GitLab

Summary of changes available at:
https://chromium.googlesource.com/skia/+log/b880d7f..e0b19d4

CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel
TBR=hcm@google.com

Review URL: https://codereview.chromium.org/1031303003

Cr-Commit-Position: refs/heads/master@{#322457}

[Icons NTP] Use GetVariationParamValue instead of experiment prefix to determine when large icons are enabled.

This is the standard pattern for adding a forcing flags in about://flags.

This also matches a google3 change that just landed which declares that parameter.

BUG=467712

Review URL: https://codereview.chromium.org/1032073003

Cr-Commit-Position: refs/heads/master@{#322456}

A non-broken but recently broken QUIC alternative service does not race for 0RTT
(because it is recently broken) but is still used (because it is not broken).
Upon successful connection, the service should be confirmed (marked as not
recently broken so that it can be used next time for 0RTT).

This CL adds this functionality and a test for it.  I locally verified that the
test fails without the rest of the change.

BUG=392575

Review URL: https://codereview.chromium.org/1035713003

Cr-Commit-Position: refs/heads/master@{#322455}

R=sergeyberezin@chromium.org
BUG=468335

Review URL: https://codereview.chromium.org/999933010

Cr-Commit-Position: refs/heads/master@{#322454}

BUG=468726

Review URL: https://codereview.chromium.org/1028933002

Cr-Commit-Position: refs/heads/master@{#322453}

This means that .localhost is always redirected to localhost in DNS
lookups, and |net::IsLocalhost| returns true for .localhost hostnames.

BUG=455825

Review URL: https://codereview.chromium.org/938093003

Cr-Commit-Position: refs/heads/master@{#322452}

This avoids a possibly janky delayed UI task for users that don't use
the app launcher.

After this, a user enabling the app list for the first time won't have a
warmed-up UI but that shouldn't matter because the education page
doesn't try to show it, and ShowForAppInstall() will warm it up.

BUG=431326, 292969

Review URL: https://codereview.chromium.org/1031223002

Cr-Commit-Position: refs/heads/master@{#322451}

Initialize the QuicDispatcher with a PacketWriter instead of an int fd.
QuicDispatcher::Initialize() used to take an fd which chromium does not make
use of. Instead, it should simply take a PacketWriter* which the caller
should be responsible for creating. Renamed Initialize() to
InitializeWithWriter() to reflect this. Removed the CreateWriter() method
from the QuicDispatcher and moved it to QuicServer.
Merge internal change: 89495510

Review URL: https://codereview.chromium.org/1031243002

Cr-Commit-Position: refs/heads/master@{#322450}

Show error message and disable Verify button if expiration date is in
the past. Logic is a direct port from CardUnmaskPromptViews.

BUG=448572

Review URL: https://codereview.chromium.org/1032193003

Cr-Commit-Position: refs/heads/master@{#322449}

Revert of Allow users to perform queries in the omnibox which begin with a search shortcut. (patchset #4 id:60001 of https://codereview.chromium.org/477873002/)

Reason for revert:
Caused a few OmniboxViewTest interactive_ui_tests failures (tests need updating too).

build:
http://build.chromium.org/p/chromium.linux/builders/Linux%20Tests/builds/22466

test output:
https://build.chromium.org/p/chromium.chromiumos/builders/Linux%20ChromiumOS%20Tests%20%281%29/builds/501/steps/interactive_ui_tests/logs/OmniboxViewTest.AcceptKeywordBySpace

Looks like try jobs weren't run, only commit queue earlier. Unsure why commit queue didn't include this test - but it's recommended to run 'git cl try' on patches and review.

