GitLab

Updates the expires_after attribute for 48 histograms that have
been used to generate alerts in the past 90 days and do not already have
a date later than or within 30 days of 2020-08-17.

Change-Id: I8116b948f4bc5a8ca791a0d320d7b552cd242d6f
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2042255
Commit-Queue: Brian White <bcwhite@chromium.org>
Reviewed-by: Brian White <bcwhite@chromium.org>
Cr-Commit-Position: refs/heads/master@{#739120}

Show a chip on the tab card if that tab has search result page in its
navigation stack. The chip contains the search term. When clicked,
the tab would visit the search result page again.

This is behind a Finch parameter "enable_search_term_chip" under
TabGridLayoutAndroid.

The CL is split into two, and this is part one, which only contains the
diff that cannot be handled by the Formal Equivalence Checker in
http://crrev.com/c/1934235/4.

See also: part two (http://crrev.com/c/2036513).

Bug: 1048255
Change-Id: I6fa83d1179cb3700b65afde86327ecbb148ed92b
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2038458
Commit-Queue: Wei-Yin Chen (陳威尹) <wychen@chromium.org>
Reviewed-by: Matthew Jones <mdjones@chromium.org>
Cr-Commit-Position: refs/heads/master@{#739119}

After app unexpectedly crashes in EG2 test, it is relaunched at -setUp
of next testMethod. Currently in this situation, |appNeedsLaunching| is
judged false and app is launched without TestCase +setUps at |activate|
method, which will cause DCHECK failures at +tearDown. This change makes
|appNeedsLaunching| judged true in this situation and run TestCase
+setUps after the relaunch to prevent DCHECK failures.

Bug: 1023911
Change-Id: I83bd06df7afdf2402fbb0593a6825910107534f9
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2037852Reviewed-by: Eugene But <eugenebut@chromium.org>
Commit-Queue: Zhaoyang Li <zhaoyangli@chromium.org>
Cr-Commit-Position: refs/heads/master@{#739118}

dom-breakpoints.js usually takes ~70 on some mac bots, but occasionally
exceeds 91s and runs into a timeout.

This CL splits the test into two, which hopefully makes it less likely
to trigger timeouts.

The test and, most importantly, the test expectations are just split;
they did not change.

Bug: chromium:1049444
Change-Id: Ie2288bafa5807726cfd1fddc5d91189bc2ad905a
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2042105
Auto-Submit: Sigurd Schneider <sigurds@chromium.org>
Reviewed-by: Yang Guo <yangguo@chromium.org>
Commit-Queue: Sigurd Schneider <sigurds@chromium.org>
Cr-Commit-Position: refs/heads/master@{#739117}

Change-Id: I8b387723819d1fd138915b507eaac4624fd34917
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2042683
Auto-Submit: Erik Chen <erikchen@chromium.org>
Commit-Queue: Ilia Samsonov <isamsonov@google.com>
Reviewed-by: Ilia Samsonov <isamsonov@google.com>
Cr-Commit-Position: refs/heads/master@{#739116}

https://chrome-internal.googlesource.com/chrome/src-internal.git/+log/e4d9391c167c..f5e5312d2154


Created with:
  gclient setdep -r src-internal@f5e5312d2154

If this roll has caused a breakage, revert this CL and stop the roller
using the controls here:
https://skia-autoroll.corp.goog/r/src-internal-chromium-autoroll
Please CC jbudorick@google.com on the revert to ensure that a human
is aware of the problem.

To report a problem with the AutoRoller itself, please file a bug:
https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug

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

Cq-Include-Trybots: luci.chrome.try:linux-chromeos-chrome
Bug: chromium:1044793
Tbr: jbudorick@google.com
Change-Id: I12401cbd0db5aa03004ed14d779f5464faa9988b
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2042654Reviewed-by: chromium-internal-autoroll <chromium-internal-autoroll@skia-corp.google.com.iam.gserviceaccount.com>
Commit-Queue: chromium-internal-autoroll <chromium-internal-autoroll@skia-corp.google.com.iam.gserviceaccount.com>
Cr-Commit-Position: refs/heads/master@{#739115}

media/base/video_frame_pool.cc: remove an unnecessary test in a while
loop.

media/filters/dav1d_video_decoder.h: remove an unneeded #include.

