GitLab

By convention, each virtual test suite should have a README.txt documenting
what directory it mirrors and what command line args are needed, so that
you can tell what an "empty" directory under LayoutTests/ is for.

This patch modifies lint-test-expectations to check that those READMEs
exist, and will help avoid the possibility that we'll delete whole directories
thinking they are unneeded and introduce weird errors when TestExpectations
still refer to nonexistent directories or files.

R=ojan@chromium.org
BUG=369385

Review URL: https://codereview.chromium.org/267803007

git-svn-id: svn://svn.chromium.org/blink/trunk@175748 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Revert of Revert of Oilpan: Prepare to make ExecutionContext GarbageCollectedMixin. (https://codereview.chromium.org/319333002/)

Reason for revert:
This was an incorrect revert--this patch was not responsible for the crash.

Original issue's description:
> Revert of Oilpan: Prepare to make ExecutionContext GarbageCollectedMixin. (https://codereview.chromium.org/318023002/)
> 
> Reason for revert:
> plugins/document-open.html is crashing on debug layout bots
> 
> Original issue's description:
> > Oilpan: Prepare to make ExecutionContext GarbageCollectedMixin.
> > 
> > We creates closures with ExecutionContexts in LocalFileSystem. The
> > ExecutionContext needs to be a CrossThreadPersistent data member of
> > closure. This CL adds ParamStorageTraits specialized to garbage-collected
> > classes.
> > 
> > This is a preparation to remove RefCounted from Node.
> > 
> > BUG=357163
> > 
> > Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=175734
> 
> TBR=oilpan-reviews@chromium.org,haraken@chromium.org,tkent@chromium.org
> NOTREECHECKS=true
> NOTRY=true
> BUG=357163
> 
> Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=175741

TBR=oilpan-reviews@chromium.org,haraken@chromium.org,tkent@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=357163

Review URL: https://codereview.chromium.org/319373002

git-svn-id: svn://svn.chromium.org/blink/trunk@175747 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Timeline uses time in milliseconds while trace event use microseconds. Do the conversion earlier.

BUG=361045

Review URL: https://codereview.chromium.org/318093002

git-svn-id: svn://svn.chromium.org/blink/trunk@175746 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Restore equality test back to being between Node* and a RawPtr<Node>,
for which there is an implicit conversion from a RawPtr<Node> to a
Node*. i.e., a RawPtr<Node> cannot be implicitly converted to a Node&
(and vice versa.)

TBR=haraken@chromium.org,oilpan-reviews@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/325573002

git-svn-id: svn://svn.chromium.org/blink/trunk@175745 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=

Review URL: https://codereview.chromium.org/317733008

git-svn-id: svn://svn.chromium.org/blink/trunk@175743 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Reason for revert:
Triggers ASSERT when running browser_test CastStreamingApiTestWithPixelOutput.RtpStreamError

Original issue's description:
> Delete CompositingUpdateAfterLayout
> 
> Rather than using a different update type, we can just mark the root as needing
> its properties updated, which will cause us to update the entire tree. This CL
> alone isn't a huge win, but it will let us remove
> RenderLayerCompositor::m_compositingLayersNeedRebuild in the next CL.
> 
> Also, move the call to enableCompositingModeIfNeeded into the new
> RenderLayerCompositor::didLayout function. This makes the sequence look a bit
> silly (because we mark the dirty bit only to clear it immediately), but that's
> what the code did before (albeit in a more roundabout way).
> 
> BUG=332248
> 
> Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=175738

TBR=esprehn@chromium.org,abarth@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=332248

Review URL: https://codereview.chromium.org/321663002

git-svn-id: svn://svn.chromium.org/blink/trunk@175742 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Revert of Oilpan: Prepare to make ExecutionContext GarbageCollectedMixin. (https://codereview.chromium.org/318023002/)

Reason for revert:
plugins/document-open.html is crashing on debug layout bots

Original issue's description:
> Oilpan: Prepare to make ExecutionContext GarbageCollectedMixin.
> 
> We creates closures with ExecutionContexts in LocalFileSystem. The
> ExecutionContext needs to be a CrossThreadPersistent data member of
> closure. This CL adds ParamStorageTraits specialized to garbage-collected
> classes.
> 
> This is a preparation to remove RefCounted from Node.
> 
> BUG=357163
> 
> Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=175734

TBR=oilpan-reviews@chromium.org,haraken@chromium.org,tkent@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=357163

Review URL: https://codereview.chromium.org/319333002

git-svn-id: svn://svn.chromium.org/blink/trunk@175741 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Revert of Generalize and refactor DeviceSensorEvent* architecture to support multi-event type targets. (https://codereview.chromium.org/315573002/)

Reason for revert:
ASAN bots are broken with this patch.

