GitLab

(original change: https://codereview.chromium.org/2042763002)

In layout tests when preferences were being overridden by specific tests they
were not being properly applied in the browser. The WebPreferences cached by the
RenderViewHost was not updated and so they would not effect the existing window.
This fixes that and changes the way spatial navigation is enabled in layout
tests to adapt to this change.

The problem the original change that caused it to be reverted was that it made
preferences leak between tests when the Shell window was reused. This version
fixes that by storing a clean copy of the test-ready WebPrecerences created with
a brand new RenderView and using that to reset existing windows' preferences.

This is an spin off change to support the moving of mixed content checks to
the browser, which requires the browser to be immediately effected by
preferences updates.

BUG=576270, 620126

Review-Url: https://codereview.chromium.org/2134723002
Cr-Commit-Position: refs/heads/master@{#404627}

Before, prefetch experiment's CSV collection was maintaining a dict
of transfered bytes per URLs as it was iterating over all resource
requests and served from memory cache URLs request would just grabe
the size from this dictionary. But we may have requests served from
memory cache before original request that filled the memory cache,
leading to a KeyError.

This CL fixes this issue by first building the dictionary by going
over resources requests, and then actually sum the bytes using this
dictionary, avoiding the KeyError.

BUG=582080

Review-Url: https://codereview.chromium.org/2134763002
Cr-Commit-Position: refs/heads/master@{#404626}

BUG=None

Review-Url: https://codereview.chromium.org/2133153002
Cr-Commit-Position: refs/heads/master@{#404625}

BUG=627009
BUG=b/30054874

Review-Url: https://codereview.chromium.org/2136933002
Cr-Commit-Position: refs/heads/master@{#404624}

Now that PageLifecycleObserver has only pageVisibilityChanged(),
it would make more sense to rename the class accordingly.

BUG=610176

Review-Url: https://codereview.chromium.org/2138663003
Cr-Commit-Position: refs/heads/master@{#404623}

Revert of Add effect node support in PaintArtifactCompositor for layer list mode [4/4] (patchset #7 id:120001 of https://codereview.chromium.org/2052763002/ )

Reason for revert:
This CL seems making webkit_unit_tests' failures.

https://findit-for-me.appspot.com/waterfall/build-failure?url=https://build.chromium.org/p/chromium.webkit/builders/WebKit%20Linux%20(dbg)/builds/8344

https://build.chromium.org/p/chromium.webkit/builders/WebKit%20Linux%20(dbg)/builds/8334

Original issue's description:
> Add effect node support in PaintArtifactCompositor for layer list mode [4/4]
>
> This CL adds a local EffectStack class similar to ClipManager, that keep track
> of effect state and assign generated cc effect node to layers. This is only
> used for layer list mode.
>
> BUG=609937
>
> Committed: https://crrev.com/889b2a8057be02fefc2fdd9e5deb289844f94040
> Cr-Commit-Position: refs/heads/master@{#404575}

TBR=jbroman@chromium.org,chrishtr@chromium.org,trchen@chromium.org
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG=609937

Review-Url: https://codereview.chromium.org/2138663002
Cr-Commit-Position: refs/heads/master@{#404622}

All .strings files that need to be in binary1 format are now
converted by using the "bundle_data_strings" template so the
conversion can be remove from "copy_bundle_data" tool.

BUG=625578

Review-Url: https://codereview.chromium.org/2130813002
Cr-Commit-Position: refs/heads/master@{#404621}

Giving more descriptive name.

BUG=624711
R=mkwst,rdsmith,clamy

Review-Url: https://codereview.chromium.org/2125523003
Cr-Commit-Position: refs/heads/master@{#404620}

BUG=620630

Review-Url: https://codereview.chromium.org/1706193002
Cr-Commit-Position: refs/heads/master@{#404619}

BUG=627008
TBR=nzolghadr

Review-Url: https://codereview.chromium.org/2137923002
Cr-Commit-Position: refs/heads/master@{#404618}

Set --enable-files-quick-view to enable quick view.

BUG=613465
TEST=manually using link.

Review-Url: https://codereview.chromium.org/2127853002
Cr-Commit-Position: refs/heads/master@{#404617}

BUG=617585

Review-Url: https://codereview.chromium.org/2130363002
Cr-Commit-Position: refs/heads/master@{#404616}

No behavioral change.

- Avoid duplicate member init in HTMLConstructionSite c-tor.
- Use Document& instead of Document* as the document can't be nullref
- Clarify that HTMLTreeBuilder can be invoked for generic Document (with a fixme comment)

BUG=None

Review-Url: https://codereview.chromium.org/2126043003
Cr-Commit-Position: refs/heads/master@{#404615}

This file contains an interface that is about to be removed
(https://codereview.webrtc.org/2034013003) since it is no longer used
by chromium (https://codereview.chromium.org/2033353002).

