GitLab

Accoding to the stack on crbug.com/410602, BlobData can be null.
However, I am not able to reproduce the bug with any test cases.
So this is just a safe speculative fix.

BUG=410602

Review URL: https://codereview.chromium.org/601043003

Cr-Commit-Position: refs/heads/master@{#297007}

R=qyearsley@chromium.org

Review URL: https://codereview.chromium.org/554283003

Cr-Commit-Position: refs/heads/master@{#297006}

BUG=417756
TEST=Bots go green
TBR=phoglund@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/609953002

Cr-Commit-Position: refs/heads/master@{#297005}

BUG=416722
R=ananta@chromium.org

Review URL: https://codereview.chromium.org/599173003

Cr-Commit-Position: refs/heads/master@{#297004}

- The image format used wasn't supported with the memory buffer framework
- The glClear on the init path causes an INVALID_FRAMEBUFFER_OPERATION error
  when executed with no framebuffer, as in the surfaceless case

BUG=380861

Review URL: https://codereview.chromium.org/592933003

Cr-Commit-Position: refs/heads/master@{#297003}

BUG=418125

Review URL: https://codereview.chromium.org/607163002

Cr-Commit-Position: refs/heads/master@{#297002}

The output java files are zipped into a srcjar archive. This
is needed for Java mojom bindings target support in the GN build.

One can still pass the existing --java_output_directory flag to get
the uncompressed java files.

BUG=417289

Review URL: https://codereview.chromium.org/605653003

Cr-Commit-Position: refs/heads/master@{#297001}

adb was moved to being owned by AndroidBrowserBackend in
crrev.com/522553002/.

Review URL: https://codereview.chromium.org/609903002

Cr-Commit-Position: refs/heads/master@{#297000}

BUG=408338

Review URL: https://codereview.chromium.org/607073002

Cr-Commit-Position: refs/heads/master@{#296999}

BUG=417339

Review URL: https://codereview.chromium.org/597313002

Cr-Commit-Position: refs/heads/master@{#296998}

Enable feedback-directed optimisation for the main library (required to
make the FDO symbols present at link time), skia, and sqlite. Update the
android_exports whitelist to allow the gcov symbols to remain public as
otherwise profiling will fail.

BUG=
TBR=michaeln@chromium.org

Review URL: https://codereview.chromium.org/591153002

Cr-Commit-Position: refs/heads/master@{#296997}

Summary of changes available at:
https://chromium.googlesource.com/chromium/blink/+log/6254fdf..a9158da

TBR=ager@chromium.org,zmo@chromium.org

Review URL: https://codereview.chromium.org/603393005

Cr-Commit-Position: refs/heads/master@{#296996}

Review URL: https://codereview.chromium.org/602283002

Cr-Commit-Position: refs/heads/master@{#296995}

R=qyearsley@chromium.org, ojan@chromium.org

Review URL: https://codereview.chromium.org/584853002

Cr-Commit-Position: refs/heads/master@{#296994}

They currently target a revised stub interface for CastSocket, but not a full implementation.
That change is witheld in a separate Git client so this CL can remain at a manageable size.

Thanks!
BUG=396345
R=mfoltz@chromium.org
CC=wez@chromium.org

Review URL: https://codereview.chromium.org/555283002

Cr-Commit-Position: refs/heads/master@{#296993}

Add retry with backoff to invalidate only for non-transient errors.

BUG=416612

Review URL: https://codereview.chromium.org/597003002

Cr-Commit-Position: refs/heads/master@{#296992}

Add nexe to isolate_driver.py.

R=vadimsh@chromium.org
BUG=415188

Review URL: https://codereview.chromium.org/574753004

Cr-Commit-Position: refs/heads/master@{#296991}

R=lcwu@chromium.org
BUG=336640

Review URL: https://codereview.chromium.org/610503002

Cr-Commit-Position: refs/heads/master@{#296990}

Revert of Clear GCM data when the user clears cookies and other site data (patchset #6 id:100001 of https://codereview.chromium.org/562423002/)

Reason for revert:
We're not longer doing this.

Original issue's description:
> Clear GCM data when the user clears cookies and other site data
>
> This is needed for the work to drop GCM sign-in enforcement. Since
> we no longer wipe out the GCM data when the user signs out, we need
> to give user a chance to clear the data if he or she wants.
>
> BUG=384041
> TEST=Manual test by selecting "Cookies and other site and plugin data"
>
> Committed: https://crrev.com/6967652cefa1e306d21e70d4a12dcc0ac572c28b
> Cr-Commit-Position: refs/heads/master@{#294754}

TBR=bauerb@chromium.org,fgorski@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=384041

Review URL: https://codereview.chromium.org/608813003

Cr-Commit-Position: refs/heads/master@{#296989}

BUG=none
TEST=none
R=tengs@chromium.org

Review URL: https://codereview.chromium.org/604213002

Cr-Commit-Position: refs/heads/master@{#296988}

List of added resources:
* PNG illustation (1x and 2x).
* Three styles of Roboto font.
* SVG icon.

optimize-png-files.sh was applied to added PNG files.
Icon was not added to 'chrome_unscaled_resources.grd' because it will be inlined
to HTML.

BUG=375191

Review URL: https://codereview.chromium.org/607473004

Cr-Commit-Position: refs/heads/master@{#296987}

For every successfully distilled article a word count is
submitted to UMA.

