Source/WebCore/ChangeLog

   1 2012-07-25  Andrey Adaikin  <aandrey@chromium.org>
   2
   3         Web Inspector: Add more jsdocs to InjectedScriptSource.js
   4         https://bugs.webkit.org/show_bug.cgi?id=92255
   5
   6         Reviewed by Pavel Feldman.
   7
   8         * inspector/InjectedScriptSource.js:
   9
  10 2012-07-24  Dana Jansens  <danakj@chromium.org>
  11
  12         [chromium] Turn the debug HUD into a layer so that it can be drawn as a WebCompositorQuad
  13         https://bugs.webkit.org/show_bug.cgi?id=92182
  14
  15         Reviewed by Adrienne Walker.
  16
  17         By changing the HUD into a quad, we are able to create a HUD and pass
  18         it to the host compositor, allowing the HUD to be used for renderers
  19         under Ubercompositor.
  20
  21         This also removes the special drawing code for the HUD from
  22         LayerRendererChromium, allowing us to move finishDrawingFrame
  23         into LayerRendererChromium as a private method and remove it
  24         from the CCRenderer interface.
  25
  26         Finally, we can remove the CCFontAtlas transport code from CCProxy
  27         and its implementations as the atlas can be synced to impl thread
  28         through the usual layer commit mechanism.
  29
  30         The debug HUD is not covered by tests.
  31
  32         * WebCore.gypi:
  33         * platform/graphics/chromium/HeadsUpDisplayLayerChromium.cpp: Added.
  34         (WebCore):
  35         (WebCore::HeadsUpDisplayLayerChromium::create):
  36         (WebCore::HeadsUpDisplayLayerChromium::HeadsUpDisplayLayerChromium):
  37         (WebCore::HeadsUpDisplayLayerChromium::~HeadsUpDisplayLayerChromium):
  38         (WebCore::HeadsUpDisplayLayerChromium::createCCLayerImpl):
  39         * platform/graphics/chromium/HeadsUpDisplayLayerChromium.h: Added.
  40         (WebCore):
  41         (HeadsUpDisplayLayerChromium):
  42         * platform/graphics/chromium/LayerRendererChromium.cpp:
  43         (WebCore::LayerRendererChromium::drawFrame):
  44         (WebCore::LayerRendererChromium::cleanupSharedObjects):
  45         * platform/graphics/chromium/LayerRendererChromium.h:
  46         (LayerRendererChromium):
  47         * platform/graphics/chromium/cc/CCDebugRectHistory.cpp:
  48         (WebCore::CCDebugRectHistory::savePropertyChangedRects):
  49         * platform/graphics/chromium/cc/CCDebugRectHistory.h:
  50         * platform/graphics/chromium/cc/CCHeadsUpDisplayLayerImpl.cpp: Renamed from Source/WebCore/platform/graphics/chromium/cc/CCHeadsUpDisplay.cpp.
  51         (WebCore):
  52         (WebCore::CCHeadsUpDisplayLayerImpl::CCHeadsUpDisplayLayerImpl):
  53         (WebCore::CCHeadsUpDisplayLayerImpl::~CCHeadsUpDisplayLayerImpl):
  54         (WebCore::CCHeadsUpDisplayLayerImpl::willDraw):
  55         (WebCore::CCHeadsUpDisplayLayerImpl::appendQuads):
  56         (WebCore::CCHeadsUpDisplayLayerImpl::didDraw):
  57         (WebCore::CCHeadsUpDisplayLayerImpl::didLoseContext):
  58         (WebCore::CCHeadsUpDisplayLayerImpl::drawHudContents):
  59         (WebCore::CCHeadsUpDisplayLayerImpl::drawFPSCounter):
  60         (WebCore::CCHeadsUpDisplayLayerImpl::drawFPSCounterText):
  61         (WebCore::CCHeadsUpDisplayLayerImpl::drawDebugRects):
  62         * platform/graphics/chromium/cc/CCHeadsUpDisplayLayerImpl.h: Renamed from Source/WebCore/platform/graphics/chromium/cc/CCHeadsUpDisplay.h.
  63         (WebCore):
  64         (CCHeadsUpDisplayLayerImpl):
  65         (WebCore::CCHeadsUpDisplayLayerImpl::create):
  66         * platform/graphics/chromium/cc/CCLayerImpl.h:
  67         (WebCore::CCLayerImpl::layerIsAlwaysDamaged):
  68         This flag is used to mark the HUD as always damaged, so that it is
  69         continually drawn on the impl thread. It is also used as a hint by
  70         the CCDebugRectHistory class to not include the layer in the display
  71         of damage, since always damaged is not interesting damage.
  72         (CCLayerImpl):
  73         * platform/graphics/chromium/cc/CCLayerTreeHost.cpp:
  74         (WebCore::CCLayerTreeHost::create):
  75         (WebCore::CCLayerTreeHost::CCLayerTreeHost):
  76         (WebCore::CCLayerTreeHost::initialize):
  77         (WebCore::CCLayerTreeHost::willCommit):
  78         (WebCore::CCLayerTreeHost::didCommit):
  79         * platform/graphics/chromium/cc/CCLayerTreeHost.h:
  80         (CCLayerTreeSettings):
  81         (WebCore::CCLayerTreeSettings::showDebugInfo):
  82         (WebCore::CCLayerTreeSettings::showDebugRects):
  83         (CCLayerTreeHost):
  84         * platform/graphics/chromium/cc/CCLayerTreeHostImpl.cpp:
  85         (WebCore::CCLayerTreeHostImpl::CCLayerTreeHostImpl):
  86         (WebCore::CCLayerTreeHostImpl::drawLayers):
  87         * platform/graphics/chromium/cc/CCLayerTreeHostImpl.h:
  88         (WebCore):
  89         (CCLayerTreeHostImpl):
  90         * platform/graphics/chromium/cc/CCProxy.h:
  91         (CCProxy):
  92         * platform/graphics/chromium/cc/CCRenderer.h:
  93         (CCRenderer):
  94         * platform/graphics/chromium/cc/CCSingleThreadProxy.cpp:
  95         * platform/graphics/chromium/cc/CCSingleThreadProxy.h:
  96         * platform/graphics/chromium/cc/CCThreadProxy.cpp:
  97         * platform/graphics/chromium/cc/CCThreadProxy.h:
  98         (CCThreadProxy):
  99
 100 2012-07-25  'Pavel Feldman'  <pfeldman@chromium.org>
 101
 102         Web Inspector: follow up to r123612, fixing CSS value suggest.
 103
 104         Not reviewed, minor JS fix.
 105
 106         * inspector/front-end/CSSCompletions.js:
 107         (WebInspector.CSSCompletions):
 108         * inspector/front-end/StylesSidebarPane.js:
 109         (WebInspector.StylePropertyTreeElement.prototype):
 110
 111 2012-07-25  Michael Saboff  <msaboff@apple.com>
 112
 113         Unreviewed build fix after r123624.
 114
 115         * inspector/InspectorPageAgent.cpp:
 116         (WebCore::InspectorPageAgent::clearGeolocationOverride):
 117         Changed ASSERT_UNUSED() to UNUSED_PARAM().
 118
 119 2012-07-25  Pavel Feldman  <pfeldman@chromium.org>
 120
 121         Web inspector auto-completes "optimizequality" but then complains
 122         https://bugs.webkit.org/show_bug.cgi?id=92262
 123
 124         Reviewed by Timothy Hatcher.
 125
 126         Fixed optimizeSpeed, optimizeLegibility, geometricPrecision.
 127
 128         * inspector/front-end/CSSKeywordCompletions.js:
 129
 130 2012-07-25  Beth Dakin  <bdakin@apple.com>
 131
 132         https://bugs.webkit.org/show_bug.cgi?id=89114
 133         REGRESSION (r112919): Setting scrollTop after setting display from none to block
 134         fails
 135         -and corresponding-
 136         <rdar://problem/11656050>
 137
 138         Reviewed by Simon Fraser.
 139
 140         ScrollAnimatorMac::immediateScrollTo() and ScrollAnimatorMac::immediateScrollBy()
 141         both have an optimization in place so that they do not call
 142         notifyPositionChanged() if the new scroll offset matches the ScrollAnimator's
 143         cached m_currentPosX and m_currentPosY. So revision 112919 caused troubled with
 144         this optimization because it allowed RenderLayers to restore a scrollOffset from
 145         the Element if there is one cached there. This caused the RenderLayer to have a
 146         scrollOffset that is improperly out-of-synch with the ScrollAnimator's
 147         currentPosition (which will just be 0,0 since it is being re-created like the
 148         RenderLayer). This fix makes sure they are in synch by calling
 149         setCurrentPosition() on the ScrollAnimator when the cached position is non-zero.
 150         * rendering/RenderLayer.cpp:
 151         (WebCore::RenderLayer::RenderLayer):
 152
 153 2012-07-25  Andreas Kling  <kling@webkit.org>
 154
 155         Make ElementAttributeData a variable-sized object to reduce memory use.
 156         <http://webkit.org/b/88240>
 157
 158         Reviewed by Ryosuke Niwa.
 159
 160         Take advantage of the fact that we know at ElementAttributeData construction time how many attributes
 161         it needs to accomodate and allocate exactly as much space as needed instead of using a Vector.
 162         For elements that never have their attribute list mutated (the vast majority), this saves a lot of
 163         memory and removes the indirection to Vector<Attribute>'s heap-allocated storage.
 164
 165         Introduced a mutability flag to ElementAttributeData and sprinkled assertions all over to make sure
 166         that nothing tries to mutate an element with a raw attribute array.
 167
 168         When an Element's attribute(s) are mutated, we reconstruct the ElementAttributeData, this time using
 169         a Vector as backing instead. This is triggered when calling Element::mutableAttributeData().
