GitLab

http://src.chromium.org/viewvc/blink?view=revision&revision=185129

BUG=377845
TBR=fmalita@chromium.org

Review URL: https://codereview.chromium.org/707753007

git-svn-id: svn://svn.chromium.org/blink/trunk@185138 bbb929c8-8fbe-4397-9dbb-9b2b20218538

V8PerIsolateData cannot own IDBPendingTransactionMonitor, because
IDBPendingTransactionMonitor is defined in modules/indexeddb and
V8PerIsolateData is defined in bindings/core. Replace the monitor
entirely by having IDBTransactions register tasks to be run once
the scopes are unwound.

BUG=358074
R=haraken,tasak,adamk

Review URL: https://codereview.chromium.org/686153003

git-svn-id: svn://svn.chromium.org/blink/trunk@185137 bbb929c8-8fbe-4397-9dbb-9b2b20218538

If a partial interface in modules extends an interface in core and uses
any union types, the generated bindings/modules/v8/V8*Partial.cpp file
needs to include UnionTypesModules.h. This did not happen, since use of
union types was only tracked per main (non-partial) interface.

BUG=240176

Review URL: https://codereview.chromium.org/684653009

git-svn-id: svn://svn.chromium.org/blink/trunk@185136 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=401424

Requires https://codereview.chromium.org/677073002

Review URL: https://codereview.chromium.org/676003002

git-svn-id: svn://svn.chromium.org/blink/trunk@185135 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This reduces the number of targets that have to be built when changing the
status of a runtime-enabled feature dramatically (for content_shell, from
1229 to 44), because these changes only affect the source file and not the
header.

In such cases, ninja notices when the modification time is left alone, and
it can propagate this "clean" (not changed) state to dependent targets.

Review URL: https://codereview.chromium.org/688443007

git-svn-id: svn://svn.chromium.org/blink/trunk@185134 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This is the only internal usage of DOMImplementation::hasFeature(), and
the only way that svg:requiredFeatures is observable, other than the
it->isEmpty() case.

If usage is low, DOMImplementation.hasFeature() and this internal usage
can be changed together, see
https://groups.google.com/a/chromium.org/d/msg/blink-dev/r2aARnIHG4k/_dN33g65mYgJ

Review URL: https://codereview.chromium.org/712813002

git-svn-id: svn://svn.chromium.org/blink/trunk@185133 bbb929c8-8fbe-4397-9dbb-9b2b20218538

http://src.chromium.org/viewvc/blink?view=revision&revision=185125

BUG=429843,432080,432079
TBR=aberent@chromium.org

Review URL: https://codereview.chromium.org/715463003

git-svn-id: svn://svn.chromium.org/blink/trunk@185132 bbb929c8-8fbe-4397-9dbb-9b2b20218538

We should wait for the V8 side operations to complete *before* tearing down the
Isolate.

BUG=431860

Review URL: https://codereview.chromium.org/716723003

git-svn-id: svn://svn.chromium.org/blink/trunk@185131 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Now that there is only one implementation it doesn't make sense to separate interface from implementation of TimelineModel.

BUG=428700

Review URL: https://codereview.chromium.org/715803002

git-svn-id: svn://svn.chromium.org/blink/trunk@185130 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Instead of storing glyph advances (only to later convert to offsets for the Skia draw calls), refactor GlyphBuffer to store offsets directly. This reduces the amount of buffer post-processing and decreases the API friction when calling into Skia.

GlyphBuffer offsets are relative to the text run origin and are either horizontal/x-only (generated by the simple shaper) or x/y interleaved (generated by the complex shaper). These correspond to the text positioning modes used by Skia.

Thanks to the new buffer format, the refactor also includes:

* zero offset processing when building text blobs (the offsets are essentially memcpy-ed)
* minimal processing (translate by destination point) for the legacy drawPosText path (we could also optimize this path, but the added complexity is probably not justified given the plans to completely switch to text blobs)
* SimpleShaper::adjustSpacing no longer needs to inject synthetic glyphs but simply adjusts the offsets for following glyps
* drawEmphasisMarks uses explicit offsets to glyph-midpoint instead of computing them based on advances

Other things to investigate:

* multi-run buffer builder (to support multi-run text blobs)
* stop storing space glyphs (simply adjust the following glyph offsets instead)

R=jbroman@chromium.org,eae@chromium.org,dominik.rottsches@intel.com
BUG=377845

Review URL: https://codereview.chromium.org/676523003

git-svn-id: svn://svn.chromium.org/blink/trunk@185129 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This patch adds a watchlist for scheduling related code reviews.

