GitLab

The @download attribute can be used to specify a suggested filename for
a download. However, this suggested name should only be used if the
referring document is in the same origin as the downloaded resource.
This CL clears the suggested name if the final download URL after
redirects is in a different origin than the original download URL.

The companion Blink CL at https://codereview.chromium.org/197033005/
ensures that the suggested name for a download initiated via a @download
attribute would only be non-empty if the referring document is allowed
to set a suggested name for the original download URL.

BUG=346744

Review URL: https://codereview.chromium.org/246893006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266441 0039d316-1c4b-4281-b951-d872f2087c98

This is in preparation of landing the unified GR
(see https://codereview.chromium.org/251543003/).

There are some cases where behavior differs in minor ways.

This is sometimes dealt with by increasing the test's margin of error
(for instance, slop handling is slightly different, so we now
exceed the slop more thoroughly).

This is sometimes dealt with by ignoring the test if the test is
run with the unified GR.

BUG=332418

Review URL: https://codereview.chromium.org/258633005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266440 0039d316-1c4b-4281-b951-d872f2087c98

TEST=TouchDispositionGestureFilterTest.TestDisallowedMultiFingerSwipe

Review URL: https://codereview.chromium.org/252753002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266439 0039d316-1c4b-4281-b951-d872f2087c98

password_form_conversion_utils.

There are Some old Blink code implemented in Webkit that predates the
Chromium/WebKit API, specifically WebPasswordFormData and WebPasswordFormUtils. At
this point the split just makes this code harder to update, as you need to wait
for a Blink roll when adding a new element to autofill::PasswordForm. This patch is
re-writing this code in components/autofill/content/renderer/password_form_conversion_util
s to make it cleaner.

BUG=None

Review URL: https://codereview.chromium.org/249153002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266438 0039d316-1c4b-4281-b951-d872f2087c98

This is a reland of 260688, which is reverted due to false charge of build failure.

Previously, we used FileEntry.getMetadata() to retrieve filesystem metadata and getDriveEntryProperties() to get Drive metadata, so we need 2 calls for 1 file on Drive.

With this patch, getDriveEntryProperties() returns not only Drive metadata but also filesystem metadata. It's enough to call only getDriveEntryProperties() and we can reduce a number of calls by half.

BUG=345196
TEST=browser_test passes.
R=asargent@chromium.org, hashimoto@chromium.org, hirono@chromium.org
TBR=asargent@chromium.org, hashimoto@chromium.org, hirono@chromium.org

Review URL: https://codereview.chromium.org/195763004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266437 0039d316-1c4b-4281-b951-d872f2087c98

We need explicitly call ProfileStartup, as it is done for usual flows from  startup_browser_creator.cc.
This will correctly initialize per-profile input method map in InputMethodManagerImpl.

R=nkostylev@chromium.org
BUG=364680

Review URL: https://codereview.chromium.org/253463002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266436 0039d316-1c4b-4281-b951-d872f2087c98

Make it possible to tab-navigate to import page and ensure tab/arrows-navigation works there.

BUG=360588

Review URL: https://codereview.chromium.org/246603004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266435 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: https://codereview.chromium.org/255983002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266434 0039d316-1c4b-4281-b951-d872f2087c98

BUG=

Review URL: https://codereview.chromium.org/257743002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266433 0039d316-1c4b-4281-b951-d872f2087c98

BUG=

Review URL: https://codereview.chromium.org/256783004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266432 0039d316-1c4b-4281-b951-d872f2087c98

If the time has not been set automatically via network syncing, the system
clock can be changed. This dialog allows the user to change the date, time
and time zone when possible. The motivation is to unbrick devices with
the wrong time that need to be rolled out into time-sensitive networks.

