GitLab

This change implements the key C++ portion of the Select-to-speak feature
by adding an event handler that takes Search plus the mouse button, and
fires accessibility events instead.

In the future we could modify or extend this event handler for use in
ChromeVox too, but for now I think building something specifically for
Select-to-speak makes sense.

Adds new accessibility events for mouse down, mouse up, and so on - allowing
the select-to-speak extension to get not only the mouse events and global
coordinates, but most importantly the proper accessibility object that's hit
for each event.

This change implements support for hit testing in Views. For web,
the extension will get the event fired on the views::WebView and it can do
its own hit testing within the accessibility tree from there. A follow-up
will improve on this by doing native hit testing in the web before firing
the accessibility event.

BUG=593887

Review-Url: https://codereview.chromium.org/2493923002
Cr-Commit-Position: refs/heads/master@{#434744}

The problem was that |route_id| is scoped to a particular child process,
but the process was not being plumbed in correctly, resulting in a
situation where we would create (1) a RenderFrameHost with a routing ID
that might later be reused, and also, (2) a RenderFrameHost with a
plumbed-in routing ID, but no corresponding extant RenderFrame in the
child process.

The solution involves plumbing in the actual SiteInstance, and only
using |route_id| if we don't change to a different SiteInstance.

BUG=627852
TBR=brettw@chromium.org

Review-Url: https://codereview.chromium.org/2521793003
Cr-Commit-Position: refs/heads/master@{#434743}

https://chromium.googlesource.com/chromiumos/chromite.git/+log/b5e9339ae484..0ab5a1cdaf39

$ git log b5e9339ae..0ab5a1cda --date=short --no-merges --format='%ad %ae %s'
2016-11-28 chrome-bot Update config settings by config-updater.
2016-11-28 drinkcat parseelf: Add comment about getting number of symbols from DT_HASH
2016-11-16 sjg crosfw: Return an exit code when we get a build error
2016-11-25 drinkcat parseelf: Ignore empty symbol_name
2016-11-25 drinkcat parseelf: Parse number of symbols from DT_HASH section
2016-11-21 sadrul deploy_chrome: Remove special handling for mash
2016-11-23 ayatane [sysmon] Track the number of running sysmon processes
2016-11-23 chrome-bot Update config settings by config-updater.
2016-11-23 nxia Revert "Add buildbucket_id index in buildTable"

BUG=none

Review-Url: https://codereview.chromium.org/2536553004
Cr-Commit-Position: refs/heads/master@{#434742}

This is a follow up to https://codereview.chromium.org/2483003003/

It introduces a factory class to create and get a CertificateReportingService
singleton. It also introduces a metrics provider class that currently doesn't
do anything, but will send pending reports in follow up CLs.

BUG=554323

Review-Url: https://codereview.chromium.org/2507253004
Cr-Commit-Position: refs/heads/master@{#434741}

GpuMemoryBufferHandle::shared_memory_handle is only valid if type is
SHARED_MEMORY_BUFFER. So try to serialize only when it's valid.

BUG=643746

Review-Url: https://codereview.chromium.org/2534533002
Cr-Commit-Position: refs/heads/master@{#434740}

Passing -S to python breaks in a virtualenv because it cannot
find the stdlib modules (e.g. base64). Only pass -S if not
already in a virtualenv.

BUG=667897
TEST=net_unittests inside a virtualenv

Review-Url: https://codereview.chromium.org/2526553002
Cr-Commit-Position: refs/heads/master@{#434739}

When download path is determined, Chrome will commit that information to history.
However, the information is not committed immediately.
As a result, if Chrome crashes before the next commit, the path information is lost.
When resuming the download later, Chrome will generate a new target path.
And the old temp file is left permanently on external storage.
This might not be a big issue for desktop, as Chrome rarely gets killed and there are ample storage.
However, this has a great impact on android.
This CL requests the path information to be committed immediately to address the issue.

BUG=664677

Review-Url: https://codereview.chromium.org/2508503002
Cr-Commit-Position: refs/heads/master@{#434738}

Reland of Disable Precompile-platform with clang-cl. (patchset #1 id:1 of https://codereview.chromium.org/2526383002/ )

Reason for revert:
Clang still asserts when building Precompile-core.obj, but it's somewhere else:
Assertion failed: AllowUnresolvedNodes && "Cannot handle unresolved nodes"...

