GitLab

https://chromium.googlesource.com/chromium/src/+/c35147b9b

BUG=626278
TBR=glebl@chromium.org

Review URL: https://codereview.chromium.org/2149433003 .

Cr-Commit-Position: refs/heads/master@{#405100}

In the top10_mobile page set google.com gets redirected to google.ca. This
sometimes causes problems in telemetry with scrolling. As a short term fix,
replace google.com with google.co.uk in the page set. We cannot change it
to google.ca because memory_top_10_mobile uses the same page set but with
a different wpr that has a recording of google.co.uk. Updated dual_browser
page set to be consistent with top_10_mobile page set.

Also recorded new wprs for memory_top_10_mobile, dual_browser and
top_10_mobile_memory.

BUG=chromium:617914
LOG=N
CQ_INCLUDE_TRYBOTS=tryserver.chromium.perf:android_s5_perf_cq;tryserver.chromium.perf:mac_retina_perf_cq

Review-Url: https://codereview.chromium.org/2138133003
Cr-Commit-Position: refs/heads/master@{#405099}

A compound Knuth-Morris-Pratt failure function is used to match URLs against the
|pattern|.

The |pattern| is split on the '*' wildcard symbol and then a failure function is
built for each subpattern by BuildFailureFunctionFuzzy or BuildFailureFunction
(depending on whether the subpattern contains separator placeholders), and
appended to a vector. Some of the subpatterns can be exempted from being
indexed. E.g., if the |pattern| has a BOUNDARY left anchor, the first subpattern
can be matched by checking if it's a prefix of a URL.

Each subpattern indexed with BuildFailureFunctionFuzzy is prepended with a value
1 (to distinguish them from the subpatterns indexed with BuildFailureFunction,
their failure functions always start with 0).

BUG=609747

Review-Url: https://codereview.chromium.org/2098233002
Cr-Commit-Position: refs/heads/master@{#405098}

The old files in the corpus had the woff signature instead of the woff2
signature and AFL could not discover any paths as a result. The new ones
have the correct signature.

BUG=539572

Review-Url: https://codereview.chromium.org/2145623002
Cr-Commit-Position: refs/heads/master@{#405097}

Merging with interfaces/KeyframeEffect/processing-a-keyframes-argument.html

BUG=623437

Review-Url: https://codereview.chromium.org/2134253004
Cr-Commit-Position: refs/heads/master@{#405096}

BUG=623437

Review-Url: https://codereview.chromium.org/2141823002
Cr-Commit-Position: refs/heads/master@{#405095}

Redundant with animation-model/keyframe-effects/the-effect-value-of-a-keyframe-effect.html

BUG=623437

Review-Url: https://codereview.chromium.org/2139683003
Cr-Commit-Position: refs/heads/master@{#405094}

[Password Generation] Checks the existence of the signin/signup form even if the password was generated

After this CL (https://codereview.chromium.org/1237543002), if the password was generated, there is no check whether the forms still exists. It was introduced to be sure Chrome saves the generated password. The disadvantage of it is that we send a positive vote to the server even if the login was failed (the vote says generation was successful).

This CL (https://codereview.chromium.org/1817483002/) introduced the presaving of generated password (i.e. save generated password right after generation and update it after every user input).

Thus, we are sure that the generated password was saved. No need in the first CL (https://codereview.chromium.org/1237543002). Let remove its code since it distorts votes statistics.

BUG=552420

Review-Url: https://codereview.chromium.org/2141513002
Cr-Commit-Position: refs/heads/master@{#405093}

BUG=615589

Review-Url: https://codereview.chromium.org/2121223002
Cr-Commit-Position: refs/heads/master@{#405092}

BUG=623437

Review-Url: https://codereview.chromium.org/2139153002
Cr-Commit-Position: refs/heads/master@{#405091}

BUG=623437

Review-Url: https://codereview.chromium.org/2137333002
Cr-Commit-Position: refs/heads/master@{#405090}

This CL fixes a couple of style nits pointed out by lint in login_database.*, and also changes
sql::Statement::ColumnBlobAsString to const.

BUG=413020

Review-Url: https://codereview.chromium.org/2133173003
Cr-Commit-Position: refs/heads/master@{#405089}

URLRequest::Delegate::OnBeforeNetworkStart is not used by
anything, despite having a lot of plumping.  It was added to
try experimenting with having the ResourceScheduler delay
requests only once they needed to go over the network, so
responses that could be served out of the cache.

