the message
image diff
---

---
also bot:
https://build.chromium.org/p/chromium.webkit/builders/WebKit%20Mac10.11%20%28dbg%29/builds/3384

http/tests/images/restyle-decode-error.html

in which case there is a renderer crash in this test
---

also some other bots for tests
fast/forms/color/color-suggestion-picker-appearance.html
fast/forms/color/color-suggestion-picker-one-row-appearance.html
fast/forms/color/color-suggestion-picker-two-row-appearance.html

Original issue's description:
> Remove platform/text/ParserUtilities.h
>
> platform/ParsingUtilities.h caters to the same needs, so transition
> users of skipString(...) to skipToken(...) and remove
> platform/text/ParserUtilities.h.
>
> Committed: https://crrev.com/39cc523b5dbf9d2ddad9483cb27fab8b97e53ec1
> Cr-Commit-Position: refs/heads/master@{#407187}

TBR=schenney@chromium.org,pdr@chromium.org,fs@opera.com
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true

Review-Url: https://codereview.chromium.org/2179563002
Cr-Commit-Position: refs/heads/master@{#407204}

https://chromium.googlesource.com/webm/libvpx.git/+log/18c7f46c125c..4b073bc39a39

$ git log 18c7f46c1..4b073bc39 --date=short --no-merges --format='%ad %ae %s'
2016-07-21 yunqingwang Add back header in threading.h
2016-07-14 yunqingwang Revert "Amend and improve VP8 multithreading implementation"

R=johannkoenig@google.com

Review-Url: https://codereview.chromium.org/2168253002
Cr-Commit-Position: refs/heads/master@{#407203}

FileReader.readAsBinaryString() was removed from the spec after typed
arrays became a thing. But all browsers kept shipping it, and it looks
risky to remove, so it was added back to the spec. Remove the use
counters and TODOs in the IDL files.

R=foolip@chromium.org

Review-Url: https://codereview.chromium.org/2171953002
Cr-Commit-Position: refs/heads/master@{#407202}

In current code, some of the DCHECKs are having multiple conditions.
It's difficult to identify the failures if DCHECK fails due to one
of those conditions. Now moved the conditions to separate DCHECKs, so
that failure can identify at correct point.

BUG=

Review-Url: https://codereview.chromium.org/2168093002
Cr-Commit-Position: refs/heads/master@{#407201}

Not enough coverage for the asserts at the moment (hopefully, we will
get them back with the newer android toolchains eventually), and this
seems like an important enough one to special case.

Context:
https://codereview.chromium.org/2165333002/#msg9

BUG=

Review-Url: https://codereview.chromium.org/2177473004
Cr-Commit-Position: refs/heads/master@{#407200}

This eliminates some accesses of ash::Shell inside the shelf code, which is
necessary for porting the shelf to mus/mustash.

* ShelfModel now owns the ShelfItemDelegates
* Explicitly delete the ShelfItemDelegates during shutdown, because some of
the subclasses reach into ShelfModel during their destructors
* Move AppListShelfItemDelegate earlier, to WmShell::Initialize phase
* Fix ShelfAppBrowserTest.LauncherContextMenuVerifyCloseItemAppearance to use
the actual shelf ID of the browser shortcut, rather than using the magic
number "1" that can sometimes mean the app list item.

This also simplifies some of the unit tests for ChromeLauncherController
because they don't need to construct their own ShelfItemDelegateManager.

BUG=629990,429870
TEST=ash_unittests, chrome unit_tests, manually test shelf pin/unpin/etc.

Review-Url: https://codereview.chromium.org/2171813004
Cr-Commit-Position: refs/heads/master@{#407199}

AcceleratorDelegate now trivially forwards to AcceleratorRouter, which
is now in terms of ash/common types.

BUG=612331
TEST=none
R=jamescook@chromium.org

Review-Url: https://codereview.chromium.org/2171983002
Cr-Commit-Position: refs/heads/master@{#407198}

We have an off-by-one error now that the mOptionLayout can have multiple children in it above the actual options.

* Use the correct index when indexing into the options layout.

* Add a test that checks for the crash and additionally checks that the "merchant doesn't accept address" string is
correctly displayed.

* Plumbs a new callback that indicates that the client has finished
checking the selection.

BUG=630348

Review-Url: https://codereview.chromium.org/2170603006
Cr-Commit-Position: refs/heads/master@{#407197}

Has never passed since it was introduced.  Disabling on low end
for now.

BUG=626289

Review-Url: https://codereview.chromium.org/2170183004
Cr-Commit-Position: refs/heads/master@{#407196}

https://chromium.googlesource.com/skia.git/+log/d50e2401787d..cbecb082d15d

$ git log d50e24017..cbecb082d --date=short --no-merges --format='%ad %ae %s'
2016-07-22 csmartdalton Allow GrReducedClip to take non-integer query bounds
2016-07-22 mtklein Add SkRasterPipeline blitter.

CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_blink_rel
TBR=jcgregorio@google.com

Review-Url: https://codereview.chromium.org/2171393002
Cr-Commit-Position: refs/heads/master@{#407195}

Raises an NSInvalidArgumentException if call setObject for NSMutableDictionary
with a nil value. Dictionary uses NSNull instead of nil.
Selected text marker range may be nil sometimes
(see BrowserAccessibilityCocoa::selectedTextMarkerRange), that leads to a crashe
in BrowserAccessibilityManagerMac::GetUserInfoForSelectedTextChangedNotification().

R=aboxhall@chromium.org, dmazzoni@chromium.org, dtseng@chromium.org
BUG=

Review-Url: https://codereview.chromium.org/2161623002
Cr-Commit-Position: refs/heads/master@{#407194}

BUG=587995

Review-Url: https://codereview.chromium.org/2175613003
Cr-Commit-Position: refs/heads/master@{#407193}

This leaves at least 1 day between the value of now() and thisUpdate/nextUpdate in OCSP responses generated by minica.py. This gives more wiggle room if there's a time differential between a test bot and a RemoteTestServer, and should help reduce HTTPSOCSPVerify test flakiness.

This also removes a reference to a non-existant variable in what should be an unreachable code path in minica.py.

BUG=630286

Review-Url: https://codereview.chromium.org/2170643004
Cr-Commit-Position: refs/heads/master@{#407192}

Reland of ll buildtools 55638fe5c3..60f7f9a8b4 (patchset #1 id:1 of https://codereview.chromium.org/2171273002/ )

Reason for revert:
GN binaries will not affects GYP's errors about
conditions.

Original issue's description:
> Revert of Roll buildtools 55638fe5c3..60f7f9a8b4 (patchset #1 id:1 of https://codereview.chromium.org/2166993002/ )
>
> Reason for revert:
> Speculative revert - this probably broke the official build - see crbug.com/630514 for more details. Will reland if the suspicion was wrong.
>
> Original issue's description:
> > Roll buildtools 55638fe5c3..60f7f9a8b4
> >
> >   In order to roll GN af92718f..1f417360 (r405547:r406705) and pick up
> >   the following changes:
> >
> >   267b1de9 Make `gn refs` handle files in data directories properly.
> >   5fc3fafd Add TaskPriority as a parameter to SequencedWorkerPool in preparation for TaskScheduler experiment.
> >   aca08753 Make `gn refs` look at script files.
> >   05316e06 Re-land r406064 "[GN] Add JSON project writer".
> >   5a1b4b17 Revert of [GN] Add JSON project writer (patchset #11 id:200001 of https://codereview.chromium.org/2064533002/ )
> >   d2edeb9a [GN] Add JSON project writer
> >   f0544fc6 Remove calls to MessageLoop::current() in tools.
> >   d93e5cf4 Document GN invoker and target name variables.
> >
> > TBR=dpranke@chromium.org
> >
> > Committed: https://crrev.com/d8a18dc3820a4d54d8fc9ad8667d242798f10d33
> > Cr-Commit-Position: refs/heads/master@{#406727}
>
> TBR=dpranke@chromium.org,brettw@chromium.org
> # Not skipping CQ checks because original CL landed more than 1 days ago.
>
> Committed: https://crrev.com/de249eccb4145727c8eee6ecb737bd50cbbf844e
> Cr-Commit-Position: refs/heads/master@{#407125}

TBR=dpranke@chromium.org,msramek@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true

Review-Url: https://codereview.chromium.org/2172323002
Cr-Commit-Position: refs/heads/master@{#407191}

This only affects test targets and Content Shell.

R=sdefresne@chromium.org,mmenke@chromium.org

Review-Url: https://codereview.chromium.org/2174853002
Cr-Commit-Position: refs/heads/master@{#407190}

All of the known bugs have been addressed; let's see what isn't known.
This affects:

  chromium.lkgr:
    - ASAN Debug
    - ASAN Release Media
    - ASAN Release
    - ASAN Debug (32-bit x86 with V8-ARM)
    - ASAN Release (32-bit x86 with V8-ARM)
    - ASAN Release Media (32-bit x86 with V8-ARM)
    - TSAN Debug
    - TSAN Release
    - UBSan Release
    - UBSan vptr Release

With this CL, the only remaining linux builders that use GYP are
the 'Closure Compilation' builders. The LKGR waterfall will still
have a few Win GYP builders that will be flipped in a follow-up CL.

R=inferno@chromium.org, mbarbella@chromium.org
BUG=621581

Review-Url: https://codereview.chromium.org/2167963002
Cr-Commit-Position: refs/heads/master@{#407189}

This is hoping to achieve better build performance.

