GitLab

Revert of Send AXLayoutComplete even when document hasn't finished loading yet. (https://codereview.chromium.org/199693003/)

Reason for revert:
Causes content_browertests to fail:

[ RUN      ] DumpAccessibilityTreeTest.AccessibilityModalDialogInIframeOpened
Testing: E:\b\build\slave\Win7_Tests\build\src\content\test\data\accessibility\modal-dialog-in-iframe-opened.html
c:\b\build\slave\win-latest-rel\build\src\content\browser\accessibility\dump_accessibility_tree_browsertest.cc(209): error: Value of: is_different
  Actual: true
Expected: false
* Line Expected
- ---- --------
     1 ROLE_SYSTEM_DOCUMENT READONLY FOCUSABLE
     2     IA2_ROLE_SECTION READONLY
     3         ROLE_SYSTEM_STATICTEXT name='Test for modal dialog opened in an iframe. '
     4         ROLE_SYSTEM_CLIENT FOCUSABLE
     5             IA2_ROLE_SCROLL_PANE
     6                 ROLE_SYSTEM_DOCUMENT READONLY FOCUSABLE
     7                     ROLE_SYSTEM_DIALOG READONLY
     8                         ROLE_SYSTEM_STATICTEXT name='Text in the dialog.'
*    9                 ROLE_SYSTEM_SCROLLBAR
    10 <-- End-of-file -->

Actual
------
ROLE_SYSTEM_DOCUMENT READONLY FOCUSABLE
    IA2_ROLE_SECTION READONLY
        ROLE_SYSTEM_STATICTEXT name='Test for modal dialog opened in an iframe. '
        ROLE_SYSTEM_CLIENT FOCUSABLE
            IA2_ROLE_SCROLL_PANE
                ROLE_SYSTEM_DOCUMENT READONLY FOCUSABLE
                    ROLE_SYSTEM_DIALOG READONLY
                        ROLE_SYSTEM_STATICTEXT name='Text in the dialog.'

Original issue's description:
> Send AXLayoutComplete even when document hasn't finished loading yet.
> 
> Because these notifications only got sent if the corresponding AXObject
> already existed, they were effectively never sent before the first
> load complete notification. This change creates the AXObject for the
> document element after the first layout, allowing future notifications too.
> 
> BUG=353067
> 
> Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=170085

TBR=eseidel@chromium.org,esprehn@chromium.org,dmazzoni@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=353067

Review URL: https://codereview.chromium.org/213883003

git-svn-id: svn://svn.chromium.org/blink/trunk@170112 bbb929c8-8fbe-4397-9dbb-9b2b20218538

We want to be able to use <select multiple> with only touch.
Mouse behavior won't change.

BUG=356499

Review URL: https://codereview.chromium.org/211093004

git-svn-id: svn://svn.chromium.org/blink/trunk@170111 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Implemented ScrollableArea on PinchViewport and hooked it up
to all the layer scrolling mechanisms. This removes the need
for DistributeScrollOffsetToViewports on the layer tree.

Also did a little cleanup with member names and namespacing.

This is a 2-sided patch, the other CL:
https://codereview.chromium.org/210543002

BUG=349941

Review URL: https://codereview.chromium.org/199253002

git-svn-id: svn://svn.chromium.org/blink/trunk@170110 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This is a followup to the discussion on
https://codereview.chromium.org/213403004.

TBR=kouhei@chromium.org
BUG=356900

Review URL: https://codereview.chromium.org/213683005

git-svn-id: svn://svn.chromium.org/blink/trunk@170109 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=320334,315606

Review URL: https://codereview.chromium.org/213703002

git-svn-id: svn://svn.chromium.org/blink/trunk@170108 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Speculative fix for a user reported crash when a worker with an active IndexedDB
transaction was terminated. This stack is very similar to the one for 345354. I
do not have a reproduction for this bug was unable to verify that it fixes the
problem, but checking for a stopped context is still a good idea.

BUG=356146, 356171

Review URL: https://codereview.chromium.org/213073004

git-svn-id: svn://svn.chromium.org/blink/trunk@170107 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Reason for revert:
Original revert was too aggressive.  Failures are just local to the test, and the original patch fixed bugs.

