GitLab

All its usages have been inlined.

BUG=361045
R=alph@chromium.org

Review URL: https://codereview.chromium.org/305133004

git-svn-id: svn://svn.chromium.org/blink/trunk@175263 bbb929c8-8fbe-4397-9dbb-9b2b20218538

R=alph, yurys

Review URL: https://codereview.chromium.org/308643003

git-svn-id: svn://svn.chromium.org/blink/trunk@175262 bbb929c8-8fbe-4397-9dbb-9b2b20218538

background-repeat serialization did not handle initial keyword values in
background-repeat value lists.
Initial values in value lists are introduced by setting the background
shorthand to multiple background values while omitting background-repeat
properties.
Example: background: url(#1), url(#2), url(#3);
The background-repeat gets stored internally as "initial, initial, initial".

This patch updates StylePropertySerializer::backgroundRepeatPropertyValue()
to handle this case correctly.

BUG=378167

Review URL: https://codereview.chromium.org/303993003

git-svn-id: svn://svn.chromium.org/blink/trunk@175261 bbb929c8-8fbe-4397-9dbb-9b2b20218538

For DocumentInit, the context document is no longer mediated
as a WeakPtr, but as a Member. It is sufficient that the
Document holds onto this reference weakly.

R=haraken@chromium.org, tkent@chromium.org
BUG=357163

Review URL: https://codereview.chromium.org/304323010

git-svn-id: svn://svn.chromium.org/blink/trunk@175260 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This is a regression caused by r174937.
After r174937, ExecutionContexts created for IDBRequestTest and IDBTransactionTest
outlive the tests (because the lifetime of the ExecutionContexts is bound to a ScriptState
associated with the tests). This is wrong, and we should clear the ExecutionContexts
when shutting down the tests.

BUG=379616

Review URL: https://codereview.chromium.org/307933011

git-svn-id: svn://svn.chromium.org/blink/trunk@175259 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This patch changes |IndentOutdentCommand::outdentParagraph()| not to call
|CompositeEditCommand::movePargrap()| with null |VisiblePositon|.

This situation can be happened in attached test case, where
|visibleStartOfParagraph| and |visibleEndOfPargraph| are second TABLE element.
Both |startOfParagraph()| and |endOfParagraph()| return null |VisiblePositon|
for them.

BUG=379058
TEST=editing/execCommand/outdent-collapse-table-crash-2.html

Review URL: https://codereview.chromium.org/305143002

git-svn-id: svn://svn.chromium.org/blink/trunk@175258 bbb929c8-8fbe-4397-9dbb-9b2b20218538

TEST=fast/events/drag-leak-document.html (run_webkit_tests with --enable-leak-detection)
BUG=364392

Review URL: https://codereview.chromium.org/302133002

git-svn-id: svn://svn.chromium.org/blink/trunk@175257 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Currently an XMLHttpRequest that fails without response is shown as "Other"
resource type. This CL fixes the behavior.

BUG=294676

Review URL: https://codereview.chromium.org/302093003

git-svn-id: svn://svn.chromium.org/blink/trunk@175256 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=369401
TEST=All tests should pass, because of just refactoring.

Review URL: https://codereview.chromium.org/304083003

git-svn-id: svn://svn.chromium.org/blink/trunk@175255 bbb929c8-8fbe-4397-9dbb-9b2b20218538

TBR=eseidel@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/307933012

git-svn-id: svn://svn.chromium.org/blink/trunk@175254 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=369451

Review URL: https://codereview.chromium.org/301113002

git-svn-id: svn://svn.chromium.org/blink/trunk@175253 bbb929c8-8fbe-4397-9dbb-9b2b20218538

http://src.chromium.org/viewvc/blink?view=revision&revision=175243

BUG=373389
TBR=ksakamoto@chromium.org

Review URL: https://codereview.chromium.org/308983002

git-svn-id: svn://svn.chromium.org/blink/trunk@175252 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This patch changes |IndentOutdentCommand::outdentParagraph()| to check whether
|enclosingBlockNode| is null or not, before passing it to
|CompositeEditCommand::splitTreeToNode()| which assumes it isn't null.

