GitLab

This CL prevents http: or https: urls from referring to file: urls as search descriptor xmls.
Allowing urls to refer to other urls with the same scheme has been considered (e.g. a file: url
referring to an OSDD xml from a file: url), but this currently doesn't work for file: urls
because of http://b/863583, so is not implemented here.

BUG=429838

Review URL: https://codereview.chromium.org/917313004

Cr-Commit-Position: refs/heads/master@{#317723}

Log events and debugging information related to HID devices through the
device event log component so that these errors are visible at
chrome://device-log. A DEVICE_PLOG macro has been added which includes
the last reported system error in the log message.

BUG=448901

Review URL: https://codereview.chromium.org/947663002

Cr-Commit-Position: refs/heads/master@{#317722}

BUG=434363
TEST=git cl presubmit -uv
R=kalman@chromium.org

Review URL: https://codereview.chromium.org/942263002

Cr-Commit-Position: refs/heads/master@{#317721}

Show menu item to view original image if the request could go over
the data reduction proxy.

BUG=456247

Review URL: https://codereview.chromium.org/944533002

Cr-Commit-Position: refs/heads/master@{#317720}

It turns out that the permissions are working fine, but the documentation
is inadequate.

BUG=434430

Review URL: https://codereview.chromium.org/949543002

Cr-Commit-Position: refs/heads/master@{#317719}

Passed in argument was not meant to be deleted upon merge:
https://codereview.chromium.org/920523002

BUG=457385
TEST=trybots

Review URL: https://codereview.chromium.org/954503003

Cr-Commit-Position: refs/heads/master@{#317718}

Review URL: https://codereview.chromium.org/918463002

Cr-Commit-Position: refs/heads/master@{#317717}

Surfaces should be feature-complete, so try enabling them on Windows, Linux, and Mac OS. They're still disabled on Chrome OS because they rely on single-threaded browser compositing, and on Android because they haven't been implemented there yet.

BUG=334090
TBR=piman@chromium.org

Review URL: https://codereview.chromium.org/923833002

Cr-Commit-Position: refs/heads/master@{#317716}

Some tests regressed in r316774 because the way edit commands were
dispatched in BridgedContentView changed from going directly to the
TextInputClient to instead go through the InputMethod.

r316774 included a check added to views::Textfield to ensure any
scheduled edit command had been processed before scheduling a new one.
Some tests failed that check: Because the textfield did not have focus,
the dispatch of the event that would commit the edit command was not
being received.

This CL ensures the textfield requests focus in the test.

Fixes (mac-only) views_unittests:
BridgedNativeWidgetTest.TextInput_MoveLeftRight
BridgedNativeWidgetTest.TextInput_DeleteBackward
BridgedNativeWidgetTest.TextInput_DeleteForward

BUG=454353

Review URL: https://codereview.chromium.org/941943003

Cr-Commit-Position: refs/heads/master@{#317715}

This value corresponds to media::VideoFrame::YV12HD which was added in a previous CL.

BUG=None

Review URL: https://codereview.chromium.org/948633002

Cr-Commit-Position: refs/heads/master@{#317714}

BUG=455586

Review URL: https://codereview.chromium.org/947613003

Cr-Commit-Position: refs/heads/master@{#317713}

This patch changes tools/perf/run_benchmark to output the list command
on stdout instead of stderr. This makes grepping for benchmarks easier.

Note: The run command will still output the benchmark list on stderr if
an invalid benchmark is specified.

Review URL: https://codereview.chromium.org/948793002

Cr-Commit-Position: refs/heads/master@{#317712}

The TextInputClient is used to validate allowable operations in a
textfield (e.g. cut/copy/paste). On a readonly textfield, Copy should
work, but Cut/Paste should present to the user as being disabled.

Currently, a read-only textfield returns null from GetTextInputClient(),
making this validation impossible, except "inside" the textfield (e.g.
Textfield::ShowContextMenuForView(..)).

The logic originated in r80226, but doesn't seem to be for a specific
reason.

