GitLab

Before on desktop, the cache backend type of the HTTP cache was
choosen by the SimpleCacheTrial on field trial. This can cause
issues with sandwich were the cache backend type might change
between runs.

This CL override the cache backend type to use in Chrome on
Desktop to not have to handle the two backend types in the cache
archive processing.

BUG=582080

Review-Url: https://codereview.chromium.org/1931523002
Cr-Commit-Position: refs/heads/master@{#390352}

BUG=607397

Review-Url: https://codereview.chromium.org/1927863003
Cr-Commit-Position: refs/heads/master@{#390351}

Move code to handle reporting of errors and notes out
into a class of its own.

R=haraken
BUG=531879

Review-Url: https://codereview.chromium.org/1926863002
Cr-Commit-Position: refs/heads/master@{#390350}

WebRTC 12519:12543

Changes: https://chromium.googlesource.com/external/webrtc/trunk/webrtc.git/+log/de1eeeb..57f37af

Libjingle 12501:12534

Changes: https://chromium.googlesource.com/external/webrtc/trunk/talk.git/+log/a2ec0b6..a3dc305

TBR=

CQ_EXTRA_TRYBOTS=tryserver.chromium.linux:linux_chromium_archive_rel_ng;tryserver.chromium.mac:mac_chromium_archive_rel_ng
BUG=

Review-Url: https://codereview.chromium.org/1926703004
Cr-Commit-Position: refs/heads/master@{#390349}

Spec: https://dom.spec.whatwg.org/#concept-event-initialize

BUG=563973

Review-Url: https://codereview.chromium.org/1925963002
Cr-Commit-Position: refs/heads/master@{#390348}

A recent regression caused the clip bounds for the
FloodFillInkDropAnimation to incorrectly be centered on the click point
for the DownloadItemViewMd buttons. This change fixes that and
suppresses the ripple for right clicks on the DownloadItemViewMd as
well.

BUG=603818
TEST=manual

Review-Url: https://codereview.chromium.org/1915983009
Cr-Commit-Position: refs/heads/master@{#390347}

Results of the following tests changed on Mac10.10 as a result of https://codereview.chromium.org/1922223003:

  fast/forms/suggestion-picker/time-suggestion-picker-appearance-rtl.html
  fast/forms/suggestion-picker/month-suggestion-picker-appearance-with-scroll-bar.html

BUG=414264
TBR=eae@chromium.org
NOTRY=true

Review-Url: https://codereview.chromium.org/1927903002
Cr-Commit-Position: refs/heads/master@{#390346}

A lot of data are being scattered in these classes.

To make it clean and their responsibilities clear, remove ShadowRootRareData,
then move data owned by it into other classes:

- Make SlotAssignment have a responsibility of managing all data related to slot assignments
- Other data are now owned by ShadowRoot directly

BUG=531990

Review-Url: https://codereview.chromium.org/1925043002
Cr-Commit-Position: refs/heads/master@{#390345}

Summary of changes available at:
https://chromium.googlesource.com/v8/v8/+log/8e49b3ef..9c51453c

Please follow these instructions for assigning/CC'ing issues:
https://github.com/v8/v8/wiki/Triaging%20issues

Please close rolling in case of a roll revert:
https://v8-roll.appspot.com/
This only works with a Google account.

TBR=hablich@chromium.org,machenbach@chromium.org,yangguo@chromium.org,vogelheim@chromium.org

Review-Url: https://codereview.chromium.org/1931803002
Cr-Commit-Position: refs/heads/master@{#390344}

Since crrev.com/390324, <meter> with -webkit-appearance:none doesn't show fancy
widget. Existing pages might remove -webkit-appearance:meter by
background/border unexpectedly. So we stop this behavior.

BUG=51182

Review-Url: https://codereview.chromium.org/1917253006
Cr-Commit-Position: refs/heads/master@{#390343}

We used to add "_apk" to all targets from this list in the internal waterfall.
With a recent change by mikecase@ we no longer do that.

It works for most targets, but not for "unit_tests", which does not exist in GN.

This change would need to be reverted if we decide to switch back to GYP, but
it is unlikely to happen.

