- 15 Aug, 2014 8 commits
-
-
dstockwell@chromium.org authored
Adds a category to ct-failure-group and enables ct-failure-stream to filter to a single category. Ports the previously broken ct-failure-stream-tests to mocha in order to add new tests. NOTRY=true BUG=400397,399967, 400387 Review URL: https://codereview.chromium.org/455263002 git-svn-id: svn://svn.chromium.org/blink/trunk@180314 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
wangxianzhu@chromium.org authored
http://src.chromium.org/viewvc/blink?view=revision&revision=180305 TBR=eae@chromium.org Review URL: https://codereview.chromium.org/464403004 git-svn-id: svn://svn.chromium.org/blink/trunk@180313 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
cbiesinger@chromium.org authored
Use label="" instead of the textContent R=ojan@chromium.org NOTRY=true Review URL: https://codereview.chromium.org/469423005 git-svn-id: svn://svn.chromium.org/blink/trunk@180312 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
dpranke@chromium.org authored
The file handler was getting confused when we would upload a full_results.json that contained a "results" key that pointed to a trie of tests. Also fix some file permissions issues that the presubmit check complained about. R=ojan@chromium.org BUG=403663 NOTRY=true Review URL: https://codereview.chromium.org/475503005 git-svn-id: svn://svn.chromium.org/blink/trunk@180310 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
trchen@chromium.org authored
This CL fixes a bug that non-composited scrolling div forgets to invalidate because the div believed all its descandants are composited, when actually some descendants paints to ancestor backing although composited. BUG=401047 Review URL: https://codereview.chromium.org/460883002 git-svn-id: svn://svn.chromium.org/blink/trunk@180309 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
tkent@chromium.org authored
BUG=403987 TBR=kouhei@chromium.org Review URL: https://codereview.chromium.org/472193002 git-svn-id: svn://svn.chromium.org/blink/trunk@180308 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
scottmg@chromium.org authored
R=jamesr@chromium.org BUG=354261 Review URL: https://codereview.chromium.org/473043002 git-svn-id: svn://svn.chromium.org/blink/trunk@180307 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
eae@chromium.org authored
Remove in-class platform ifdefs from platform/fonts/FontPlatformData.h as the entire file is wrapped in a OS(MACOSX) macro. R=leviw@chromium.org, pdr@chromium.org Review URL: https://codereview.chromium.org/471693002 git-svn-id: svn://svn.chromium.org/blink/trunk@180306 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
- 14 Aug, 2014 32 commits
-
-
eae@chromium.org authored
Implement support for the CSS font-stretch property on platforms using SkFontMgr. This allows the width/stretch of a given font to be specified and matches the IE and Firefox implementations. Currently SkFontMgr is only used on Windows however as other platforms migrate to SkFontMgr they'll get this support for free. Entry in Chromium Dashboard: http://www.chromestatus.com/admin/features/edit/4598830058176512 Intent to Implement and Ship: https://groups.google.com/a/chromium.org/forum/#!msg/blink-dev/N24I2zJemvI/ZClPu585uAsJ TEST=fast/text/font-stretch.html BUG=331119 R=dglazkov@chromium.org Review URL: https://codereview.chromium.org/476943003 git-svn-id: svn://svn.chromium.org/blink/trunk@180305 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
oysteine@chromium.org authored
R=abarth BUG= Review URL: https://codereview.chromium.org/472943002 git-svn-id: svn://svn.chromium.org/blink/trunk@180303 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
adamk@chromium.org authored
Now also timing out on Mac/Debug, and it's consistently timing out, not flakily. BUG=356957 TBR=mkwst@chromium.org Review URL: https://codereview.chromium.org/476843004 git-svn-id: svn://svn.chromium.org/blink/trunk@180301 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
rob.buis@samsung.com authored
Fix some leftovers after the WebCore -> blink namespace changes. NOTRY=true Review URL: https://codereview.chromium.org/477743003 git-svn-id: svn://svn.chromium.org/blink/trunk@180300 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
ch.dumez@samsung.com authored
Use new match Functor ElementTraversal API in DOMSiblingTraversalStrategy to simplify the code a bit. R=adamk@chromium.org, esprehn@chromium.org Review URL: https://codereview.chromium.org/475023002 git-svn-id: svn://svn.chromium.org/blink/trunk@180299 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