Note that TextInputClient::GetTextInputType() returns
TEXT_INPUT_TYPE_NONE for readonly textfields, and a test is updated to
check.

BUG=454353

Review URL: https://codereview.chromium.org/919543003

Cr-Commit-Position: refs/heads/master@{#317711}

R=tsepez@chromium.org
BUG=336640

Review URL: https://codereview.chromium.org/942683003

Cr-Commit-Position: refs/heads/master@{#317710}

Mips builds hit this warning when linking the mesa library with libc++.
Temporarily disable the warning until a permanent solution can be found.

BUG=456380

Review URL: https://codereview.chromium.org/898403003

Cr-Commit-Position: refs/heads/master@{#317709}

This changelist primarily changes HostProcess::OnPolicyError, so that it
doesn't anymore shut down the host, but instead restarts the host and prevents
the host from starting again until a valid policy has been read (the
"preventing" functionality has already been present in the form of
StartHostIfReady method for some time now).

This changelist also simplifies HostSignalingManager's functionality related to
sending host-offline-reason.  Previously HostProcess was destroyed while
HostSignalingManager was trying to send host-offline-reason.  Things are
simpler if HostProcess waits until HostSignalingManager sends
host-offline-reason or times out.  This synchronization is especially helpful
when restarting the host (to avoid having to deal with an old
HostSignalingManager from previous SendHostOfflineReason attempt, while we
construct a new HostSignalingManager when starting a host after a reset).

This changelist also introduces a separate host-offline-reason for policy
errors (and updates WebApp resources strings to reflect that).  Note that since
we are not shutting down the host, we probably cannot use one of HostExitCodes
as the host-offline-reason.

Testing done:
1. On Linux manually introduce policy errors at startup, verify host doesn't
   start and sends host-offline-reason, fix errors, verify host starts,
   reintroduce errors, verify host suspends and sends host-offline-reason, fix
   errors, verify host starts again.
2. Verify WebApp UI shows correct tooltip text for the new host-offline-reason.

BUG=455903

Review URL: https://codereview.chromium.org/910403002

Cr-Commit-Position: refs/heads/master@{#317708}

(e.g. scrolling, letterboxing the plugin, resizing the host) from
remoting.DesktopConnectedView into smaller single-purposed classes.

Summary of changes:
1. Moves bumpScrolling implementation into the remoting.BumpScroller.
2. Moves all viewport related implementation from remoting.DesktopConnectedView
   into remoting.DesktopViewport

BUG=457890

Review URL: https://codereview.chromium.org/918783002

Cr-Commit-Position: refs/heads/master@{#317707}

Summary of changes available at:
https://chromium.googlesource.com/chromium/third_party/ffmpeg/+log/d4b1674..f5698b4

BUG=344767

Review URL: https://codereview.chromium.org/938223003

Cr-Commit-Position: refs/heads/master@{#317706}

Reboot and brain freeze screens can lead be followed by seeing the
browser compositor's background colour, which defaults to white.

BUG=internal b/19379351

Review URL: https://codereview.chromium.org/950123002

Cr-Commit-Position: refs/heads/master@{#317705}

Actually, this creates two libraries:
ppapi_simple
ppapi_simple_cpp

The second must be used if you want to use ppapi_cpp. If you use the C
interfaces exclusively, you can use ppapi_simple instead.

BUG=

Review URL: https://codereview.chromium.org/914983003

Cr-Commit-Position: refs/heads/master@{#317704}

Removing the isolate.py functionality and requiring tasks to be created from an archive hash rather than an isolate file.

Also adding back the controller_test.py and isolate files which were missed in the refactor cl due to a bad move. The only change to these files are to support isolate hashes rather than isolate files.

BUG=460340

Review URL: https://codereview.chromium.org/943923003

Cr-Commit-Position: refs/heads/master@{#317703}

To bring in:

6658243  Update the GN build with the cpu_arch -> current_cpu rename.

TBR=rtoy@chromium.org
BUG=344767

Review URL: https://codereview.chromium.org/947323002

Cr-Commit-Position: refs/heads/master@{#317702}

Revert of Whitespace change to test gnumbd mirror. (patchset #1 id:1 of https://codereview.chromium.org/928073005/)

Reason for revert:
This was only a test, cleaning up.