The build files have to be updated before the file is removed from the
webrtc repo or else the webrtc roll will cause chromium to break.

BUG=webrtc:5708, webrtc:5707
NOTRY=true

Review-Url: https://codereview.chromium.org/2120493002
Cr-Commit-Position: refs/heads/master@{#404614}

The previous src tag collides with the one used by desktop NTP (in the
remote NTP).

BUG=626235

Review-Url: https://codereview.chromium.org/2125133004
Cr-Commit-Position: refs/heads/master@{#404613}

Follow-up to https://chromium.googlesource.com/chromium/src/+/3854a2102cfd8a10f28cf5123f593b87aaac3af0

BUG=None

Review-Url: https://codereview.chromium.org/2134043002
Cr-Commit-Position: refs/heads/master@{#404612}

WebURLResponse can directly point to a ResourceResponse object instead.

BUG=625529

Review-Url: https://codereview.chromium.org/2133693002
Cr-Commit-Position: refs/heads/master@{#404611}

https://chromium.googlesource.com/external/github.com/catapult-project/catapult.git/+log/6cad5da3c39e..4160831d2082

$ git log 6cad5da3c..4160831d2 --date=short --no-merges --format='%ad %ae %s'

TBR=catapult-sheriff@chromium.org

Review-Url: https://codereview.chromium.org/2135893002
Cr-Commit-Position: refs/heads/master@{#404610}

There are two cases when cache is flushed:

1) MEMORY_PRESSURE_LEVEL_MODERATE: flush expired sessions only.
2) MEMORY_PRESSURE_LEVEL_CRITICAL: flush everything.

BUG=622069

Review-Url: https://codereview.chromium.org/2131883002
Cr-Commit-Position: refs/heads/master@{#404609}

It's now unused.

BUG=610176

Review-Url: https://codereview.chromium.org/2134033002
Cr-Commit-Position: refs/heads/master@{#404608}

Review-Url: https://codereview.chromium.org/2129083002
Cr-Commit-Position: refs/heads/master@{#404607}

URLRequest::Delegate::OnBeforeNetworkStart is not used by
anything, despite having a lot of plumping.  It was added to
try experimenting with having the ResourceScheduler delay
requests only once they needed to go over the network, so
responses that could be served out of the cache.

BUG=479994

Review-Url: https://codereview.chromium.org/2123093002
Cr-Commit-Position: refs/heads/master@{#404606}

This symbol no longer appears in the codebase.

BUG=27837

Review-Url: https://codereview.chromium.org/2134023002
Cr-Commit-Position: refs/heads/master@{#404605}

The code is simpler without this extra method.

Review-Url: https://codereview.chromium.org/1798743002
Cr-Commit-Position: refs/heads/master@{#404604}

Using optinList() instead of listItems() simplifies the code.
This CL should not have any behavior changes.

Review-Url: https://codereview.chromium.org/2141483002
Cr-Commit-Position: refs/heads/master@{#404603}

BUG=

Review-Url: https://codereview.chromium.org/2136913002
Cr-Commit-Position: refs/heads/master@{#404602}

It was blocked on https://crbug.com/274855 which is now fixed.

BUG=387045

Review-Url: https://codereview.chromium.org/2136833003
Cr-Commit-Position: refs/heads/master@{#404601}

The symbol no longer appears in the codebase.

BUG=403544

Review-Url: https://codereview.chromium.org/2140483002
Cr-Commit-Position: refs/heads/master@{#404600}

Removes crash keys and temporary instrumentations to investigate
the issue crbug.com/621730 .  The crash keys and instrumentations
were introduced at
    https://crrev.com/2107773002
    https://crrev.com/2101283002

BUG=621730

Review-Url: https://codereview.chromium.org/2133743002
Cr-Commit-Position: refs/heads/master@{#404599}

This CL introduces Document::unthrottledTaskRunner() and move one task in
HTMLCanvasElement to the unthrottled task runner. We'll move more tasks to
the unthrottled task runner.

BUG=624696

Review-Url: https://codereview.chromium.org/2133623002
Cr-Commit-Position: refs/heads/master@{#404598}

https://chromium.googlesource.com/skia.git/+log/3ab14ca950d9..6af677ff8f54

$ git log 3ab14ca95..6af677ff8 --date=short --no-merges --format='%ad %ae %s'
2016-07-10 dvonbeck Changed SkLightingShader API to take in a shader as color source

CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_blink_rel
TBR=msarett@google.com

Review-Url: https://codereview.chromium.org/2141473002
Cr-Commit-Position: refs/heads/master@{#404597}

To show the up-to-date storage information, this CL modifies storage manager
to update the information every 5 seconds while the manager overlay is visible.
Also, avoiding updating storage size while computing size is ongoing.