BUG=607455

Review-Url: https://codereview.chromium.org/1924893004
Cr-Commit-Position: refs/heads/master@{#390342}

Revert of [Cronet] Avoid crashing when CronetEngines are created on multiple threads (patchset #2 id:20001 of https://codereview.chromium.org/1926683003/ )

Reason for revert:
CronetUrlRequestContextTest#testSkipLibraryLoading fails on public cronet bots:

http://crbug.com/607464

Original issue's description:
> [Cronet] Avoid crashing when CronetEngines are created on multiple threads
>
> Creating a CronetEngine on one thread calls
> CronetLibraryLoader.ensureInitialized() which posts to the main thread, but
> if a CronetEngine is created immediately after this on the main thread without
> first returning to the message loop, that initialization won't have happened
> and the initialization of the CronetEngine being created on the main thread
> could crash when it tries to get the current native MessageLoop.
>
> BUG=607178

TBR=mef@chromium.org,kapishnikov@chromium.org,pauljensen@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=607178

Review-Url: https://codereview.chromium.org/1927863004
Cr-Commit-Position: refs/heads/master@{#390341}

R=aizatsky@chromium.org, davidben@chromium.org, ochang@chromium.org
BUG=569578

Review-Url: https://codereview.chromium.org/1885493002
Cr-Commit-Position: refs/heads/master@{#390340}

This makes it easier to write tests, since each test doesn't have to
enable the field trial (which is especially hard from Java).

Part of a series of patches:
1. https://codereview.chromium.org/1832833002 adds InstanceIDWithSubtype
2. https://codereview.chromium.org/1830983002 adds JNI bindings
3. https://codereview.chromium.org/1829023002 adds fake and test
4. https://codereview.chromium.org/1899753002 fixes strict mode violations
5. this patch
6. https://codereview.chromium.org/1923953002 adds crypto integration
7. https://codereview.chromium.org/1851423003 switches Push to InstanceIDs

BUG=589461, 477084

Review-Url: https://codereview.chromium.org/1854093002
Cr-Commit-Position: refs/heads/master@{#390339}

Sometimes GlobalFetch::fetch is called with an invalid ScriptState.
Do not start fetching in such a case.

BUG=607156

Review-Url: https://codereview.chromium.org/1924193002
Cr-Commit-Position: refs/heads/master@{#390338}

Remove the static FontCache::s_useSubpixelPositioning field and the bool
FontPlatformData::m_useSubpixelPositioning field. They are only ever set
explicitly on Windows and always to true since the GDI path was removed.

BUG=579678
R=kojii@chromium.org

Review-Url: https://codereview.chromium.org/1930503003
Cr-Commit-Position: refs/heads/master@{#390337}

This patch locks down copying of PassOwnPtr<T> to facilitate migration
of OwnPtr to std::unique_ptr. This is the first step of filling the
functionality gap between OwnPtr and std::unique_ptr.

After this change, copying of PassOwnPtrs won't be possible; to express
ownership transfer, you need to explicitly use std::move() if the
PassOwnPtr is an lvalue. Also, you may need to use passed() wrapper
function if you want to bind a PassOwnPtr to a function and want the
pointer auto-passed to the function on the functor's first invocation.

This patch also includes fixes for the code landed after my earlier mass
fixes of compile errors.

BUG=582349

Review-Url: https://codereview.chromium.org/1917193006
Cr-Commit-Position: refs/heads/master@{#390336}

Size:
- before: 167MB
- after:  45MB

Review-Url: https://codereview.chromium.org/1923823003
Cr-Commit-Position: refs/heads/master@{#390335}

Makes trace_to_chrome_trace.py go from >1m to 20s.

Review-Url: https://codereview.chromium.org/1926513003
Cr-Commit-Position: refs/heads/master@{#390334}

Remove unneeded restoring of FBO bindings after creating
CopyTextureCHROMIUM helper instance.

CopyTextureCHROMIUMResourceManager::Initialize() does not change the FBO
bindings at all, so the removed call
GLES2DecoderImpl::RestoreCurrentFramebufferBindings should not be needed.

This is a work aiming to improve decoder MakeCurrent performance by
making more of the state updated in lazy manner. This works towards
updating the logic for FBOs. Removing the extra FBO restore simplifies
this.

BUG=603407
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_optional_gpu_tests_rel;tryserver.chromium.mac:mac_optional_gpu_tests_rel;tryserver.chromium.win:win_optional_gpu_tests_rel

Review-Url: https://codereview.chromium.org/1927503002
Cr-Commit-Position: refs/heads/master@{#390333}

Do not call glGetError for debug info after context has been lost.