BUG=432040

Review URL: https://codereview.chromium.org/715793002

git-svn-id: svn://svn.chromium.org/blink/trunk@185128 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This inspector app is intended to be run inside an iframe.
InspectorAppHost exposes an API between inspector app and hosting app.

Created DevTools app which is only embedding inspector app in iframe.
Next patches will define API and move hosting functionality from
inspector app to devtools app.

BUG=431697

Review URL: https://codereview.chromium.org/710033003

git-svn-id: svn://svn.chromium.org/blink/trunk@185127 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Use ContainerNode more in DOMEditor, specifically
this allows us to call replaceChild/removeChild etc.
on ContainerNode directly, skipping the checks in
Node::replaceChild/removeChild.

Review URL: https://codereview.chromium.org/716603002

git-svn-id: svn://svn.chromium.org/blink/trunk@185126 bbb929c8-8fbe-4397-9dbb-9b2b20218538

We want the overlay cast icon to be visible on all backgrounds, so its
background has to be at least semi-opaque. The non-overlay cast icon,
however, should have the same background as the media controls, which
is easiest to achive by making its background transparent. As such we
need different icons for the two cases.

Note that this also needs a Chromium CL for the new versions of the
icons.

BUG=426373

Review URL: https://codereview.chromium.org/716613002

git-svn-id: svn://svn.chromium.org/blink/trunk@185125 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Once https://codereview.chromium.org/614373007/ lands, the following code doesn't work as expected (It confuses reference counting).

class X : public RefCountedGarbageCollected<X> {
  static PassRefPtr<X> create() {
    return adoptRef(new X);
  }
}

Instead we have to write:

class X : public RefCountedGarbageCollected<X> {
  static X* create() {
    return new X;
  }
}

This CL adds a compile-time verification to detect adoptRef(X*) where X is RefCountedGarbageCollected.

BUG=420515

Review URL: https://codereview.chromium.org/647393002

git-svn-id: svn://svn.chromium.org/blink/trunk@185124 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Gracefully handle cancellation of a parser-blocking script load that
was initiated while already executing scripts. That is, quietly stop
waiting for it to load without notifying the host.

R=marja@chromium.org,kouhei@chromium.org,haraken@chromium.org
BUG=398090

Review URL: https://codereview.chromium.org/710933003

git-svn-id: svn://svn.chromium.org/blink/trunk@185122 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=363976

Review URL: https://codereview.chromium.org/622843002

git-svn-id: svn://svn.chromium.org/blink/trunk@185121 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This reverts commit cf26e70f1858fec58da5a214468b4252d1917310 aka

https://src.chromium.org/viewvc/blink?view=rev&revision=185102

an auto rebaseline that's waiting on a failing mac 10.8 bot build slave.

Reason for revert: green the blink tree while mac 10.8 build slave has no
space left on device error, issue 432058, chrome troopers mailed. Will do
the auto rebaseline again when the 10.8 bot is fixed.

TBR=yhirano@chromium.org
BUG=432058
NOTRY=true

Review URL: https://codereview.chromium.org/694123004

git-svn-id: svn://svn.chromium.org/blink/trunk@185120 bbb929c8-8fbe-4397-9dbb-9b2b20218538

media/video-overlay-cast-light-rendering.html
media/video-overlay-cast-dark-rendering.html
media/video-controls-with-cast-rendering.html

BUG=429843, 432079, 432080
TBR=senorblanco@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/709123006

git-svn-id: svn://svn.chromium.org/blink/trunk@185119 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This patch switches to lazily allocating transforms in RenderSVGShape
because local transforms are relatively rare.

BUG=429551

Review URL: https://codereview.chromium.org/701433002

git-svn-id: svn://svn.chromium.org/blink/trunk@185118 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Byte order mark detection needs to be done on the Blink side, since it also
affects encoding detection.

This also enables the preparse data for streamed scripts again, because the
corrupt data was caused by byte order marks not being detected properly.