Bug: N/A
Change-Id: I0014cee024c2321382f018bfbbe9065a9e135b55
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2042675Reviewed-by: Dale Curtis <dalecurtis@chromium.org>
Commit-Queue: Dale Curtis <dalecurtis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#739114}

These files currently rely on getting these includes
transitively via inspector_protocol generated code.
Would like to make the dependency direct so it's easier
to change the generated code.

Change-Id: I84cb8889c6fd5c0d2245a91cb1b667f3e79f2a12
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2037966Reviewed-by: Leonard Grey <lgrey@chromium.org>
Commit-Queue: Johannes Henkel <johannes@chromium.org>
Cr-Commit-Position: refs/heads/master@{#739113}

Tbr: ender@google.com
Bug: None
Change-Id: Iba7a0e122310d7f7ba286dfd4c4296610a4f9bb0
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2042296Reviewed-by: Peter Wen <wnwen@chromium.org>
Commit-Queue: Peter Wen <wnwen@chromium.org>
Auto-Submit: Peter Wen <wnwen@chromium.org>
Cr-Commit-Position: refs/heads/master@{#739112}

This fixes timeouts that were occurring when FlexNG was enabled:
https://test-results.appspot.com/data/layout_results/linux-rel/292466/webkit_layout_tests%20%28with%20patch%29/layout-test-results/results.html
(*/text-with-display-style.html)

These tests timeout if they don't receive all the expected performance
entries, causing the failure (the test never completes, waiting for the
performance entries).

The specific failure for FlexNG was occurring as we didn't have a
ScopedPaintTimingDetectorBlockPaintHook for the flexbox on the stack,
only one for the anonymous block-flow. This meant that the performance
entry was never recorded for the flexbox element.

Previously the ScopedPaintTimingDetectorBlockPaintHook would get
emplaced in two different places -
 1) Within BlockPainter
 2) Within the inline painter

This changes the EmplaceIfNeeded call to walk up the layout object
tree to the first non-anonymous object which has a DOM node associated
with it.

This walk up the layout object tree should be relatively cheap. Most of
the time it will be only walking up zero or one objects.

Additionally with the layout object walk we don't need to EmplaceIfNeeded
on nested <div>s, only when we actually encounter some text to be
painted.

Bug: 845235
Change-Id: I80e18539f70d2e1df56d3d85dd928dc10a35e28e
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2032428
Commit-Queue: Ian Kilpatrick <ikilpatrick@chromium.org>
Reviewed-by: Xianzhu Wang <wangxianzhu@chromium.org>
Reviewed-by: Steve Kobes <skobes@chromium.org>
Cr-Commit-Position: refs/heads/master@{#739111}

The current test case for gfx::TransformBetweenRects only verifies
the function on the source rectangle whose origin is at (0,0).
This CL provides a more general test case.

Bug: 1046983
Change-Id: I06a451ce7bcbad1344139792d6265d5313fe782d
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2036272Reviewed-by: Ian Vollick <vollick@chromium.org>
Commit-Queue: Andrew Xu <andrewxu@chromium.org>
Cr-Commit-Position: refs/heads/master@{#739110}

This change removes the watcher process.  The watcher process is
responsible for waiting for the browser process to exit and logs the exit
code to a known registry location. Subsequent launches read from the
registry location and log the value to a histogram.  The watcher process
was originally created prior to the creation of crashpad.  Since all the
watcher process is doing is waiting for the browser process to exit it
makes sense to move this functionality to the crashpad process and get
rid of the watcher process and code entirely.  Use of the registry is
now removed and the histogram is logged directly on shutdown.
This change removes the chrome_watcher.dll which gives back ~ 1MB of
disk space.

Bug: 1043844
Change-Id: I666819e64817197e8950a9c5749c921fcebdb964
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2035054
Commit-Queue: Chris Davis <chrdavis@microsoft.com>
Reviewed-by: Scott Violet <sky@chromium.org>
Reviewed-by: Sigurður Ásgeirsson <siggi@chromium.org>
Cr-Commit-Position: refs/heads/master@{#739109}

This reverts commit c2f07278.