Original issue's description:
> Allow users to perform queries in the omnibox which begin with a search
> shortcut.
>
> Previously, if a user had (for example) configured a search shortcut of 'cs' which
> went to cs.chromium.org, they would be unable to (easily) start a query in the
> omnibox starting with 'cs'. Now, backspacing the keyword leaves the space after
> the keyword, allowing the user to continue typing without reenabling keyword
> mode.
>
> BUG=393371
> NOTRY=true
> R=pkasting@chromium.org
> TEST=type 'google.com ' into the omnibar and it should enter keyword mode.
> Press backspace and 'google.com ' will be left in the omnibar, rather than
> 'google.com'.
>
> Committed: https://crrev.com/709649607f13acb04e3850d5fd7fc1903c1becae
> Cr-Commit-Position: refs/heads/master@{#322429}

TBR=pkasting@chromium.org,sky@chromium.org,lenny.khazan@gmail.com
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=393371

Review URL: https://codereview.chromium.org/1030413002

Cr-Commit-Position: refs/heads/master@{#322448}

This CL removes:
* Unused support for logging in environment.py. The two current diagnostics messages will be replaced later if needed, after the code gets rearranged. The command-line support for log levels was not used (and not working last time I tried).
* Some unused code was removed, including WebsiteTest's SendEnterTo action.
* It simplified the way to obtain all tests names -- no longer needed to instantiate Environment.
* It removes adding ../../../../third_party/webdriver/pylib/ to sys.path -- this has not worked in most cases and was not needed (webdriver path has been added, from various sources, outside of the scripts in the framework we used).

It also removes some part of the unused ability to run all tests, which has recently been simulated by running all tests one by one outside of Python. That was to make the tests run in parallel. This is just a transitional state, and is described in the TODOs.

BUG=369521

Review URL: https://codereview.chromium.org/1022703004

Cr-Commit-Position: refs/heads/master@{#322447}

Improve PhoneFieldTest to test this.

Review URL: https://codereview.chromium.org/1031923003

Cr-Commit-Position: refs/heads/master@{#322446}

BUG=None
TEST=./build/gyp_chromium still works
R=sievers@chromium.org

Review URL: https://codereview.chromium.org/1038743003

Cr-Commit-Position: refs/heads/master@{#322445}

BUG=none

Review URL: https://codereview.chromium.org/1038003002

Cr-Commit-Position: refs/heads/master@{#322444}

Second step of a a three sided patch to move IdleTask to WebThread.
This is needed because we plan to introduce a scheduler for worker
threads, and they will use the WebTherad interface for posting idle
tasks.

Other patches https://codereview.chromium.org/1031843002/
and https://codereview.chromium.org/1029803002

BUG=463143

Review URL: https://codereview.chromium.org/1029343002

Cr-Commit-Position: refs/heads/master@{#322443}

Summary of changes available at:
https://chromium.googlesource.com/chromium/blink/+log/9737cee..1bb46ea

TBR=loislo@chromium.org,pkasting@chromium.org

Review URL: https://codereview.chromium.org/1037683003

Cr-Commit-Position: refs/heads/master@{#322442}

Adds red permanent error message banner between title and instructions.
Error messages passed to
CardUnmaskPromptViewBridge::GotVerificationResult() with
allow_retry == false are displayed. The banner is hidden until an error
message is passed to GotVerificationResult().

Note that the red banner does not extend to the full width of the dialog
(see screenshot attached to bug). Extending to the full dialog width
requires significant refactoring of the layout code, so I'm doing it in
two parts. This CL adds the banner functionality, and a separate CL will
fix the layout.

Dialog will re-layout to fit entire message if a multiline message is
set.

Also adds an ASCII art diagram of the dialog layout.

Screenshot attached to bug: https://code.google.com/p/chromium/issues/detail?id=448572#c15

BUG=448572

Review URL: https://codereview.chromium.org/1038503003

Cr-Commit-Position: refs/heads/master@{#322441}

BUG=470230

Review URL: https://codereview.chromium.org/997153004

Cr-Commit-Position: refs/heads/master@{#322440}

BUG=

Review URL: https://codereview.chromium.org/1027343002

Cr-Commit-Position: refs/heads/master@{#322439}

nsimage_cache.* and the unittest were all purged in r179484

BUG=None

Review URL: https://codereview.chromium.org/1032413002

Cr-Commit-Position: refs/heads/master@{#322438}

App Delegate conversion:
* Removed Application.Delegate
* Convert DesktopRemoting and AppRemoting to be subclasses of Application
* Add ApplicationInterface so that jscompile can verify all required overrides are provided.

