GitLab

BUG=397889
TBR=pfeldman@chromium.org

Review URL: https://codereview.chromium.org/564763002

Cr-Commit-Position: refs/heads/master@{#294375}

https://chromium.googlesource.com/chromium/blink/+log/c8efcc05b7acf878b1c94ad94d969c08e4d8e9dc..a69fea182204b349e347b767a610d9eb201a191c
TBR=jianli@chromium.org,loislo@chromium.org

Review URL: https://codereview.chromium.org/563583003

Cr-Commit-Position: refs/heads/master@{#294374}

Previously, WebSocket has no handshake timeout, i.e. It waited unlimitedly
when the server accepts the TCP connection but doesn't send any handshake
response. This CL fixes the issue.

BUG=391263
R=ricea@chromium.org

Review URL: https://codereview.chromium.org/565573002

Cr-Commit-Position: refs/heads/master@{#294373}

Previously the ServiceWorkerCacheStorage called ServiceWorkerCache::CreateBackend on Get() and Create().  Now the ServiceWorkerCache does it itself lazily.  This is simpler for the ServiceWorkerCache's caller and should improve performance as backends are created only when needed.  This makes the downstream refcounting CL simpler.

Related CLs:
1. https://crrev.com/542703002: Change ownership of the parameters to ServiceWorkerCache:: Put and Match.
* 2. https://crrev.com/545533002: Move ServiceWorkerCache backend creation to a lazy init function.
3. https://crrev.com/548533002: Make ServiceWorkerCacheStorage::CacheLoader::LoadCache synchronous
4. https://crrev.com/549493002: Expose ServiceWorkerCache objects to ServiceWorkerCacheStorageManager clients.

BUG=392621

Review URL: https://codereview.chromium.org/545533002

Cr-Commit-Position: refs/heads/master@{#294372}

https://chromium.googlesource.com/chromium/blink/+log/4ed3f7079546a9ab0414a43783a5b76a806d1ccd..c8efcc05b7acf878b1c94ad94d969c08e4d8e9dc
TBR=jianli@chromium.org,loislo@chromium.org

Review URL: https://codereview.chromium.org/559393002

Cr-Commit-Position: refs/heads/master@{#294371}

This reverts commit 484d29dc.

This patch made media/media-controller-playbackrate.html test crash on DEBUG.

it crashes at content/browser/web_contents/web_contents_impl.cc:4296 on DCHECK
Looks like AddMediaPlayerEntry was called twice for the same player_cookie

It is not clear how it happens due to async nature of the notification.
The author is sleeping at the moment, so I decided to roll back the patch.

BUG=
TBR=dalecurtis@chromium.org

Review URL: https://codereview.chromium.org/562863002

Cr-Commit-Position: refs/heads/master@{#294370}

BUG=399333
TEST=manually
1. Download .exe file.
2. Click the item in Downloads panel.

Review URL: https://codereview.chromium.org/561133004

Cr-Commit-Position: refs/heads/master@{#294369}

during Telemetry tests consists of sending an "adb force-stop" command
to the device, in order to force kill any previous browsers which did
not close as expected.

Sending a kill -9 to the process is, however, more efficient.
This CL also adds some efficiency improvements to DeviceUtils.KillAll.
The change amounts to a 6-7% speedup when running all of telemetry
unittests.

BUG=379378

Review URL: https://codereview.chromium.org/536343003

Cr-Commit-Position: refs/heads/master@{#294368}

Soon the storage::WatcherManager interface instantiations will want to be an
observer for operations. However, not all backends which have
storage::WatcherManager have storage::QuotaUtil.

To resolve this issue, this CL moves the observer lists one level up to the
storage::FileSystemBackend.

TEST=Just refactoring. Current tests should pass.
BUG=261491

Review URL: https://codereview.chromium.org/549413003

Cr-Commit-Position: refs/heads/master@{#294367}

These functions used to take unowned raw pointers as arguments.  Now the functions own them and they're scoped_ptrs.  This both makes ownership clearer and allows for the functions to terminate early without calling their callbacks (needed for a downstream change to run CreateBackend lazily).

