GitLab

When the browser controls (URL bar) are hidden/shown, they change the visible
area of the viewport, allowing more or less scrolling. Until a resize occurs and
the FrameView/LayoutView is resized, we keep an "adjustment" value that we use to
fudge the maximumScrollOffset in the interim.

When a scroller is set as the document.rootScroller, we want it to behave in
the same way. This patch extends the behavior to all FrameViews and
PaintLayerScrollableAreas when they become the global root scroller. The
adjustment value is no longer stored in the FrameView. Instead,
TopDocumentRootScrollerController uses the value stored in VisualViewport to
calculate the current visible area of the viewport.

BUG=505516
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_layout_tests_slimming_paint_v2

Review-Url: https://codereview.chromium.org/2622103002
Cr-Commit-Position: refs/heads/master@{#443079}

Changes:
Bug happens when v2app like Files is pinned, the ash::ShelfItemType for this case is TYPE_APP_SHORTCUT.

BUG=675032
TEST=emulator test saw bug fixed

Review-Url: https://codereview.chromium.org/2628023002
Cr-Commit-Position: refs/heads/master@{#443078}

On MacViews, the tests ComboboxTest.NotifyOnClickWithSpaceKey and
ComboboxTest.NotifyOnClickWithReturnKey do not terminate currently when run
individually. This only happens on Mac, because on other platforms the combobox
dropdown menu is async (non-blocking). However on Mac, a native NSMenu is used
to show the combobox dropdown, which runs a nested message loop and hence is
blocking.

To solve, install a TestMenuRunner for all the tests using the ComboboxTest
fixture. This ensures that an actual menu is not created.

BUG=679980
TEST= Run out/Default/views_unittests
--gtest_filter="ComboboxTest.NotifyOnClickWithSpaceKey" on Mac. Ensure that the
test terminates normally.

Review-Url: https://codereview.chromium.org/2620963002
Cr-Commit-Position: refs/heads/master@{#443077}

media::PipelineMetadata::natural_size could be either the orignal
video natural size or the rotated one when VideoRotation indicates a
90 degree rotation. To avoid confusion, this Cl let it always store
the video natural size after rotation.

BUG=679574

Review-Url: https://codereview.chromium.org/2622803002
Cr-Commit-Position: refs/heads/master@{#443076}

https://skia.googlesource.com/skia.git/+log/a5494f117086..8d80bb5f20d8

$ git log a5494f117..8d80bb5f2 --date=short --no-merges --format='%ad %ae %s'
2017-01-11 herb Revert "Introduce SkArenaAlloc - should be fast for POD types and RAII for types with dtors."
2017-01-11 halcanary SkUTF8_CountUnichars(s,l) and SkUTF8_NextUnichar(s,l) now safe.
2017-01-11 bsalomon Fix undefined GrIORef test method on Chrome win bot
2017-01-11 herb Introduce SkArenaAlloc - should be fast for POD types and RAII for types with dtors.
2017-01-11 jvanverth More fixes for distance field paths
2017-01-11 bsalomon Make GrPaints move their GrProcessor ownership into GrPipelineBuilder.
2017-01-11 reed move SkTRegister.h into tools
2017-01-11 bungeman Remove SK_LEGACY_FONTMGR_FACTORY.
2017-01-11 reed remove dead file: PathOpsSkpClipTest.cpp
2017-01-11 mtklein SkSplicer: start on Windows support
2017-01-11 halcanary SkTestTypeface: correct encoding
2017-01-11 halcanary SkTypes.h : move SkAutoMalloc into SkAutoMalloc.h
2017-01-11 jvanverth Revert "More fixes for distance field paths"
2017-01-11 scroggo Make SkColorToHSV and SkHSVToColor "perfect" inverses
2017-01-11 jvanverth More fixes for distance field paths
2017-01-11 kjlubick Revert "SkTypes.h : move SkAutoMalloc into SkAutoMalloc.h"
2017-01-11 bsalomon Add test for processor->resource ref/io counts

BUG=677889,677889,b/33737498,677889

Documentation for the AutoRoller is here:
https://skia.googlesource.com/buildbot/+/master/autoroll/README.md

If the roll is causing failures, see:
http://www.chromium.org/developers/tree-sheriffs/sheriff-details-chromium#TOC-Failures-due-to-DEPS-rolls

CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel
TBR=kjlubick@google.com

Review-Url: https://codereview.chromium.org/2623123003
Cr-Commit-Position: refs/heads/master@{#443075}

BUG=678276

Review-Url: https://codereview.chromium.org/2628983003
Cr-Commit-Position: refs/heads/master@{#443074}

Since std::deque is memory inefficient for small numbers of elements,
especially on Android, (see http://crbug.com/674287), this CL replaces
the std::deque in DataReductionProxyEventStore with a simple circular
buffer.