Original issue's description:
> Revert of Move Event to oilpan's heap (https://codereview.chromium.org/208173004/)
> 
> Reason for revert:
> Seems likely to have caused a number of flaky failures in layout tests in oilpan builds:
> 
> Flakiness dashboard link: http://test-results.appspot.com/dashboards/flakiness_dashboard.html#group=%40ToT%20Blink&tests=http%2Ftests%2Fmedia%2Fmedia-source%2Fwebkitmediasource-closed-on-htmlmediaelement-destruction.html%2Csvg%2Fanimations%2Fsmil-leak-element-instances-noBaseValRef.svg%2Csvg%2Fanimations%2Fsmil-leak-element-instances.svg%2Csvg%2Fanimations%2Fsmil-leak-elements.svg
> 
> Example failure bot: http://build.chromium.org/p/chromium.webkit/builders/WebKit%20Linux%20Oilpan/builds/2770
> 
> Example failure:
> --- /mnt/data/b/build/slave/WebKit_Linux_Oilpan/build/layout-test-results/svg/animations/smil-leak-element-instances-noBaseValRef-expected.txt
> +++ /mnt/data/b/build/slave/WebKit_Linux_Oilpan/build/layout-test-results/svg/animations/smil-leak-element-instances-noBaseValRef-actual.txt
> @@ -1 +1 @@
> -PASS
> +FAIL: -3 extra live node(s)
> 
> 
> 
> 
> Original issue's description:
> > Move Event to oilpan's heap
> > 
> > Now there is no test failing when I change RefCounted<Event> to RefCountedGarbageCollected<Event>. Thus this CL does the change.
> > 
> > BUG=340522
> > 
> > Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=169941
> 
> TBR=oilpan-reviews@chromium.org,ager@chromium.org,wibling@chromium.org,haraken@chromium.org
> NOTREECHECKS=true
> NOTRY=true
> BUG=340522
> 
> Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=170092

TBR=oilpan-reviews@chromium.org,ager@chromium.org,wibling@chromium.org,haraken@chromium.org,sigbjornf@opera.com
NOTREECHECKS=true
NOTRY=true
BUG=340522

Review URL: https://codereview.chromium.org/213803002

git-svn-id: svn://svn.chromium.org/blink/trunk@170106 bbb929c8-8fbe-4397-9dbb-9b2b20218538

NOTE: this requires Skia change
https://codereview.chromium.org/211103006/ to land and roll into
Chromium.

Pixel-moving filters (e.g., blur, morphology) don't work well with
matrices applying anything more than scale and translate. (We do a
best-effort, but it's really wrong.) To fix, we separate out only
the scaling and translation components from the CTM, process the filters
with that, and apply the remainder of the CTM as a final matrix
transform image filter.

Since the matrix image filter is a generalization of the resize image
filter, we switch the filterRes implementation over to use that as well.

Covered by svg/batik/filters/feTile.svg and svg/filters/filterRes.svg
when run with --enable-deferred-filters.

R=fmalita@chromium.org
BUG=353630

Review URL: https://codereview.chromium.org/208683008

git-svn-id: svn://svn.chromium.org/blink/trunk@170104 bbb929c8-8fbe-4397-9dbb-9b2b20218538

The logic for handling 2-finger swipes should be:
1. If a 2-finger swipe can scroll the content (or iframe), it should do so.
2. If the 2-finger swipe can cause history navigation, it should do so.
3. The 2-finger swipe should rubber-band the content view.

(2) is handled by the embedder, and (1 & 3) by the renderer. Right now, the
renderer is passed an IPC, and it tries to do (1), and then it tries to do (3).
I've added a new IPC so that there are flags that prevent the renderer from
performing (3) under specific circumstances.

There was an existing mechanism to determine whether the renderer should try
(3), but it was fragile, incorrectly used, and insufficient for newer features
in chrome - the user can cancel a history swipe without cancelling the gesture,
at which point rubber-banding should be possible again. I've simplified and
fixed the logic inside ScrollElasticityController.mm to reflect the desired
behavior.

I modified the IPC InputMsg_HandleInputEvent to include 2 additional bools:
can_rubberband_left and can_rubberband_right. Blink will attempt to consume the
event to scroll the content view, and will only attempt to rubberband if bools
allow it too.

This is part 1 of a two-part change. It only includes the IPC change. The
changes to ScrollElasticityController will land after I've updated Chromium to
use the new API.

BUG=321437

Review URL: https://codereview.chromium.org/197213011

git-svn-id: svn://svn.chromium.org/blink/trunk@170102 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Got rid of extensive argument lists for invalidation set features by
introducing a struct for them.