BUG=625071
TEST=manually tested
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2137463002
Cr-Commit-Position: refs/heads/master@{#404596}

Before this CL, Document::baseElementURL was used to initialize
PreloadScanner. However, this URL can be !isValid().

This CL introduces Document::validBaseElementURL() which returns
baseElementURL only when its valid, end returns empty URL when its invalid.
This can safely be used to initialize preload scanner.

BUG=623365

Review-Url: https://codereview.chromium.org/2132283002
Cr-Commit-Position: refs/heads/master@{#404595}

ScopedCrashKey(blah, blah) should be ScopedCrashKey key(blah, blah)

BUG=625040

Review-Url: https://codereview.chromium.org/2134003002
Cr-Commit-Position: refs/heads/master@{#404594}

https://chromium.googlesource.com/skia.git/+log/4c42eebc5551..3ab14ca950d9

$ git log 4c42eebc5..3ab14ca95 --date=short --no-merges --format='%ad %ae %s'
2016-07-10 robertphillips Fix misplaced guard
2016-07-10 reed Revert of change mapRectScaleTranslate to pass args/ret by value (patchset #1 id:1 of https://codereview.chromium.org/2137853002/ )
2016-07-10 reed change mapRectScaleTranslate to pass args/ret by value

CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_blink_rel
TBR=msarett@google.com

Review-Url: https://codereview.chromium.org/2136883002
Cr-Commit-Position: refs/heads/master@{#404593}

BUG=3333
TEST=None

Review-Url: https://codereview.chromium.org/2129373002
Cr-Commit-Position: refs/heads/master@{#404592}

Remove listItems() usage from:
 - HTMLOptionElement::index()
 - HTMLSelectElement::saveListboxActiveSelection()
 - HTMLSelectElement::selectedOption()
 - HTMLSelectElement::deselectItemsWithoutValidation()
We use optionList() instead.

These changes don't improve computational complexity. However we can avoid to
update m_listItems when we add/remove OPTIONs.

Other changes:
 - HTMLSelectElement::updateListBoxSelectioN() needs to be updated to synchronize
  with saveListboxActiveSelection().  m_cachedStateForActiveSelection now
  represents OPTIONs instead of listItems, which includes OPTGROUP and HR.
 - Change the arguments of OptionList and OptionListIterator to |const| so that
  we can create them in const member functions of HTMLSelectElement.
 - Remove the optimized fast path in HTMLSelectElement::setRecalcListItems(). We
  don't use m_listItems in critical paths any longer.  So the optimization isn't
  necessary.

This CL improves performance because of the removal of the fast path.
blink_perf.dom:
select-multiple-add: 1,772 runs/s ->  1,912 runs/s
select-single-add: 867.9 runs/s -> 944.7 runs/s
select-single-remove: 137.9 runs/s -> 145.5 runs/.s

BUG=577989

Review-Url: https://codereview.chromium.org/2131073002
Cr-Commit-Position: refs/heads/master@{#404591}

Propagate force_use_new_methods_for_test and SetDecoderHeaderTableDebugVisitor to the nested SpdyFramerDecoderAdapter, if present. This fixes several test failures, if --use_nested_spdy_framer_decoder=true.

This CL lands server change 126441874 by jamessynge.

BUG=488484

Review-Url: https://codereview.chromium.org/2130193002
Cr-Commit-Position: refs/heads/master@{#404590}

This item is only shown when in fullscreen, and allows users to exit
fullscreen using the mouse (for users who don't have a keyboard
attached).

This is still not foolproof (as it won't work on a site that captures
the right-click menu) but should work in the majority of cases.

BUG=594868
TEST=https://permission.site. Click Fullscreen. Right-click background; should
     see "Exit full screen ... Esc".
TEST=F11 to go fullscreen. Right-click background; should see
     "Exit full screen ... F11".

Review-Url: https://codereview.chromium.org/2116973002
Cr-Commit-Position: refs/heads/master@{#404589}

We issue a fine-grained runtimeEnabled/runtimeDiabled notification.

Introduced a new setting forceMainWorldInitialization, which inspector
uses to force contexts in all frames to be able to evaluate there.

Moved scriptBlockedByCSP pause reason to DOMDebugger agent,
similar to scriptFirstStatement.

BUG=580337

Review-Url: https://codereview.chromium.org/2112593003
Cr-Commit-Position: refs/heads/master@{#404588}