BUG=679603

Review-Url: https://codereview.chromium.org/2619373003
Cr-Commit-Position: refs/heads/master@{#443073}

I suspect that host.options was originally a simple dictionary, and so the cast was safe. When it got upgraded to an object with methods on the prototype it broke, but because the window of opportunity for failure is small (you have to connect without a call to HostList.refresh happening in the interim), it was difficult to track down.

BUG=680302

Review-Url: https://codereview.chromium.org/2621403004
Cr-Commit-Position: refs/heads/master@{#443072}

This should hopefully get the bots back to green while individual bugs
are being investigated.

BUG=680201, 661725, 680249, 680307
NOTRY=true

Review-Url: https://codereview.chromium.org/2625233002
Cr-Commit-Position: refs/heads/master@{#443071}

When moving from checkboxes -> toggles, I unwrapped the label text of
UIs using

  <paper-checkbox hidden=[[...]]>...</paper-checkbox>

to

  <div>...</div>
  <paper-toggle-button hidden=[[...]]></paper-toggle-button>

I forgot to move the hidden=[[]] bindings, and only the toggle would get
hidden on platforms that don't support the data type.

R=tommycli@chromium.org
BUG=none
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2624203003
Cr-Commit-Position: refs/heads/master@{#443070}

Before:

<include src="...">
<if expr="chromeos">
    chromeOsOnlyCode();
</if>

After:

// <includes src="...">
// <if expr="chromeos">
    chromeOsOnlyCode();
// </if>

This is to unblock running clang-format on JavaScript and allow other
tools to run on more expected input inside of .js files (i.e. closure
compiler).

BUG=678778
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2624503002
Cr-Commit-Position: refs/heads/master@{#443069}

R=dpranke@chromium.org

Review-Url: https://codereview.chromium.org/2630463002
Cr-Commit-Position: refs/heads/master@{#443068}

This is a refactoring change related to, but separate from,
the issue of replacing wdiff and PrettyPatch.rb with a Python
diff module (http://crbug.com/672651), and a follow-up to
adding html_diff.py (http://crrev.com/2580143002).

Reason: The functionality of getting a text diff shouldn't be platform
dependent, so it doesn't really seem to belong in the Port class. (It
seems more similar to the html_diff module in webkitpy/common/.)

Review-Url: https://codereview.chromium.org/2595983002
Cr-Commit-Position: refs/heads/master@{#443067}

chrome/browser/android/data_usage has a few uses of std::deque that
would work just about as well using std::vector, and since std::deque is
memory inefficient for small numbers of elements (see
http://crbug.com/674287), this CL replaces those uses of std::deque with
std::vector.

BUG=679568

Review-Url: https://codereview.chromium.org/2624663002
Cr-Commit-Position: refs/heads/master@{#443066}

Adds BluetoothAdapter::Observer callbacks to Adapter implementation
for tracking changes in Adapter state including:
  AdapterDiscoverableChanged
  AdapterPoweredChanged
  AdapterPresentChanged

Adds adapter page to display details about the current state of the adapter.
Adds ObjectFieldSet interface component for displaying properties of a JavaScript object.

Screenshot: https://goo.gl/photos/dCbsULiydMbiAtiJ9

BUG=651282
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2567983007
Cr-Commit-Position: refs/heads/master@{#443065}

This is a mechanical change to convert i18n-content to $i18n{} tags.

BUG=677338

Review-Url: https://codereview.chromium.org/2628093002
Cr-Commit-Position: refs/heads/master@{#443064}

https://chromium.googlesource.com/angle/angle.git/+log/98e3407..008450b

BUG=None

TBR=geofflang@chromium.org

TEST=bots

CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.win:win_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.android:android_optional_gpu_tests_rel

Review-Url: https://codereview.chromium.org/2627713006
Cr-Commit-Position: refs/heads/master@{#443063}

Debian 8 restricts use of CLONE_NEWUSER to only processes with
CAP_SYS_ADMIN. (https://github.com/semplice/linux/blob/master/debian/patches/debian/add-sysctl-to-disallow-unprivileged-CLONE_NEWUSER-by-default.patch)
Chrome was previously checking if the kernel supported CLONE_NEWUSER
by running clone(CLONE_NEWUSER, ...) with the same capabilities chrome
was launched with.  This leads to 2 scenarios:

1. If Chrome was run as root:
   The check for CLONE_NEWUSER will succeed.  Chrome will then set up
   the namespace sandbox by clone()'ing and dropping CAP_SYS_ADMIN.
   Subsequent clone()'s with CLONE_NEWUSER will then fail.