Original issue's description:
> Generalize and refactor DeviceSensorEvent* architecture to support
> multi-event type targets.
> 
> This patch performs a significant refactoring of the
> DeviceSensorEvent{Controller,Dispatcher} and related classes.
> It provides new Base classes to allow simple implementation of
> APIs involving multi-event type non-window targets (e.g. Battery
> Status API). Additionally it makes the implementation of existing
> event-based APIs (device_orientation, device_light) simpler and
> more compact.
> 
> In particular the following new classes have been added:
> * DeviceEventControllerBase and DeviceEventDispatcherBase
> (derived from by the Battery Status API).
> * DeviceSingleWindowEventController is more specific and
> contains functionality necessary for single-event type window
> target APIs (used by Device Motion/Orientation and Device Light
> APIs).
> 
> BUG=122593, 360068
> 
> Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=175661
> 
> Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=175686

TBR=ch.dumez@samsung.com,abarth@chromium.org,jochen@chromium.org,tkent@chromium.org,timvolodine@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=122593, 360068

Review URL: https://codereview.chromium.org/321653003

git-svn-id: svn://svn.chromium.org/blink/trunk@175740 bbb929c8-8fbe-4397-9dbb-9b2b20218538

R=haraken@chromium.org
BUG=340522

Review URL: https://codereview.chromium.org/319183007

git-svn-id: svn://svn.chromium.org/blink/trunk@175739 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Rather than using a different update type, we can just mark the root as needing
its properties updated, which will cause us to update the entire tree. This CL
alone isn't a huge win, but it will let us remove
RenderLayerCompositor::m_compositingLayersNeedRebuild in the next CL.

Also, move the call to enableCompositingModeIfNeeded into the new
RenderLayerCompositor::didLayout function. This makes the sequence look a bit
silly (because we mark the dirty bit only to clear it immediately), but that's
what the code did before (albeit in a more roundabout way).

BUG=332248

Review URL: https://codereview.chromium.org/323653002

git-svn-id: svn://svn.chromium.org/blink/trunk@175738 bbb929c8-8fbe-4397-9dbb-9b2b20218538

The IDB spec defines that transactions are created 'active' and become
'inactive' when control returns to the event loop. The way this was
implemented, transactions created within microtasks (e.g. Promise
callbacks) were not deactivated.

Flip the order of microtasks and deactivation, and add tests.

BUG=380910
R=adamk@chromium.org

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=175600

Review URL: https://codereview.chromium.org/321433002

git-svn-id: svn://svn.chromium.org/blink/trunk@175737 bbb929c8-8fbe-4397-9dbb-9b2b20218538

We creates closures with ExecutionContexts in LocalFileSystem. The
ExecutionContext needs to be a CrossThreadPersistent data member of
closure. This CL adds ParamStorageTraits specialized to garbage-collected
classes.

This is a preparation to remove RefCounted from Node.

BUG=357163

Review URL: https://codereview.chromium.org/318023002

git-svn-id: svn://svn.chromium.org/blink/trunk@175734 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Chromium's checkdeps.py has moved but the
script was never updated.

TEST=Manually run the script and it works!

Review URL: https://codereview.chromium.org/325473003

git-svn-id: svn://svn.chromium.org/blink/trunk@175733 bbb929c8-8fbe-4397-9dbb-9b2b20218538

To avoid needing put all the children into a NodeVector
inside removeChildren() we can reverse the order of childrenChanged
and the associated child notifications (insertedInto/removedFrom).

This seems fine, it means that a parent now is not told it has new
children until they've already been told they're in the tree.
Inversely it does mean that when a child is told it's in the tree
the parent doesn't know it has new children yet, but that's less
crazy since it doesn't result in things like children without
inDocument() bits being set existing inside ::childrenChanged.

For removal I swapped the calls to notifyNodeRemoved with
the calls to childrenChanged.

For insertion I moved the call to childrenChanged into the
notifyNodeInserted function. This is necessary since it can
run script and we need to call childrenChanged before running
the script.

I also added some more RefPtr<Node> protect(this) guards
to methods which can runs script and more ScriptForbiddenScopes
to parts of the code that now can't run script.

Review URL: https://codereview.chromium.org/306233005

git-svn-id: svn://svn.chromium.org/blink/trunk@175732 bbb929c8-8fbe-4397-9dbb-9b2b20218538

GraphicsLayerUpdater::rebuildTree and GraphicsLayerUpdater::update no longer
have anything to do with each other. This CL move rebuildTree into a separate
file, called GraphicsLayerTreeBuilder.