 170
 171         This reduces memory consumption by 3.2MB when viewing the full HTML5 spec at <http://whatwg.org/c/>.
 172         That is a ~35% reduction in DOM attribute memory use.
 173
 174         Furthermore, that page ends up promoting 27% of the elements to mutable attribute storage due to dynamic
 175         adding of "class" attributes. For more static pages, savings are even greater.
 176
 177         Also did away with ElementAttributeData::removeAttribute(name) and do separate index lookup where
 178         needed. Not a big deal but avoids double lookup of attributes when removing them.
 179
 180         * dom/Element.cpp:
 181         (WebCore::Element::detachAttribute):
 182         (WebCore::Element::removeAttribute):
 183         (WebCore::Element::attributes):
 184         (WebCore::Element::setAttributeInternal):
 185         (WebCore::Element::parserSetAttributes):
 186         (WebCore::Element::hasEquivalentAttributes):
 187         (WebCore::Element::setAttributeNode):
 188         (WebCore::Element::removeAttributeNode):
 189         (WebCore::Element::getAttributeNode):
 190         (WebCore::Element::getAttributeNodeNS):
 191         (WebCore::Element::hasAttribute):
 192         (WebCore::Element::hasAttributeNS):
 193         (WebCore::Element::cloneAttributesFromElement):
 194
 195             Sprinkle const on ElementAttributeData* pointers.
 196             Switch to using mutableAttributeData() in code paths that modify attributes.
 197
 198         (WebCore::Element::normalizeAttributes):
 199
 200             Cleaner iteration over attributes, I removed ElementAttributeData::attributeVector() since that
 201             was just incurring a bunch of extra Vector copying. Since ElementAttributeData already exposes
 202             length() and attributeItem(index), use those to iterate directly over the attributes.
 203
 204         (WebCore::Element::createMutableAttributeData):
 205
 206             Added, converts existing ElementAttributeData to a mutable object. Otherwise creates a new
 207             ElementAttributeData (mutable and empty.)
 208
 209         * dom/Element.h:
 210         (WebCore::Element::attributeData):
 211         (WebCore::Element::updatedAttributeData):
 212         (WebCore::Element::ensureAttributeData):
 213         (WebCore::Element::ensureUpdatedAttributeData):
 214
 215             Made all of these return "const ElementAttributeData*" to ensure at compile-time that nothing
 216             attempts to modify an ElementAttributeData that may be immutable.
 217
 218             ensureUpdatedAttributeData() is still const, despite possibly calling mutableAttributeData().
 219             The goal of having methods be const is to prevent conversion from one ElementAttributeData
 220             object to another, so that pointers remain intact and no unnecessary work gets done.
 221
 222         (WebCore::Element::mutableAttributeData):
 223
 224             Added, returns a guaranteed mutable ElementAttributeData* pointer.
 225
 226         * dom/ElementAttributeData.cpp:
 227         (WebCore::ElementAttributeData::createImmutable):
 228
 229             Creates a new ElementAttributeData tailored to fit the provided Vector<Attribute>.
 230
 231         (WebCore::ElementAttributeData::ElementAttributeData):
 232
 233             Moved constructors out-of-line, getting too complex for header files.
 234
 235             ElementAttributeData(const Vector<Attribute>&) creates an immutable ElementAttributeData
 236             containing the provided attributes in an array tacked onto the end of the object.
 237
 238         (WebCore::ElementAttributeData::~ElementAttributeData):
 239
 240             Free the attribute vector if mutable.
 241             For immutable objects, manually invoke destructor on each Attribute object,
 242             since these won't be destroyed automatically by ~ElementAttributeData.
 243
 244         (WebCore::ElementAttributeData::attrIfExists):
 245
 246         (WebCore::ElementAttributeData::ensureAttr):
 247         (WebCore::ElementAttributeData::setAttr):
 248         (WebCore::ElementAttributeData::removeAttr):
 249
 250             Make these const, as they should always have been- they don't affect the ElementAttributeData,
 251             only the global DOM Attr <-> ElementAttributeData mappings.
 252
 253         (WebCore::ElementAttributeData::ensureInlineStyle):
 254         (WebCore::ElementAttributeData::ensureMutableInlineStyle):
 255
 256             Sprinkle ASSERT(isMutable()). This doesn't mean that all Elements with inline styles automatically
 257             have mutable attribute data. There's still inlineStyle() for that, which may return null.
 258             These are only for use where you need a valid StylePropertySet*, even if there is no inline style
 259             attribute, e.g in editing code that wants to add style.
 260
 261         (WebCore::ElementAttributeData::updateInlineStyleAvoidingMutation):
 262         (WebCore::ElementAttributeData::destroyInlineStyle):
 263
 264             Make these const. While destroyInlineStyle() doesn't sound very const-y, immutable objects that
 265             contain a style attribute will have a valid inlineStyle() with that style data. This is just
 266             an interface for ~StyledElement to destroy the style object.
 267             It'd be nice to do that in ~ElementAttributeData(), but we need a StyledElement* pointer to
 268             clean up properly and we don't store one in this class.
 269
 270         (WebCore::ElementAttributeData::addAttribute):
 271         (WebCore::ElementAttributeData::removeAttribute):
 272         (WebCore::ElementAttributeData::clearAttributes):
 273         (WebCore::ElementAttributeData::replaceAttribute):
 274
 275             Sprinkle ASSERT(isMutable()).
 276             Always go straight for m_mutableAttributeVector since we know that's the storage being used.
 277
 278         (WebCore::ElementAttributeData::isEquivalent):
 279         (WebCore::ElementAttributeData::detachAttrObjectsFromElement):
 280         (WebCore::ElementAttributeData::getAttributeItemIndexSlowCase):
 281         (WebCore::ElementAttributeData::removeAttribute):
 282         (WebCore::ElementAttributeData::getAttributeItem):
 283         (WebCore::ElementAttributeData::getAttributeItemIndex):
 284
 285             Use length() and attributeItem(index) to iterate over the attributes.
 286
 287         (WebCore::ElementAttributeData::cloneDataFrom):
 288
 289             Sprinkle ASSERT(isMutable()). Added a FIXME that cloning elements could create immutable
 290             attribute data. I'm not sure this optimization is worthwhile, as cloning elements is already
 291             a semi-rare occurrence.
 292
 293             Updated code to handle both immutable and mutable source objects. This could
 294
 295         (WebCore::ElementAttributeData::getAttributeNode):
 296
 297             Const correctness.
 298
 299         * dom/ElementAttributeData.h:
 300         (ElementAttributeData):
 301
 302             Turn attribute storage into what's effectively union { OwnPtr<Vector<Attribute>>; Attribute[]; }
 303             The new m_isMutable bit determines which union member should be used for access.
 304
 305         (WebCore::ElementAttributeData::create):
 306         (WebCore::ElementAttributeData::createImmutable):
 307
 308             Added createImmutable(const Vector<Attribute>&) as a complement to create().
 309
 310         (WebCore::ElementAttributeData::setClass):
 311         (WebCore::ElementAttributeData::setIdForStyleResolution):
 312         (WebCore::ElementAttributeData::inlineStyle):
 313         (WebCore::ElementAttributeData::setAttributeStyle):
 314
 315             Make these const, and their data member counterparts 'mutable'.
 316             An immutable ElementAttributeData object still has m_classNames, m_idForStyleResolution,
 317             m_inlineStyleDecl and m_attributeStyle.
 318
 319         (WebCore::ElementAttributeData::reportMemoryUsage):
 320
 321             Updated for isMutable().
 322
 323         (WebCore::ElementAttributeData::makeMutable):
 324
 325             Returns a mutable clone of itself.
 326
 327         (WebCore::ElementAttributeData::isEmpty):
 328         (WebCore::ElementAttributeData::length):
 329         (WebCore::ElementAttributeData::attributeItem):
 330
 331             Check isMutable() to know how to access attribute storage.
 332
 333         * dom/StyledElement.cpp:
 334         (WebCore::StyledElement::style):
 335         (WebCore::StyledElement::setInlineStyleProperty):
 336
 337             Simplify by using StyledElement::ensureInlineStyle().
 338
 339         (WebCore::StyledElement::classAttributeChanged):
 340
 341             Use mutableAttributeData() if the attribute is being removed completely.
 342             In other cases, tiptoe around without causing the attribute data to go mutable.
 343
 344         (WebCore::StyledElement::removeInlineStyleProperty):
 345
 346             Use mutableAttributeData() if/when actually removing something.
 347
 348         (WebCore::StyledElement::addSubresourceAttributeURLs):
 349
 350             Const correctness.
 351
 352         * dom/StyledElement.h:
 353         (WebCore::StyledElement::ensureInlineStyle):
 354
 355             This now implies conversion to mutable attribute data. Used by codepaths that add/remove
 356             properties, so conversion is inevitable.