Related CLs:
* 1. https://crrev.com/542703002: Change ownership of the parameters to ServiceWorkerCache:: Put and Match.
2. https://crrev.com/545533002: Move ServiceWorkerCache backend creation to a lazy init function.
3. https://crrev.com/548533002: Make ServiceWorkerCacheStorage::CacheLoader::LoadCache synchronous
4. https://crrev.com/549493002: Expose ServiceWorkerCache objects to ServiceWorkerCacheStorageManager clients.

BUG=392621

Review URL: https://codereview.chromium.org/542703002

Cr-Commit-Position: refs/heads/master@{#294366}

RunLoop::Run() is needed if any of the operation is implemented using
asynchronous I/O in OS layer, or performed in a separate thread.
Currently, only Copy needs RunLoop::Run().

This CL replace all RunUntilIdle() with Run() to keep all test consistent.

BUG=411153
TEST=content_unittest

Review URL: https://codereview.chromium.org/557273002

Cr-Commit-Position: refs/heads/master@{#294365}

Revert of Use changed_button_flags() when converting MouseEvent to blink::WebMouseEvent. (patchset #2 id:20001 of https://codereview.chromium.org/302593002/)

Reason for revert:
Aura builds fail DCHECKs:

ASSERTION FAILED: event.type() == PlatformEvent::MouseMoved || event.button() != NoButton
third_party/WebKit/Source/core/events/MouseEvent.cpp(54)

blink::MouseEvent::create()
blink::Node::dispatchMouseEvent()
blink::EventHandler::dispatchMouseEvent()
blink::EventHandler::handleMousePressEvent()
blink::PageWidgetEventHandler::handleMouseDown()
blink::WebViewImpl::handleMouseDown()
blink::WebViewImpl::handleMouseDown()
blink::PageWidgetDelegate::handleInputEvent()
blink::WebViewImpl::handleInputEvent()
content::RenderWidget::OnHandleInputEvent()

Original issue's description:
> Use changed_button_flags() when converting MouseEvent to blink::WebMouseEvent.
>
> Event::flags() contains all the buttons that are being held down during the event, while MouseEvent::changed_button_flags() contains the buttons that are actually pressed/released during the event (and this is what we really are interested in).
>
> This fixes the issue with LMB being pressed while RMB is being held down. Without the fix, we always get blink::WebMouseEvent::ButtonRight in blink::WebMouseEvent::button.
>
> Committed: https://crrev.com/87e6b0f6d72b12e4b0308b035d1c06f2beb90feb
> Cr-Commit-Position: refs/heads/master@{#293487}

TBR=joleksy@opera.com
NOTREECHECKS=true
NOTRY=true

Review URL: https://codereview.chromium.org/563713002

Cr-Commit-Position: refs/heads/master@{#294364}

*src/athena
Change ShellAppActivity to accept both AppWindow and ShellAppWindow.
Implement AppsClient to create NativeAppWindow and AppDelegate instances.

*src/extensions/shell
Add a new method DefaultShellBrowserMainDelegate::CreateAppsClient() to inject athena's AppsClient.

*src/extensions/browser
Add missing dtor to extensions::AppsClient to destroy an instance with a base class pointer in DefaultShellBrowserMainDelegate.

BUG=387288
TBR=benwells@chromium.org for the addition of the missing dtor of extensions::AppsClient.

Review URL: https://codereview.chromium.org/552133003

Cr-Commit-Position: refs/heads/master@{#294363}

https://chromium.googlesource.com/chromium/blink/+log/a54014ac471a9d5980ca18a582f1f4a1a767efa4..4ed3f7079546a9ab0414a43783a5b76a806d1ccd
TBR=jianli@chromium.org,loislo@chromium.org

Review URL: https://codereview.chromium.org/566513002

Cr-Commit-Position: refs/heads/master@{#294362}

- Remove the global unstub_client_mask_ and accordingly make DBusThreadManager::IsUsingStub a non-static function. This makes it obvious when it's ok to call this function (and prevents misuse), namely only after the Manager is initialized.

- By moving the mask to the DBusClientBundle, the cyclic dependency between bundle and manager is broken. The Bundle does not know about the Manager anymore.

- Fix the difference between DBusClientTypeMask ("A set of DBusClients") and DBusClientType ("A single DBusClient").