The histogram ranges from 1->4000 words with 50 buckets.

Changes rolled in from the DOM Distiller repo:

bbf7c01 Add StatisticsInfo to DomDistillerResult proto for number of words.
fc1a5c1 treat non-breaking space as whitespace
de38c78 Expand usage of SimilarSiblingContentExpansion
444a55e reorder table tests
5ff895b add and fix missing tests to suite
970a419 Add SimilarSiblingContentExpansion
da76b1e add new table classification heuristic

BUG=417049

Review URL: https://codereview.chromium.org/608583003

Cr-Commit-Position: refs/heads/master@{#296986}

This CL changes the return value of
PasswordManagerClient::PromptUserToSavePassword to bool, indicating if
the prompt was indeed displayed, and logs that appropriately through |logger|

BUG=393138

Review URL: https://codereview.chromium.org/604293002

Cr-Commit-Position: refs/heads/master@{#296985}

This is the same issue as in 8d01729a. MSVC 2013 now requires including
<algorithm> to use std::{min, max}.

BUG=none
R=sbc@chromium.org

Review URL: https://codereview.chromium.org/607923002

Cr-Commit-Position: refs/heads/master@{#296984}

BUG=412006

Review URL: https://codereview.chromium.org/599263002

Cr-Commit-Position: refs/heads/master@{#296983}

Review URL: https://codereview.chromium.org/591383002

Cr-Commit-Position: refs/heads/master@{#296982}

This missing dependency could cause intermittent broken builds.

BUG=417583

Review URL: https://codereview.chromium.org/605243002

Cr-Commit-Position: refs/heads/master@{#296981}

BUG=133828

Review URL: https://codereview.chromium.org/491163002

Cr-Commit-Position: refs/heads/master@{#296980}

BUG=418019
TEST=linux bots, content_browsertests
TBR=viettrungluu@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/612483002

Cr-Commit-Position: refs/heads/master@{#296979}

Summary of changes available at:
https://chromium.googlesource.com/chromium/blink/+log/e48d3d4..6254fdf

TBR=ager@chromium.org,zmo@chromium.org

Review URL: https://codereview.chromium.org/607133002

Cr-Commit-Position: refs/heads/master@{#296978}

AwContents are deleted asynchronously, so to stop receiving
callbacks we unregister delegates / observers / listeners
from AwContents.destroy() and ContentViewCore.destroy().
Receiving callbacks after AwContents.destroy() is called is
problematic because Java refs have already been set to null.

BUG=415666

Review URL: https://codereview.chromium.org/594103002

Cr-Commit-Position: refs/heads/master@{#296977}

BUG=None

Review URL: https://codereview.chromium.org/607953002

Cr-Commit-Position: refs/heads/master@{#296976}

Previously, setShape({'rects': []}) or
setShape({'rects': [{left:0, width:0, top:0, height:1}]})
would make the window completely transparent, but still capture input.

This was because ShapedAppWindowTargeter::GetHitTestMask returned
false if the shape was empty. It now returns false only if the shape
is NULL.

BUG=417014

Review URL: https://codereview.chromium.org/602013002

Cr-Commit-Position: refs/heads/master@{#296975}

Currently the DOM Distiller does not work in incognito, since the source
for the chrome-distiller:// scheme has not been registered.

This CL moves the functionality for registering the scheme to a generic place
and it is now used by both ProfileImpl and by OffTheRecordProfile during
profile creation.

It also makes both the normal profile and the incognito profile use the same
instance of the DomDistillerService.

BUG=380036

Review URL: https://codereview.chromium.org/608653002

Cr-Commit-Position: refs/heads/master@{#296974}

BUG=413744

Review URL: https://codereview.chromium.org/611543002

Cr-Commit-Position: refs/heads/master@{#296973}

After we're sure we have fixed 317138, let's drop these CHECKs. Also,
celebrate!

R=ttuttle@chromium.org
BUG=317138

Review URL: https://codereview.chromium.org/509993002

Cr-Commit-Position: refs/heads/master@{#296972}

R=lcwu@chromium.org,byungchul@chromium.org
BUG=None

Review URL: https://codereview.chromium.org/603243003

Cr-Commit-Position: refs/heads/master@{#296971}

Previously a focus ring was only drawn around web elements; this uses a
new API to draw the focus ring around native views, too, by adding a new
field to the existing private accessibility API with the bounding rect of
the focused view.

BUG=314889

Review URL: https://codereview.chromium.org/600063002

Cr-Commit-Position: refs/heads/master@{#296970}

We were passing a 'parent_window' parameter to the network_connect::ConnectToNetwork() in an effort to open network configuration dialogs from the desktop they were triggered from. This was never a good idea since gfx::NativeWindow is actually a pointer which may become invalid before the callback is triggered.

In practice this is unnecessary since we decided that all network configuration applies to the primary user, so we should always use the currently active desktop to host the dialog (which is what GetNativeWindow() does in ash_system_tray_delegate.cc).

BUG=413925

Review URL: https://codereview.chromium.org/607613002

Cr-Commit-Position: refs/heads/master@{#296969}

sync_initialized() does not account for backup/rollback modes. For now we
just manually check the backend mode. In the future the PSS will expose
a better method for denoting when its safe to reconfigure.

BUG=417400

Review URL: https://codereview.chromium.org/605483002

Cr-Commit-Position: refs/heads/master@{#296968}