GitLab

http://src.chromium.org/viewvc/blink?view=revision&revision=176016

BUG=335075
TBR=habib.virji@samsung.com

Review URL: https://codereview.chromium.org/334813002

git-svn-id: svn://svn.chromium.org/blink/trunk@176076 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Comment-only change.

An existing comment explains some code as saying "necessary
for MSVC", but it's actually required by the C++ standard.

This CL corrects the comment and explains.

R=haraken

Review URL: https://codereview.chromium.org/331783003

git-svn-id: svn://svn.chromium.org/blink/trunk@176075 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=341119

Review URL: https://codereview.chromium.org/328933002

git-svn-id: svn://svn.chromium.org/blink/trunk@176074 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Use ScriptWrappable::init less, it is only needed for most derived classes usually, but these SVG classes are not exposed as JS type.

Review URL: https://codereview.chromium.org/328703005

git-svn-id: svn://svn.chromium.org/blink/trunk@176073 bbb929c8-8fbe-4397-9dbb-9b2b20218538

RuntimeEnabledFeatures::overlayFullscreenVideoEnabled is true and have a touch
handler on the document or body.

We had an optimization in the cc touch hit rect calculation that if there was a
handler on the document or body then we'd short circuit the computation and
report a single rect at the root layer covering the whole page. Now when making
an HTML5 video element fullscreen with this flag turned on the
RenderLayerCompositor removes the root cc::layer and reattaches the video layer
causing the compositor to think that there are no handlers registered.


BUG=372314

Review URL: https://codereview.chromium.org/317373007

git-svn-id: svn://svn.chromium.org/blink/trunk@176071 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This CL remove the RenderGeometryMap from updateLayerPositionsAfterLayout. The
only remaining use of the geometry map was to compute an argument for
RenderLayerScrollableArea::positionOverflowControls. However, it turns out it
doesn't matter what value we pass to positionOverflowControls at this point. By
removing the RenderGeometryMap, this CL makes updateLayerPositionsAfterLayout
twice as fast.

Historically, the overflow controls were backed by NSViews, which meant they
needed to be positioned in absolute space by the widget tree. Now, however,
they're either drawn by Blink or composited by cc. If the overflow controls
are painted by Blink, we're reposition them immediately before painting. If the
overflow controls are composited, we'll subtract out the offsetFromRoot before
passing the values to cc. In both cases, it doesn't matter care what value we
pass here.

We should eventually remove this function call from this codepath. This isn't
the correct codepath in which to position the overflow controls. However, we
can save that work for a future CL.

R=esprehn@chromium.org
BUG=383636,382548

Review URL: https://codereview.chromium.org/335743002

git-svn-id: svn://svn.chromium.org/blink/trunk@176070 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This is as per the the coding guidelines for blink, which disallows the
use of "using" declarations of any kind to import names in the C++ Standard
library.

This is in continuation of the CL https://codereview.chromium.org/329183002/
Effected Folders: Source/core/html/

BUG=NONE

Review URL: https://codereview.chromium.org/334593005

git-svn-id: svn://svn.chromium.org/blink/trunk@176069 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This CL takes care the dom, editing, events subdirs under core.

BUG=381876

Review URL: https://codereview.chromium.org/332933002

git-svn-id: svn://svn.chromium.org/blink/trunk@176068 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Removes uses of 192-bit AES keys from the LayoutTests. Since once the corresponding Chromium change lands they will no longer work.

BUG=381829

Review URL: https://codereview.chromium.org/325323004

git-svn-id: svn://svn.chromium.org/blink/trunk@176066 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Don't expose the user agent shadow tree used for the cloned <use> tree. Add one test for the attribute event listener code path (onclick) and one for addEventListener API. To make sure the shadow tree is not exposed, make sure we always wrap non-shadow tree SVG Elements.

BUG=313438

Review URL: https://codereview.chromium.org/312423002

git-svn-id: svn://svn.chromium.org/blink/trunk@176065 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This CL removes WTF_MAKE_FAST_ALLOCATED in TimerBase for Oilpan because
new/delete defined by it conflict with new/delete defined by GarbageCollected in
DOMWindowEventQueueTimer.

BUG=

Review URL: https://codereview.chromium.org/331623002

git-svn-id: svn://svn.chromium.org/blink/trunk@176064 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Also, remove remaining RefPtr<MIDIPort>, the last user of one of the converters.

BUG=357163

Review URL: https://codereview.chromium.org/331493007

git-svn-id: svn://svn.chromium.org/blink/trunk@176063 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=377941,245025

Review URL: https://codereview.chromium.org/310513004

git-svn-id: svn://svn.chromium.org/blink/trunk@176062 bbb929c8-8fbe-4397-9dbb-9b2b20218538

dllexport/dllimport aren't inherited by nested classes like visibility
attributes are on Posix, and this class has some non-inline methods that
need to get exported.