 357
 358         * html/parser/HTMLConstructionSite.cpp:
 359         (WebCore::HTMLConstructionSite::mergeAttributesFromTokenIntoElement):
 360         (WebCore::HTMLConstructionSite::createHTMLElementFromSavedElement):
 361         * svg/properties/SVGAnimatedPropertyMacros.h:
 362         (WebCore::SVGSynchronizableAnimatedProperty::synchronize):
 363
 364             Use mutableAttributeData() as appropriate.
 365
 366         * xml/parser/XMLDocumentParserQt.cpp:
 367         (WebCore::XMLDocumentParser::XMLDocumentParser):
 368
 369             Const correctness.
 370
 371 2012-07-25  Andrew Wilson  <atwilson@chromium.org>
 372
 373         Unreviewed, rolling out r123560.
 374         http://trac.webkit.org/changeset/123560
 375         https://bugs.webkit.org/show_bug.cgi?id=90321
 376
 377         Breaks chromium valgrind tests.
 378
 379         * html/FTPDirectoryDocument.cpp:
 380         (WebCore::FTPDirectoryDocumentParser::append):
 381         * html/parser/HTMLEntityParser.cpp:
 382         (WebCore::HTMLEntityParser::consumeNamedEntity):
 383         * html/parser/HTMLParserIdioms.cpp:
 384         (WebCore::stripLeadingAndTrailingHTMLSpaces):
 385         * html/parser/HTMLSourceTracker.cpp:
 386         (WebCore::HTMLSourceTracker::sourceForToken):
 387         * loader/cache/CachedScript.cpp:
 388         (WebCore::CachedScript::script):
 389         * platform/text/SegmentedString.cpp:
 390         (WebCore::SegmentedString::SegmentedString):
 391         (WebCore::SegmentedString::operator=):
 392         (WebCore::SegmentedString::clear):
 393         (WebCore::SegmentedString::append):
 394         (WebCore::SegmentedString::prepend):
 395         (WebCore::SegmentedString::advanceSubstring):
 396         (WebCore::SegmentedString::advance):
 397         (WebCore::SegmentedString::advanceSlowCase):
 398         (WebCore::SegmentedString::advanceAndUpdateLineNumberSlowCase):
 399         * platform/text/SegmentedString.h:
 400         (WebCore::SegmentedSubstring::SegmentedSubstring):
 401         (WebCore::SegmentedSubstring::clear):
 402         (WebCore::SegmentedSubstring::appendTo):
 403         (SegmentedSubstring):
 404         (WebCore::SegmentedString::SegmentedString):
 405         (WebCore::SegmentedString::push):
 406         (WebCore::SegmentedString::isEmpty):
 407         (WebCore::SegmentedString::lookAhead):
 408         (WebCore::SegmentedString::lookAheadIgnoringCase):
 409         (WebCore::SegmentedString::advance):
 410         (WebCore::SegmentedString::advanceAndASSERT):
 411         (WebCore::SegmentedString::advanceAndASSERTIgnoringCase):
 412         (WebCore::SegmentedString::advancePastNewlineAndUpdateLineNumber):
 413         (SegmentedString):
 414         (WebCore::SegmentedString::advancePastNonNewline):
 415         (WebCore::SegmentedString::advanceAndUpdateLineNumber):
 416         (WebCore::SegmentedString::operator*):
 417         (WebCore::SegmentedString::operator->):
 418         (WebCore::SegmentedString::current):
 419         (WebCore::SegmentedString::equalsLiterally):
 420         (WebCore::SegmentedString::equalsIgnoringCase):
 421         (WebCore::SegmentedString::lookAheadInline):
 422         (WebCore::SegmentedString::lookAheadSlowCase):
 423         * xml/parser/CharacterReferenceParserInlineMethods.h:
 424         (WebCore::consumeCharacterReference):
 425         * xml/parser/MarkupTokenizerBase.h:
 426         (WebCore::MarkupTokenizerBase::InputStreamPreprocessor::nextInputCharacter):
 427         (WebCore::MarkupTokenizerBase::InputStreamPreprocessor::peek):
 428         (WebCore::MarkupTokenizerBase::InputStreamPreprocessor::advance):
 429
 430 2012-07-25  Konrad Piascik  <kpiascik@rim.com>
 431
 432         Web Inspector: Geolocation override
 433         https://bugs.webkit.org/show_bug.cgi?id=89365
 434
 435         Reviewed by Pavel Feldman.
 436
 437         Re-added the m_page member variable for GeolocationController since
 438         it's now being used by InspectorInstrumentaiton.
 439
 440         Added a way to override the GeolocationPosition and GeolocationError
 441         using Web Inspector. Curretnly only PositionError can be set.
 442
 443         This will be hidden behind the experiments flag for now.
 444
 445         Tests: inspector/geolocation-error.html
 446                inspector/geolocation-success.html
 447                inspector/geolocation-watchPosition.html
 448
 449         * English.lproj/localizedStrings.js:
 450         * Modules/geolocation/GeolocationController.cpp:
 451         (WebCore::GeolocationController::GeolocationController):
 452         (WebCore::GeolocationController::positionChanged):
 453         * Modules/geolocation/GeolocationController.h:
 454         (GeolocationController):
 455         * inspector/Inspector.json:
 456         * inspector/InspectorInstrumentation.cpp:
 457         (WebCore):
 458         (WebCore::InspectorInstrumentation::overrideGeolocationPositionImpl):
 459         * inspector/InspectorInstrumentation.h:
 460         (WebCore):
 461         (InspectorInstrumentation):
 462         (WebCore::InspectorInstrumentation::overrideGeolocationPosition):
 463         * inspector/InspectorPageAgent.cpp:
 464         (WebCore::InspectorPageAgent::InspectorPageAgent):
 465         (WebCore::InspectorPageAgent::setGeolocationOverride):
 466         (WebCore):
 467         (WebCore::InspectorPageAgent::clearGeolocationOverride):
 468         (WebCore::InspectorPageAgent::canOverrideGeolocation):
 469         (WebCore::InspectorPageAgent::overrideGeolocationPosition):
 470         * inspector/InspectorPageAgent.h:
 471         * inspector/front-end/Settings.js:
 472         * inspector/front-end/SettingsScreen.js:
 473         (WebInspector.UserAgentSettingsTab):
 474         (WebInspector.UserAgentSettingsTab.prototype._createDeviceMetricsElement):
 475         (WebInspector.UserAgentSettingsTab.prototype._onGeolocationOverrideCheckboxClicked):
 476         (WebInspector.UserAgentSettingsTab.prototype._applyGeolocationUserInput):
 477         (WebInspector.UserAgentSettingsTab.prototype._setGeolocationPosition.set if):
 478         (WebInspector.UserAgentSettingsTab.prototype._setGeolocationPosition):
 479         (WebInspector.UserAgentSettingsTab.prototype._createGeolocationOverrideElement.createInput):
 480         (WebInspector.UserAgentSettingsTab.prototype._createGeolocationOverrideElement):
 481         * inspector/front-end/UserAgentSupport.js:
 482         (WebInspector.UserAgentSupport.GeolocationPosition):
 483         (WebInspector.UserAgentSupport.GeolocationPosition.prototype.toSetting):
 484         (WebInspector.UserAgentSupport.GeolocationPosition.parseSetting):
 485         (WebInspector.UserAgentSupport.GeolocationPosition.parseUserInput):
 486         (WebInspector.UserAgentSupport.GeolocationPosition.clearGeolocationOverride):
 487         * inspector/front-end/inspector.js:
 488         (WebInspector.doLoadedDone):
 489
 490 2012-07-25  Shawn Singh  <shawnsingh@chromium.org>
 491
 492         [chromium] Build fix for clang builds after r123628
 493         https://bugs.webkit.org/show_bug.cgi?id=92266
 494
 495         Reviewed by Adrienne Walker.
 496
 497         An un-used variable was causing build failures for clang builds
 498         after the refactoring in r123628.  It is correct to simply remove
 499         this value.
 500
 501         * platform/graphics/chromium/cc/CCLayerTreeHostCommon.cpp:
 502         (WebCore::calculateDrawTransformsInternal):
 503
 504 2012-07-25  Hans Wennborg  <hans@chromium.org>
 505
 506         Speech JavaScript API: extract function from SpeechRecognitionResult::emma()
 507         https://bugs.webkit.org/show_bug.cgi?id=92232
 508
 509         Reviewed by Adam Barth.
 510
 511         Extract code for creating emma-qualified names to a helper function.
 512
 513         Just a refactoring; covered by the existing test.
 514
 515         * Modules/speech/SpeechRecognitionResult.cpp:
 516         (WebCore::emmaQualifiedName):
 517         (WebCore):
 518         (WebCore::SpeechRecognitionResult::emma):
 519
 520 2012-07-25  Adrienne Walker  <enne@google.com>
 521
 522         [chromium] Unify compositor quadTransform/drawTransform/originTransform
 523         https://bugs.webkit.org/show_bug.cgi?id=91417
 524
 525         Reviewed by Kenneth Russell.
 526
 527         drawTransform used to operate on centered layer rects. To unify quad
 528         transforms, draw transforms, and origin transforms, make draw
 529         transforms operate on content rects with the origin in the top left.
 530
 531         This essentially eliminates the need for surface origin transforms,
 532         which will be eliminated in a future patch.
 533
 534         Also, remove the unused CCLayerImpl::getDrawRect function.
 535
 536         No-op refactoring. Tested by existing tests.
 537
 538         * platform/graphics/chromium/LayerRendererChromium.cpp:
 539         (WebCore::LayerRendererChromium::drawRenderPassQuad):
 540         * platform/graphics/chromium/RenderSurfaceChromium.cpp:
 541         (WebCore::RenderSurfaceChromium::drawableContentRect):
 542         * platform/graphics/chromium/cc/CCDamageTracker.cpp:
 543         (WebCore::CCDamageTracker::extendDamageForLayer):
 544         * platform/graphics/chromium/cc/CCLayerImpl.cpp:
 545         (WebCore::CCLayerImpl::createSharedQuadState):
 546         * platform/graphics/chromium/cc/CCLayerImpl.h:
 547         (CCLayerImpl):
 548         * platform/graphics/chromium/cc/CCLayerSorter.cpp:
 549         (WebCore::CCLayerSorter::LayerShape::LayerShape):
 550         (WebCore::CCLayerSorter::createGraphNodes):
 551         * platform/graphics/chromium/cc/CCLayerTreeHostCommon.cpp:
 552         (WebCore::calculateVisibleContentRect):
 553         (WebCore::calculateDrawTransformsInternal):
 554         * platform/graphics/chromium/cc/CCOcclusionTracker.cpp:
 555         (WebCore::::markOccludedBehindLayer):
 556         (WebCore::::occluded):
 557         (WebCore::::unoccludedContentRect):
 558         * platform/graphics/chromium/cc/CCRenderPass.cpp:
 559         (WebCore::CCRenderPass::appendQuadsToFillScreen):
 560         * platform/graphics/chromium/cc/CCRenderSurface.cpp:
 561         (WebCore::CCRenderSurface::drawableContentRect):
 562
 563 2012-07-25  Tommy Widenflycht  <tommyw@google.com>
 564
 565         MediaStream API: Update MediaStreamTrack to match the specification
 566         https://bugs.webkit.org/show_bug.cgi?id=90180
 567
 568         Reviewed by Adam Barth.
 569
 570         MediaStreamTracks are now required to show the status of the underlying source,
 571         and trigger events when that status changes.