BUG=430890

Review URL: https://codereview.chromium.org/708093002

git-svn-id: svn://svn.chromium.org/blink/trunk@185117 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Move the v8::Isolate* parameter to the first parameter of various binding methods in third_party/WebKit/Source/bindings

This step is to move the v8::Isolate* parameter to the first parameter of various binding methods in third_party/WebKit/Source/bindings
(Ref: binding-methods changed:
static v8::Local<v8::Object> createWrapper(v8::Handle<v8::Object> creationContext, const WrapperTypeInfo*, ScriptWrappableBase* internalPointer, v8::Isolate*)
PassRefPtrWillBeRawPtr<ScriptCallStack> createScriptCallStack(v8::Handle<v8::StackTrace>, size_t maxStackSize, v8::Isolate*))

BUG=424446

Review URL: https://codereview.chromium.org/717703002

git-svn-id: svn://svn.chromium.org/blink/trunk@185116 bbb929c8-8fbe-4397-9dbb-9b2b20218538

R=vsevik
BUG=424085

Review URL: https://codereview.chromium.org/701153002

git-svn-id: svn://svn.chromium.org/blink/trunk@185115 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Remove unnecessary explicit keywords from a constructor with more than one parameters in css directory.

BUG=None
TEST=None; No behavior change

Review URL: https://codereview.chromium.org/710293002

git-svn-id: svn://svn.chromium.org/blink/trunk@185114 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=428700,387546

Review URL: https://codereview.chromium.org/713913002

git-svn-id: svn://svn.chromium.org/blink/trunk@185113 bbb929c8-8fbe-4397-9dbb-9b2b20218538

PerformanceTests/ShadowDOM/LargeDistributionWithoutLayout allocates a lot of
collection backings with 32KB size and 16KB size. allocateFromFreeList failed
to allocate memory for them frequently even if there were FreeListEntry
with enough sizes because:
 - Actual required size is 32KB+8B or 16KB+8B because of object headers
 - allocateFromFreeList only checked the minimum size of a bucket.  For example,
   a 32KB+8B free slot was listed in the 32KB bucket, and 32KB is smaller than
   32KB+8B.

This CL reduces the peak number of HeapPages in LargeDistributionWithoutLayout
test by 200.

BUG=420515

Review URL: https://codereview.chromium.org/711173004

git-svn-id: svn://svn.chromium.org/blink/trunk@185112 bbb929c8-8fbe-4397-9dbb-9b2b20218538

As getPropertyCSSValue is no longer web-exposed we shouldn't require it
in tests so we can delete the interface.

BUG=428595

Review URL: https://codereview.chromium.org/714693002

git-svn-id: svn://svn.chromium.org/blink/trunk@185111 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Move the v8::Isolate* parameter to the first parameter of various binding methods in third_party/WebKit/Source/bindings

This step is to move the v8::Isolate* parameter to the first parameter of various binding methods in third_party/WebKit/Source/bindings
(Ref: binding-methods changed:
static void installAttributes(v8::Handle<v8::ObjectTemplate>, v8::Handle<v8::ObjectTemplate>, const AttributeConfiguration*, size_t attributeCount, v8::Isolate*)
static void installConstants(v8::Handle<v8::FunctionTemplate>, v8::Handle<v8::ObjectTemplate>, const ConstantConfiguration*, size_t constantCount, v8::Isolate*)
static void installConstant(v8::Handle<v8::FunctionTemplate>, v8::Handle<v8::ObjectTemplate>, const char* name, v8::AccessorGetterCallback, v8::Isolate*)
static void installMethods(v8::Handle<v8::ObjectTemplate>, v8::Handle<v8::Signature>, v8::PropertyAttribute, const MethodConfiguration*, size_t callbackCount, v8::Isolate*)
static void installAccessors(v8::Handle<v8::ObjectTemplate>, v8::Handle<v8::Signature>, const AccessorConfiguration*, size_t accessorCount, v8::Isolate*))

BUG=424446

Review URL: https://codereview.chromium.org/709023003

git-svn-id: svn://svn.chromium.org/blink/trunk@185109 bbb929c8-8fbe-4397-9dbb-9b2b20218538

According to the spec[1,2,3], named getter of HTMLAllCollection, HTMLFormControlsCollection and HTMLOptionsCollection should be nullable.