Original issue's description:
> Whitespace change to test gnumbd mirror.
>
> This will be manually committed to test for the replication lag from
> MTV.
>
> TBR=iannucci@chromium.org
> BUG=None
>
> Committed: https://chromium.googlesource.com/chromium/src/+/824fb9f7c83330b7470608d71bb6cff63b5f858a

TBR=iannucci@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=None

Review URL: https://codereview.chromium.org/951743004

Cr-Commit-Position: refs/heads/master@{#317701}

TBR=rockot@chromium.org
BUG=

Review URL: https://codereview.chromium.org/921033004

Cr-Commit-Position: refs/heads/master@{#317700}

Simply hide the Autofill/Wallet preferences. The correct fix will involve checking the state of the sync experiment, but for M42 it's enough to just disable this preference page.

Will require merge to m42.

BUG=460900
R=aurimas@chromium.org

Review URL: https://codereview.chromium.org/947253002

Cr-Commit-Position: refs/heads/master@{#317699}

This will be manually committed to test for the replication lag from
MTV.

TBR=iannucci@chromium.org
BUG=None

Review URL: https://codereview.chromium.org/928073005

Cr-Commit-Position: refs/heads/master@{#317698}

BUG=454009
TBR=newt@chromium.org

Review URL: https://codereview.chromium.org/948953002

Cr-Commit-Position: refs/heads/master@{#317697}

BUG=454908

Review URL: https://codereview.chromium.org/941133002

Cr-Commit-Position: refs/heads/master@{#317696}

BUG=

Review URL: https://codereview.chromium.org/945903005

Cr-Commit-Position: refs/heads/master@{#317695}

R=erikchen@chromium.org, rockot@chromium.org
TBR=erikchen@chromium.org, rockot@chromium.org

Review URL: https://codereview.chromium.org/950043003

Cr-Commit-Position: refs/heads/master@{#317694}

When the bookmark bar is hidden, BookmarkBarToolbarView's height
is set to 0.
Before https://crrev.com/924143002/, when BookmarkButton calls its
drawRect through cr_drawUsingAncestor, it was filling its bounds,
effectively drawing nothing.
After the change, it was filling |dirtyRect|, so it was drawing the SHOW
state even though its bounds has 0 height.
The whole thing is fragile and can probably be done better, but we should
not ask the ancestor to draw outside its bounds anyway, thus this fix.

BUG=459992

Review URL: https://codereview.chromium.org/951063002

Cr-Commit-Position: refs/heads/master@{#317693}

The performance is horrendous, so let's disable the animations for each of the
menu items for now as it looks 90% as good as a well performing animation on
newer builds of android.

BUG=428477

Review URL: https://codereview.chromium.org/948183003

Cr-Commit-Position: refs/heads/master@{#317692}

By default Skia blends the source and destination buffers when drawing
an image. This forces it to copy the source buffer to the modeset
buffer.

BUG=455605

Review URL: https://codereview.chromium.org/948903003

Cr-Commit-Position: refs/heads/master@{#317691}

BUG=417532

Review URL: https://codereview.chromium.org/938453002

Cr-Commit-Position: refs/heads/master@{#317690}

With new appcontainer token in picture, it fails to create shared memory with empty name like "", it explicity seem to need NULL for anonymous.

BUG=455496
R=cpu

Review URL: https://codereview.chromium.org/950993003

Cr-Commit-Position: refs/heads/master@{#317689}

This is a resubmit of https://codereview.chromium.org/904083002
which was reverted due to an unrelated flaky test. There are no
changes compared to the original patch, hence skipping the code
review (zea@ was the original reviewer).