This situation can be happened in attach test case, where
|enclosingBlockNode| is calculated from |visibleStartOfParagraph|, which is
second TABLE element, by |enclosingBlock()|.

BUG=378758
TEST=editing/execCommand/outdent-collapse-table-crash.html

Review URL: https://codereview.chromium.org/308023003

git-svn-id: svn://svn.chromium.org/blink/trunk@175251 bbb929c8-8fbe-4397-9dbb-9b2b20218538

The TestExpectations objects we were creating in a loop would consume memory
and never return it. I'm not sure why the memory wasn't getting freed, but reparsing
TestExpectations over and over again is really slow and by far the longest pole
in rebaseline performance. So, fix both issues by not doing this.

Instead, grab all the test_configurations that we want to remove up front and then
remove them all in one go from TestExpectations. This makes the work and memory
cost be O(number of ports) instead of O(number of tests * number of ports).

Also, fix a bug where we would not remove lines for passing tests that were being
rebaselined. I had previously fixed this incorrectly in auto-rebaseline. Instead, fix
it correctly in AbstractParallelRebaselineCommand so that the rebaseline-expectations,
rebaseline-json and rebaseline commands also get the fix.

BUG=379484

Review URL: https://codereview.chromium.org/302003009

git-svn-id: svn://svn.chromium.org/blink/trunk@175250 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=379616
TBR=oilpan-reviews@chromium.org

Review URL: https://codereview.chromium.org/308963003

git-svn-id: svn://svn.chromium.org/blink/trunk@175249 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This CL changes generated bindings so that both toV8(RawPtr<X>) and toV8(X*) are generated. Otherwise, if someone uses RawPtr<X> instead of X* in the code base, it can be bound to unexpected toV8() and a wrong wrapper can be returned to V8. For more details, see my comment in the patch set 3.

BUG=340522

Review URL: https://codereview.chromium.org/257503003

git-svn-id: svn://svn.chromium.org/blink/trunk@175248 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This finishes qualifying include paths in bindings!
There are some remaining unqualified paths into bindings from core
generated files, so we can't yet remove the include directories from
core.gyp, but bindings is sorted!

TBR=haraken
BUG=358074
BUG=377364

Review URL: https://codereview.chromium.org/306673002

git-svn-id: svn://svn.chromium.org/blink/trunk@175247 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Before this patch, |SVGAnimateElement::m_animatedPropertyType|
was maintained along |SVGAnimatedTypeAnimator::m_type|,
and there were many ASSERTs checking that they were same.

In this patch, I removed
|SVGAnimateElement::m_animatedPropertyType| and changed
their references to point to
|SVGAnimatedTypeAnimator::m_type| instead.

BUG=378670

Review URL: https://codereview.chromium.org/303093002

git-svn-id: svn://svn.chromium.org/blink/trunk@175245 bbb929c8-8fbe-4397-9dbb-9b2b20218538

The indirection plumbing obscures what is really happening at the callsites
and should be removed. Also, strengthen up the assertions, since
now HTMLImportChild has HTMLImportLoader for (observable) life.

R=morrita@chromium.org
BUG=

Review URL: https://codereview.chromium.org/299313013

git-svn-id: svn://svn.chromium.org/blink/trunk@175244 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This fixes a bug where a bullet character (U+25CF) becomes tofu when it
is a fallback for a webfont and italic style. The fallback font
(Helvetica) has a glyph for U+25CF, but its italic variation (Helvetica
Oblique) does not. In such a case, we should use synthetic oblique.

BUG=373389
TEST=fast/text/fallback-for-custom-font.html

Review URL: https://codereview.chromium.org/300393003

git-svn-id: svn://svn.chromium.org/blink/trunk@175243 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This CL broke worker-reload-repeated.html in oilpan bots.

> Oilpan: Move ThreadableWebSocketChannelClientWrapper to Oilpan's heap
> 
> This is a preparation for moving WebSocketChannelClient to the heap.
> 
> This CL adds a template specialization to CrossThreadCopier.h so that ThreadableWebSocketChannelClientWrapper* can be passed to createCallbackTask().
> 
> BUG=340522
> 
> Review URL: https://codereview.chromium.org/267323004

TBR=haraken@chromium.org

Review URL: https://codereview.chromium.org/309503013

git-svn-id: svn://svn.chromium.org/blink/trunk@175242 bbb929c8-8fbe-4397-9dbb-9b2b20218538

- Basically, this CL replaces V8AbstractEventListener::m_world with V8AbstractEventListener::m_scriptState.