R=esprehn@chromium.org,chrishtr@chromium.org
BUG=

Review URL: https://codereview.chromium.org/212603002

git-svn-id: svn://svn.chromium.org/blink/trunk@170101 bbb929c8-8fbe-4397-9dbb-9b2b20218538

http://src.chromium.org/viewvc/blink?view=revision&revision=170066

BUG=353933
TBR=jchaffraix@chromium.org

Review URL: https://codereview.chromium.org/213553002

git-svn-id: svn://svn.chromium.org/blink/trunk@170100 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Can't preemptively mark Mac as flaky, because it needs a rebaseline.  :(

TBR=jchaffraix@chromium.org
BUG=356834

Review URL: https://codereview.chromium.org/212623005

git-svn-id: svn://svn.chromium.org/blink/trunk@170099 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Switch to throwing TypeError instead of TypeMismatchError when encountering
non-finite doubles, following what WebIDL requires

 http://heycam.github.io/webidl/#es-double

R=acolwell@chromium.org
BUG=353245

Review URL: https://codereview.chromium.org/212953004

git-svn-id: svn://svn.chromium.org/blink/trunk@170098 bbb929c8-8fbe-4397-9dbb-9b2b20218538

    
For rects/ellipses with width or height of 0, getBBox() returns the
bounding box as being empty. This patch fixes this bug.
    
It ports and extends one initial patch, fixing the isEmpty() check,
by Changhun Kang <temoochin@company100.net>
    
On top of Changhun's patch, this patch addresses the case where
rects/ellipses fall back to renderSVGShape and their fill bounding box
should not fall back.
    
In detail, this patch involves the following:
    
- Patch for the bug (RenderSVGEllipse, RenderSVGRect and SVGPathData)
- Three new test cases for rect, ellipse and circle
- Current layout tests updated under platform/linux
- Updated TestExpectations

BUG=140472

Review URL: https://codereview.chromium.org/208323007

git-svn-id: svn://svn.chromium.org/blink/trunk@170097 bbb929c8-8fbe-4397-9dbb-9b2b20218538

SaveLayer() can optionally apply a clip. In that case we need to make
sure that we're saving the current clip, to avoid tainting it.

This was discovered while experimenting with SaveFlags removal and
requires a minor test rebaseline.

R=schenney@chormium.org,senorblanco@chromium.org

Review URL: https://codereview.chromium.org/212993004

git-svn-id: svn://svn.chromium.org/blink/trunk@170096 bbb929c8-8fbe-4397-9dbb-9b2b20218538

TBR=kouhei@chromium.org
BUG=356828

Review URL: https://codereview.chromium.org/212953009

git-svn-id: svn://svn.chromium.org/blink/trunk@170095 bbb929c8-8fbe-4397-9dbb-9b2b20218538

We've already shipped '::content'. It's time to unship '::-webkit-distributed'.

The usage of CSSPseudoElementPrefixedDistributed is '~0%'.
http://www.chromestatus.com/metrics/feature/timeline/popularity/99

BUG=268236

Review URL: https://codereview.chromium.org/212123002

git-svn-id: svn://svn.chromium.org/blink/trunk@170094 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Blink supported :host-context without a functional notation as an equivalent
to :host. That is not allowed according to the CSS Scoping spec.

BUG=356582

Review URL: https://codereview.chromium.org/212513002

git-svn-id: svn://svn.chromium.org/blink/trunk@170093 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Reason for revert:
Seems likely to have caused a number of flaky failures in layout tests in oilpan builds:

Flakiness dashboard link: http://test-results.appspot.com/dashboards/flakiness_dashboard.html#group=%40ToT%20Blink&tests=http%2Ftests%2Fmedia%2Fmedia-source%2Fwebkitmediasource-closed-on-htmlmediaelement-destruction.html%2Csvg%2Fanimations%2Fsmil-leak-element-instances-noBaseValRef.svg%2Csvg%2Fanimations%2Fsmil-leak-element-instances.svg%2Csvg%2Fanimations%2Fsmil-leak-elements.svg

Example failure bot: http://build.chromium.org/p/chromium.webkit/builders/WebKit%20Linux%20Oilpan/builds/2770

Example failure:
--- /mnt/data/b/build/slave/WebKit_Linux_Oilpan/build/layout-test-results/svg/animations/smil-leak-element-instances-noBaseValRef-expected.txt
+++ /mnt/data/b/build/slave/WebKit_Linux_Oilpan/build/layout-test-results/svg/animations/smil-leak-element-instances-noBaseValRef-actual.txt
@@ -1 +1 @@
-PASS
+FAIL: -3 extra live node(s)




Original issue's description:
> Move Event to oilpan's heap
> 
> Now there is no test failing when I change RefCounted<Event> to RefCountedGarbageCollected<Event>. Thus this CL does the change.
> 
> BUG=340522
> 
> Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=169941