For example, if ~BitVector gets inlined, we might try calling
OutOfLineBits::destroy across a dll boundry.

BUG=82385
TEST=attempt to build blink_web.dll with Clang on Windows

Review URL: https://codereview.chromium.org/338463002

git-svn-id: svn://svn.chromium.org/blink/trunk@176061 bbb929c8-8fbe-4397-9dbb-9b2b20218538

The call to setMayNeedPaintInvalidation was compensating for a bug in RenderBox::clippedOverflowRectForPaintInvalidation.
Previously, that code considered only whether the current RenderLayer had visible
contents instead of asking whether any descendant was visible.  This change will
cause us to issue more invalidations in cases where some parts of the subtree are
visible, but they'll be the same ones we issue in the normal case when elements
are visible.

Removing this call will let us remove the RenderGeometryMap in
updateLayerPositionsAfterLayout because it's now almost entirely unused.

R=esprehn@chromium.org
BUG=383636

Review URL: https://codereview.chromium.org/337593002

git-svn-id: svn://svn.chromium.org/blink/trunk@176060 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=379271
R=rbyers@chromium.org

Review URL: https://codereview.chromium.org/329393006

git-svn-id: svn://svn.chromium.org/blink/trunk@176059 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This function was crashing if the FrameTree name lookup failed to find
a matching frame.

TBR=japhet@chromium.org
BUG=383845

Review URL: https://codereview.chromium.org/331633004

git-svn-id: svn://svn.chromium.org/blink/trunk@176058 bbb929c8-8fbe-4397-9dbb-9b2b20218538

They don't really belong in their current headers, and it's confusing.
For example, RenderInfo is defined in LineBreaker.h and its constructor
and destructor are defined in RenderBlockLineLayout.cpp.

Review URL: https://codereview.chromium.org/329853006

git-svn-id: svn://svn.chromium.org/blink/trunk@176057 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=341119

Review URL: https://codereview.chromium.org/324183002

git-svn-id: svn://svn.chromium.org/blink/trunk@176056 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This is in preparation for a fix to V8

BUG=383596

Review URL: https://codereview.chromium.org/329403002

git-svn-id: svn://svn.chromium.org/blink/trunk@176055 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Adjust the width of each column for the fast/multicol/newmulticol/
breaks-3-columns-3.html test so to allow the slightly wider text
on Windows with DirectWrite.

TBR=dglazkov@chromium.org
BUG=378610

Review URL: https://codereview.chromium.org/330693002

git-svn-id: svn://svn.chromium.org/blink/trunk@176054 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Change the fast/text/international/complex-text-missing-glyph-fallback
.html test to have a fixed line height to account for differences
between fonts.

BUG=378610
TBR=leviw@chromium.org

Review URL: https://codereview.chromium.org/336603004

git-svn-id: svn://svn.chromium.org/blink/trunk@176053 bbb929c8-8fbe-4397-9dbb-9b2b20218538

LayoutState keeps track of a handful of fields, but the most
important part is providing a fast-mapping between renderers
and their repaint container. For parts where the logic doesn't
work, we disable LayoutState, which causes us to fall back on
walking up the tree and slowly calculating our rects.

Renaming LayoutStateDisabler, which sounds relatively innocuous
and which most people have no idea of the implications of to
ForceHorriblySlowRectMapping, with the hopes of striking fear
into any dev or reviewer who hopes to introduce one.

Review URL: https://codereview.chromium.org/333743002

git-svn-id: svn://svn.chromium.org/blink/trunk@176052 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Text layout in SVG is assumed to happen in unzoomed coordinates.
Prior to this patch, there was a special SVG mode in the
StyleResolver that disregarded zoom when resolving lengths for
SVG elements. That broke down during inheritance though, so a
better way to handle the SVG zoom model is for the SVG module to
remove zoom from style values itself.

BUG=382366

Review URL: https://codereview.chromium.org/323053002

git-svn-id: svn://svn.chromium.org/blink/trunk@176051 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Removed StyleResolverChange class by:

- Marking Document scope for recalc in the stylesheet collection class like
  we do for shadow scopes instead of propagating it up to Document.

- Move the check for doing full repaint after all stylesheets have been
  loaded to Document::styleResolverChanged.

Introduced a shouldUpdateDocumentStyleSheetCollection helper method.

R=esprehn@chromium.org,ojan@chromium.org

Review URL: https://codereview.chromium.org/330043002

git-svn-id: svn://svn.chromium.org/blink/trunk@176050 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Clean up SVGNames usage in core/rendering, sometimes the header is included but it is not used.
In some other spots, prefer "using namespace SVG".