jl@opera.com authored
Representing array and sequence types this way, rather than via auxiliary flags on the member type, means exotic types like array-of-arrays, array-of-sequences, arrays-of-nullables and similar can be supported. More importantly, this also means that an array or sequence type can't be mistaken for its member type if you forget to check the "is array" and "is sequence" flags, and thus lets us remove a bunch of "is this an array or sequence type" checks from code that has nothing to do with supporting arrays or sequences. This patch doesn't change code generation. Review URL: https://codereview.chromium.org/470063003 git-svn-id: svn://svn.chromium.org/blink/trunk@180298 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
zerny@chromium.org authored
This CL adds GC memory profiling using the about:tracing infrastructure. The profiling can be time consuming, so the GC_PROFILE_* macros are used to selectively enable various profiling features for the time being. The GC_TRACING macro has been renamed to GC_PROFILE_MARKING to avoid confusion with about:tracing. The tracing event group has been renamed from blink to blink_gc to allow collection of GC related events independent of other blink events. BUG= Review URL: https://codereview.chromium.org/383743002 git-svn-id: svn://svn.chromium.org/blink/trunk@180297 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
timvolodine@chromium.org authored
Set Battery Status API status to stable in blink. We have implementations for all platforms supported by blink now, i.e. on Android, Windows, Linux, ChromeOS and Mac. BUG=122593 TEST=http://jsbin.com/battery-status-test Review URL: https://codereview.chromium.org/451583003 git-svn-id: svn://svn.chromium.org/blink/trunk@180296 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
skyostil@chromium.org authored
Shut down the scheduler before terminating the heap (i.e., PartitionAlloc). This is needed because the scheduler makes heap allocations that need to be freed at destruction time. BUG=391005 Review URL: https://codereview.chromium.org/475033002 git-svn-id: svn://svn.chromium.org/blink/trunk@180295 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
ojan@chromium.org authored
-A bunch of minor UI tweaks to get things to line up better. -Align the examine/snooze buttons vertically so they take up less space. -Make the tabs in ct-results-by-builder black and more tab-looking. -Change the name to Sheriff-o-matic now that it's clear we're sticking with it. -Add a useful links link to the right of the revisions. BUG=400411,400398 NOTRY=true Review URL: https://codereview.chromium.org/462363003 git-svn-id: svn://svn.chromium.org/blink/trunk@180294 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
timloh@chromium.org authored
The files CSSGrammar.y, CSSParserMode.{h,cpp}, CSSTokenizer{.h,-in.cpp}, CSSParserValues{.h,.cpp,Test.cpp} are all primarily for the parser so should live in core/css/parser (making a lexing vs parsing distinction here doesn't seem very useful). BUG=330389 Review URL: https://codereview.chromium.org/469183002 git-svn-id: svn://svn.chromium.org/blink/trunk@180293 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
vollick@chromium.org authored
This layout test demonstrates a case where they do not. BUG=403819 Review URL: https://codereview.chromium.org/471073002 git-svn-id: svn://svn.chromium.org/blink/trunk@180292 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
jsbell@chromium.org authored
The interface was given a less generic name in a recent spec update: https://github.com/slightlyoff/ServiceWorker/issues/350 BUG=403536 R=falken@chromium.org,haraken@chromium.org Review URL: https://codereview.chromium.org/463423003 git-svn-id: svn://svn.chromium.org/blink/trunk@180291 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
wangxianzhu@chromium.org authored
http://src.chromium.org/viewvc/blink?view=revision&revision=180279 BUG=394692 TBR=eae@chromium.org Review URL: https://codereview.chromium.org/471163002 git-svn-id: svn://svn.chromium.org/blink/trunk@180290 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
jl@opera.com authored
Some duplication had sneaked in as part of the IDL dictionary work, with IdlTypeBase.impl_should_use_nullable_container meaning almost the same thing as IdlTypeBase.is_explicit_nullable, except the latter also implies that the type is nullable. Common up the actual logic in the new IdlTypeBase.cpp_type_has_null_value property, and use it for both impl_should_use_nullable_container and is_{implicit,explicit}_nullable. Review URL: https://codereview.chromium.org/472803002 git-svn-id: svn://svn.chromium.org/blink/trunk@180287 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