BUG=479994

Committed: https://crrev.com/0af4009f7ce7a5a00091070e85bfd36745bba02c
Review-Url: https://codereview.chromium.org/2123093002
Cr-Original-Commit-Position: refs/heads/master@{#404606}
Cr-Commit-Position: refs/heads/master@{#405088}

Multi-iframe NTP continues to use it, but no longer logs statistics for
load time and number of tiles. This happens only for external NTPs,
where the load time and number of tiles are not fully under our control
anyway.

Single-iframe NTP now uses NTP_ALL_TILES_LOADED. When this is logged, we
emit statistics, ensuring that we log a load time and number of tiles
exactly once per NTP (well, unless the user closes the page or navigates
away without the page having finished loading).

BUG=625990
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2141933002
Cr-Commit-Position: refs/heads/master@{#405087}

This patch adds a heuristic check to see if the font has invalid
avgCharWidth.

The spec is silent about how avgCharWidth is computed for CJK fonts,
and some fonts selected to match to CJK full-width characters.

Since in most fonts, avgCharWidth is close to the width of "0", when
it's too big compared to the width of "0" can be considered as invalid.

BUG=625058

Review-Url: https://codereview.chromium.org/2145903002
Cr-Commit-Position: refs/heads/master@{#405086}

BUG=586161

Review-Url: https://codereview.chromium.org/2124503002
Cr-Commit-Position: refs/heads/master@{#405085}

Now hard-coded the new Blended behaviour, which ranks app list results
by their score, rather than stratifying them into groups.

BUG=487494

Review-Url: https://codereview.chromium.org/1136363003
Cr-Commit-Position: refs/heads/master@{#405084}

Need to call LabelButton::Layout since this requires updated preferred
width of the label and that needs up to date text.

BUG=624612

Review-Url: https://codereview.chromium.org/2143473002
Cr-Commit-Position: refs/heads/master@{#405083}

This test is redundant with existing tests in the W3C web-platform-tests.

BUG=623437

Review-Url: https://codereview.chromium.org/2141723005
Cr-Commit-Position: refs/heads/master@{#405082}

Before, Sandwich was using the patched WPR archive with headers
ensure the resource get cached. But that could lead to non realistic
page load time in theory if multiple requests are issued on uncachable
resources.

This CL fixes this issue by using the original WPR archive when running
the benchmarks for NoState-Prefetch.

BUG=582080

Review-Url: https://codereview.chromium.org/2144633002
Cr-Commit-Position: refs/heads/master@{#405081}

The root property nodes weren't applied because we expect the PaintLayer of
the root LayoutView to apply them before painting. However in debug builds
the GraphicsLayer could paint debug background before we ever reach the
PaintLayer. Also the scrollbar layers (not implemented in SPv2. Upcoming.)
are not backed by PaintLayer.

Review-Url: https://codereview.chromium.org/2140173004
Cr-Commit-Position: refs/heads/master@{#405080}

This has been broken since offline pages caused the refresh of the NTP
MostVisitedSites to be asynchronous.

BUG=612160

Review-Url: https://codereview.chromium.org/2105933002
Cr-Commit-Position: refs/heads/master@{#405079}

https://codereview.chromium.org/2008873004 removed the Element argument
for Fullscreen::didExitFullScreenForElement, so now remove it from the
method name as well.

Remove the dummy boolean return values for WebFrameClient.

Also lowercase the S in {will,did}{Enter,Exit}FullScreen everywhere.

R=alexmos@chromium.org,jochen@chromium.org
BUG=402421

Review-Url: https://codereview.chromium.org/2139303002
Cr-Commit-Position: refs/heads/master@{#405078}

(namely those in https://github.com/w3c/web-platform-tests/tree/master/web-animations/interfaces/AnimationEffectTiming).

BUG=623437

Review-Url: https://codereview.chromium.org/2142593003
Cr-Commit-Position: refs/heads/master@{#405077}

This bug was found by the IPC fuzzer.

BUG=627354

Review-Url: https://codereview.chromium.org/2139913003
Cr-Commit-Position: refs/heads/master@{#405076}

When the selection includes a directory, no ARC tasks should be returned.

BUG=627679
TEST=Cannot open a directory with ARC apps

Review-Url: https://codereview.chromium.org/2141373002
Cr-Commit-Position: refs/heads/master@{#405075}

BUG=618232
TEST=manuall confirmed that the view renders as intended

Review-Url: https://codereview.chromium.org/2143873002
Cr-Commit-Position: refs/heads/master@{#405074}

Speculative crash fix for non-reproducible clusterfuzz case.

BUG=624903

Review-Url: https://codereview.chromium.org/2145003002
Cr-Commit-Position: refs/heads/master@{#405073}

In EmbeddedWorkerDispatcher::OnStartWorker,
start_data.pauseAfterDownloadMode used not to be initialized, but this was
already fixed at http://crrev.com/1713863002, so now we can remove the suppression.