 572
 573         Test: fast/mediastream/MediaStreamTrack.html
 574
 575         * GNUmakefile.list.am:
 576         * Modules/mediastream/MediaStream.cpp:
 577         (WebCore::MediaStream::MediaStream):
 578         (WebCore::MediaStream::addTrack):
 579         * Modules/mediastream/MediaStreamTrack.cpp:
 580         (WebCore::MediaStreamTrack::create):
 581         (WebCore::MediaStreamTrack::MediaStreamTrack):
 582         (WebCore::MediaStreamTrack::~MediaStreamTrack):
 583         (WebCore::MediaStreamTrack::setEnabled):
 584         (WebCore::MediaStreamTrack::readyState):
 585         (WebCore):
 586         (WebCore::MediaStreamTrack::sourceChangedState):
 587         (WebCore::MediaStreamTrack::stop):
 588         (WebCore::MediaStreamTrack::interfaceName):
 589         (WebCore::MediaStreamTrack::scriptExecutionContext):
 590         (WebCore::MediaStreamTrack::eventTargetData):
 591         (WebCore::MediaStreamTrack::ensureEventTargetData):
 592         * Modules/mediastream/MediaStreamTrack.h:
 593         * Modules/mediastream/MediaStreamTrack.idl:
 594         * WebCore.gypi:
 595         * dom/EventNames.h:
 596         (WebCore):
 597         * dom/EventTargetFactory.in:
 598         * platform/chromium/support/WebMediaStreamDescriptor.cpp:
 599         * platform/chromium/support/WebMediaStreamSource.cpp:
 600         (WebKit::WebMediaStreamSource::setReadyState):
 601         (WebKit):
 602         (WebKit::WebMediaStreamSource::readyState):
 603         * platform/mediastream/MediaStreamSource.cpp: Copied from Source/WebCore/platform/mediastream/MediaStreamSource.h.
 604         (WebCore):
 605         (WebCore::MediaStreamSource::create):
 606         (WebCore::MediaStreamSource::MediaStreamSource):
 607         (WebCore::MediaStreamSource::setReadyState):
 608         (WebCore::MediaStreamSource::addObserver):
 609         (WebCore::MediaStreamSource::removeObserver):
 610         * platform/mediastream/MediaStreamSource.h:
 611         (Observer):
 612         (WebCore::MediaStreamSource::Observer::~Observer):
 613         (MediaStreamSource):
 614         (WebCore::MediaStreamSource::readyState):
 615         * platform/mediastream/chromium/MediaStreamCenterChromium.cpp:
 616         (WebCore::MediaStreamCenterChromium::queryMediaStreamSources):
 617         (WebCore::MediaStreamCenterChromium::didCreateMediaStream):
 618
 619 2012-07-25  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
 620
 621         Incorrect rounding in ceiledLayoutUnit and roundedLayoutUnit.
 622         https://bugs.webkit.org/show_bug.cgi?id=91061
 623
 624         Reviewed by Adam Barth.
 625
 626         The two functions are currently both returning the flooredLayoutUnit when subpixel
 627         layout is enabled. This patch changes them to return the correct fractional rounded
 628         value.
 629
 630         * platform/FractionalLayoutUnit.h:
 631         (WebCore::FractionalLayoutUnit::fromFloatCeil):
 632         (FractionalLayoutUnit):
 633         (WebCore::FractionalLayoutUnit::fromFloatRound):
 634         * rendering/LayoutTypes.h:
 635         (WebCore::roundedLayoutUnit):
 636         (WebCore::ceiledLayoutUnit):
 637         * rendering/RenderBlock.cpp:
 638         (WebCore::updatePreferredWidth):
 639
 640 2012-07-25  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
 641
 642         Add *explicit* keyword to constructors in WebCore/platform
 643         https://bugs.webkit.org/show_bug.cgi?id=92238
 644
 645         Reviewed by Kentaro Hara.
 646
 647         As a step to let *explicit* keyword add to constructors which have a parameter,
 648         *explicit* keyword is added to Source/WebCore/platform in order to avoid implicit
 649         type conversion.
 650
 651         Some files in platform directory can't use explicit keyword because of build breaks.
 652
 653         No new tests. Covered by existing tests.
 654
 655         * platform/AutodrainedPool.h:
 656         (AutodrainedPool):
 657         * platform/ContentType.h:
 658         (ContentType):
 659         * platform/ContextMenu.h:
 660         (ContextMenu):
 661         * platform/ContextMenuItem.h:
 662         (ContextMenuItem):
 663         * platform/Cursor.h:
 664         (WebCore::SharedCursor::SharedCursor):
 665         (Cursor):
 666         * platform/KURL.h:
 667         (WebCore::KURL::KURL):
 668         (KURL):
 669         * platform/KURLGooglePrivate.h:
 670         (KURLGooglePrivate):
 671         * platform/PODRedBlackTree.h:
 672         (WebCore::PODRedBlackTree::PODRedBlackTree):
 673         * platform/Pasteboard.h:
 674         (Pasteboard):
 675         * platform/PlatformEvent.h:
 676         (WebCore::PlatformEvent::PlatformEvent):
 677         * platform/PlatformKeyboardEvent.h:
 678         (PlatformKeyboardEvent):
 679         * platform/PlatformMouseEvent.h:
 680         (PlatformMouseEvent):
 681         * platform/PlatformPasteboard.h:
 682         (PlatformPasteboard):
 683         * platform/PlatformTouchEvent.h:
 684         (PlatformTouchEvent):
 685         * platform/PlatformWheelEvent.h:
 686         (PlatformWheelEvent):
 687         * platform/RunLoop.h:
 688         (TimerBase):
 689         * platform/ScrollAnimator.h:
 690         (ScrollAnimator):
 691         * platform/ScrollAnimatorNone.h:
 692         (ScrollAnimatorNone):
 693         * platform/SharedBuffer.h:
 694         (SharedBuffer):
 695         * platform/WheelFlingPlatformGestureCurve.h:
 696         (WheelFlingPlatformGestureCurve):
 697         * platform/Widget.h:
 698         (Widget):
 699
 700 2012-07-25  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
 701
 702         NodesFromRect and area-based hit-testing can not handle CSS transforms.
 703         https://bugs.webkit.org/show_bug.cgi?id=85792
 704
 705         Reviewed by Eric Seidel.
 706
 707         To support the combination of CSS transforms and rect based hit testing,
 708         we need to test against the transformed rect, instead of the original rect.
 709
 710         This patch makes HitTestPoint store the exact transformed FloatPoint and
 711         FloatQuad, and modifies the intersection methods so that they will use a
 712         new FloatQuad based intersection when transforms requires it.
 713
 714         Tests: fast/dom/nodesFromRect/nodesFromRect-rotate.html
 715                fast/dom/nodesFromRect/nodesFromRect-scale.html
 716
 717         * platform/graphics/FloatQuad.cpp:
 718         (WebCore::determinant):
 719         (WebCore::rightMostCornerToVector):
 720         (WebCore::FloatQuad::intersectsRect):
 721         (WebCore::FloatQuad::isCounterclockwise):
 722         * platform/graphics/FloatQuad.h:
 723         (FloatQuad):
 724         * rendering/HitTestResult.cpp:
 725         (WebCore::HitTestPoint::HitTestPoint):
 726         (WebCore::HitTestPoint::operator=):
 727         (WebCore::HitTestPoint::move):
 728         (WebCore::HitTestPoint::intersectsRect):
 729         (WebCore::HitTestPoint::intersects):
 730         * rendering/HitTestResult.h:
 731         (HitTestPoint):
 732         (WebCore::HitTestPoint::isRectilinear):
 733         (WebCore::HitTestPoint::transformedPoint):
 734         (WebCore::HitTestPoint::transformedRect):
 735         * rendering/HitTestingTransformState.cpp:
 736         (WebCore::HitTestingTransformState::flattenWithTransform):
 737         (WebCore::HitTestingTransformState::mappedArea):
 738         (WebCore::HitTestingTransformState::boundsOfMappedArea):
 739         * rendering/HitTestingTransformState.h:
 740         (WebCore::HitTestingTransformState::create):
 741         (WebCore::HitTestingTransformState::HitTestingTransformState):
 742         * rendering/RenderFlowThread.cpp:
 743         (WebCore::RenderFlowThread::hitTestRegion):
 744         * rendering/RenderLayer.cpp:
 745         (WebCore::RenderLayer::hitTest):
 746         (WebCore::RenderLayer::createLocalTransformState):
 747         (WebCore::RenderLayer::hitTestLayer):
 748         (WebCore::RenderLayer::hitTestChildLayerColumns):
 749         * rendering/RenderLayer.h:
 750         * rendering/svg/RenderSVGText.cpp:
 751         (WebCore::RenderSVGText::nodeAtFloatPoint):
 752
 753 2012-07-25  Kwang Yul Seo  <skyul@company100.net>
 754
 755         Add HTMLStackItem.h to project files
 756         https://bugs.webkit.org/show_bug.cgi?id=92221
 757
 758         Reviewed by Adam Barth.
 759
 760         Added HTMLStackItem.h introduced in r123399 to project files.
 761
 762         Project file changes only, no change in behavior thus no new tests.
 763
 764         * GNUmakefile.list.am:
 765         * WebCore.gypi:
 766         * WebCore.vcproj/WebCore.vcproj:
 767         * WebCore.xcodeproj/project.pbxproj:
 768
 769 2012-07-25  Sheriff Bot  <webkit.review.bot@gmail.com>
 770
 771         Unreviewed, rolling out r123606.
 772         http://trac.webkit.org/changeset/123606
 773         https://bugs.webkit.org/show_bug.cgi?id=92247
 774
 775         broke Windows build of Chromium (Requested by tomhudson on
 776         #webkit).