Review URL: https://codereview.chromium.org/319183006

git-svn-id: svn://svn.chromium.org/blink/trunk@175731 bbb929c8-8fbe-4397-9dbb-9b2b20218538

We don't appear to need this incremental update anymore.

BUG=343756

Review URL: https://codereview.chromium.org/317373004

git-svn-id: svn://svn.chromium.org/blink/trunk@175730 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This CL teaches the callers of CompositingUpdateOnScroll how to do partial
compositing updates. Now that they're smarter, we can use
CompositingUpdateAfterCompositingInputChange and we don't need
CompositingUpdateOnScroll anymore.

Review URL: https://codereview.chromium.org/326443002

git-svn-id: svn://svn.chromium.org/blink/trunk@175729 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=380054

Review URL: https://codereview.chromium.org/325483002

git-svn-id: svn://svn.chromium.org/blink/trunk@175728 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This patch fixes the deleteSelection callsites in CompositeEditCommand
after [1] refactored them to remove the bool 'replace' parameter.

[1] http://src.chromium.org/viewvc/blink?revision=164025&view=revision

BUG=381069

Review URL: https://codereview.chromium.org/321583004

git-svn-id: svn://svn.chromium.org/blink/trunk@175727 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Previously TouchEvents had their co-ordinates truncated to integers.
By supporting sup-pixel precision we get smoother dragging and line 
drawing at slow speeds on high-dpi devices or when zoomed.

Intent to ship thread: https://groups.google.com/a/chromium.org/forum/#!topic/blink-dev/J0Lca9fCuNw

Tests depend on EventSender change here: https://src.chromium.org/viewvc/chrome?view=rev&revision=275340

BUG=323935

Review URL: https://codereview.chromium.org/298133003

git-svn-id: svn://svn.chromium.org/blink/trunk@175726 bbb929c8-8fbe-4397-9dbb-9b2b20218538

These have been removed in Chrome and do nothing.

R=jochen@chromium.org
BUG=none

Review URL: https://codereview.chromium.org/298803009

git-svn-id: svn://svn.chromium.org/blink/trunk@175725 bbb929c8-8fbe-4397-9dbb-9b2b20218538

We haven't supported Core Graphics in some time.
Only one of these functions is implemented at all, and it  is unused.

BUG=381789

Review URL: https://codereview.chromium.org/315403007

git-svn-id: svn://svn.chromium.org/blink/trunk@175724 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=350571

Review URL: https://codereview.chromium.org/318923004

git-svn-id: svn://svn.chromium.org/blink/trunk@175723 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Use ElementTraversal API in ContainerNode::checkForChildrenAdjacentRuleChanges()
as we are only interested in child Elements.

R=esprehn@chromium.org

Review URL: https://codereview.chromium.org/316173007

git-svn-id: svn://svn.chromium.org/blink/trunk@175722 bbb929c8-8fbe-4397-9dbb-9b2b20218538

http://src.chromium.org/viewvc/blink?view=revision&revision=175704

TBR=abarth@chromium.org

Review URL: https://codereview.chromium.org/311393004

git-svn-id: svn://svn.chromium.org/blink/trunk@175721 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Removes <style scoped>.

This was discussed and approved with LGTMs in blink-dev:
https://groups.google.com/a/chromium.org/d/msg/blink-dev/R1x18ZLS5qQ/Bjuh_cENhlQJ

Chrome status dashboard entry:
http://www.chromestatus.com/features/5374137958662144

Note:
this is based on esprehn's original cl:
https://codereview.chromium.org/214693002/

BUG=379096
TEST=pass all layout tests

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=175555

Review URL: https://codereview.chromium.org/310443002

git-svn-id: svn://svn.chromium.org/blink/trunk@175720 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Blink r175423 fixed non-composited nested RenderView repainting
for repaint-after-layout, but in the process broke repainting of
composited nested RenderViews for all code paths. The fix is to
correctly issue repaints to the backing store of composited
RenderViews regardless of whether they're nested.

BUG=381022
R=esprehn@chromium.org

Review URL: https://codereview.chromium.org/318183002

git-svn-id: svn://svn.chromium.org/blink/trunk@175718 bbb929c8-8fbe-4397-9dbb-9b2b20218538

http://src.chromium.org/viewvc/blink?view=revision&revision=175702

BUG=369375
TBR=dcheng@chromium.org

Review URL: https://codereview.chromium.org/321613003

git-svn-id: svn://svn.chromium.org/blink/trunk@175717 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This call isn't need anymore. Now setNeedsToUpdateAncestorDependentProperties
does this work.

BUG=332248

Review URL: https://codereview.chromium.org/319863004

git-svn-id: svn://svn.chromium.org/blink/trunk@175716 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Revert of Implementing mixed content for forms posting to insecure location from secure ones (https://codereview.chromium.org/311033003/)

Reason for revert:
This CL incorrectly integrates with the inspector.