Reason for revert: This change is causing regression. Bounding box
calculation of a given element requires the |kScrollX| and |kScrollY|
positions. In case of PDF there is a scroll element which lives outside
the PDF node. The scroll element also reports scroll positions. During
calculation of bounding boxes in AXTree::RelativeToTreeBoundsInternal
|kScrollX| and |kScrollY| are applied twice.

Original change's description:
> Add viewport scroll information to PDFAccessibilityTree
>
> This CL adds following information to PDFAccessibilityTree
> - ax::mojom::IntAttribute::kScrollXMin
> - ax::mojom::IntAttribute::kScrollXMax
> - ax::mojom::IntAttribute::kScrollX
> - ax::mojom::IntAttribute::kScrollYMin
> - ax::mojom::IntAttribute::kScrollYMax
> - ax::mojom::IntAttribute::kScrollY
>
> It enables partial implementation of IScrollProvider, following methods
> will work with this change
> - IScrollProvider::get_HorizontallyScrollable
> - IScrollProvider::get_VerticallyScrollable
> - IScrollProvider::get_HorizontalScrollPercent
> - IScrollProvider::get_VerticalScrollPercent
>
> This information is required by Screen readers to announce the current
> scroll position and extent of scroll to the user.
>
> Tests: The best way to test this change will be through browser tests.
> However, it requires the second part of the change, detailed in the
> associated bug, to be completed.
>
> Bug: 1034521
> Change-Id: I629b2773c75f2b6b222d339cd0ae01799d695295
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1985821
> Reviewed-by: Daniel Cheng <dcheng@chromium.org>
> Reviewed-by: Lei Zhang <thestig@chromium.org>
> Commit-Queue: Virender Singh <virens@microsoft.com>
> Cr-Commit-Position: refs/heads/master@{#728843}

TBR=dcheng@chromium.org,thestig@chromium.org,mohitb@microsoft.com,virens@microsoft.com

# Not skipping CQ checks because original CL landed > 1 day ago.

Bug: 1034521
Change-Id: I15675f9ba604be96e9c50c9bea1dbe84ef8c9a0a
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2041714Reviewed-by: Lei Zhang <thestig@chromium.org>
Commit-Queue: Lei Zhang <thestig@chromium.org>
Cr-Commit-Position: refs/heads/master@{#739108}

Bug: 1049281
Change-Id: Ib9d4ada81de8568cbfc5f7dc7f69ef3eb6470daa
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2042192
Auto-Submit: Yulun Wu <yulunwu@chromium.org>
Reviewed-by: Toni Baržić <tbarzic@chromium.org>
Commit-Queue: Yulun Wu <yulunwu@chromium.org>
Cr-Commit-Position: refs/heads/master@{#739107}

The original method declared the argument as OnceCallback:
https://cs.chromium.org/chromium/src/chrome/browser/offline_pages/background_loader_offliner.h?rcl=f1e0ecaf7d175cf0f4eb923daa10511be4b11931&l=75

This is part of the base::Callback migration.

Context: https://cs.chromium.org/chromium/src/docs/callback.md?rcl=9fcc3764aea8f97e9f6de4a9ee61d554e67edcda&l=40
Change-Id: I4f477760f4cf423d6352fd82041da75b2f0e1903
Bug: 714018
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2031405Reviewed-by: danakj <danakj@chromium.org>
Reviewed-by: Carlos Knippschild <carlosk@chromium.org>
Commit-Queue: Carlos Knippschild <carlosk@chromium.org>
Auto-Submit: Jose Lopes <jabolopes@google.com>
Cr-Commit-Position: refs/heads/master@{#739106}

Views with backing Layers that set a |clip_path_| will now create
a ViewMaskLayer object that masks the area designated by the clip.
The View Layer and all child Layers will be masked by the
ViewLayerMask.

Made some small changes to the original set_clip_path() setter and
calling functions.

Bug: None
Change-Id: Id746a72042d9f114144db01e289a603abe5afe9a
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2031640Reviewed-by: Scott Violet <sky@chromium.org>
Reviewed-by: Peter Kasting <pkasting@chromium.org>
Commit-Queue: Thomas Lukaszewicz <tluk@chromium.org>
Cr-Commit-Position: refs/heads/master@{#739105}

This will simplify the task of changing WebLayer to take advantage of
the newly-componentized ContentMetricsHelper in a followup CL.