TBR=oilpan-reviews@chromium.org,ager@chromium.org,wibling@chromium.org,haraken@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=340522

Review URL: https://codereview.chromium.org/213403004

git-svn-id: svn://svn.chromium.org/blink/trunk@170092 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=353406
TEST=fast/dom/shadow/style-and-shadow-element.html

Review URL: https://codereview.chromium.org/210953003

git-svn-id: svn://svn.chromium.org/blink/trunk@170091 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Prior to this CL, we were recomputing the 3D rendering context root for every
RenderLayer every frame. That work accounted for 13.5% of
GraphicsLayerUpdater::rebuildTree and 2.4% of the entire compositing update.
This work is almost entirely wasted because the 3D rendering context is
completely determined by style information and therefore can change only when
style changes.

Elliott and I looked at integrating this computation with recalc style, but
upon studying the code closely, I realize that this function doesn't actually
walk up the tree very far. Instead, it looks only at the enclosing render
layer of the containing block. That means we don't need to cache this data on
RenderLayer at all. Instead, we can recompute it when needed by the one caller:
CompositedLayerMapping::updateRenderingContext. That caller only cares about
RenderLayers that actually have a composited layer mapping, which is vastly
fewer than all RenderLayers. Moreover, that caller already understands
dirtiness caused by style changes.

Review URL: https://codereview.chromium.org/209003014

git-svn-id: svn://svn.chromium.org/blink/trunk@170090 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Review URL: https://codereview.chromium.org/197283042

git-svn-id: svn://svn.chromium.org/blink/trunk@170089 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Merge FontPlatformDataWin.h with FontPlatformDataHarfBuzz.h and use the
same version of the FontPlatformData class defintion/header on Windows
as on linux, chromeos and android.

R=dglazkov@chromium.org
BUG=355806

Review URL: https://codereview.chromium.org/205343010

git-svn-id: svn://svn.chromium.org/blink/trunk@170088 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This macro is defined but never used, and appears to be a vestige of
landing https://bugs.webkit.org/show_bug.cgi?id=109369 that was never
cleaned up.

BUG=

Review URL: https://codereview.chromium.org/212733002

git-svn-id: svn://svn.chromium.org/blink/trunk@170087 bbb929c8-8fbe-4397-9dbb-9b2b20218538

    
    Fix a ChickenAndEgg issue by delaying scroll until the
    next paint when the compositing layers are up to date.
    The amount scrolled but not painted is stored in
    ScrollView::m_pendingScrollDelta.
    
    Rearrange the code such that if there are fixed position
    elements, do the following before painting:
    
    1. Walk the tree and do updateLayerPosition
    2. Mark that we need to update compositing layers since
       layers may have changed as a result of scrolling if
         there are fixed elements.
    3. Update compositing layers.
    4. Invalidate fixed position elements.
    5. Update repaint rects for fixed position elements to
       reflect the new scroll position.
    
    Also, when calculating repaint rects for fixed elements,
    take into account both the scroll position and the
    scroll delta. This way we can invalidate the previous
    scroll position for the fixed element if it is
    promoted to a layer as a result of scrolling.

BUG=343766
BUG=343767

Review URL: https://codereview.chromium.org/191693002

git-svn-id: svn://svn.chromium.org/blink/trunk@170086 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Because these notifications only got sent if the corresponding AXObject
already existed, they were effectively never sent before the first
load complete notification. This change creates the AXObject for the
document element after the first layout, allowing future notifications too.

BUG=353067

Review URL: https://codereview.chromium.org/199693003

git-svn-id: svn://svn.chromium.org/blink/trunk@170085 bbb929c8-8fbe-4397-9dbb-9b2b20218538

TBR=caseq@chromium.org
BUG=344238

Review URL: https://codereview.chromium.org/213263003

git-svn-id: svn://svn.chromium.org/blink/trunk@170084 bbb929c8-8fbe-4397-9dbb-9b2b20218538

TBR=japhet@chromium.org
BUG=356777

Review URL: https://codereview.chromium.org/213173004

git-svn-id: svn://svn.chromium.org/blink/trunk@170082 bbb929c8-8fbe-4397-9dbb-9b2b20218538

The <compound-selector> argument may not be empty according to spec.

BUG=356578

Review URL: https://codereview.chromium.org/212383002

git-svn-id: svn://svn.chromium.org/blink/trunk@170081 bbb929c8-8fbe-4397-9dbb-9b2b20218538

mediaPlayerRequestFullscreen() is only ever called from Chromium's
BrowserMediaPlayerManager::OnProtectedSurfaceRequested() when the
--disable-gesture-requirement-for-media-fullscreen flag is used.