 777
 778         * Resources/colorSuggestionPicker.css: Removed.
 779         * Resources/colorSuggestionPicker.js: Removed.
 780         * WebCore.gyp/WebCore.gyp:
 781
 782 2012-07-25  Pavel Feldman  <pfeldman@chromium.org>
 783
 784         Web Inspector: disabling property should remove "overriden" decoration from overriden UA shorthand.
 785         https://bugs.webkit.org/show_bug.cgi?id=92253
 786
 787         Reviewed by Vsevolod Vlasov.
 788
 789         Bug fix with a test and more compilation markup.
 790
 791         Test: inspector/styles/styles-disable-then-enable-overriden-ua.html
 792
 793         * inspector/front-end/StylesSidebarPane.js:
 794         (WebInspector.StylePropertiesSection.prototype.update):
 795         (WebInspector.StylePropertyTreeElement):
 796         (WebInspector.StylePropertyTreeElement.prototype):
 797
 798 2012-07-25  Vsevolod Vlasov  <vsevik@chromium.org>
 799
 800         Web Inspector: Improve ObjectPropertiesSection UI by adding a tooltip for the value.
 801         https://bugs.webkit.org/show_bug.cgi?id=92225
 802
 803         Reviewed by Pavel Feldman.
 804
 805         Added tooltip with property value description on the property's valueElement.
 806
 807         * inspector/front-end/ObjectPropertiesSection.js:
 808         (WebInspector.ObjectPropertyTreeElement.prototype.update):
 809
 810 2012-07-25  Pavel Feldman  <pfeldman@chromium.org>
 811
 812         https://bugs.webkit.org/show_bug.cgi?id=91931
 813
 814         Reviewed by Vsevolod Vlasov.
 815
 816         - Synthetic shorthand values are now only returned for source-less styles
 817         - CSSAgents reports longhands along with the property names in getSupportedCSSProperties
 818         - StylesSidebarPane now has two paths: for source-enabled styles and for CSSOM-based ones (UA styles)
 819         - CSSStyleModel uses style metainfo as a source of shorthand / longhand relationship now
 820
 821         Test: inspector/styles/styles-overriden-properties.html
 822
 823         * inspector/Inspector.json:
 824         * inspector/InspectorCSSAgent.cpp:
 825         (WebCore::InspectorCSSAgent::getSupportedCSSProperties):
 826         * inspector/InspectorCSSAgent.h:
 827         (InspectorCSSAgent):
 828         * inspector/InspectorStyleSheet.cpp:
 829         (WebCore::InspectorStyle::styleWithProperties):
 830         * inspector/front-end/CSSCompletions.js:
 831         (WebInspector.CSSCompletions):
 832         (WebInspector.CSSCompletions.requestCSSNameCompletions):
 833         (WebInspector.CSSCompletions.prototype._closest):
 834         (WebInspector.CSSCompletions.prototype.longhands):
 835         (WebInspector.CSSCompletions.prototype.shorthands):
 836         * inspector/front-end/CSSStyleModel.js:
 837         (WebInspector.CSSStyleDeclaration):
 838         (WebInspector.CSSStyleDeclaration.prototype.longhandProperties):
 839         (WebInspector.CSSStyleDeclaration.prototype.shorthandValue):
 840         (WebInspector.CSSStyleDeclaration.prototype.newBlankProperty):
 841         (WebInspector.CSSProperty):
 842         (WebInspector.CSSProperty.parsePayload):
 843         * inspector/front-end/SourceCSSTokenizer.js:
 844         (WebInspector.SourceCSSTokenizer):
 845         * inspector/front-end/SourceCSSTokenizer.re2js:
 846         * inspector/front-end/StylesSidebarPane.js:
 847         (WebInspector.StylePropertiesSection.prototype.isPropertyOverloaded):
 848         (WebInspector.StylePropertiesSection.prototype.update):
 849         (WebInspector.StylePropertiesSection.prototype.onpopulate):
 850         (WebInspector.StylePropertyTreeElement.prototype):
 851
 852 2012-07-25  Keishi Hattori  <keishi@webkit.org>
 853
 854         Implement ColorSuggestionPicker page popup
 855         https://bugs.webkit.org/show_bug.cgi?id=92109
 856
 857         Reviewed by Kent Tamura.
 858
 859         Implements ColorSuggestionPicker page popup. This will be used in
 860         certain ports as the datalist UI for <input type=color>.
 861
 862         No new tests because there is no way to open the popup yet.
 863
 864         * Resources/colorSuggestionPicker.css: Added.
 865         (body):
 866         (#main):
 867         (.color-swatch):
 868         (.color-swatch:hover):
 869         (.color-swatch-container):
 870         (.other-color):
 871         * Resources/colorSuggestionPicker.js: Added.
 872         (getScrollbarWidth):When there are more than 20 colors we need to
 873         increase the window width to accommodate the scroll bar.
 874         (createElement):
 875         (handleMessage):
 876         (initialize):
 877         (handleArgumentsTimeout):
 878         (validateArguments):
 879         (submitValue): Calls setValueAndClosePopup with a numValue of 0.
 880         (handleCancel): Calls setValueAndClosePopup with a numValue of -1.
 881         (chooseOtherColor): Calls setValueAndClosePopup with a numValue of -2.
 882         (ColorPicker):
 883         (ColorPicker.prototype._layout): Lays out the color swatches in a 5x4
 884         grid.
 885         (ColorPicker.prototype._handleSwatchClick):
 886         * WebCore.gyp/WebCore.gyp:
 887
 888 2012-07-25  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
 889
 890         Fix arithmetic composite filter for auto-vectorization
 891         https://bugs.webkit.org/show_bug.cgi?id=92123
 892
 893         Reviewed by Nikolas Zimmermann.
 894
 895         Since only clamping of the result prevents GCC from auto-vectorizing the inner loop, this patch
 896         adds a faster version of the inner loop to handle cases where clamping is unnecessary,
 897
 898         * platform/graphics/filters/FEComposite.cpp:
 899         (WebCore::computeArithmeticPixelsUnclamped):
 900         (WebCore::arithmeticSoftware):
 901
 902 2012-07-25  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
 903
 904         Fix blend filter for autovectorizing
 905         https://bugs.webkit.org/show_bug.cgi?id=91398
 906
 907         Reviewed by Nikolas Zimmermann.
 908
 909         Get rid of wrapper classes, and use functions directly as template argument.
 910
 911         * platform/graphics/filters/FEBlend.cpp:
 912         (WebCore::feBlendNormal):
 913         (WebCore::feBlendMultiply):
 914         (WebCore::feBlendScreen):
 915         (WebCore::feBlendDarken):
 916         (WebCore::feBlendLighten):
 917         (WebCore::feBlendUnknown):
 918         (WebCore::platformApply):
 919         (WebCore::FEBlend::platformApplyGeneric):
 920
 921 2012-07-25  Keishi Hattori  <keishi@webkit.org>
 922
 923         Add methods to ColorChooserClient so the color chooser can show suggestions
 924         https://bugs.webkit.org/show_bug.cgi?id=92114
 925
 926         Reviewed by Kent Tamura.
 927
 928         Added methods to ColorChooserClient so the color chooser can ask the
 929         client for suggestions and current color.
 930
 931         No new tests. The methods are not used yet.
 932
 933         * html/ColorInputType.cpp:
 934         (WebCore::ColorInputType::currentColor): Returns what color the color
 935         chooser should display.
 936         (WebCore):
 937         (WebCore::ColorInputType::shouldShowSuggestions): Returns if the UI for
 938         suggestions should be shown.
 939         (WebCore::ColorInputType::suggestions): Returns a list of colors to be
 940         suggested to the user.
 941         * html/ColorInputType.h:
 942         (ColorInputType):
 943         * html/HTMLInputElement.h:
 944         (HTMLInputElement): Made dataList() public.
 945         * platform/ColorChooserClient.h:
 946         (ColorChooserClient):
 947
 948 2012-07-25  Kentaro Hara  <haraken@chromium.org>
 949
 950         Replace int with long in test IDL files
 951         https://bugs.webkit.org/show_bug.cgi?id=92181
 952
 953         Reviewed by Adam Barth.
 954
 955         int is not a valid type in WebIDL. It should be long.