Original issue's description:
> Revert of Disable Precompile-platform with clang-cl. (patchset #1 id:1 of https://codereview.chromium.org/2520833005/ )
>
> Reason for revert:
> The crash is now hopefully fixed in clang-cl.
>
> Original issue's description:
> > Disable Precompile-platform with clang-cl.
> >
> > It currently asserts while trying to build that PCH file.
> >
> > BUG=667891,495697
> > NOTRY=true
> >
> > Committed: https://crrev.com/5b7b04e7fa59d53b8b7917d4c0cf4081abdfb0e3
> > Cr-Commit-Position: refs/heads/master@{#434028}
>
> TBR=rnk@chromium.org
> # Not skipping CQ checks because original CL landed more than 1 days ago.
> BUG=667891,495697
>
> Committed: https://crrev.com/a0eb8ed36a1c3e2c5becc95ef03d42655c7cd95c
> Cr-Commit-Position: refs/heads/master@{#434534}

TBR=thakis@chromium.org
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG=667891,495697

Review-Url: https://codereview.chromium.org/2533853003
Cr-Commit-Position: refs/heads/master@{#434737}

Previously ash would make a direct connection to Chrome (content_browser) to
handle showing system tray menu UI. This is a layering violation and makes
it more difficult to run ash without chrome.

Chrome's SystemTrayClient now explicitly registers itself with ash on startup.
It no longer attempts to reconnect on error; my understanding is that chrome
tolerating an ash crash is not a goal.

BUG=none
TEST=manual, can still open settings via system tray menu

Review-Url: https://codereview.chromium.org/2525813004
Cr-Commit-Position: refs/heads/master@{#434736}

BUG=667373

Review-Url: https://codereview.chromium.org/2530633003
Cr-Commit-Position: refs/heads/master@{#434735}

BUG=666616

Review-Url: https://codereview.chromium.org/2522313003
Cr-Commit-Position: refs/heads/master@{#434734}

Adding EmbeddedTestServer handlers after the server has started isn't
thread safe.

This CL fixes the threading issues by either starting the server after
the handlers are added or by splitting the server start into two phases.
This is needed when the server's base URL is required for registering
the handlers.

BUG=546060

Review-Url: https://codereview.chromium.org/2522283002
Cr-Commit-Position: refs/heads/master@{#434733}

BUG=652332

Review-Url: https://codereview.chromium.org/2528553002
Cr-Commit-Position: refs/heads/master@{#434732}

Use an int instead of a StaticAtomicSequenceNumber to generate the ids for
the gpu memory buffers. This should be sufficient since this is used only
in the same thread.

BUG=none

Review-Url: https://codereview.chromium.org/2532763003
Cr-Commit-Position: refs/heads/master@{#434731}

This adds signature validation for the component cloud policy
(e.g. policy for extensions). The signature is validated against
the same key that is used for the "superior" policy (e.g. the
user policy, the device local account policy, etc.).

This CL also adds keeping a copy of the most recent component
cloud policy and rechecking it when some of the credentials
change. This allows to handle key rotations gracefully: even
though the component cloud policy may fail the validation
immediately after the cloud policy refresh with the rotated key,
it will be re-validated and applied when the superior policy gets
processed and the credentials get propagated.

BUG=644632
TEST=existing tests (now with the signature checks enabled),
     new unit tests and new browser test

Review-Url: https://codereview.chromium.org/2493603002
Cr-Commit-Position: refs/heads/master@{#434730}

https://pdfium.googlesource.com/pdfium.git/+log/e421327725da..c92e26defaa6

$ git log e42132772..c92e26def --date=short --no-merges --format='%ad %ae %s'
2016-11-28 npm Merge CPDF_ImageLoader with its handle
2016-11-28 npm Nits in fx_ge_linux
2016-11-28 npm Fix leak in cpdf_document_unittest

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/2539513002
Cr-Commit-Position: refs/heads/master@{#434729}

Both type name and stack deduplicators need to estimate their memory
overhead. Previously they were using simplistic models and estimated
only half of their actual memory usage. This CL changes estimation
methods to use EstimateMemoryUsage(), resulting in better accuracy.

Review-Url: https://codereview.chromium.org/2514913002
Cr-Commit-Position: refs/heads/master@{#434728}

The DCHECK was mostly benign, though Chrome would close the socket
without waiting for a response to its QUIT messages, which has unclear
consequences.

BUG=668759

Review-Url: https://codereview.chromium.org/2532873003
Cr-Commit-Position: refs/heads/master@{#434727}

* When making a function to get offline pages using query public, we
  missed making it run only when the model is loaded.
  This patch fixes that by deferring the actually execution after the
  model is loaded.

BUG=668174

Review-Url: https://codereview.chromium.org/2536573003
Cr-Commit-Position: refs/heads/master@{#434726}

BUG=546953

Review-Url: https://codereview.chromium.org/2534873002
Cr-Commit-Position: refs/heads/master@{#434725}

It is no longer possible to explicitly construct and destroy surfaces. A
SurfaceFactory instance now handles only one surface at a time. Once the
local frame id passed to SubmitCompositorFrame changes, the factory gets
rid of the old surface and creates a new one.