- This CL uses the ScriptState in a lot of places about event listeners and cleans up the code (for example, this CL removes an empty creation context from toV8()).

- The complicated part is that V8LazyEventListener does not know the ScriptState when it's created. Thus, V8AbstractEventListener::m_scriptState can be null if the listener is a V8LazyEventListener until the V8LazyEventListener is called back from Blink (i.e., either of handleEvent(), callListenerFunction() and getEventListener() is called).

- This CL changes web-exposed behavior. In a case where the frame that registered an onload event handler (say Frame1) is different from the frame that triggers the onload event (Frame2), before this CL, the onload event handler is invoked in the context of Frame2. After this CL, the onload event handler is not triggered if Frame1 != Frame2. This behavior is not speced but aligns with Firefox. See comment #17 and #19 for more details.

BUG=357144

Review URL: https://codereview.chromium.org/293053007

git-svn-id: svn://svn.chromium.org/blink/trunk@175241 bbb929c8-8fbe-4397-9dbb-9b2b20218538

> The reason for reverting is: broke interactive_ui_tests:
> 
> Referrer
> SearchReusesInstantTab
> TypedSearchURLDoesntReuseInstantTab.
I fixed these tests in https://src.chromium.org/viewvc/chrome?view=rev&revision=273996 .

Original Description:

Initializing |v8::Context| is a heavy operation, and should be avoided if necessary. This is meant to be delayed until first time |toV8Context| is called, but |ScriptController::updateDocument| was forcing creation.

This patch fixes the if branch so that it would not create a |v8::Context| if it does not exist.

BUG=368548, 368555

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=173044

Review URL: https://codereview.chromium.org/263583002

git-svn-id: svn://svn.chromium.org/blink/trunk@175240 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This belongs on the HTML elements, not on the renderer which is torn
down when the fallback content is attached.

BUG=379594

Review URL: https://codereview.chromium.org/301353005

git-svn-id: svn://svn.chromium.org/blink/trunk@175239 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Automatically generated using:
webkit-patch update-flaky-tests

R=eseidel
BUG=

Review URL: https://codereview.chromium.org/306293002

git-svn-id: svn://svn.chromium.org/blink/trunk@175238 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Spec: https://slightlyoff.github.io/ServiceWorker/spec/service_worker/#navigator-service-worker-controller

BUG=363967

Review URL: https://codereview.chromium.org/301103003

git-svn-id: svn://svn.chromium.org/blink/trunk@175237 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This CL contains a number of minor cleanups to RenderLayerStackingNode. The
most interesting change is merging ancestorStackingContextNode and
ancestorStackingNode, which do exactly the same thing.

Review URL: https://codereview.chromium.org/301233008

git-svn-id: svn://svn.chromium.org/blink/trunk@175236 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Unreviewed rebaseline of virtual/gpu/compositedscrolling/overflow/overflow-positioning.html

TBR=ojan@chromium.org

Review URL: https://codereview.chromium.org/309753002

git-svn-id: svn://svn.chromium.org/blink/trunk@175235 bbb929c8-8fbe-4397-9dbb-9b2b20218538

- write-while-waiting.html was fixed by r175127.
- Tighten terminate-during-sync-operation-*.html expectations

BUG=357427
TBR=sigbjornf@opera.com

Review URL: https://codereview.chromium.org/308673008

git-svn-id: svn://svn.chromium.org/blink/trunk@175234 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Remove unnecessary Persistent reference (to an object contained within
another Persistent) + arrange for manual GCing of resources to avoid
test objects being reported as not destructed and leaking with
Oilpan enabled.

R=
BUG=

Review URL: https://codereview.chromium.org/306273002

git-svn-id: svn://svn.chromium.org/blink/trunk@175233 bbb929c8-8fbe-4397-9dbb-9b2b20218538

For :hover, :active, :focus, and :drag, childrenAffectedBy really means
that descendants, or siblings, or elements in the siblings' subtrees may be
affected. Methods and constants are renamed to reflect that.