BUG=

Review URL: https://codereview.chromium.org/332773002

git-svn-id: svn://svn.chromium.org/blink/trunk@176049 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=383594

Review URL: https://codereview.chromium.org/327243005

git-svn-id: svn://svn.chromium.org/blink/trunk@176048 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=371611

Review URL: https://codereview.chromium.org/320393003

git-svn-id: svn://svn.chromium.org/blink/trunk@176047 bbb929c8-8fbe-4397-9dbb-9b2b20218538

We only need to update the ScrollingCoordinator once at the root of the frame
tree. We shouldn't be poking at it at every level of the recursion.

This CL finally achieves my dream of controlling
RenderLayerCompositor::updateIfNeeded with just the CompositingUpdateType enum.

Review URL: https://codereview.chromium.org/330513002

git-svn-id: svn://svn.chromium.org/blink/trunk@176046 bbb929c8-8fbe-4397-9dbb-9b2b20218538

V8 has GN build files now, we don't need this stub any more.

Review URL: https://codereview.chromium.org/329813002

git-svn-id: svn://svn.chromium.org/blink/trunk@176045 bbb929c8-8fbe-4397-9dbb-9b2b20218538

TBR=yurys@chromium.org, scroggo@google.com
BUG=361045

Review URL: https://codereview.chromium.org/336513006

git-svn-id: svn://svn.chromium.org/blink/trunk@176044 bbb929c8-8fbe-4397-9dbb-9b2b20218538

When resolving a protocol-agnostic URL against a base of 'about:blank',
the result would be an invalid (and empty) URL, because of missing special-
casing of 'about:blank' - which is present in
Document::completeURLWithOverride.
Fix by modifying SVGImageLoader::sourceURI to always resolve the URL using
Document::completeURLWithOverride to not miss out on this special-casing
logic for 'about:blank' base URLs.

BUG=379100

Review URL: https://codereview.chromium.org/329943003

git-svn-id: svn://svn.chromium.org/blink/trunk@176043 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Blink side of https://codereview.chromium.org/299043003/

BUG=354697

Review URL: https://codereview.chromium.org/310623004

git-svn-id: svn://svn.chromium.org/blink/trunk@176042 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=383749
TEST=none
R=dglazkov@chromium.org, hayato@chromium.org

Review URL: https://codereview.chromium.org/330543007

git-svn-id: svn://svn.chromium.org/blink/trunk@176041 bbb929c8-8fbe-4397-9dbb-9b2b20218538

The behavior once invoked by the SkPicture::kUsePathBoundsForClip_RecordingFlag is now the default so this flag is going away.
Skia 9f1c241e (Remove SkPicture::kUsePathBoundsForClip_RecordingFlag - https://codereview.chromium.org/316143003/)

Review URL: https://codereview.chromium.org/326953002

git-svn-id: svn://svn.chromium.org/blink/trunk@176040 bbb929c8-8fbe-4397-9dbb-9b2b20218538

These are always equivalent since the WinCE case was dropped over a year ago:
https://chromiumcodereview.appspot.com/13901012

BUG=383594

Review URL: https://codereview.chromium.org/325283002

git-svn-id: svn://svn.chromium.org/blink/trunk@176039 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=none
R=ojan@chromium.org,atwilson@chromium.org

Review URL: https://codereview.chromium.org/329873002

git-svn-id: svn://svn.chromium.org/blink/trunk@176037 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This CL takes care the accessibility, animation and css subdirs under core. There's one exception that css/CSSTokenizer-in.cpp didn't change because of a coding style issue.

BUG=381876

Review URL: https://codereview.chromium.org/331633003

git-svn-id: svn://svn.chromium.org/blink/trunk@176036 bbb929c8-8fbe-4397-9dbb-9b2b20218538

The height of text and form control can vary slightly depending on
alignment and baseline. Explicitly set height for CSS tests that do not
test nor depend on text rendering.

BUG=378610
R=dglazkov@chromium.org, eseidel@chromium.org

Review URL: https://codereview.chromium.org/323293002

git-svn-id: svn://svn.chromium.org/blink/trunk@176035 bbb929c8-8fbe-4397-9dbb-9b2b20218538

RenderStyle objects are not on the heap and they keep the CSSFontSelector
and its associated ResourceFetcher alive. We turned raw pointers to the
document in CSSFontSelector and ResourceFetcher into traced Members. That
will leak until we can also trace through the RenderStyle structure.

See crbug.com/383860 for details.

This fixed the leaks by using WeakMembers to the document in these two
places. We really should trace through all of this but that is going
to be a larger change and we should fix this leak now.

R=haraken@chromium.org, oilpan-reviews@chromium.org, sigbjornf@opera.com
BUG=383860

Review URL: https://codereview.chromium.org/335633002

git-svn-id: svn://svn.chromium.org/blink/trunk@176034 bbb929c8-8fbe-4397-9dbb-9b2b20218538