BUG=627637

Review-Url: https://codereview.chromium.org/2170813003
Cr-Commit-Position: refs/heads/master@{#407188}

platform/ParsingUtilities.h caters to the same needs, so transition
users of skipString(...) to skipToken(...) and remove
platform/text/ParserUtilities.h.

Review-Url: https://codereview.chromium.org/2176623003
Cr-Commit-Position: refs/heads/master@{#407187}

Some preliminary work to remove PlatformEvent abstractions.

Rename a constant to pass check-webkit-style.

BUG=625684
TBR=rbyers@chromium.org

Review-Url: https://codereview.chromium.org/2173073003
Cr-Commit-Position: refs/heads/master@{#407186}

The font cascade reconfiguration which was introduced as fix for AAT
shaping performance regressions in crbug.com/547912 seems to
occasionally cause CoreText crashes on OS X 10.9. We don't have a better
way of detecting this than by OS or CoreText API version number. This is
one of our top Mac crashers on Mac OS 10.9 with Chrome across versions [1].

This crash does not occur in newer versions of OS X and we can keep this
important performance optimization enabled there.

A big thanks to Robert Sesek (rsesek@) for the patient and thorough
initial investigation. Discussing and working together on this issue we
were able to identify the crash triggering code in HarfBuzz in this
case.

[1] https://bugs.chromium.org/p/chromium/issues/detail?id=576941#c74

BUG=576941, 625902

Review-Url: https://codereview.chromium.org/2173883002
Cr-Commit-Position: refs/heads/master@{#407185}

I wrote this while investigating a Promise issue. I think it wouldn't
hurt to land it :)

BUG=None

Review-Url: https://codereview.chromium.org/2178463002
Cr-Commit-Position: refs/heads/master@{#407184}

libyuv offers a new API, Android420ToI420, to convert the
Android YUV_420_888 frame (which has interleaved UV planes)
to normal I420 frame.
This can significantly reduce the de-interlacing time on
Android, up to dozens of milliseconds, which can help a lot
on the overall end-to-end video delay.

BUG=629342

Review-Url: https://codereview.chromium.org/2156003006
Cr-Commit-Position: refs/heads/master@{#407183}

Android lint tool pointed out that ... should really be … (U+2026).
Tested that this glyph renders well on Mac + Win as well, so
applying to all OS's by default.

TBR=flackr
BUG=621772

Review-Url: https://codereview.chromium.org/2112653003
Cr-Commit-Position: refs/heads/master@{#407182}

WindowServer::GetTreeWithId can be null in the case of
|in_nonclient_area|. It looks like |tree| was used without check in
4d3588cb.

BUG=630402

Review-Url: https://codereview.chromium.org/2170053003
Cr-Commit-Position: refs/heads/master@{#407181}

BUG=

Review-Url: https://codereview.chromium.org/2148293005
Cr-Commit-Position: refs/heads/master@{#407180}

This patch arranges the payment instruments in the following order:
> Non-autofill payment apps
> Complete credit cards from Autofill.
> Incomplete credit card from Autofill (need editing before sending to
  the merchant).

BUG=629882

Review-Url: https://codereview.chromium.org/2165163002
Cr-Commit-Position: refs/heads/master@{#407179}

Both Chromium and Blink code can use std::unique_ptr<> recently, so now
we can express the ownership passing on the type of WebTaskRunner::clone.
This CL replaces the return type of WebTaskRunner::clone from a raw pointer
to a std::unique_ptr<>, removes adoptClone, and updates the call-sites.

Review-Url: https://codereview.chromium.org/2148363004
Cr-Commit-Position: refs/heads/master@{#407178}

When escape is pressed to hide the spectrum, use commitEdit to return to
the original color text.

R=lushnikov

Review-Url: https://codereview.chromium.org/2171003002
Cr-Commit-Position: refs/heads/master@{#407177}

Revert of mus: Rename .*PointerWatcher.* to .*PointerDownWatcher.*. (patchset #5 id:80001 of https://codereview.chromium.org/2158273003/ )

Reason for revert:
As discussed (https://codereview.chromium.org/2163453002/), changing PointerWatcher to watch for all events instead.

Original issue's description:
> mus: Rename .*PointerWatcher.* to .*PointerDownWatcher.* to avoid confusion with PointerEventWatcher (https://codereview.chromium.org/2163453002/).
>
> Committed: https://crrev.com/b2db88d57e677ab3ff4b3fc7f58896603816053c
> Cr-Commit-Position: refs/heads/master@{#406584}

Review-Url: https://codereview.chromium.org/2167643005
Cr-Commit-Position: refs/heads/master@{#407176}