2. If Chrome was run as a normal user:
   The check for CLONE_NEWUSER will fail.  Chrome will fallback to
   using the setuid sandbox.

The solution is to simply drop CAP_SYS_ADMIN before the check.

In addition, this CL disallows running Chromium as root unless launched
with --no-sandbox.

BUG=638180

Review-Url: https://codereview.chromium.org/2578483002
Cr-Commit-Position: refs/heads/master@{#443062}

We had previously added ForceReclaimResources to BeginCommit to throttle
frame output and prevent starving the display compositor's rasterization
tasks. Unfortunately, this introduces significant overhead as well.

With the addition of the Display Scheduler, this change no longer
appears to be needed. The scheduler has independent controls for
limiting the number of pending frames, and these should be used instead
(if further tweaking is necessary).

Removing this logic does not regress the benchmarks which it was
initially added for, and seems unneeded at this point. Removing.

R=reveman@chromium.org
BUG=676852,489515,617268
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel

Review-Url: https://codereview.chromium.org/2609253003
Cr-Commit-Position: refs/heads/master@{#443061}

BUG=388780

Review-Url: https://codereview.chromium.org/2608073002
Cr-Commit-Position: refs/heads/master@{#443060}

Previously scheduled cache mutation operations would begin the
size update calculation and then return to potentially begin
another mutation operation in parallel with the running size
update calculation. This could result in size held by CacheStorage
becoming mismatched with the actual cache size if Chrome were to
crash before the index could be written.

This change completes the size calculation before calling the
closure to begin the next scheduled operation.

BUG=678455

Review-Url: https://codereview.chromium.org/2613183003
Cr-Commit-Position: refs/heads/master@{#443059}

For device with high DPR, the image could be loaded from srcset. When
the img element is serialized, srcset attribute is skipped. We need to
set width and height attributes to retain the original size of the
image.

BUG=674999
TEST=new tests added

Review-Url: https://codereview.chromium.org/2624953002
Cr-Commit-Position: refs/heads/master@{#443058}

https://pdfium.googlesource.com/pdfium.git/+log/8fa82794ffc2..96f482c9cd3c

$ git log 8fa82794f..96f482c9c --date=short --no-merges --format='%ad %ae %s'
2017-01-11 dsinclair Convert FDE CSS enums to enum classes.
2017-01-11 npm Really fix m_nb_mct_records calculation in opj_j2k_read_mct

BUG=678461,680102

Documentation for the AutoRoller is here:
https://skia.googlesource.com/buildbot/+/master/autoroll/README.md

If the roll is causing failures, see:
http://www.chromium.org/developers/tree-sheriffs/sheriff-details-chromium#TOC-Failures-due-to-DEPS-rolls

TBR=dsinclair@chromium.org

Review-Url: https://codereview.chromium.org/2623103004
Cr-Commit-Position: refs/heads/master@{#443057}

(And also use it for positioned boxes)

We want to use this for layoutng/legacy interop. For legacy layout,
there should be no change in behavior because flexbox is currently
the only caller of this function:
https://cs.chromium.org/search/?q=setOverrideLogicalContentWidth&sq=package:chromium&type=cs

Interestingly, the height version did not have this limitation already.

This patch also deleted an outdated FIXME comment.

R=atotic@chromium.org,eae@chromium.org
BUG=635619

Review-Url: https://codereview.chromium.org/2624143005
Cr-Commit-Position: refs/heads/master@{#443056}

This patch removes two functions in InvalidatableInterpolation that are
unreached: apply and getCachedValueForTesting.

Since InvalidatableInterpolation has cached values, I had originally intended to
supply a meaningful implementation for getCachedValueForTesting. However, after
further consultation, I'm instead removing the virtual function from the
Interpolation superclass and modifying the tests that use it to make it explicit
that they will be executing LegacyStyleInterpolation::getCachedValueForTesting.

BUG=678875

Review-Url: https://codereview.chromium.org/2627793002
Cr-Commit-Position: refs/heads/master@{#443055}

Also removes an oddity with CertVerifyResult::operator== by handling the null case instead of documenting it as invalid.

operator== was only used by a unit-test (components/cronet/android/cert/cert_verifier_cache_serializer_unittest.cc), so this generalization does not weaken nullability guarantees.

Review-Url: https://codereview.chromium.org/2624283002
Cr-Commit-Position: refs/heads/master@{#443054}

Using update-w3c-deps in Chromium 2e36b96a.