Make this behavior explicit, limiting it to just this case instead of
disabling it even for script-initiated fullscreen requests.

This will allow removing the mediaFullscreenRequiresUserGesture setting
once Chromium has been updated.

BUG=355439
TEST=LayoutTests/media/media-player-request-fullscreen.html

Review URL: https://codereview.chromium.org/209693003

git-svn-id: svn://svn.chromium.org/blink/trunk@170080 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Generate a user gesture when drop event is fired. This is useful for extensions
to call some user gesture based APIs during a drop event. Please see the bug
for a sample extension.

R=adamk@chromium.org,abarth@chromium.org
BUG=355269

Review URL: https://codereview.chromium.org/208853006

git-svn-id: svn://svn.chromium.org/blink/trunk@170079 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Clean up blink_platform.gypi and remove references to files that are
no longer part of the build.

TBR=leviw@chromium.org
BUG=313423

Review URL: https://codereview.chromium.org/213163002

git-svn-id: svn://svn.chromium.org/blink/trunk@170078 bbb929c8-8fbe-4397-9dbb-9b2b20218538

https://groups.google.com/a/chromium.org/forum/#!topic/blink-dev/GzMh_HfjUVc

This method is removed is from SVG2, uses the deprecated CSSValue and use counter data indicates it is little or not used.

Review URL: https://codereview.chromium.org/196173020

git-svn-id: svn://svn.chromium.org/blink/trunk@170077 bbb929c8-8fbe-4397-9dbb-9b2b20218538

TBR=dgozman
BUG=356743

Review URL: https://codereview.chromium.org/212883003

git-svn-id: svn://svn.chromium.org/blink/trunk@170076 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Enabling DirectWrite requires the sandbox to have knowledge of whether
DirectWrite or GDI is used for font rendering, so it's now decided
in the browser. Don't do fallback behavior in the renderer any more.

Browser side is https://codereview.chromium.org/209163002/ .

R=jamesr@chromium.org
BUG=333029

Review URL: https://codereview.chromium.org/210243004

git-svn-id: svn://svn.chromium.org/blink/trunk@170075 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Since there's no reason that the MediaQueryTokenizer's codepoints array would be created dynamically, I've created it here as an initialized array.
I used the build scripts to do that.



BUG=

Review URL: https://codereview.chromium.org/209503003

git-svn-id: svn://svn.chromium.org/blink/trunk@170074 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=285976

Review URL: https://codereview.chromium.org/210833004

git-svn-id: svn://svn.chromium.org/blink/trunk@170073 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This CL adds an intra-priority level sorting value for finer grained
control of resource loading. The blink side of this patch sorts images by
estimated on screen real-estate, preferring to download images in order of
decreasing visual impact. Ie. On a page with a large banner, and dozens of
small icons, the large banner will most likely be scheduled ahead of the
others.

In the future, we can also use this to experiment with other signals, ie.
prioritize out of viewport images by their distance to viewport.

Will be followed up by https://codereview.chromium.org/146333004/.

In our tests, this was a nice win on speed-index for desktop, with a gain of 
about ~2-3%. 

This is first of 3 sided patch. Following pattern suggested on chromium-dev:

"
1) Add the new WebKit interface, preserving the old.
2) Change Chrome to use the new interface.
3) Modify WebKit to remove the old interface.
"

Review URL: https://codereview.chromium.org/153983004

git-svn-id: svn://svn.chromium.org/blink/trunk@170072 bbb929c8-8fbe-4397-9dbb-9b2b20218538

WebGL spec requires that getExtension "Returns an object if, and only if,
name is an ASCII case-insensitive match [HTML] for one of the names
returned from getSupportedExtensions; otherwise, returns null." Respect
this requirement by always returning both unprefixed and prefixed names
from getSupportedExtensions in case they are supported by getExtension.

Functionality that is common to both getExtension and
getSupportedExtensions is refactored to separate helper functions so that
maintaining this becomes easier.

BUG=354454
TEST=WebGL conformance tests
R=bajones@chromium.org, kbr@chromium.org

Review URL: https://codereview.chromium.org/205903004

git-svn-id: svn://svn.chromium.org/blink/trunk@170071 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=356742
TBR=fs@opera.com

Review URL: https://codereview.chromium.org/213153004

git-svn-id: svn://svn.chromium.org/blink/trunk@170070 bbb929c8-8fbe-4397-9dbb-9b2b20218538