Original patch description:

The fix improves matching of nodes in BookmarkModelAssociator in
situations where there are multiple bookmarks or folders with the
same titles or URLs.
This will address one particular scenario leading to bookmark
duplication (see crbug.com/118105).

1) In BookmarkModelAssociator::BuildAssociations, when there are
multiple native model nodes with matching title / URL, a secondary
match on external ID is used to pick a preferred one; otherwise
the first matching node is returned.
The preferred match on external ID should be applicable in most
situations except when the native model has been rebuilt from scratch.
Picking a wrong folder during the association process results in
duplicating the entire subtree within the wrong folder. This issue
should be addressed now.
2) In BookmarkModelAssociator::ApplyDeletesFromSyncJournal the external
ID match is now the primary criteria for selecting a native model
node to be deleted. The previous implementation would pick an arbitrary
native model node based on just the title / URL match anywhere in the
node hierarchy. That would happen every time after deleting a bookmark
or folder and recreating it in another place.
Since external IDs might be reused, there is a secondary match on
title and URL to ensure that the right node gets deleted. To avoid
costly O(N*M) algorithm (where N is number of bookmarks and M is
number of entries in delete journal), the implementation uses a set
of external IDs to reduce the cost to O(N*logM).

BUG=456228
TBR=zea@chromium.org

Review URL: https://codereview.chromium.org/912693002

Cr-Commit-Position: refs/heads/master@{#317688}

DRM devices such as VGEM do not have resources, so we want to ignore
them rather than crash.

BUG=460982

Review URL: https://codereview.chromium.org/947293002

Cr-Commit-Position: refs/heads/master@{#317687}

Revert of ui: Add another presubmit check to catch two more scoped_ptr usages. (patchset #6 id:100001 of https://codereview.chromium.org/919253002/)

Reason for revert:
Presubmit in ui/ failing with

...
File "<string>", line 38, in CheckScopedPtr
NameError: global name 're' is not defined

Original issue's description:
> ui: Add another presubmit check to catch two more scoped_ptr usages.
>
> This should catch the following usages:
>
> 1- return scoped_ptr<T>(foo)
> 2- bar = scoped_ptr<T>(foo)
>
> And recommend the solo usage of make_scoped_ptr().
>
> The entries were found with the following command line:
>
> $ git grep -E '(=|\breturn)\s*scoped_ptr<.*?>([^)]+)'
>
> BUG=None
> TEST=g cl presubmit -uv
> R=sky@chromium.org,maruel@chromium.org
>
> Committed: https://crrev.com/63f25bef305bd8143766e2e1a98bf01548c78376
> Cr-Commit-Position: refs/heads/master@{#317669}

TBR=maruel@chromium.org,sky@chromium.org,maruel@google.com,sadrul@chromium.org,tfarina@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=None

Review URL: https://codereview.chromium.org/954473003

Cr-Commit-Position: refs/heads/master@{#317686}

* Stop calling GetIconID() as a proxy for having RTTI to detect
  ThreeDAPIInfoBarDelegates.  Just add a downcast function.
* Slightly more reasonable base class function order.
* Fix declaration/definition orders to match base class (and each other) in
  other instances where it was noticeably wrong.
* Keep infobar code consistent about not inlining method definitions in class
  declarations, and making overridden functions private unless otherwise
  necessary.

BUG=none
TEST=none

Review URL: https://codereview.chromium.org/922953002

Cr-Commit-Position: refs/heads/master@{#317685}

This brings in:

28ddd48  Replace second set of #defines with templates in JS_Define.h
94dd28e  Update GN build files w/ cpu_arch -> current_cpu changes.
dce5730  Add test for PDF's JS "global".
a96cc9b  Replace ugly JS_Define macros with templates.
2da7bd4  Tidy up JS_Defines.h
7435e8e  Run pixel tests via automated script.
83c87e5  run_javascript_tests.py: Be more flexible about directory layout.

TBR=jam@chromium.org, tsepez@chromium.org
BUG=344767

Review URL: https://codereview.chromium.org/948933003

Cr-Commit-Position: refs/heads/master@{#317684}