BUG=581634
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_optional_gpu_tests_rel;tryserver.chromium.mac:mac_optional_gpu_tests_rel;tryserver.chromium.win:win_optional_gpu_tests_rel

Review-Url: https://codereview.chromium.org/1922703004
Cr-Commit-Position: refs/heads/master@{#390332}

Currently IsGarbageCollectedType<void> and PointerParamStorageTraits<void*> can't be compiled.
This CL makes them successfully compiled, and removes ParamStorageTraits<void*> specialization.

BUG=597856

Review-Url: https://codereview.chromium.org/1923183002
Cr-Commit-Position: refs/heads/master@{#390331}

Review-Url: https://codereview.chromium.org/1929803002
Cr-Commit-Position: refs/heads/master@{#390330}

BUG=b/28338159

Review-Url: https://codereview.chromium.org/1923363002
Cr-Commit-Position: refs/heads/master@{#390329}

BUG=

Review-Url: https://codereview.chromium.org/1918693007
Cr-Commit-Position: refs/heads/master@{#390328}

Formerly this event was needed to eliminate old WebContents leak on swap
(crbug.com/103966). Recently with heroic refactoring the
WebContentsDelegate::SwappedOut() stopped being delivered.

This makes it safe to remove OnCloseWebContentsDeleter::SwappedOut().

The old WebContents does not leak because the way it is replaced is:
1. Subscribe to CloseContents() with our WebContentsDelegate
2. old_web_contents->DispatchBeforeUnload()
3. After 3 seconds, destroy the old_web_contents regardless of beforeunload
   dispatching
4. In CloseContents() we also destroy old_web_contents, but maybe sooner

.. modulo some deduplication of destroying and other cleanup happening around
asynchronously.

BUG=600693

Review-Url: https://codereview.chromium.org/1896943002
Cr-Commit-Position: refs/heads/master@{#390327}

Further, and hopefully the last step for refactoring needed for the feature.
 1) We need mime types on the task running time as well. Added a cycle for it.
 2) Split the task enumeration function for the placeholder for ARC stuff and
   post processing part.
 3) For the efficiency and clarity for the split of 2), changed std::vector
   value copying to movement by std::unique_ptr.

BUG=607059
TEST=unit_tests --gtest_filter='*FileManagerFileTask*'
TEST=browser_tests --gtest_filter='FileManagerJsTest.FileTasks'

Review-Url: https://codereview.chromium.org/1923953005
Cr-Commit-Position: refs/heads/master@{#390326}

R=
BUG=
NOTRY=true

Review-Url: https://codereview.chromium.org/1928743002
Cr-Commit-Position: refs/heads/master@{#390325}

This CL changes the behavior of METER element with -webkit-appearance:none.

Before this CL:
  -webkit-appearance:none had almost no effect because -webkit-appearance:meter
  has no theme painting.

After this CL:
  -webkit-appearance:none disables all of UA-provided rendering of METER elements.
  meter::-webkit-meter-* pseudo elements don't work, and web authors need to
  render METER elements by themselves.
  This behavior follows appearance:none defined by css-ui-4 [1].

Implementation:
Because we can't update METER Shadow DOM structure in response to CSS style
resolution, a METER has the following structure.

  METER
    #shadow-root
      MeterInnerElement pseudo=-webkit-meter-inner-element
        DIV pseudo=-webkit-meter-bar
         DIV pseudo=-webkit-meter-*-value
      MeterFallbackElement
        CONTENT

When the METER has -webkit-appearance:meter, MeterFallbackElement has display:
none. Otherwise, MeterInnerElement has display:none.

[1] https://www.w3.org/TR/css-ui-4/#appearance-switching

BUG=51182

Review-Url: https://codereview.chromium.org/1877763002
Cr-Commit-Position: refs/heads/master@{#390324}

This dummy class was accidentally re-introduced in PersistentNode.cpp during the
r389469 revert; remove it.

R=
BUG=

Review-Url: https://codereview.chromium.org/1925973002
Cr-Commit-Position: refs/heads/master@{#390323}

if a class does not have exposed attributes.

Currently, all script wrappable classes have definition of
installConditionallyEnabledProperties(), and all the definition
is an empty function.
Beside it, WrapperTypeInfo checks if the method exists and
calls it.

This CL removes such definition and replaces the function
pointers in WrapperTypeInfo with nullptrs to avoid calling
empty functions.

BUG=503508

Review-Url: https://codereview.chromium.org/1924763002
Cr-Commit-Position: refs/heads/master@{#390322}

This CL adds support in Blink and content/ for capturing
VideoFrames via ImageCapture.grabFrame() method.
For that, it adds a web/WebImageCaptureFrameGrabber
interface implemented in a new content class
ImageCaptureFrameGrabber.