This CL also adds layout tests for nullable arguments.

[1] https://html.spec.whatwg.org/multipage/infrastructure.html#the-htmlallcollection-interface
[2] https://html.spec.whatwg.org/multipage/infrastructure.html#the-htmlformcontrolscollection-interface
[3] https://html.spec.whatwg.org/multipage/infrastructure.html#htmlallcollection

BUG=430337

Review URL: https://codereview.chromium.org/692993003

git-svn-id: svn://svn.chromium.org/blink/trunk@185108 bbb929c8-8fbe-4397-9dbb-9b2b20218538

If the split-off heap had free-list entries lager than the biggest one in the
main heap, they were not recycled.

BUG=420515

Review URL: https://codereview.chromium.org/712293002

git-svn-id: svn://svn.chromium.org/blink/trunk@185107 bbb929c8-8fbe-4397-9dbb-9b2b20218538

It is wrong to call EscapableHandleScope::Escape() multiple times
in executeScriptInIsolatedWorld(). Since the callers of executeScriptInIsolatedWorld()
already have HandleScopes, we can just simply remove a EscapableHandleScope
from executeScriptInIsolatedWorld().

Review URL: https://codereview.chromium.org/713743002

git-svn-id: svn://svn.chromium.org/blink/trunk@185106 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=
NOTRY=true

Review URL: https://codereview.chromium.org/711293002

git-svn-id: svn://svn.chromium.org/blink/trunk@185105 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Track attribute uses object type, but we should use
(VideoTrack or AudioTrack or TextTrak)? as the spec says[1].

[1] https://html.spec.whatwg.org/multipage/embedded-content.html#the-trackevent-interface

BUG=430337

Review URL: https://codereview.chromium.org/690923004

git-svn-id: svn://svn.chromium.org/blink/trunk@185104 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=431996
TBR=senorblanco@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/692283003

git-svn-id: svn://svn.chromium.org/blink/trunk@185103 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Delete mac result, rebaseline to grab any new mac result, which should
match the windows result and be optimized out if things are working as
expected per issue 405803 and issue 409435.

BUG=405803
TBR=dstockwell@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/697843005

git-svn-id: svn://svn.chromium.org/blink/trunk@185102 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This function was removed from the spec:
https://slightlyoff.github.io/ServiceWorker/spec/service_worker/index.html#install-event-section

BUG=423600
TEST=n/a

Review URL: https://codereview.chromium.org/690393003

git-svn-id: svn://svn.chromium.org/blink/trunk@185101 bbb929c8-8fbe-4397-9dbb-9b2b20218538

- Remove PreloadRequest::defer()
- Remove ResourceFetcher::requestPreload(). The only user is preload()
  and is just calling requestPreload().
- Move setDefer() call in HTMLResourcePreloader::preload() into
  resourceRequest() call
- Make HTMLResourcePreloader::preload() private

R=japhet
BUG=none

Review URL: https://codereview.chromium.org/717453002

git-svn-id: svn://svn.chromium.org/blink/trunk@185100 bbb929c8-8fbe-4397-9dbb-9b2b20218538

R=haraken@chromium.org
BUG=

Review URL: https://codereview.chromium.org/711053002

git-svn-id: svn://svn.chromium.org/blink/trunk@185099 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This is a follow-up CL of https://codereview.chromium.org/661453004/.

I don't see why the call sites of getOwnPropertyNames() can't use
getPropertyNames() instead. The reason why they need property names
is that we don't support OpenEndedDictionary and IDL dictionary
which contains an union type at this point. In both cases we should
be able to use getPropertyNames().

Review URL: https://codereview.chromium.org/715583008

git-svn-id: svn://svn.chromium.org/blink/trunk@185098 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This test was flaky because the iframe sent a postMessage to the parent
frame in its load handler. However, the parent frame may or may not have
run its load handler. If the parent frame's load handler hadn't run yet,
then there would be no handler for a postMessage. To prevent this race,
don't load the iframe until the parent frame is fully initialized.

BUG=409579

Review URL: https://codereview.chromium.org/713013005

git-svn-id: svn://svn.chromium.org/blink/trunk@185097 bbb929c8-8fbe-4397-9dbb-9b2b20218538