 956         We should replace int with long in bindings/scripts/test/*.idl.
 957         This is a follow-up patch for r123550.
 958
 959         Test: bindings/scripts/test/TestObj.idl
 960
 961         * bindings/scripts/test/CPP/WebDOMTestObj.cpp:
 962         (WebDOMTestObj::readOnlyLongAttr):
 963         (WebDOMTestObj::Attr):
 964         (WebDOMTestObj::setAttr):
 965         (WebDOMTestObj::voidMethodWithArgs):
 966         (WebDOMTestObj::Method):
 967         (WebDOMTestObj::MethodWithArgs):
 968         (WebDOMTestObj::objMethodWithArgs):
 969         * bindings/scripts/test/CPP/WebDOMTestObj.h:
 970         * bindings/scripts/test/GObject/WebKitDOMTestObj.cpp:
 971         (webkit_dom_test_obj_set_property):
 972         (webkit_dom_test_obj_get_property):
 973         (webkit_dom_test_obj_class_init):
 974         (webkit_dom_test_obj_void_method_with_args):
 975         (webkit_dom_test_obj_method):
 976         (webkit_dom_test_obj_method_with_args):
 977         (webkit_dom_test_obj_obj_method_with_args):
 978         (webkit_dom_test_obj_get_read_only_long_attr):
 979         (webkit_dom_test_obj_get_attr):
 980         (webkit_dom_test_obj_set_attr):
 981         (webkit_dom_test_obj_get_strawberry):
 982         (webkit_dom_test_obj_set_strawberry):
 983         * bindings/scripts/test/GObject/WebKitDOMTestObj.h:
 984         * bindings/scripts/test/JS/JSTestObj.cpp:
 985         (WebCore):
 986         (WebCore::jsTestObjReadOnlyLongAttr):
 987         (WebCore::jsTestObjConstructorStaticReadOnlyLongAttr):
 988         (WebCore::jsTestObjAttr):
 989         (WebCore::setJSTestObjAttr):
 990         (WebCore::setJSTestObjStrawberry):
 991         (WebCore::jsTestObjPrototypeFunctionVoidMethodWithArgs):
 992         (WebCore::jsTestObjPrototypeFunctionMethod):
 993         (WebCore::jsTestObjPrototypeFunctionMethodWithArgs):
 994         (WebCore::jsTestObjPrototypeFunctionObjMethodWithArgs):
 995         (WebCore::jsTestObjPrototypeFunctionMethodReturningSequence):
 996         (WebCore::jsTestObjPrototypeFunctionOverloadedMethod2):
 997         (WebCore::jsTestObjPrototypeFunctionOverloadedMethod4):
 998         * bindings/scripts/test/JS/JSTestObj.h:
 999         (WebCore):
1000         * bindings/scripts/test/ObjC/DOMTestObj.h:
1001         * bindings/scripts/test/ObjC/DOMTestObj.mm:
1002         (-[DOMTestObj readOnlyLongAttr]):
1003         (-[DOMTestObj Attr]):
1004         (-[DOMTestObj setAttr:]):
1005         (-[DOMTestObj voidMethodWithArgs:strArg:objArg:]):
1006         (-[DOMTestObj Method]):
1007         (-[DOMTestObj MethodWithArgs:strArg:objArg:]):
1008         (-[DOMTestObj objMethodWithArgs:strArg:objArg:]):
1009         (-[DOMTestObj customMethodWithArgs:strArg:objArg:]):
1010         * bindings/scripts/test/TestObj.idl:
1011         * bindings/scripts/test/V8/V8TestObj.cpp:
1012         (WebCore::TestObjV8Internal::readOnlyLongAttrAttrGetter):
1013         (WebCore::TestObjV8Internal::staticReadOnlyLongAttrAttrGetter):
1014         (WebCore::TestObjV8Internal::AttrAttrGetter):
1015         (WebCore::TestObjV8Internal::AttrAttrSetter):
1016         (WebCore::TestObjV8Internal::strawberryAttrSetter):
1017         (WebCore::TestObjV8Internal::voidMethodWithArgsCallback):
1018         (WebCore::TestObjV8Internal::MethodCallback):
1019         (WebCore::TestObjV8Internal::MethodWithArgsCallback):
1020         (WebCore::TestObjV8Internal::objMethodWithArgsCallback):
1021         (WebCore::TestObjV8Internal::methodReturningSequenceCallback):
1022         (WebCore::TestObjV8Internal::overloadedMethod2Callback):
1023         (WebCore::TestObjV8Internal::overloadedMethod4Callback):
1024         (WebCore):
1025         (WebCore::ConfigureV8TestObjTemplate):
1026
1027 2012-07-25  Shinya Kawanaka  <shinyak@chromium.org>
1028
1029         A validation bubble should not be modifiable.
1030         https://bugs.webkit.org/show_bug.cgi?id=92217
1031
1032         Reviewed by Kent Tamura.
1033
1034         A user can set -webkit-user-modify: read-write to validation message elements.
1035         This might cause a undesired crash like Bug 65738. They should be unmodifiable.
1036
1037         Test: fast/forms/validation-user-modify.html
1038
1039         * css/html.css:
1040         Added -webkit-user-modify: read-only !important; for these elements so that they won't be
1041         changed by a user.
1042         (::-webkit-validation-bubble):
1043         (::-webkit-validation-bubble-message):
1044         (::-webkit-validation-bubble-text-block):
1045         (::-webkit-validation-bubble-heading):
1046         (::-webkit-validation-bubble-arrow):
1047         (::-webkit-validation-bubble-arrow-clipper):
1048
1049 2012-07-25  Zeno Albisser  <zeno@webkit.org>
1050
1051         [Qt][WK2] Disable CopyToTexture for GraphicsSurfaceMac.
1052         https://bugs.webkit.org/show_bug.cgi?id=92159
1053
1054         Disable CopyToTexture feature for GraphicsSurface on Mac.
1055         While this is actually working, it is currently used for Tiles
1056         that are single buffered, and therefore requires a call to glFlush.
1057         This call blocks the GPU for about 40ms which would make smooth animations impossible.
1058
1059         Reviewed by Noam Rosenthal.
1060
1061         * platform/graphics/surfaces/mac/GraphicsSurfaceMac.cpp:
1062         (WebCore::GraphicsSurface::platformCreate):
1063         (WebCore::GraphicsSurface::platformImport):
1064
1065 2012-07-25  Kenichi Ishibashi  <bashi@chromium.org>
1066
1067         [Chromium] Fix up includes in HarfBuzzNGFace.h
1068         https://bugs.webkit.org/show_bug.cgi?id=92127
1069
1070         Reviewed by Hajime Morita.
1071
1072         Move constructors of FontPlatformData from headers to corresponding cpp files
1073         so that we can use forward declaration for HarfBuzzNGFace instead of including HarfBuzzNGFace.h.
1074         This way we can include hb.h in HarfBuzzNGFace.h.
1075
1076         No new tests. No changes in behavior.
1077
1078         * platform/graphics/FontPlatformData.cpp:
1079         (WebCore::FontPlatformData::FontPlatformData): Moved from header file.
1080         (WebCore):
1081         * platform/graphics/FontPlatformData.h:
1082         (WebCore):
1083         (FontPlatformData): Moved to cpp file.
1084         * platform/graphics/cocoa/FontPlatformDataCocoa.mm: Include HarfBuzzNGFace.h
1085         * platform/graphics/harfbuzz/FontPlatformDataHarfBuzz.cpp:
1086         (WebCore::FontPlatformData::FontPlatformData): Moved from header file.
1087         (WebCore):
1088         * platform/graphics/harfbuzz/FontPlatformDataHarfBuzz.h:
1089         (WebCore):
1090         (FontPlatformData): Moved to cpp file.
1091         * platform/graphics/harfbuzz/ng/HarfBuzzNGFace.h: Include hb.h.
1092
1093 2012-07-24  Benjamin Poulain  <bpoulain@apple.com> && Joseph Pecoraro  <pecoraro@apple.com>
1094
1095         QualifiedName's HashSet should be big enough to hold at least all the static names
1096         https://bugs.webkit.org/show_bug.cgi?id=91891
1097
1098         Reviewed by Darin Adler.
1099
1100         QualifiedName's table has a standard size of 64 buckets. When initializing WebKit,
1101         we create 850 static QualifiedName for the standard names (HTMLNames, SVGNames etc).
1102
1103         The small base size forces us to grow and rehash the table several time on startup.
1104
1105         This patch solves the issue by defining the initial table size to the minimum size that
1106         can hold all the static QualifiedName.
1107
1108         * dom/QualifiedName.cpp:
1109         (QualifiedNameHashTraits):
1110         * dom/make_names.pl:
1111         (printNamesHeaderFile):