Notice that in Element::didModifyAttribute, InspectorInstrumentation::didModifyDOMAttr will be called with the original value, not the mutated value.

Original issue's description:
> Implementing mixed content for forms posting to insecure location from secure ones.
> 
> When a form "action" attribute is pointing to insecure location from a secure one, this is flagged as mixed content. This is even checked if the pages dynamically changes the "action" attribute after the initial loading. The mixed content is non-blocking by default, however, this can be overridden in the preferences, and this will cause a submit to silently fail, other than a console message.
> 
> Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=175714

TBR=jww@chromium.org,mkwst@chromium.org,mohammed@chromium.org
NOTREECHECKS=true
NOTRY=true

Review URL: https://codereview.chromium.org/319183008

git-svn-id: svn://svn.chromium.org/blink/trunk@175715 bbb929c8-8fbe-4397-9dbb-9b2b20218538

When a form "action" attribute is pointing to insecure location from a secure one, this is flagged as mixed content. This is even checked if the pages dynamically changes the "action" attribute after the initial loading. The mixed content is non-blocking by default, however, this can be overridden in the preferences, and this will cause a submit to silently fail, other than a console message.

Review URL: https://codereview.chromium.org/311033003

git-svn-id: svn://svn.chromium.org/blink/trunk@175714 bbb929c8-8fbe-4397-9dbb-9b2b20218538

> Eliminate MediaPlayer & MediaPlayerClient abstractions(play/pause, other APIs)
> 
> BUG=350571
> 
> Review URL: https://codereview.chromium.org/298093004

TBR=srirama.m@samsung.com

Review URL: https://codereview.chromium.org/325513002

git-svn-id: svn://svn.chromium.org/blink/trunk@175713 bbb929c8-8fbe-4397-9dbb-9b2b20218538

> Eliminate MediaPlayer & MediaPlayerClient abstractions(seek, duration, poster APIs)
> 
> Bug=350571
> 
> Review URL: https://codereview.chromium.org/306123003

TBR=srirama.m@samsung.com

Review URL: https://codereview.chromium.org/314413003

git-svn-id: svn://svn.chromium.org/blink/trunk@175712 bbb929c8-8fbe-4397-9dbb-9b2b20218538

> Eliminate MediaPlayer & MediaPlayerClient abstractions(rate, setRate and other APIs)
> 
> BUG=350571
> 
> Chromium side patch for this is https://codereview.chromium.org/303403002/
> 
> Review URL: https://codereview.chromium.org/304353005

TBR=srirama.m@samsung.com

Review URL: https://codereview.chromium.org/320053003

git-svn-id: svn://svn.chromium.org/blink/trunk@175711 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=376194

Review URL: https://codereview.chromium.org/312143003

git-svn-id: svn://svn.chromium.org/blink/trunk@175709 bbb929c8-8fbe-4397-9dbb-9b2b20218538

There's no connection between being visible and z-order lists. There's no
reason to mark the z-order list as dirty when changing visibility status.

Review URL: https://codereview.chromium.org/320593004

git-svn-id: svn://svn.chromium.org/blink/trunk@175708 bbb929c8-8fbe-4397-9dbb-9b2b20218538

http://src.chromium.org/viewvc/blink?view=revision&revision=175681

BUG=313438
TBR=rob.buis@samsung.com

Review URL: https://codereview.chromium.org/311193011

git-svn-id: svn://svn.chromium.org/blink/trunk@175707 bbb929c8-8fbe-4397-9dbb-9b2b20218538

The previous version of this test generated non-deterministic results
due to the way console messages are generated. Specifically, the test
case was constructed while parsing might not yet be complete, but the
act of constructing the test case generated console messages. If a
console message is generated with no specified line number,
Document::internalAddMessage attempts to attach a line number, but one
of the conditionals for doing so is whether the Document is still
parsing making the output non-deterministic.

To get around that, the test  has been rewritten to only run once the
document is finished loading.

BUG=381451
R=japhet@chromium.org

Review URL: https://codereview.chromium.org/314393002

git-svn-id: svn://svn.chromium.org/blink/trunk@175706 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This code evolved and grew over the years, but it seems to be trying to compute
something very simple.

Review URL: https://codereview.chromium.org/307933010

git-svn-id: svn://svn.chromium.org/blink/trunk@175704 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=369375
TBR=sandersd@chromium.org

Review URL: https://codereview.chromium.org/321603002

git-svn-id: svn://svn.chromium.org/blink/trunk@175702 bbb929c8-8fbe-4397-9dbb-9b2b20218538