In order to pass ImageBitmaps in WebCallbacks,
ImageBitmap gets a method take().

It also splits the current LayoutTests into extended
-creation.html and -grabFrame.html.

BUG=518807

TEST= run demo html in
https://cdn.rawgit.com/Miguelao/demos/master/imagecapture.html
with flag --enable-blink-features=ImageCapture
(click on "Open Camera 320x240", then on
"Create Image Capturer" and finally, repeatedly,
on "grabFrame").

Review-Url: https://codereview.chromium.org/1899403002
Cr-Commit-Position: refs/heads/master@{#390321}

Previously, the update job would use the script url provided when the
job was scheduled.

This change fixes the crash when the update job starts after the registration
has no versions, and also matches the spec.

BUG=604991

Review-Url: https://codereview.chromium.org/1928443003
Cr-Commit-Position: refs/heads/master@{#390320}

We are going to make WTF::bind() to depend less on ParamStorageTraits and
simplify ParamStorageTraits, for WTF::bind() refactoring and merging
WTF::bind() and base::Bind().
This CL removes uses of ParamStorageTraits from outside WTF::bind() to make
those activities easier.

BUG=597856

Review-Url: https://codereview.chromium.org/1920213002
Cr-Commit-Position: refs/heads/master@{#390319}

The file only contains one owner, who no longer works on this part of
chrome.

BUG=NONE

Review-Url: https://codereview.chromium.org/1925533005
Cr-Commit-Position: refs/heads/master@{#390318}

In some cases, we would navigate the main frame when going back/forward
in a subframe.  This was because the FTN ID had changed from what was
stored in the FrameNavigationEntry.  We now use the frame's unique name
for subframes and the position in the tree for main frames, as is done
in default Chrome (in HistoryController).

This also relaxes the restriction that a FrameNavigationEntry's item
and document sequence numbers don't change, at least until
https://crbug.com/596707 is fixed.

This is a second attempt, after r386756 caused issue 603245.

BUG=586324, 568768
TEST=See bug for repro steps.
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_site_isolation

Review-Url: https://codereview.chromium.org/1906213003
Cr-Commit-Position: refs/heads/master@{#390317}

Change the LayoutInline::linesBoundingBox method, and subclass overrides
in LayoutText and LayoutSVGInlineText, to return a LayoutRect instead of
an IntRect. Thereby exposing more correct and consistent offsetWidth and
offsetHeight values through the DOM. Move enclosing int logic to callers
as needed, particularly for PaintLayer, editing, and spatial navigation.

TEST=fast/sub-pixel/width-of-inline-in-float.html
BUG=414264
R=drott@chromium.org

Review-Url: https://codereview.chromium.org/1922223003
Cr-Commit-Position: refs/heads/master@{#390316}

In this CL, we implement the changes to the platform independent and BlueZ
specific class changes to support local GATT services. The major changes
from the code in trunk are,

.) Cleaning up of the classes and implementing all missing parts.
.) Adding code to properly take ownership of attributes of all three kinds.
.) Adding structure to the attributes to correctly setup the hierarchy
expected (services <-- characteristics <-- descriptors).
.) Addition of an AttributeDelegate to handle calls from DBus and wrappers
to translate them to the BluetoothLocalGattService::Delegate.
.) Cleanup in all the code touched.

Part 1 of a 3 patch series:
https://crrev.com/1915803002 Bluetooth class changes <<<
https://crrev.com/1914893002 DBus class changes
https://crrev.com/1919683002 Adapter changes + tests

R=ortuno@chromium.org, scheib@chromium.org
BUG=601935

Review-Url: https://codereview.chromium.org/1915803002
Cr-Commit-Position: refs/heads/master@{#390315}

BUG=606944

Review-Url: https://codereview.chromium.org/1917223004
Cr-Commit-Position: refs/heads/master@{#390314}

Summary of changes available at:
https://chromium.googlesource.com/v8/v8/+log/b86a9f9c..8e49b3ef

Please follow these instructions for assigning/CC'ing issues:
https://github.com/v8/v8/wiki/Triaging%20issues

Please close rolling in case of a roll revert:
https://v8-roll.appspot.com/
This only works with a Google account.

TBR=hablich@chromium.org,machenbach@chromium.org,yangguo@chromium.org,vogelheim@chromium.org

Review-Url: https://codereview.chromium.org/1932643002
Cr-Commit-Position: refs/heads/master@{#390313}