wangxianzhu@chromium.org authored
http://src.chromium.org/viewvc/blink?view=revision&revision=180269 BUG=382809 TBR=yoichio@chromium.org Review URL: https://codereview.chromium.org/470253002 git-svn-id: svn://svn.chromium.org/blink/trunk@180285 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
hayato@chromium.org authored
BUG=402749 Review URL: https://codereview.chromium.org/473833002 git-svn-id: svn://svn.chromium.org/blink/trunk@180284 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
adamk@chromium.org authored
BUG=403818 TBR=tkent@chromium.org Review URL: https://codereview.chromium.org/471063003 git-svn-id: svn://svn.chromium.org/blink/trunk@180283 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
zerny@chromium.org authored
Oilpan: Ensure that left-most vtable is the first thing to be initialized for polymorphic classes with trace methods. This is done by ensuring one of two properties for any polymorphic class with a trace method. 1. If the trace method is virtual, then the left-most base class must define a virtual trace method too. 2. If the trace method is non-virtual, then the left-most base class must define some virtual method. The Blink GC plugin statically checks these properties. Review URL: https://codereview.chromium.org/463543002 git-svn-id: svn://svn.chromium.org/blink/trunk@180281 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
adamk@chromium.org authored
This newly-added test always crashes when run on Android. BUG=403804 TBR=abarth@chromium.org Review URL: https://codereview.chromium.org/472013002 git-svn-id: svn://svn.chromium.org/blink/trunk@180280 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
eae@chromium.org authored
Implement support for numeric font-weights (i.e. 100..900) on platforms using SkFontMgr by switching to the new matchFamilyStyle API instead of the old legacyCreateTypeface one. Currently SkFontMgr is only used on Windows. R=dglazkov@chromium.org BUG=394692 TEST=fast/text/font-weight.html Review URL: https://codereview.chromium.org/470443004 git-svn-id: svn://svn.chromium.org/blink/trunk@180279 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
timloh@chromium.org authored
The class CSSArrayFunctionValue was introduced a while back for custom filters which have since been removed, so the class is now unused. Review URL: https://codereview.chromium.org/472753002 git-svn-id: svn://svn.chromium.org/blink/trunk@180278 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
haraken@chromium.org authored
Currently, in oilpan builds, AudioNodeOutput::disconnectAll() is called in AudioNode::dispose(). In non-oilpan builds, AudioNodeOutput::disconnectAll() is called in AudioNode::finishDeref(). We don't want to have different call paths between oilpan builds and non-oilpan builds. Given that AudioNodeOutput::disconnectAll() should be called when the AudioNode is destructed (but before the associated AudioContext is gone), AudioNodeOutput::disconnectAll() should always be called in AudioNode::dispose(). BUG=340522 Review URL: https://codereview.chromium.org/467683002 git-svn-id: svn://svn.chromium.org/blink/trunk@180277 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
jl@opera.com authored
In v8_types.py and elsewhere, we extend the classes defined in idl_types.py by adding properties or methods. Usually, these properties or methods are supposed to be available on all type objects. Since all type objects now have a common base class, IdlTypeBase, prefer to set new properties and methods on it instead of on IdlType. This doesn't change CG, and is done to simplify the process of adding new IdlTypeBase sub-classes in the future. Review URL: https://codereview.chromium.org/469243002 git-svn-id: svn://svn.chromium.org/blink/trunk@180276 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
haraken@chromium.org authored
IdlDefinition, IdlOperation, IdlAttribute etc should know the name of an IDL file where they are defined. We want to pass these names to PrivateScriptRunner's methods so that PrivateScriptRunner can understand the names of JS files of the DOM attributes/methods that the PrivateScriptRunner is running. Binding componentization for partial interfaces also need this change. BUG=341031 Review URL: https://codereview.chromium.org/473893002 git-svn-id: svn://svn.chromium.org/blink/trunk@180275 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
sigbjornf@opera.com authored