Documentation of the changes to Chrome OS and Ash for this CL is here:
https://docs.google.com/a/google.com/drawings/d/1T3demthtROnXf1iE31p5aIPcQzK9SmjfE9EjnJbZ4zs

System Time Manual Update UI design doc:
https://docs.google.com/a/google.com/document/d/1djzhBrtbx-52Gctp3Fd5MIosARbTwQh_lMmd_qUnqgo

Screenshot with timezone:
https://drive.google.com/a/google.com/file/d/0B6HSBrih6pNUd3p2SFBoVktjVzQ

Screenshot from settings page, no timezone:
https://drive.google.com/a/google.com/file/d/0B6HSBrih6pNUXzk0TjNiT0tKMTQ

BUG=232066
TEST=SetTimeWebUITest, DateTimeOptionsWebUITest
R=stevenjb@chromium.org,nkostylev@chromium.org,dbeam@chromium.org,derat@chromium.org,asvitkine@chromium.org
TBR=sky@chromium.org
# TBR for adding resources to chrome/browser/chrome_resources.grd

Please review:
stevenjb@chromium.org
- ash/system
- chromeos/dbus

derat@chromium.org:
- chrome/browser/chromeos
- chrome/browser/ui/ash
- chromeos/dbus (optional)

nkostylev@chromium.org:
- chrome/app
- chrome/browser/chromeos
- chrome/browser/resources/chromeos
- chrome/browser/ui/webui/chrome_web_ui_controller_factory.cc
- chrome/browser/ui/webui/chromeos
- chrome/browser/ui/webui/options/chromeos
- chrome/browser/browser_resources.grd
- chrome/chrome_*.gypi
- chrome/common

dbeam@chromium.org:
- chrome/browser/resources/options
- chrome/browser/ui/webui/options/*.cc

sky@chromium.org:
- chrome/browser/browser_resources.grd

Review URL: https://codereview.chromium.org/247663003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266431 0039d316-1c4b-4281-b951-d872f2087c98

R=tbreisacher@chromium.org
BUG=366972
NOTRY=true

Review URL: https://codereview.chromium.org/251713002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266430 0039d316-1c4b-4281-b951-d872f2087c98

If executeHardwareAction fails in onDetachedFromWindow, then
HardwareRenderer destructor will deadlock later when being destroyed on
UI thread. In this case, run the destructor on UI without deadlock but
leak GL resources instead.

Achieve this by running always running release gl even on UI thread, and
making the allow_gl global to be a true thread local, which can be set
to true on UI thread.

BUG=
TBR=benm

Review URL: https://codereview.chromium.org/255783007

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266429 0039d316-1c4b-4281-b951-d872f2087c98

BUG=None
TEST=None
TBR=yoshiki@chromium.org

Review URL: https://codereview.chromium.org/251933002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266426 0039d316-1c4b-4281-b951-d872f2087c98

NOTRY=true
NOTREECHECKS=true
BUG=367600
TBR=bartfab@chromium.org, jianli@chromium.org

Review URL: https://codereview.chromium.org/251093002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266423 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266421 0039d316-1c4b-4281-b951-d872f2087c98

Revert of Set kCanMaximizeKey/kCanResizeKey before adding to parent. (https://codereview.chromium.org/251733004/)

Reason for revert:
NativeWidgetAuraTest.TestPropertiesWhenAddedToLayout triggering lsan errors
link: http://build.chromium.org/p/chromium.memory/builders/Linux%20Chromium%20OS%20ASan%20LSan%20Tests%20%281%29/builds/1042

errors like 
Direct leak of 24 byte(s) in 1 object(s) allocated from:
..
    #1 0x97042a in views::(anonymous namespace)::NativeWidgetAuraTest_TestPropertiesWhenAddedToLayout_Test::TestBody() ui/views/widget/native_widget_aura_unittest.cc:238


Original issue's description:
> Set kCanMaximizeKey/kCanResizeKey before adding to parent.
> 
> BUG=None
> TEST=NativeWidgetAuraTest.TestPropertiesWhenAddedToLayout
> R=ben@chromium.org
> 
> Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=266386

TBR=ben@chromium.org,oshima@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=None

Review URL: https://codereview.chromium.org/257823004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266419 0039d316-1c4b-4281-b951-d872f2087c98