Build: https://build.chromium.org/p/chromium.infra.cron/builders/w3c-test-autoroller/builds/7867

TBR=qyearsley@chromium.org
NOEXPORT=true

Review-Url: https://codereview.chromium.org/2623223003
Cr-Commit-Position: refs/heads/master@{#443053}

 - Adds the touch calibration setting to settings > display
 - Adds a command line switch to enable the settings.
 - Updates system_display.idl to add a new parameter.

BUG=634166
COMPONENT=System Display API, ChromeOS Switch, md settings
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2620073002
Cr-Commit-Position: refs/heads/master@{#443052}

The initial value for scale is none not 1
https://drafts.csswg.org/css-transforms-2/#individual-transforms

BUG=679873

Review-Url: https://codereview.chromium.org/2620263002
Cr-Commit-Position: refs/heads/master@{#443051}

This removes the access to x509_chain in Chromium which crept in.

BUG=671420

Review-Url: https://codereview.chromium.org/2626113002
Cr-Commit-Position: refs/heads/master@{#443050}

This method has callsites inside of compositing, in which it's illegal
to try to update layout and style because compositing comes after.
However, it's also pointless to do so from such callsites.

The other call sites are: FrameView::scrollBehaviorStyle
Various scrolling methods in Element. All of the Element callsites already
update style and layout; this CL adds it to the former (which has a very
complicated set of ways it can be called, making it very hard to determine
if style or layout could be dirty before calling it).

BUG=679099
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_layout_tests_slimming_paint_v2

Review-Url: https://codereview.chromium.org/2618323004
Cr-Commit-Position: refs/heads/master@{#443049}

Review-Url: https://codereview.chromium.org/2621403002
Cr-Commit-Position: refs/heads/master@{#443048}

This causes validation failures breaking the test expectations
when the new asm.js -> wasm pipeline is turned on.

BUG=v8:4203
R=danno@chromium.org,machenbach@chromium.org,binji@chromium.org

Review-Url: https://codereview.chromium.org/2624863003
Cr-Commit-Position: refs/heads/master@{#443047}

BUG=680276,680278,680282
TEST=FYI waterfall bots
TBR=kbr@chromium.org,yunchao.he@intel.com
NOTRY=true
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;master.tryserver.chromium.android:android_optional_gpu_tests_rel

Review-Url: https://codereview.chromium.org/2626903006
Cr-Commit-Position: refs/heads/master@{#443046}

This is my review comment from https://codereview.chromium.org/2616093004/

R=atotic@chromium.org
BUG=635619

Review-Url: https://codereview.chromium.org/2628983002
Cr-Commit-Position: refs/heads/master@{#443045}

Set up a mojom connection for content-browser to talk to content-gpu
through the mojom.GpuMain interface. Subsequent CLs will add code to
allow chrome-browser to set up a mojom.GpuService connection through
GpuMain to the gpu process. This GpuService will replace the existing
chrome-ipc messages between chrome-browser and chrome-gpu.

BUG=643746, 630895

Review-Url: https://codereview.chromium.org/2625933002
Cr-Commit-Position: refs/heads/master@{#443044}

Also introduce a new wrapper class for testing ResourceHandlers,
which will be useful to help make an upcoming API chance a bit simpler.

Also add support for on ResourceThrottle calling Resume() after another
does an out-of-band-cancel.  We have nothing to prevent that case,
but a DCHECK could be triggered in the (unlikely) case it happened.

BUG=672581

Review-Url: https://codereview.chromium.org/2563163002
Cr-Commit-Position: refs/heads/master@{#443043}

Makes case sensitivity and option on highlightRanges, which becomes
a static method on FilteredListWidget.

BUG=none

Review-Url: https://codereview.chromium.org/2627783002
Cr-Commit-Position: refs/heads/master@{#443042}

Renamed EUnicodeBidi to just UnicodeBidi to match the name of the file,
UnicodeBidi.h. This is pre-work to allow UnicodeBidi to be used by
generated fields in ComputedStyleBase, and also better matches the
style guidelines (class names should have the same name as the file).

BUG=628043

Review-Url: https://codereview.chromium.org/2620873002
Cr-Commit-Position: refs/heads/master@{#443041}

Previously, an oscillator would start at the nearest sample frame
boundary.  This can produce noticeable effects.

Instead, make the oscillator start at the requested sample time and
sample the curve appropriately.

BUG=631576
TEST=Oscillator/start-sampling.html

Review-Url: https://codereview.chromium.org/2186813003
Cr-Original-Commit-Position: refs/heads/master@{#442669}
Committed: https://chromium.googlesource.com/chromium/src/+/7909df464ca46a1ea121b0c54e5d04cfa521e38c
Review-Url: https://codereview.chromium.org/2186813003
Cr-Commit-Position: refs/heads/master@{#443040}