Bug: 1030692
Change-Id: Iad9d8e557b73837ffd9c4506cec4ed63424b63cf
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2041626
Commit-Queue: Colin Blundell <blundell@chromium.org>
Reviewed-by: Clark DuVall <cduvall@chromium.org>
Cr-Commit-Position: refs/heads/master@{#739104}

Bug: 1028410
Change-Id: I99d415261b48e05d401148dcf6b1165c0f020f16
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2042311
Auto-Submit: Sébastien Marchand <sebmarchand@chromium.org>
Commit-Queue: Ilya Sherman <isherman@chromium.org>
Reviewed-by: Ilya Sherman <isherman@chromium.org>
Cr-Commit-Position: refs/heads/master@{#739103}

The Crostini new terminal app is a system web app, although the app type
in AppService is still Crostini. So update IsOpenedInBrowser to return
true for the new terminal app id kCrostiniTerminalSystemAppId, so that
AppServiceAppWindowLauncherController can skip that, and leave it to
BrowserStatusMonitor to handle all windows related requests.

BUG=1048906

Change-Id: Ie06f0616cf3a3ff6f25d2ad878151c8f71c98e84
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2038930Reviewed-by: Xiyuan Xia <xiyuan@chromium.org>
Commit-Queue: Nancy Wang <nancylingwang@chromium.org>
Cr-Commit-Position: refs/heads/master@{#739102}

Bug: 1048225
Change-Id: I39b1e3d99fd092d95ad9dc9361e4912f92d6461a
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2036375
Auto-Submit: Ganggui Tang <gogerald@chromium.org>
Reviewed-by: Ted Choc <tedchoc@chromium.org>
Reviewed-by: Jesse Doherty <jwd@chromium.org>
Commit-Queue: Jesse Doherty <jwd@chromium.org>
Cr-Commit-Position: refs/heads/master@{#739101}

Change-Id: I7225bb27624db7abe7c95a13df9b79830c81ef9c
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2035954
Commit-Queue: Jesse Doherty <jwd@chromium.org>
Auto-Submit: Matthias Körber <koerber@google.com>
Reviewed-by: Jesse Doherty <jwd@chromium.org>
Cr-Commit-Position: refs/heads/master@{#739100}

This CL changes the IPC interface for prerendering by introducing Mojo
interfaces PrerenderHandle, PrerenderHandleClient & PrerenderProcessor.

PrerenderHandleClient is a rename of PrerenderDispatcher, but now with
that name indicates how it relates to a requested prerender.

PrerenderHandle corresponds to a requested prerender. This name is
chosen intentionally to match prerender::PrerenderHandle and blink::
PrerenderHandle. These all correspond one-to-one.

PrerenderProcessor is the interface used by a renderer to make requests
for prerendering. It is instantiated per-frame, which enables the
browser to get the render_view_id without it having to be passed over
IPC as a parameter.

A future CL will be able to move these interfaces into Blink, and then
eliminate some of chrome/renderer/prerender, WebRenderer and other Blink
layers of plumbing.

As part of this change, PrerenderLinkManager's list of LinkPrerender
objects is changed to be a list of std::unique_ptr<..> instead since
LinkPrerender objects are no longer copyable since they have to store
a mojo::Remote to the PrerenderHandleClient. This impacted a fair chunk
of code.

The concept of prerender_id is also removed. That is no longer needed
as each PrerenderHandle instance is effectively a unique ID. This in
turn caused a good amount of surgery to prerender_unittest.cc. It makes
some tests unnecessary since they were testing codepaths related to
no longer valid prerender IDs and such.

Change-Id: I1d1911a3a44fe8b4478b9a109048c3081e78822d
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2036906
Commit-Queue: Darin Fisher <darin@chromium.org>
Reviewed-by: Daniel Cheng <dcheng@chromium.org>
Reviewed-by: Ryan Sturm <ryansturm@chromium.org>
Cr-Commit-Position: refs/heads/master@{#739099}

This is a mechanical change to remove implicit conversions of
ScopedBstr to BSTR.