Access to SessionConnector was a big motivation for having the apps as a subclass:
* Create the SessionConnector in the constructor rather than in the getter.
* Remove some references to remoting.app.getSessionConnector

Test updates:
* Updated base.inherits unittests to verify calling superclass methods with arguments.
* Removed desktopDelegateForTesting since we can use remoting.app for that now.

BUG=465878

Review URL: https://codereview.chromium.org/1016373003

Cr-Commit-Position: refs/heads/master@{#322437}

R=rockot@chromium.org

Review URL: https://codereview.chromium.org/1033003003

Cr-Commit-Position: refs/heads/master@{#322436}

The web app frame is experimental and should only be used by bookmark
apps.

This change also moves tests for bookmark app browser functionality out
of the general browser browser tests file into a separate file.

BUG=368372

Review URL: https://codereview.chromium.org/1037003002

Cr-Commit-Position: refs/heads/master@{#322435}

As part of a refactoring, the code to detect if a new live region
was added was accidentally modified to trigger if a live region
was changed in any way. It should only trigger on a brand-new
live region.

BUG=469276

Review URL: https://codereview.chromium.org/1030373003

Cr-Commit-Position: refs/heads/master@{#322434}

All leveldb uses now set the reuse_logs option to trigger a new fast open path.
For db's with prior changes (i.e. non-empty logs) this avoids three fsync calls
(on average) for every open as well as the need to compact (rewrite) the db.

More info at:

https://github.com/google/leveldb/commit/251ebf5dc70129ad3c38193fe6c99a5b0ec6b9fa

BUG=468810

Review URL: https://codereview.chromium.org/1025433003

Cr-Commit-Position: refs/heads/master@{#322433}

Calling the InfoBarContainerIOS destructor leads to the removal of infobars via
|InfoBarContainerIOS::PlatformSpecificRemoveInfoBar|, which leads to the BVC
wanting to change the InfoBarContainerIOS's height (in |infoBarContainerStateChanged|)
This leads to a crash because the InfoBarContainerIOS is in the process of
being deleted.

This CL makes the flow abort early.

BUG=448093

Review URL: https://codereview.chromium.org/1038903003

Cr-Commit-Position: refs/heads/master@{#322432}

The MouseEventPreconnect field trial has ended, but there is still a
trace of it remaining in ChromeBrowserFieldTrials. Remove it.

BUG=240959
TEST=chrome compiles

Review URL: https://codereview.chromium.org/1037033002

Cr-Commit-Position: refs/heads/master@{#322431}

GCD support is off by default.

BUG=470744

Review URL: https://codereview.chromium.org/1007153003

Cr-Commit-Position: refs/heads/master@{#322430}

shortcut.

Previously, if a user had (for example) configured a search shortcut of 'cs' which
went to cs.chromium.org, they would be unable to (easily) start a query in the
omnibox starting with 'cs'. Now, backspacing the keyword leaves the space after
the keyword, allowing the user to continue typing without reenabling keyword
mode.

BUG=393371
NOTRY=true
R=pkasting@chromium.org
TEST=type 'google.com ' into the omnibar and it should enter keyword mode.
Press backspace and 'google.com ' will be left in the omnibar, rather than
'google.com'.

Review URL: https://codereview.chromium.org/477873002

Cr-Commit-Position: refs/heads/master@{#322429}

  StoragePartitionHttpCacheDataRemover.

This will enable us to to clear http cache from a StoragePartition from
  chrome/ layer. <webview>s with their own storage partitions can use
  this to clear patitions' cache.

This will be used to implement clear cache for <webview>, complete
  CL is: https://chromiumcodereview.appspot.com/1021073002/

BUG=406437
Test=None, internal only change.