BUG=658607

Committed: https://crrev.com/b902fd366ba91757dbee99d5f821f00d2de1d181
Cr-Commit-Position: refs/heads/master@{#432312}
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel

Review-Url: https://codereview.chromium.org/2485473003
Cr-Commit-Position: refs/heads/master@{#434724}

The Bar has main bar text that can either show the Search Term or the
Context.  When the user taps we show the Context first, and the Search
Term is not yet visible, so calculations based on its size are invalid.

We now check to see if the Search Term text is visible before using any
attribute of it.

BUG=658771

Review-Url: https://codereview.chromium.org/2518503004
Cr-Commit-Position: refs/heads/master@{#434723}

This CL creates a content_renderer service manifest overly for android_webview
to enable autofill mojo interfaces to be accessed well from content_browser
service.

BUG=662040

Review-Url: https://codereview.chromium.org/2523193002
Cr-Commit-Position: refs/heads/master@{#434722}

Just as with the client group the window_manager group should include
most of the client interfaces.

BUG=none
TEST=none
R=sadrul@chromium.org, tsepez@chromium.org

Review-Url: https://codereview.chromium.org/2528983002
Cr-Commit-Position: refs/heads/master@{#434721}

It was removed upstream in a few months ago in rL268670
(https://reviews.llvm.org/D19892)

No behavior change.

BUG=none

Review-Url: https://codereview.chromium.org/2530413003
Cr-Commit-Position: refs/heads/master@{#434720}

Revert of Improve fallback for Burmese with leading punctuation + spacing mark (patchset #3 id:30009 of https://codereview.chromium.org/2530153002/ )

Reason for revert:
This seems to be causing test failures for fast/text/international/text-spliced-font.html on https://build.chromium.org/p/chromium.webkit/builders/WebKit%20Win10/builds/17622

Original issue's description:
> Improve fallback for Burmese with leading punctuation + spacing mark
>
> Issue 618178 describes an example where a Burmese text run starts with a
> leading punctuation character followed by a combining spacing mark. This
> grapheme cannot be shaped with the default font, since Times for example
> cannot display the combination of a left quote with a Burmese combining
> mark. Our fallback code attempts to find a fallback font based on the
> first character at the beginning of an extracted unshaped sub-run, which
> does not lead to finding a font suitable for Myanmar text in this case.
> So in a way it runs into a fallback trap, where no fallback hint helps
> to find the right fallback font and the whole run ends up as notdef
> glyphs.
>
> This CL attempts to resolve this by looking for a better fallback hint
> character, which is not script common or inherited, if such is
> available. This improves the situation for the Burmese text from the
> issue report.
>
> In addition, as a better fix we should give higher importance to the
> locale information in font fallback, filed as issue 668706.
>
> BUG=618178
> R=eae,kojii,behdad
>
> Committed: https://crrev.com/d9280a5e951415e1b2c6c7958adda19b731a99fa
> Cr-Commit-Position: refs/heads/master@{#434665}

TBR=behdad@chromium.org,eae@chromium.org,kojii@chromium.org,drott@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=618178

Review-Url: https://codereview.chromium.org/2532253002
Cr-Commit-Position: refs/heads/master@{#434719}

This patch fixes an issue where the first enumeration of USB devices on
Linux would only return a single device and observers would not be
notified of additional devices. This was because the logic to determine
if enumeration callbacks should be called was checking only that all
previously seen devices (of which there was only one at that point) were
ready and not that the sysfs tree had been completely walked.

BUG=662486

Review-Url: https://codereview.chromium.org/2523403002
Cr-Commit-Position: refs/heads/master@{#434718}

BUG=666450

Review-Url: https://codereview.chromium.org/2535773004
Cr-Commit-Position: refs/heads/master@{#434717}

These are from two recent CLs:
https://codereview.chromium.org/2463953002
https://codereview.chromium.org/2494553002

Note: This CL is a test of the WPT export process. Do not make changes
in LayoutTests/imported/wpt/ without coordination or changes may be
lost.

BUG=657117

Review-Url: https://codereview.chromium.org/2509983007
Cr-Commit-Position: refs/heads/master@{#434716}

StreamTextureHost is always created with a valid |route_id|, and it also
does not handle OnChannelError().

This change removes the useless |route_id| checks, sets |channel_| to
nullptr when OnChannelError() is called, and adds appropriate |channel_|
null checks.

This change also removes raw pointers from StreamTextureProxy, to fix
style conformance and StreamTextureHost ownership clearer.