BUG=1034666
TBR=dtseng@chromium.org,ganesh@chromium.org,grt@chromium.org,jamiewalch@chromium.org,pmonette@chromium.org,proberge@chromium.org,sky@chromium.org

Change-Id: I9fa7d62bddeecd81d1944bb289bad2a5758fe7b7
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2038171Reviewed-by: Robert Liao <robliao@chromium.org>
Reviewed-by: S. Ganesh <ganesh@chromium.org>
Reviewed-by: Scott Violet <sky@chromium.org>
Reviewed-by: proberge <proberge@chromium.org>
Reviewed-by: Patrick Monette <pmonette@chromium.org>
Reviewed-by: Will Harris <wfh@chromium.org>
Commit-Queue: Robert Liao <robliao@chromium.org>
Cr-Commit-Position: refs/heads/master@{#739098}

Currently, if a DialogDelegate's Accept() or Cancel() method leads back
into its Close(), two of the registered callbacks can be run, which is a
violation of DialogDelegate's contract and causes confusing workarounds
in client code. This often happens in code like this:

void SomeBubble::OnDialogAccepted() {
  NavigateTo(kSomeURL);
}

where NavigateTo causes a page navigation synchronously, which causes
the involved bubble to be closed.

Instead:

* In the near term, DialogDelegate should ensure that it never delivers
  more than one of these callbacks
* In the far term, DialogDelegate should enforce that callers do not
  re-enter Close() from inside Accept() or Cancel()

The latter will likely need to wait for the Accept/Cancel/Close callback
refactor to be finished.

This change also removes the remnants of the GlobalErrorBubbleView unit test
suite, which were exercising this behavior deliberately. These tests
were providing very little actual coverage and I have been gradually
deleting them as I refactor GlobalError anyway.

Bug: 1011446
Change-Id: Ifd76330c0f79d04ce2e5741c5a1977159ee598d5
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2042252
Commit-Queue: Elly Fong-Jones <ellyjones@chromium.org>
Reviewed-by: Peter Boström <pbos@chromium.org>
Cr-Commit-Position: refs/heads/master@{#739097}

BUG=1006999

Change-Id: I2a85e86059c617e7676e9b9319fd8a245a197b35
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2040639Reviewed-by: Xiyuan Xia <xiyuan@chromium.org>
Reviewed-by: Nancy Wang <nancylingwang@chromium.org>
Commit-Queue: Nancy Wang <nancylingwang@chromium.org>
Cr-Commit-Position: refs/heads/master@{#739096}

These tests need to perform cross-origin navigations as they test
security restrictions that only apply to cross-origin.

Bug: 1049624
Change-Id: I241962c44eaa1e4862e3df38be4b614491818b64
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2042402Reviewed-by: Robert Ma <robertma@chromium.org>
Commit-Queue: Nick Burris <nburris@chromium.org>
Cr-Commit-Position: refs/heads/master@{#739095}

Now that Preview Tab is working on top of ThinWebView and BottomSheet,
this CL deletes the old implementation that based it on OverlayPanel.

Bug: 1043677
Change-Id: Icccc93cca3f09e9ae122b1e0dcb7537331136ac0
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2032466
Commit-Queue: Jinsuk Kim <jinsukkim@chromium.org>
Reviewed-by: David Trainor <dtrainor@chromium.org>
Reviewed-by: Donn Denman <donnd@chromium.org>
Cr-Commit-Position: refs/heads/master@{#739094}

This is recorded in addition to the histogram so that we can use the user action sequencing dashboard to see what users do on the chrome://extensions page, and use the histogram to see which messaging is most effective.

Design Doc: http://go/extensionscheckupmetrics

Bug: 1034541
Change-Id: I490eeec2a666e7fac094f7ac0872540c50c0dc8d
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2031534Reviewed-by: Moe Ahmadi <mahmadi@chromium.org>
Reviewed-by: Devlin <rdevlin.cronin@chromium.org>
Reviewed-by: Ilya Sherman <isherman@chromium.org>
Commit-Queue: Archana Simha <archanasimha@chromium.org>
Cr-Commit-Position: refs/heads/master@{#739093}

[Problem] When Remote Web Inspector is connected to a page doing a lot
of XHR requests (for example
https://www.youtube.com/watch?v=Bey4XXJAqS8_) renderer's memory usage
starts to grow with no limit.