Revert of Stop passing chrome.test.fail as ErrorCallback (https://codereview.chromium.org/258523004/)

Reason for revert:
Suspected for making OpenAudioFiles/FileManagerBrowserTest.Test/{0,1} flaky/racy since http://build.chromium.org/p/chromium.chromiumos/builders/Linux%20ChromiumOS%20Tests%20%28dbg%29%282%29/builds/10706

Errors like: 

[26400:26400:0426/004939:INFO:CONSOLE(0)] "[FAIL] undefined: API Test Error in undefined: Displayed data of 1st file is wrong.
Actual: Beautiful Song.ogg
Expected: Beautiful Song
...
    at chrome-extension://oobinhbdbiehknkpbpejbbpdbkdjmoco/open_audio_files.js:90:21", source: chrome-extension://oobinhbdbiehknkpbpejbbpdbkdjmoco/_generated_background_page.html (0)
[26400:26400:0426/004939:INFO:CONSOLE(0)] "[FAIL] undefined: chrome.test.failure

Original issue's description:
> Stop passing chrome.test.fail as ErrorCallback
> 
> chrome.test.fail cannot be used as ErrorCallback because it throws exception when the number of arguments doesn't match.
> This wrong use of chrome.test.fail makes the error output useless when something goes wrong.
> 
> BUG=None
> TEST=browser_tests
> 
> Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=266271

TBR=hirono@chromium.org,hashimoto@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=None

Review URL: https://codereview.chromium.org/254943002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266416 0039d316-1c4b-4281-b951-d872f2087c98

It is failing and I suspect the problem is in the unittest not in the code
under test, so I'm disabling rather than reverting the other change.

File "E:\b\build\slave\XP_Tests__2_\build\src\tools\telemetry\telemetry\page\page_runner_unittest.py", line 240, in testPagesetRepeat
os.remove(output_file)
WindowsError: [Error 32] The process cannot access the file because it is being used by another process: 'c:\\docume~1\\chrome~2\\locals~1\\temp\\tmpiwpnmk'

TBR=dalecurtis@chromium.org
NOTRY=True
NOTREECHECKS=True
BUG=

Review URL: https://codereview.chromium.org/257983003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266411 0039d316-1c4b-4281-b951-d872f2087c98

BUG=254066

Review URL: https://codereview.chromium.org/255903002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266403 0039d316-1c4b-4281-b951-d872f2087c98

CC touch hit testing now (as of http://crrev.com/264395 does a ray cast
through all layers under the touched point.  We included an optimization
to stop searching once you reach a layer we know is opaque to hit testing.
This breaks the common case of sites with a touch handler on the document
because blink has an optimization to effectively disable touch hit testing
in that scenario (since there's no benefit to be had to expensive rect
computation) by marking just the entire root layer with a handler region.

This change removes this opaque-to-hit-testing optimization.  There's some
risk that we'll block on main more often, but it should be rare (you'd
need a scrolling layer without a handler over top of a layer with a
handler).

This also fixes the case where a layer has a region that is supposed to
be transparent to hit-testing with the CSS pointer-events: none property.

BUG=366034, 269598

Review URL: https://codereview.chromium.org/250663003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266402 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: https://codereview.chromium.org/254803007

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266401 0039d316-1c4b-4281-b951-d872f2087c98

BUG=110814

Review URL: https://codereview.chromium.org/251833004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266400 0039d316-1c4b-4281-b951-d872f2087c98

This functionality does not exist on aura or android anymore, and is
going away soon on Mac. RWHViews are always composited now on all
platforms, so the active tab is always able to be copied in the
browser process with CopyFromBackingStore().

This removes the ViewMsg_Snapshot and ViewHostMsg_Snapshot IPCs.

R=piman, thestig
BUG=251966, 251960

Review URL: https://codereview.chromium.org/259523008

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266399 0039d316-1c4b-4281-b951-d872f2087c98

In ForeignSessionHelper class, there were two places
we allocate ScopedJavaLocalRef<jstring> inside a loop and
doesn't delete. The Java local references are cleaned upon
returning to Java, but there is 512 local reference hard-limit 
so we should delete it every loop to avoid hitting the limit.