R=esprehn@chromium.org

Review URL: https://codereview.chromium.org/308683007

git-svn-id: svn://svn.chromium.org/blink/trunk@175232 bbb929c8-8fbe-4397-9dbb-9b2b20218538

No one reads this state so we can remove it.

Review URL: https://codereview.chromium.org/306123008

git-svn-id: svn://svn.chromium.org/blink/trunk@175231 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Right now when it exceeds the appengine storage limit for a single
entry it 500s. Instead, just create a new entry. This isn't totally
ideal. Ideally we'd create a new entry but link it to the old
one so we can show it correctly in the UI, but that's too much work.

NOTRY=true
BUG=378490

Review URL: https://codereview.chromium.org/304303003

git-svn-id: svn://svn.chromium.org/blink/trunk@175230 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=326139
TBR=mkwst@chromium.org

Review URL: https://codereview.chromium.org/305323002

git-svn-id: svn://svn.chromium.org/blink/trunk@175229 bbb929c8-8fbe-4397-9dbb-9b2b20218538

I accidentally already landed part of this support
in an earlier change, but this makes --upload actually
work.  You can see an example change here:
https://codereview.chromium.org/308743011/
(which again accidentally includes this change).

This also adds a hack-and-slash version of sheriff-calendar
code which I and Eric Boren originally wrote for the AutoRollBot
but that lives in a different repository:
https://chromium.googlesource.com/chromium/tools/build/+/master/scripts/tools/blink_roller/auto_roll.py

This combined with a cron job should make
it easy for gardeners to keep FlakyTests up to date.

Looking at 308743011, we clearly need to address some
of this marauding flakiness as FlakyTests should be
a very stable file.  Right now it is changing by the hour.

BUG=242366

Review URL: https://codereview.chromium.org/307183002

git-svn-id: svn://svn.chromium.org/blink/trunk@175228 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Windows XP specific rebaselines for r175038 as the XP > Win7 fallback is
no longer valid for text tests.

TBR=dglazkov@chromium.org

Review URL: https://codereview.chromium.org/308653008

git-svn-id: svn://svn.chromium.org/blink/trunk@175227 bbb929c8-8fbe-4397-9dbb-9b2b20218538

The showSecurityErrorMessage wasn't called by anything. So I killed it.

R=morrita
BUG=

Review URL: https://codereview.chromium.org/305643009

git-svn-id: svn://svn.chromium.org/blink/trunk@175226 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Prior to this CL, we computed RenderLayer::isUnclippedDescendant in a separate
O(n^2) walk of the out-of-flow RenderLayers. On workloads like famo.us, this
walk is quite expensive because everything is out-of-flow positioned, and
O(n^2) is large.

After this CL, we compute this information as we walk down the tree in the
CompositingPropertyUpdater, which means we can compute the information in
linear time and we can entirely remove the separate tree walk. Moving this
computation also lets us rip out a large amount of machinery that existed to
prune the extra tree walk that this CL removes.

Now that we compute this information with the CompositingPropertyUpdater,
we compute it even when composited overflow scrolling is off.  This confused
some tests that used Internals to poke this state.  I've added a workaround to
keep theses tests working identically.  I'll remove the work around and update
the tests in the next CL.

Review URL: https://codereview.chromium.org/309743002

git-svn-id: svn://svn.chromium.org/blink/trunk@175225 bbb929c8-8fbe-4397-9dbb-9b2b20218538

The notion of "importedDocument" only exists at the LinkImport level
(because its used, HTMLLinkElement has to deal with both importer
and the importee), so we can just call the imported documents simply
"documents" inside of the imports plumbing.

No functional changes, plumbotron.

R=morrita
BUG=

Review URL: https://codereview.chromium.org/298413004

git-svn-id: svn://svn.chromium.org/blink/trunk@175224 bbb929c8-8fbe-4397-9dbb-9b2b20218538

We can just test root-ness by the absence of parent.

No behavior changes, just plumbotron.

R=morrita
BUG=

Review URL: https://codereview.chromium.org/302433016

git-svn-id: svn://svn.chromium.org/blink/trunk@175223 bbb929c8-8fbe-4397-9dbb-9b2b20218538