BUG=408617

Review URL: https://codereview.chromium.org/556833003

Cr-Commit-Position: refs/heads/master@{#294361}

Prior to this change we only checked the polymorphic requirements on classes
that declared a trace method. Also, this fixes a typo in condition (2) where we
checked on the derived class instead of the left-most class.

R=ager@chromium.org
BUG=

Review URL: https://codereview.chromium.org/564453002

Cr-Commit-Position: refs/heads/master@{#294360}

AwURLRequestContextGetter now owns an instance of net::NetLog.

BUG=277075
R=boliu@chromium.org

Review URL: https://codereview.chromium.org/557733002

Cr-Commit-Position: refs/heads/master@{#294359}

Followup to CL 469993007, which added a c++11 style enum scope,
but we haven't officially switched to c++11 yet.

BUG=360088

Review URL: https://codereview.chromium.org/562513005

Cr-Commit-Position: refs/heads/master@{#294358}

This patch resolves very old todos about missing descriptions of arguments
of methods in the metadata directory.

TEST=No change. Comments only.
BUG=413100

Review URL: https://codereview.chromium.org/562763006

Cr-Commit-Position: refs/heads/master@{#294357}

BUG=402684

Review URL: https://codereview.chromium.org/546803002

Cr-Commit-Position: refs/heads/master@{#294356}

TBR=machenbach@chromium.org

Review URL: https://codereview.chromium.org/562693004

Cr-Commit-Position: refs/heads/master@{#294355}

R=sdefresne@chromium.org,cmasone@chromium.org

Review URL: https://codereview.chromium.org/548343005

Cr-Commit-Position: refs/heads/master@{#294354}

Currently there is not enough unit test cases available to cover
the functionality of Paste Popup menu behavior. This patch covers
essential unit test cases for the same.

These tests originally landed in r293222, but were incidentally reverted
due to some pre-existing test failures in the same file.

BUG=

Review URL: https://codereview.chromium.org/560883002

Cr-Commit-Position: refs/heads/master@{#294353}

https://skia.googlesource.com/skia/+log/97bf60ff7f3dd7d6e782b706ec94efb2b414b27e..26a4b514999dcbd8c6a4a2e85bcc5dbd55f02236

CQ_EXTRA_TRYBOTS=tryserver.blink:linux_blink_rel,linux_blink_dbg
TBR=fmalita@google.com

Review URL: https://codereview.chromium.org/561133003

Cr-Commit-Position: refs/heads/master@{#294352}

The whole purpose of FileManagerPrivateAPI class was to hold an
EventRouter instance. It's simpler to just make EventRouter a
KeyedService and get rid of FileManagerPrivateAPI.

Note that BrowserContextKeyedAPI could not be used for EventRouter as
EventRouter needs a Profile rather than a BrowserContext.

In the next patch, EventRouter will be renamed to something like
FileManagerService.

BUG=412643
TEST=Files.app works as before

Review URL: https://codereview.chromium.org/553393002

Cr-Commit-Position: refs/heads/master@{#294351}

Before, the specified playback position was ignored so this patch fixes the bug.

BUG=395868

Review URL: https://codereview.chromium.org/556493002

Cr-Commit-Position: refs/heads/master@{#294350}

When onDraw is called, everytime RectF is created which allocates new object. To compute the rect, it needs to be created only once and re-used every time when onDraw() is called altering its values, else too many RectF objects could affect performance.

This issue is found from below lint warning:
../../../tmp/tmppzTHqO/0/PopupZoomer.java:466 Avoid object allocations during draw/layout operations (preallocate and reuse instead): DrawAllocation [warning]
            RectF rect = new RectF();

Bug=None.

Review URL: https://codereview.chromium.org/549163002

Cr-Commit-Position: refs/heads/master@{#294349}

Blink GC plugin: Add plugin flags for generating raw-pointer errors and unneeded-finalizer warnings.

R=haraken@chromium.org
BUG=334149
NOTRY=true

Review URL: https://codereview.chromium.org/551343002

Cr-Commit-Position: refs/heads/master@{#294348}

Previously, the maximize button would be hidden if the window is
created with a maxWidth or maxHeight, or non-resizable. Once the size
constraint is removed, the maximize button remains hidden. Similarly,
a size constraint that is applied after the window is created would
not hide the maximize button.