R=haraken BUG=397473 Review URL: https://codereview.chromium.org/476663002 git-svn-id: svn://svn.chromium.org/blink/trunk@180274 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
eustas@chromium.org authored
Fix UI regression. See: https://codereview.chromium.org/474453002/ BUG= NOTRY=true Review URL: https://codereview.chromium.org/471703003 git-svn-id: svn://svn.chromium.org/blink/trunk@180273 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
sunil.ratnu@samsung.com authored
Cleanup blink:: prefix usage in Source/core/modules/[battery/*.cpp to indexeddb/*.cpp] - Merge multiple |namespace blink {}| blocks - Remove unnecessary blink:: prefixes - Make blank lines consistent BUG=None Review URL: https://codereview.chromium.org/464353002 git-svn-id: svn://svn.chromium.org/blink/trunk@180272 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
tkent@chromium.org authored
These are the last instances of namespace indentation in public/. BUG=none TEST=none; no behavior changes. Review URL: https://codereview.chromium.org/476633002 git-svn-id: svn://svn.chromium.org/blink/trunk@180271 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
yurys@chromium.org authored
All places where raw TimelineAgent.Record was used are fixed. BUG=403684,398785 Review URL: https://codereview.chromium.org/461323003 git-svn-id: svn://svn.chromium.org/blink/trunk@180270 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
yoichio@chromium.org authored
We do not need update caret rectangle synchronously because we just need updated caret in painting. Thus this CL delays computing new caret rectangle. Source/core/editing/FrameSelection.cpp: - Add ResetCaretBlinkOption to updateAppearance(). The option is set to ResetCaretBlink when it is called from setSelection. If ResetCaretBlink, we reset caret blinking. If we need to repaint caret, set |m_caretRectDirty| flag. If |m_caretRectDirty| is set, we call PageAnimator::scheduleVisualUpdate to trigger repaint. For range, create VisibleSelection without validation like HTMLTextFormControlElement::setSelectionRange. - Add the setCaretRectNeedsUpdate function to just set |m_caretRectDirty| flag and call new scheduleVisualUpdate function, which calls PageAnimator::scheduleVisualUpdate. - FrameSelection::invalidateCaretRect does 1. Checks the dirty flag. 2. Gets new caret rectangle and node which has the caret renderer. 3. if caret is changed, invalidate the new caret rect and the old caret rect. 4. Caches the new caret rect and node. Sets dirty flag off. This function is similar to old recomputeCaretRect(deleted). - Delete unused updateRenderTreeIfneeded() from notifyRenderOfSelectionChange(). In old days, notifyRenderOfSelectionChange() used renderer. - Refactor FrameSelection::absoluteCaretBounds() to call updateCaretRect directly. Since we update layout, ASSERT that document's lifecycle is not in InPaintInvalidation. Source/core/editing/Caret.cpp: - Remove document->updateRenderTreeIfNeeded() from CaretBase::updateCaretRect and call it from caller. - In FrameSelection::paintCaret, call without updateRenderTreeIfNeeded because the tree must be updated on the painting phase. Source/core/frame/FrameView.cpp: - FrameView::invalidateTreeIfNeeded calls FrameSelection::InvalidateCaretRect to invalidate caret rect for each frame. Source/core/html/HTMLTextFormControlElementTest.cpp: - Add new test to confirm setSelectionRange with another start/end not cause layout. - Delete FrameSelectionLocalCaretRectDoesNotCauseLayout because localCaretRect is removed. Layouttests: - This cl reduces redundant RenderObject::invalidatePaintRectangle calls so we need to delete lines from expected.txt but result images are still same. BUG=382809, 397303, 403317 TEST=Layouttests/fast/forms/textarea-scrollbar.html, textarea-scrolled-type.html Review URL: https://codereview.chromium.org/431983005 git-svn-id: svn://svn.chromium.org/blink/trunk@180269 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-
eustas@chromium.org authored
Source/devtools/front_end/inspectorStyle.css:1891 hide-disclosure-buttons Never used. Source/devtools/front_end/timelinePanel.css:358 timeline-loading There is no aggregates stats in popover anymore. Source/devtools/front_end/elementsPanel.css:648 css-font-value https://codereview.chromium.org/59993003 Source/devtools/front_end/inspectorStyle.css:1821 console-filter-top https://bugs.webkit.org/show_bug.cgi?id=68155 BUG= Review URL: https://codereview.chromium.org/471873002 git-svn-id: svn://svn.chromium.org/blink/trunk@180268 bbb929c8-8fbe-4397-9dbb-9b2b20218538
-