1112
1113 2012-07-24  Kwang Yul Seo  <skyul@company100.net>
1114
1115         Remove anonymous namespace and make functions static.
1116         https://bugs.webkit.org/show_bug.cgi?id=92214
1117
1118         Reviewed by Adam Barth.
1119
1120         Mark functions as static instead of enclosing them in anonymous namespaces.
1121         One exception to this rule is when a function is passed as a template argument.
1122         In this case, the function must have an external linkage.
1123
1124         No functional change. Covered by existing tests.
1125
1126         * html/parser/HTMLConstructionSite.cpp:
1127         (WebCore::hasImpliedEndTag):
1128         (WebCore::causesFosterParenting):
1129         (WebCore::isAllWhitespace):
1130         * html/parser/HTMLDocumentParser.cpp:
1131         (WebCore):
1132         (WebCore::tokenizerStateForContextElement):
1133         * html/parser/HTMLElementStack.cpp:
1134         (WebCore::isNumberedHeaderElement):
1135         (WebCore::isRootNode):
1136         (WebCore):
1137         * html/parser/HTMLEntityParser.cpp:
1138         (WebCore::isAlphaNumeric):
1139         * html/parser/HTMLEntitySearch.cpp:
1140         (WebCore::halfway):
1141         * html/parser/HTMLPreloadScanner.cpp:
1142         (WebCore):
1143         * html/parser/HTMLTokenizer.cpp:
1144         (WebCore::toLowerCase):
1145         (WebCore::vectorEqualsString):
1146         (WebCore::isEndTagBufferingState):
1147         * html/parser/HTMLTreeBuilder.cpp:
1148         (WebCore):
1149         (WebCore::uninitializedPositionValue1):
1150         (WebCore::isAllWhitespace):
1151         (WebCore::isAllWhitespaceOrReplacementCharacters):
1152         (WebCore::isNumberedHeaderTag):
1153         (WebCore::isCaptionColOrColgroupTag):
1154         (WebCore::isTableCellContextTag):
1155         (WebCore::isTableBodyContextTag):
1156         (WebCore::isSpecialNode):
1157         (WebCore::isNonAnchorNonNobrFormattingTag):
1158         (WebCore::isNonAnchorFormattingTag):
1159         (WebCore::isFormattingTag):
1160         (WebCore::closestFormAncestor):
1161         (WebCore::mapLoweredLocalNameToName):
1162         (WebCore::adjustSVGTagNameCase):
1163         (WebCore::adjustAttributes):
1164         (WebCore::adjustSVGAttributes):
1165         (WebCore::adjustMathMLAttributes):
1166         (WebCore::addNamesWithPrefix):
1167         (WebCore::adjustForeignAttributes):
1168
1169 2012-07-24  Vsevolod Vlasov  <vsevik@chromium.org>
1170
1171         Unreviewed r123494 follow-up: fixed inspector undock icon.
1172
1173         * inspector/front-end/Images/statusbarButtonGlyphs.png:
1174
1175 2012-07-24  Kwang Yul Seo  <skyul@company100.net>
1176
1177         Read tag names and attributes from the saved tokens where possible.
1178         https://bugs.webkit.org/show_bug.cgi?id=92079
1179
1180         Reviewed by Adam Barth.
1181
1182         Read tag names and attributes from the saved tokens, not from the DOM.
1183         Also added convenient methods such as hasLocalName, hasTagName, localName,
1184         isElementNode and isDocumentFragmentNode to HTMLStackItem class.
1185
1186         No new tests, covered by existing tests.
1187
1188         * html/parser/HTMLConstructionSite.cpp:
1189         (WebCore::HTMLNames::hasImpliedEndTag):
1190         (WebCore::HTMLNames::causesFosterParenting):
1191         (WebCore::HTMLConstructionSite::generateImpliedEndTagsWithExclusion):
1192         (WebCore::HTMLConstructionSite::generateImpliedEndTags):
1193         (WebCore::HTMLConstructionSite::shouldFosterParent):
1194         * html/parser/HTMLConstructionSite.h:
1195         (WebCore::HTMLConstructionSite::currentStackItem):
1196         * html/parser/HTMLElementStack.cpp:
1197         (WebCore::HTMLNames::isNumberedHeaderElement):
1198         (WebCore::HTMLNames::isRootNode):
1199         (WebCore::HTMLNames::isScopeMarker):
1200         (WebCore::HTMLNames::isListItemScopeMarker):
1201         (WebCore::HTMLNames::isTableScopeMarker):
1202         (WebCore::HTMLNames::isTableBodyScopeMarker):
1203         (WebCore::HTMLNames::isTableRowScopeMarker):
1204         (WebCore::HTMLNames::isForeignContentScopeMarker):
1205         (WebCore::HTMLNames::isButtonScopeMarker):
1206         (WebCore::HTMLNames::isSelectScopeMarker):
1207         (WebCore::HTMLElementStack::ElementRecord::replaceElement):
1208         (WebCore::HTMLElementStack::popUntilNumberedHeaderElementPopped):
1209         (WebCore::HTMLElementStack::popUntilTableScopeMarker):
1210         (WebCore::HTMLElementStack::popUntilTableBodyScopeMarker):
1211         (WebCore::HTMLElementStack::popUntilTableRowScopeMarker):
1212         (WebCore::HTMLElementStack::isMathMLTextIntegrationPoint):
1213         (WebCore::HTMLElementStack::isHTMLIntegrationPoint):
1214         (WebCore::HTMLElementStack::popUntilForeignContentScopeMarker):
1215         (WebCore::HTMLElementStack::pushRootNode):
1216         (WebCore::HTMLElementStack::oneBelowTop):
1217         (WebCore::HTMLElementStack::topmost):
1218         (WebCore):
1219         (WebCore::inScopeCommon):
1220         (WebCore::HTMLElementStack::hasNumberedHeaderElementInScope):
1221         (WebCore::HTMLElementStack::inScope):
1222         * html/parser/HTMLElementStack.h:
1223         (WebCore::HTMLElementStack::topStackItem):
1224         (HTMLElementStack):
1225         (WebCore::isInHTMLNamespace):
1226         * html/parser/HTMLStackItem.h:
1227         (WebCore::HTMLStackItem::isDocumentFragmentNode):
1228         (WebCore::HTMLStackItem::isElementNode):
1229         (HTMLStackItem):
1230         (WebCore::HTMLStackItem::hasLocalName):
1231         (WebCore::HTMLStackItem::hasTagName):
1232         (WebCore::HTMLStackItem::HTMLStackItem):
1233         * html/parser/HTMLTreeBuilder.cpp:
1234         (WebCore):
1235         (WebCore::HTMLTreeBuilder::constructTreeFromAtomicToken):
1236         (WebCore::HTMLTreeBuilder::processFakeEndTag):
1237         (WebCore::HTMLTreeBuilder::processCloseWhenNestedTag):
1238         (WebCore::HTMLTreeBuilder::processStartTagForInBody):
1239         (WebCore::HTMLTreeBuilder::processStartTag):
1240         (WebCore::HTMLTreeBuilder::processAnyOtherEndTagForInBody):
1241         (WebCore::HTMLTreeBuilder::furthestBlockForFormattingElement):
1242         (WebCore::HTMLTreeBuilder::callTheAdoptionAgency):
1243         (WebCore::HTMLTreeBuilder::processEndTagForInTableBody):
1244         (WebCore::HTMLTreeBuilder::processEndTagForInCell):
1245         (WebCore::HTMLTreeBuilder::processEndTagForInBody):
1246         (WebCore::HTMLTreeBuilder::processCaptionEndTagForInCaption):
1247         (WebCore::HTMLTreeBuilder::processTrEndTagForInRow):
1248         (WebCore::HTMLTreeBuilder::processEndTag):
1249         (WebCore::HTMLTreeBuilder::processCharacterBuffer):
1250         (WebCore::HTMLTreeBuilder::processEndOfFile):
1251         (WebCore::HTMLTreeBuilder::shouldProcessTokenInForeignContent):
1252         (WebCore::HTMLTreeBuilder::processTokenInForeignContent):
1253         * html/parser/HTMLTreeBuilder.h:
1254
1255 2012-07-24  Hironori Bono  <hbono@chromium.org>
1256
1257         Avoid moving child objects multiple times when vertical scrollbar are shown at the left side.
1258         https://bugs.webkit.org/show_bug.cgi?id=91756
1259
1260         Reviewed by Tony Chang.
1261
1262         My r123067 moves the top-left origin of an RTL element right when its vertical
1263         scrollbar is shown at its left side. (That is, r123067 moves all child objects
1264         in the RTL element right.) This change also increases RenderBox::clientLeft()
1265         at the same time, i.e. it also moves child objects right. Furthermore, my r109512
1266         moves positioned objects in an RTL element right at the same time. This makes
1267         WebKit move objects in an RTL element up to three times by the scrollbar width.
1268         (Moving an absolute object right increases the scrollWidth value and it causes
1269         this bug.) This change removes unnecessary code that moves objects right in my
1270         r109512 and RenderBox::clientLeft().
1271
1272         Test: scrollbars/rtl/div-absolute.html
1273               fast/block/float/026.html
1274               fast/block/float/028.html
1275               fast/overflow/unreachable-overflow-rtl-bug.html
1276
1277         * dom/Element.cpp:
1278         (WebCore::Element::clientLeft): Increase clientLeft value by the width of a vertical scrollbar as written in the CSSOM specification.
1279         * rendering/RenderBlock.cpp:
1280         (WebCore::RenderBlock::addOverflowFromPositionedObjects): Removed unnecessary code.
1281         (WebCore::RenderBlock::determineLogicalLeftPositionForChild): Removed unnecessary code.
1282         * rendering/RenderBox.h:
1283         (WebCore::RenderBox::clientLeft): Removed unnecessary code.
1284
1285 2012-07-24  Dan Bernstein  <mitz@apple.com>
1286
1287         RenderBlock::positionForPoint can fail when the block or its children have a vertical writing mode
1288         https://bugs.webkit.org/show_bug.cgi?id=92202
1289
1290         Reviewed by Beth Dakin.
1291
1292         Test: fast/writing-mode/positionForPoint.html
1293
1294         * rendering/RenderBlock.cpp:
1295         (WebCore::RenderBlock::positionForPoint): Fixed two issues, each of which are covered by
1296         one of the sub-tests in the new regression test: (1) changed to compare children’s logical
1297         bounds against pointInLogicalContents, i.e. compare childern’s horizontal bounds to the
1298         point’s horizontal component; and (2) changed to use logicalTopForChild and
1299         logicalHeightForChild, which account for this block’s writing mode, instead of logicalTop
1300         and logicalBottom, which are based on the child’s writing mode.
1301
1302 2012-07-24  MORITA Hajime  <morrita@google.com>
1303
1304         https://bugs.webkit.org/show_bug.cgi?id=89179
1305         Rendering LI having Shadow DOM seems weird
1306
1307         Reviewed by Dimitri Glazkov.
1308
1309         This incorrect rendering happend because HTMLLIElement::attach()
1310         wasn't aware of composed shadow subtree. This change makes it
1311         aware of that using ComposedShadowTreeParentWalker.