[Cause] There is a corner case in NetworkResourcesData when the limit
of maximum_resources_content_size_ (100.000.000 bytes) is almost
reached. When the next request finishes its raw data is replaced with
decoded form. The decoded form can be larger that the raw data.
Sometimes the difference is so big that it causes the
NetworkResourcesData::content_size_ to go above the
maximum_resources_content_size_.
From now on there is basically no limit in appending requests in
NetworkResourcesData as maximum_resources_content_size_ - content_size_
in NetworkResourcesData::EnsureFreeSpace() will give a very large number
(unsigned overflow).

[Solution] Ensure that after request data is decoded
NetworkResourcesData has enough space to store the decoded form. It
needs to have at least "decoded size" - "raw size" bytes free.

Bug: 1019862
Change-Id: I5adf3aa4a91d01012f8620628f1a442697668105
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1946476Reviewed-by: Yang Guo <yangguo@chromium.org>
Reviewed-by: Sigurd Schneider <sigurds@chromium.org>
Reviewed-by: Andrey Kosyakov <caseq@chromium.org>
Commit-Queue: Andrey Kosyakov <caseq@chromium.org>
Cr-Commit-Position: refs/heads/master@{#739092}

Before this patch, BobPay native Android payment app could be displayed
side-by-side with BobPay service worker based payment app, even if the
web app manifest was configured to prefer the native payment app.

This patch adds a
PaymentInstrument.getApplicationIdentifiersThatHideThisApp() override in
ServiceWorkerPaymentApp, which was accidentally deleted in
https://crrev.com/c/1988215 as it was being renamed from
PaymentApp.getPreferredRelatedApplicationIds().

After this patch, BobPay native Android payment hides the service worker
based payment app.

Bug: 1049217
Change-Id: Id439daf9435317be30885e1e0adafce1284481ed
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2039632
Auto-Submit: Rouslan Solomakhin <rouslan@chromium.org>
Commit-Queue: Danyao Wang <danyao@chromium.org>
Reviewed-by: Danyao Wang <danyao@chromium.org>
Cr-Commit-Position: refs/heads/master@{#739091}

This change removes almost all the remaining overrides of ::Close() in cbuiv,
and some of the overrides of other methods. Replacements are either:

1) Deleted outright
2) Replaced with lambdas inlined in the constructor
3) Replaced with void OnDialog*() methods that are used for the new callback
   system

Bug: 1011446
Change-Id: Idfe315895958f7cc92db17a30fc4cf591e4ffec8
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2036943
Commit-Queue: Elly Fong-Jones <ellyjones@chromium.org>
Reviewed-by: Peter Boström <pbos@chromium.org>
Cr-Commit-Position: refs/heads/master@{#739090}

Change-Id: I286fe466283e329c5e71fdc4454bd521a32ee2e8
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2039810Reviewed-by: Dominic Mazzoni <dmazzoni@chromium.org>
Commit-Queue: David Tseng <dtseng@chromium.org>
Cr-Commit-Position: refs/heads/master@{#739089}

Add the output of the origin table to the
WebUI for debugging.

BUG=1024353

Change-Id: I2eb8e99d94bf8792bd459529ae5aae3bacffe89d
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2037642Reviewed-by: Daniel Cheng <dcheng@chromium.org>
Reviewed-by: Tommy Steimel <steimel@chromium.org>
Commit-Queue: Becca Hughes <beccahughes@chromium.org>
Cr-Commit-Position: refs/heads/master@{#739088}

The setTimeout with 0 delay caused notifyDone() to sometimes be called
too early. Instead add an event listener in the non-isolated world that
should be triggered when the media query changes and add an extra rAF()
in case the event listener is called first for the non-isolated world
handler.

The test still fails though, so the *expected.txt, which is now the same
for all platforms, needs to be kept.

The tests were marked with Timeout. If it turns out it is still slow on
some bots, we can add it to SlowTests.

The media query event handler argument had changed since the test was
written, so fixed the test itself as well.

Bug: 390452
Change-Id: I31c370601d932152adeb16bdffc59e4b68a2907d
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2041476Reviewed-by: Kentaro Hara <haraken@chromium.org>
Commit-Queue: Rune Lillesveen <futhark@chromium.org>
Cr-Commit-Position: refs/heads/master@{#739087}

This CL moves GLHelper from viz into gpu in preparation for it to be
used by RasterImplementationGLES in a future patch. This will allow us
to remove calls to GLHelper from CanvasCaptureHandler which is necessary
as part of the OOPR Canvas project.