BUG=667162

Review-Url: https://codereview.chromium.org/2530443002
Cr-Commit-Position: refs/heads/master@{#434715}

Usually custom tabs force-load URLs or redirects to URLs that can be
handled by apps inside custom tabs. For instant-apps enabled clients
override the default behaviour and launch the instant app, if possible.

BUG=659301

Review-Url: https://codereview.chromium.org/2532223002
Cr-Commit-Position: refs/heads/master@{#434714}

Neither dvh@ nor grv@ still work on Chromium.

BUG=None

Review-Url: https://codereview.chromium.org/2533033002
Cr-Commit-Position: refs/heads/master@{#434713}

Currently, when we create the Physical Web data source, we do not
populate it with any initial data, which can lead to onLosts occuring
when there has not even been an onFound.  This change call onFound for
currently known URLs when first creating the data source.

BUG=667754

Review-Url: https://codereview.chromium.org/2521333002
Cr-Commit-Position: refs/heads/master@{#434712}

BUG=666630

Review-Url: https://codereview.chromium.org/2517693002
Cr-Commit-Position: refs/heads/master@{#434711}

This patch fixes spellcheck caching in the following case:
1. paragraph is edited by removing text from the back and
2. paragraph does not contain misspelled words.

Currently this case results in a cache miss and consequently
spellcheck requests being sent to the spellchecking service
while completely unnecessary. This fix applies to all
platforms.

This patch also adds a few unit tests for the relevant
caching specific behavior.

BUG=664247,629609

Review-Url: https://codereview.chromium.org/2494123002
Cr-Commit-Position: refs/heads/master@{#434710}

Also fixed and reenabled the fullscreen test in BrowserWindowControllerBrowserTest

BUG=599119, 643418

Review-Url: https://codereview.chromium.org/2516803002
Cr-Commit-Position: refs/heads/master@{#434709}

Remove superfluous closeUI call from onDismiss, this is
already done in disconnectFromClientWithDebugMessage.

BUG=587995

Review-Url: https://codereview.chromium.org/2536783002
Cr-Commit-Position: refs/heads/master@{#434708}

Move struct VideoFrameInfo from content/common/video_capture.mojom to
media/capture/mojo/video_capture_types.mojom. This will allow us to use the
struct in code that transports frames through the browser-process. This will,
in turn, allow us to (in a later CL) switch between the existing in-process
implementation and the new Mojo service.

This CL is part of the Mojo Video Capture work. For the bigger picture,
see [1] CL1.9.6

BUG=584797
TEST=video_capture_unittests, content_unittests, capture_unittests

[1] https://docs.google.com/a/chromium.org/document/d/1Qw7rw1AJy0QHXjha36jZNiEuxsxWslJ_X-zpOhijvI8/edit?usp=sharing

Review-Url: https://codereview.chromium.org/2524103002
Cr-Commit-Position: refs/heads/master@{#434707}

Remove the restriction that, under certain conditions, we require
the protocol of an auto-created custom search engine be HTTP.  The
condition has no connection with a need for privacy.  There is no
problem with creating custom search engines for secure sites.  In
fact, we should be encouraging more sites to be secure, not penalizing
those that do so by making it more difficult for them to create a
custom search engine.

BUG=642848
TEST=amazon.com now gets a custom search engine created

Review-Url: https://codereview.chromium.org/2505933005
Cr-Commit-Position: refs/heads/master@{#434706}

Revert of ImageCapture: move image_capture.mojom from media/mojo/interfaces/ to media/capture/mojo/ (patchset #2 id:40001 of https://codereview.chromium.org/2526953002/ )

Reason for revert:
imagecapture layout tests started failing in https://build.chromium.org/p/chromium.webkit/builders/WebKit%20Mac10.9/builds/40056

Original issue's description:
> ImageCapture: move image_capture.mojom to media/capture/mojo/
>
> This CL moves image_capture.mojom from media/mojo/interfaces/ to
> media/capture/mojo/ where it belongs logically speaking, and
> updates BUILD.gns and include paths.
>
> image_capture.mojom was in media/mojo/interfaces because
> when it landed, media/capture/mojo was not existing.
>
> BUG=518807
> TEST=No code touched, if compile passes, it works
> (but there's also content_browsertests).
>
> TBR=xhwang@chromium.org for the removal in media/mojo/interfaces/BUILD.gn
>
> Committed: https://crrev.com/d7c801bb7ed0393d1634e547a696ced5040bc93c
> Cr-Commit-Position: refs/heads/master@{#434675}

TBR=mkwst@chromium.org,chfremer@chromium.org,jochen@chromium.org,tsepez@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/2537573002
Cr-Commit-Position: refs/heads/master@{#434705}