This adds Widget::OnSizeConstraintsChanged which causes the native
widget to update the window's properties, and the non client view to
update the visibility of the maximize button.

BUG=351181,412241

Review URL: https://codereview.chromium.org/554183002

Cr-Commit-Position: refs/heads/master@{#294347}

* NotificationRegister and NotificationObserver is need to collect
  file information.

* This CL is related to https://codereview.chromium.org/530553006

BUG=412691

Review URL: https://codereview.chromium.org/566433004

Cr-Commit-Position: refs/heads/master@{#294346}

DevTools code needs access to platform thread id in order to collect Worker thread ids and based on them differentiate worker threads in recorded trace events.

Blink part of the change: https://codereview.chromium.org/514193002/

BUG=401895

Review URL: https://codereview.chromium.org/515243002

Cr-Commit-Position: refs/heads/master@{#294345}

https://chromium.googlesource.com/chromium/blink/+log/ac6fda4a504cc8837dcd763d22aceb9d7935980f..a54014ac471a9d5980ca18a582f1f4a1a767efa4
TBR=keishi@chromium.org,jianli@chromium.org

Review URL: https://codereview.chromium.org/567443002

Cr-Commit-Position: refs/heads/master@{#294344}

Cr-Commit-Position: refs/heads/master@{#294343}

Review URL: https://codereview.chromium.org/549273002

Cr-Commit-Position: refs/heads/master@{#294342}

This is a very minor change to the motion of pages (fast start and slow
end), but brings the launcher in line with the material design spec.

BUG=406222

Review URL: https://codereview.chromium.org/543403004

Cr-Commit-Position: refs/heads/master@{#294341}

Review URL: https://codereview.chromium.org/554563002

Cr-Commit-Position: refs/heads/master@{#294340}

Visulaizer ==> Visualizer

R=dmikurube@chromium.org
BUG=405447

Review URL: https://codereview.chromium.org/494593002

Cr-Commit-Position: refs/heads/master@{#294339}

Currently 'Bookmark all tabs' and 'Bookmark Edit' window have different minimum and default sizes. Make them consistent.

XIB Changes:
BookmarkAllTabs & Bookmark Editor window's minimum and default sizes are set in .xib file located at
chrome/app/nibs/BookmarkAllTabs.xib
chrome/app/nibs/BookmarkEditor.xib

Their window's default and minimum width and height values are decided by NSWindowRect & NSWindowContentMinSize key strings repectively. Set the correct values such that BookmarkAllTabs & Bookmark Editor windows are consistent

BUG=401491
R=asvitkine@chromium.org

TEST=
1) Launch chrome, open 2-3 tabs, right click on last tab and select 'Bookmark all tabs'.
2) On opened window increase the size by dragging towards R.H.S, and then try to decrease the size by dragging towards L.H.S and Observe.
3) The bookmark window should be minimize upto default size as of "Bookmark Edit" window.

Review URL: https://codereview.chromium.org/537553002

Cr-Commit-Position: refs/heads/master@{#294338}

closure_linter: temporarily add support for @suppress {suspiciousCode} to unblock compiling settings.

R=vitalyp@chromium.org
TBR=tbreisacher@chromium.org
BUG=393873
NOTRY=true

Review URL: https://codereview.chromium.org/561873004

Cr-Commit-Position: refs/heads/master@{#294337}

XIB Changes:
Font style for 'New folder' title in BookmarksAllTabs and BookmarkEditor window is set in .xib files placed at following locations.
chrome/app/nibs/BookmarkAllTabs.xib
chrome/app/nibs/BookmarkEditor.xib
Removing the 'fontBold' key string for the 'newFolder' object.

BUG=397086

R=asvitkine@chromium.org

TEST=
1) Launch chrome and bookmark any page.
2) Now in the Bookmarks bar right click on the bookmarked page and click on edit.
3) In the edit box click on 'New folder' and observe.
4) The 'New folder' name font style should not be bold.

Review URL: https://codereview.chromium.org/535083002

Cr-Commit-Position: refs/heads/master@{#294336}