TBR=bsalomon@google.com

Bug: 1023262
Change-Id: I9f7f98cfc4cc73acabcda275d2eee27488c992de
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2023282
Commit-Queue: Khushal <khushalsagar@chromium.org>
Reviewed-by: Yuri Wiitala <miu@chromium.org>
Reviewed-by: Andrew Grieve <agrieve@chromium.org>
Reviewed-by: Daniel Cheng <dcheng@chromium.org>
Reviewed-by: J Kardatzke <jkardatzke@chromium.org>
Reviewed-by: Zhenyao Mo <zmo@chromium.org>
Reviewed-by: Khushal <khushalsagar@chromium.org>
Cr-Commit-Position: refs/heads/master@{#739086}

Previously OfflineAudioContext::resumeContext() method did not check
if the context is cleared by ExecutionContext::ContextDestroyed().
Such case is possible when the audio context is a part of a detached
iframe.

This CL changes the check so we can verify if the context's resources
is still available. Otherwise, we can reject the resume promise
resolver.

Test: Locally confirmed ASAN does not crash with the repro case.
Bug: 1048373
Change-Id: I96a601dcf63963525d95cfc5089fd4a3b0176687
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2042409
Commit-Queue: Hongchan Choi <hongchan@chromium.org>
Commit-Queue: Kentaro Hara <haraken@chromium.org>
Auto-Submit: Hongchan Choi <hongchan@chromium.org>
Reviewed-by: Kentaro Hara <haraken@chromium.org>
Cr-Commit-Position: refs/heads/master@{#739085}

This is a reland of 359bdf76

Proguard rules now extended to ensure Fragments aren't removed in
release: https://crrev.com/c/2036872.

There is no change to this patch from the original.

Original change's description:
> [Settings Refactor] Move autofill settings
>
> Move autofill settings alongside autofill code.
>
> Bug: 1047357
> Change-Id: Id5679d4392e23cff775dfcefa41930af9ab0f602
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2031538
> Reviewed-by: Rouslan Solomakhin <rouslan@chromium.org>
> Reviewed-by: Theresa  <twellington@chromium.org>
> Commit-Queue: Natalie Chouinard <chouinard@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#737874}

Bug: 1047357
Change-Id: Ic87a7b18859e51fd9683910e3707a1ffb2a1da47
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2036871
Commit-Queue: Natalie Chouinard <chouinard@chromium.org>
Reviewed-by: Theresa  <twellington@chromium.org>
Reviewed-by: Rouslan Solomakhin <rouslan@chromium.org>
Reviewed-by: Natalie Chouinard <chouinard@chromium.org>
Cr-Commit-Position: refs/heads/master@{#739084}

TBR=dtseng@chromium.org

Change-Id: I299c2905122c06a0cb5492545a8d7a695fc81e18
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2042275Reviewed-by: David Tseng <dtseng@chromium.org>
Commit-Queue: David Tseng <dtseng@chromium.org>
Cr-Commit-Position: refs/heads/master@{#739083}

... and places adjustment within:
LayoutNGMixin<Base>::ComputeIntrinsicLogicalWidths

Previously ComputeMinMaxSize could work in two different modes,
one to return the border-box min/max sizes, and one for the content-box.

This was primarily to support ComputeIntrinsicLogicalWidths.

This patch removes this enum, and instead places the adjustment logic
within ComputeIntrinsicLogicalWidths.

Change-Id: I1bf839e5afc5f7a7241749b9e9eeb34d50fc77ce
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2031533Reviewed-by: Christian Biesinger <cbiesinger@chromium.org>
Reviewed-by: Morten Stenshorne <mstensho@chromium.org>
Commit-Queue: Ian Kilpatrick <ikilpatrick@chromium.org>
Cr-Commit-Position: refs/heads/master@{#739082}

Split out webgpu.

Bug: 1048907
Change-Id: I8313c7cb6757d040deb3cd77d7825c6158ab8492
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2040615
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Daniel Cheng <dcheng@chromium.org>
Cr-Commit-Position: refs/heads/master@{#739081}