BUG=366659

Review URL: https://codereview.chromium.org/257513007

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266398 0039d316-1c4b-4281-b951-d872f2087c98

BUG=365822
R=oshima@chromium.org, ben@chromium.org
TEST=manually

Review URL: https://codereview.chromium.org/246073009

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266397 0039d316-1c4b-4281-b951-d872f2087c98

We cannot guarantee that urls will not be redirected when android mediaplayer requests video streams.
Always treat media urls as cross-origin for now.

BUG=334204

Review URL: https://codereview.chromium.org/247573004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266396 0039d316-1c4b-4281-b951-d872f2087c98

TBR=jochen@chromium.org for chrome/browser/chrome_notification_types.h (*removing* a notification type).
BUG=358710

Review URL: https://codereview.chromium.org/252593003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266395 0039d316-1c4b-4281-b951-d872f2087c98

On the chromium.perf waterfall, we have some devices that are not going offline
(N4 + N5) and some that are (N7, N10, GN). Diffing the results of the
provision_device step pointed out some configuration differences between the
working and non-working devices. Most notably, wifi_on was set on all the
devices that are going offline and not on the ones that are staying on.

I hope making these settings deterministic will get these devices into better
shape.

BUG=366697

Review URL: https://codereview.chromium.org/251793003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266394 0039d316-1c4b-4281-b951-d872f2087c98

Also update channel to stable.

BUG=348085

Review URL: https://codereview.chromium.org/184483005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266393 0039d316-1c4b-4281-b951-d872f2087c98

BUG=333078

Review URL: https://codereview.chromium.org/247283006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266392 0039d316-1c4b-4281-b951-d872f2087c98

BUG=362271
TEST=AutofillDialogControllerTest
NOTRY=True

Review URL: https://codereview.chromium.org/258543005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266391 0039d316-1c4b-4281-b951-d872f2087c98

This patch makes the anchor text no bold in 'Port forwarding' dialog.

BUG=366576

Review URL: https://codereview.chromium.org/250433003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266390 0039d316-1c4b-4281-b951-d872f2087c98

I add the followings to histograms.xml.
  SharedWorker.RendererSurviveForWorkerTime
  SharedWorker.TimeToDeleted
  SharedWorker.TimeToScriptLoaded
  SharedWorker.TimeToScriptLoadFailed

They were introduced in http://crrev.com/259790 and http://crrev.com/260376.

BUG=327256

Review URL: https://codereview.chromium.org/245963004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266389 0039d316-1c4b-4281-b951-d872f2087c98

BUG=359672
TEST=WindowSelectorTest.DisplayOrientationChanged
TEST=While overview is active, rotate or resize the display. Windows should reposition / relayout in response.

Review URL: https://codereview.chromium.org/256413004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266388 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: https://codereview.chromium.org/250833003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266387 0039d316-1c4b-4281-b951-d872f2087c98

BUG=None
TEST=NativeWidgetAuraTest.TestPropertiesWhenAddedToLayout
R=ben@chromium.org

Review URL: https://codereview.chromium.org/251733004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266386 0039d316-1c4b-4281-b951-d872f2087c98

Failing to do so confuses the omnibox_edit_model. A visible bug appears when using ctrl+enter after having switched tab, which can cause the text of the previous tab to be used.

BUG=338448

Review URL: https://codereview.chromium.org/232023004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266385 0039d316-1c4b-4281-b951-d872f2087c98

BUG=

Review URL: https://codereview.chromium.org/255833002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266384 0039d316-1c4b-4281-b951-d872f2087c98

Software delegated rendering uses one fd per tile, so with a bunch of tiles that means it can run out of fds. If it seems close to hitting the limit the browser should throw away old frames to avoid this from happening.

BUG=362603

Review URL: https://codereview.chromium.org/248193003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266383 0039d316-1c4b-4281-b951-d872f2087c98