BUG=387373

Review-Url: https://codereview.chromium.org/2139983002
Cr-Commit-Position: refs/heads/master@{#405072}

Revert of ImageCapture: Implement takePhoto() for Mac AVFoundation (patchset #4 id:220001 of https://codereview.chromium.org/2129733004/ )

Reason for revert:
Most likely breaks WebRtcWebcamBrowserTests/WebRtcWebcamBrowserTest.MANUAL_TestAcquiringAndReacquiringWebcam/0 on the WebRTC in Chrome bots.

See https://build.chromium.org/p/chromium.webrtc/builders/Mac%20Tester/builds/56889/steps/browser_tests/logs/stdio for details.

Original issue's description:
> ImageCapture: Implement takePhoto() for Mac AVFoundation
>
> By pulling the necessary symbols out of AVFoundation
> and using them; most notably (Cr)AVCaptureStillImageOutput.
>
> TakePhoto is already implemented for Android (both APIs).
>
> Also adding VideoCaptureDeviceTest::MAYBE_TakePhoto
> test case, enabled only for Mac.
>
> BUG=518807
> TEST=Run build with flag --enable-blink-features=ImageCapture,
> navigate to [1] and push buttons
> - Open Camera ...
> - Create ImageCapturer
> - takePhoto()  (N times!) --> profit
>
> [1] https://rawgit.com/Miguelao/demos/master/imagecapture.html
>
> Committed: https://crrev.com/ad75c34fbb7940127a99be04ed6019ab09d73880
> Cr-Commit-Position: refs/heads/master@{#404832}

TBR=rsesek@chromium.org,emircan@chromium.org,sandersd@chromium.org,mcasas@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=518807

Review-Url: https://codereview.chromium.org/2147873002
Cr-Commit-Position: refs/heads/master@{#405071}

This doesn't yet fix the bug but stops recreating mask layers.

BUG=624431

Review-Url: https://codereview.chromium.org/2126873005
Cr-Commit-Position: refs/heads/master@{#405070}

Skip -webkit-input-placeholder elements when we compute direction for dir=auto in
HTMLElement::directionality.

- Using <bdi> for placeholders doesn't work. We want to inherit placeholder direction
  from the parent element.
- Using unicode-bidi CSS property doesn't work. It doesn't affect dir=auto
  computation.

BUG=623270,625739

Review-Url: https://codereview.chromium.org/2145823003
Cr-Commit-Position: refs/heads/master@{#405069}

This CL fixes an issue where the query state was not correctly bound to
the list container. This caused the loading spinner and message to break.

BUG=619799
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2126173002
Cr-Commit-Position: refs/heads/master@{#405068}

animation-playbackRate.html~362094f2f56a0a4152c3399f032312f93990c256 is a merge conflict marker.

animationtimeline-consistent.html was deleted in http://crrev.com/2137403002 and became animation-playbackRate.html.

BUG=623437

Review-Url: https://codereview.chromium.org/2145893002
Cr-Commit-Position: refs/heads/master@{#405067}

This patch fixes the problem when a text appears chopped in HTML select element with -webkit-appearance:none. If appearance:none then we should use the height that is calculated based using platform-native styling based on the users' operating system's theme.

R=eae@chromium.org
BUG=626278
TEST=third_party/WebKit/LayoutTests/fast/forms/select/menulist-appearance-none.html

Review-Url: https://codereview.chromium.org/2131303003
Cr-Commit-Position: refs/heads/master@{#405066}

This CL updates Blink HTML Parser to better match the spec.
re: http://html5.org/tools/web-apps-tracker?from=8330&to=8331

The LayoutTest is based on WebKit change by rniwa@webkit.org:
https://trac.webkit.org/changeset/160182

Bonus:
- HTMLConstructionSite::setForm is merged into initFragmentParsing

BUG=326058

Review-Url: https://codereview.chromium.org/2143833003
Cr-Commit-Position: refs/heads/master@{#405065}

Cr-Commit-Position: refs/heads/master@{#405064}

ServiceRegistry has been eliminated and cannot thus be used any more.

BUG=242107,623393

Review-Url: https://codereview.chromium.org/2132003002
Cr-Commit-Position: refs/heads/master@{#405063}

BUG=622125

Review-Url: https://codereview.chromium.org/2127413002
Cr-Commit-Position: refs/heads/master@{#405062}

Remove SVGZoomEvent(s) interface and its corresponding
onzoom attribute.

It has been removed from SVG 2.0 specification.

BUG=367890

Review-Url: https://codereview.chromium.org/2133983003
Cr-Commit-Position: refs/heads/master@{#405061}