Review URL: https://codereview.chromium.org/1033013003

Cr-Commit-Position: refs/heads/master@{#322428}

Now that the image resources that blink depends on are in the blink
repository, we depend on the blink resources directly, delete the ones
in the chrome tree, and use these resources in the html_viewer.

[This reland also contains a fix to the generate_blink_resource_map.py
file so that its generated code uses unsigned chars, which should
hopefully fix the Win8 gn bot.]

BUG=467176
First Review URL: https://codereview.chromium.org/1033733004

Review URL: https://codereview.chromium.org/1038723002

Cr-Commit-Position: refs/heads/master@{#322427}

BUG=405038
TEST=
1. Sign in to Chrome and make sure there are no Auth Errors
2. Navigate to chrome://signin-internals
3. In the Basic Information section, observe that there is no ErrorController Username or ErrorController Status section
4. Trigger an Auth Error (for example by revoking Chrome's access token at accounts.google.com)
5. In chrome://signin-internals, observe that the Basic Information section was updated with the ErrorController Username and ErrorController Status fields.

Review URL: https://codereview.chromium.org/1025323004

Cr-Commit-Position: refs/heads/master@{#322426}

Currently media code uses a separate logging system (MediaLog) and some important messages are reported there, but are not reported in the main browser log. Currently media log messages can only be seen in the chrome://media-internals, but since they are very useful, we should add them to the main browser log as well.

BUG=453180

Review URL: https://codereview.chromium.org/877273002

Cr-Commit-Position: refs/heads/master@{#322425}

BUG=469582

Review URL: https://codereview.chromium.org/1036663002

Cr-Commit-Position: refs/heads/master@{#322424}

Currently, ChromeVox Next uses the programmatic string when whenever $role gets used. This cl makes it so $role resolves to a msg id, if one exists. Additionally, the role info object also lists the arcon associated with that role.

This allows us to remove a few of the output rules (button, textField).

Review URL: https://codereview.chromium.org/1011913002

Cr-Commit-Position: refs/heads/master@{#322423}

Set Fixedsource image unscaled to make it really fixed.

BUG=None

Review URL: https://codereview.chromium.org/1038523006

Cr-Commit-Position: refs/heads/master@{#322422}

Review URL: https://codereview.chromium.org/1036513008

Cr-Commit-Position: refs/heads/master@{#322421}

BUG=470232

Review URL: https://codereview.chromium.org/1034873003

Cr-Commit-Position: refs/heads/master@{#322420}

Revert of Send autocapitalize flags to the Android virtual keyboard based on Blink hints. (patchset #6 id:100001 of https://codereview.chromium.org/1003143002/)

Reason for revert:
compile fail on
http://build.chromium.org/p/chromium.linux/builders/Android%20Arm64%20Builder%20%28dbg%29/builds/13039

../content/public/android/java/src/org/chromium/content/browser/input/AdapterInputConnection.java:128:error: cannot find symbol
        if ((inputFlags & WebTextInputFlags.AutocapitalizeCharacters) != 0) {

Original issue's description:
> Send autocapitalize flags to the Android virtual keyboard based on Blink hints.
>
> This changes the behaviour of AdapterInputConnection to pass a
> InputType.TYPE_TEXT_FLAG_CAP_* based on the
> WebTextInputFlags.Autocapitalize* instead of doing it purely based on the
> element's type.
>
> This CL requires the following Blink CL:
> https://codereview.chromium.org/995363002
>
> BUG=466930
>
> Committed: https://crrev.com/d1421a5faf9dc2d3b3cad10640576b24a092d9ba
> Cr-Commit-Position: refs/heads/master@{#322414}

TBR=aurimas@chromium.org,shuchen@chromium.org,mlamouri@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=466930

Review URL: https://codereview.chromium.org/1008153003

Cr-Commit-Position: refs/heads/master@{#322419}

Summary of changes available at:
https://chromium.googlesource.com/chromium/blink/+log/12736d7..9737cee

TBR=loislo@chromium.org,pkasting@chromium.org

Review URL: https://codereview.chromium.org/1035993002

Cr-Commit-Position: refs/heads/master@{#322418}