1312
1313         Test: fast/dom/shadow/shadow-and-list-elements.html
1314
1315         * html/HTMLLIElement.cpp:
1316         (WebCore::HTMLLIElement::attach):
1317
1318 2012-07-24  Alec Flett  <alecflett@chromium.org>
1319
1320         IndexedDB: fix #include dependencies so IDBRequest isn't an include root
1321         https://bugs.webkit.org/show_bug.cgi?id=92167
1322
1323         Reviewed by Tony Chang.
1324
1325         Remove implicit dependencies due to using IDBRequest.h, and fix all
1326         places that break as a result.
1327
1328         * Modules/indexeddb/IDBCallbacks.h:
1329         (WebCore):
1330         * Modules/indexeddb/IDBObjectStore.cpp:
1331         * Modules/indexeddb/IDBRequest.h:
1332         * inspector/InspectorIndexedDBAgent.cpp:
1333
1334 2012-07-24  Keishi Hattori  <keishi@webkit.org>
1335
1336         Move PagePopupClient helper functions so they can be shared
1337         https://bugs.webkit.org/show_bug.cgi?id=92096
1338
1339         Reviewed by Kent Tamura.
1340
1341         Move PagePopupClient helper functions to a separate file,
1342         so they can be shared with new PagePopupClients.
1343
1344         No new tests. Covered by fast/forms/date/calendar-picker-appearance.html
1345
1346         * WebCore.gypi: Added PagePopupClient.cpp
1347         * html/shadow/CalendarPickerElement.cpp:
1348         (WebCore::CalendarPickerElement::writeDocument): Modified to use PagePopupClient.
1349         * page/PagePopupClient.cpp: Added.
1350         (WebCore):
1351         (WebCore::PagePopupClient::addJavaScriptString):
1352         (WebCore::PagePopupClient::addProperty):
1353         * page/PagePopupClient.h:
1354         (WebCore::PagePopupClient::addString):
1355
1356 2012-07-24  Jer Noble  <jer.noble@apple.com>
1357
1358         setting playbackRate on a MediaController doesn't change the playbackRate for slaved media
1359         https://bugs.webkit.org/show_bug.cgi?id=92035
1360
1361         Reviewed by Darin Adler.
1362
1363         Allow updatePlaybackRate() to change the MediaPlayer rate when the media element has a
1364         current media controller.
1365
1366         Test: media/media-controller-playbackrate.html
1367
1368         * html/HTMLMediaElement.cpp:
1369         (WebCore::HTMLMediaElement::updatePlaybackRate):
1370
1371 2012-07-24  Hayato Ito  <hayato@chromium.org>
1372
1373         Rename FocusScope to FocusNavigationScope.
1374         https://bugs.webkit.org/show_bug.cgi?id=92062
1375
1376         Reviewed by Dimitri Glazkov.
1377
1378         FocusNavagationScope is better name than FocusScope since the
1379         class is only used for controlling focus navigation.
1380
1381         No new tests - no functional changes.
1382
1383         * page/FocusController.cpp:
1384         (WebCore::FocusNavigationScope::FocusNavigationScope):
1385         (WebCore::FocusNavigationScope::rootNode):
1386         (WebCore::FocusNavigationScope::owner):
1387         (WebCore::FocusNavigationScope::focusNavigationScopeOf):
1388         (WebCore::FocusNavigationScope::focusNavigationScopeOwnedByShadowHost):
1389         (WebCore::FocusNavigationScope::focusNavigationScopeOwnedByIFrame):
1390         (WebCore::FocusController::findFocusableNodeDecendingDownIntoFrameDocument):
1391         (WebCore::FocusController::advanceFocusInDocumentOrder):
1392         (WebCore::FocusController::findFocusableNodeAcrossFocusScope):
1393         (WebCore::FocusController::findFocusableNodeRecursively):
1394         (WebCore::FocusController::findFocusableNode):
1395         (WebCore::FocusController::nextFocusableNode):
1396         (WebCore::FocusController::previousFocusableNode):
1397         * page/FocusController.h:
1398         (FocusNavigationScope):
1399         (FocusController):
1400
1401 2012-07-24  Michael Saboff  <msaboff@apple.com>
1402
1403         Convert HTML parser to handle 8-bit resources without converting to UChar*
1404         https://bugs.webkit.org/show_bug.cgi?id=90321
1405
1406         Reviewed by Geoffrey Garen.
1407
1408         No new tests, no new functionality therefore covered by existing tests.
1409
1410         Changed the parsing of HTML to handle 8-bit strings without converting the
1411         whole string to 16-bits.  Primary change was to use the appropriate
1412         character pointer type (LChar* or UChar*) depending on the source string.
1413         This access is abstracted in SegmentedSubstring by getCurrentChar{8,16} and
1414         incrementAndGetCurrentChar{8,16} methods.  In SegmentedString, the advance()
1415         and advanceAndUpdateLineNumber() methods have been refactored into a
1416         state machine managed pair of function pointers.  The possible functions
1417         have not only the 8 and 16 bitness factored out, they also have the
1418         "exclude line numbers" logic factored out for advanceAndUpdateLineNumber().
1419         This provides a net speed-up even with the 8/16 bit testing.
1420         Other changes involve using String methods to access and compare with the
1421         source data.
1422
1423         * html/FTPDirectoryDocument.cpp:
1424         (WebCore::FTPDirectoryDocumentParser::append): Changed to use currentChar()
1425         instead of SegmentedString '*' operator.
1426         * html/parser/HTMLEntityParser.cpp: Changed to use currentChar() instead of
1427         SegmentedString '*' operator.
1428         * html/parser/HTMLParserIdioms.cpp:
1429         (WebCore::stripLeadingAndTrailingHTMLSpaces): Added 8-bit fast path.
1430         * html/parser/HTMLSourceTracker.cpp:
1431         (WebCore::HTMLSourceTracker::sourceForToken): Changed to use currentChar() instead of
1432         SegmentedString '*' operator.
1433         * loader/cache/CachedScript.cpp:
1434         (WebCore::CachedScript::script): Updated to use new String::dataSize() method.
1435         * platform/text/SegmentedString.cpp:
1436         (WebCore::SegmentedString::SegmentedString):
1437         (WebCore::SegmentedString::operator=):
1438         (WebCore::SegmentedString::clear):
1439         (WebCore::SegmentedString::append):
1440         (WebCore::SegmentedString::prepend):
1441         (WebCore::SegmentedString::advanceSubstring):
1442         (WebCore::SegmentedString::advance): Changed to be a wrapper that invokes current
1443         advance function.
1444         (WebCore::SegmentedString::advance8): Fast path advance function for 8 bit data.
1445         (WebCore::SegmentedString::advance16): Fast path advance function for 16 bit data.
1446         (WebCore::SegmentedString::advanceAndUpdateLineNumber8IncludeLineNumbers): Fast
1447         path advanceAndUpdateLineNumber for 8 bit data when we need to update the line number.
1448         (WebCore::SegmentedString::advanceAndUpdateLineNumber16IncludeLineNumbers): Fast
1449         path advanceAndUpdateLineNumber for 16 bit data when we need to update the line number.
1450         (WebCore::SegmentedString::advanceSlowCase): Slow case advance function for both
1451         8 and 16 bit advance.
1452         (WebCore::SegmentedString::advanceAndUpdateLineNumberSlowCase): Slow case
1453         advanceAndUpdateLineNumber function for both 8 and 16 bit data.
1454         (WebCore::SegmentedString::advanceEmpty): Advance function when input has been exhausted.
1455         * platform/text/SegmentedString.h:
1456         (WebCore::SegmentedSubstring::SegmentedSubstring):
1457         (WebCore::SegmentedSubstring::clear):
1458         (SegmentedSubstring):
1459         (WebCore::SegmentedSubstring::is8Bit): New getter.
1460         (WebCore::SegmentedSubstring::appendTo):
1461         (WebCore::SegmentedSubstring::getCurrentChar8): New getter for the 8 bit case.
1462         (WebCore::SegmentedSubstring::getCurrentChar16): New getter for the 16 bit case.
1463         (WebCore::SegmentedSubstring::incrementAndGetCurrentChar8): New wrapper functions
1464         that pre-increments the 8 bit pointer and returns the next character.
1465         (WebCore::SegmentedSubstring::incrementAndGetCurrentChar16): New wrapper functions
1466         that pre-increments the 16 bit pointer and returns the next character.
1467         (WebCore::SegmentedSubstring::currentSubString):
1468         (WebCore::SegmentedSubstring::getCurrentChar):
1469         (WebCore::SegmentedSubstring::incrementAndGetCurrentChar):
1470         (WebCore::SegmentedString::SegmentedString):
1471         (WebCore::SegmentedString::push):
1472         (WebCore::SegmentedString::isEmpty):
1473         (WebCore::SegmentedString::lookAhead):
1474         (WebCore::SegmentedString::lookAheadIgnoringCase):
1475         (WebCore::SegmentedString::advance):
1476         (WebCore::SegmentedString::advanceAndUpdateLineNumber):
1477         (WebCore::SegmentedString::advanceAndASSERT):
1478         (WebCore::SegmentedString::advanceAndASSERTIgnoringCase):
1479         (WebCore::SegmentedString::advancePastNonNewline):
1480         (WebCore::SegmentedString::advancePastNewlineAndUpdateLineNumber):
1481         (WebCore::SegmentedString::currentChar):
1482         (WebCore::SegmentedString::setSlowCase): Set advance function pointers to slow case
1483         functions.
1484         (WebCore::SegmentedString::decrementAndCheckLength): Decrement substring length and
1485         change to slow case functions when only one character left.
1486         (WebCore::SegmentedString::setAdvanceFunctionPointers): Select appropriate advance
1487         functions based on current substring.
1488         (WebCore::SegmentedString::lookAheadInline): Changed to use String::startsWith().
1489         (WebCore::SegmentedString::lookAheadSlowCase):
1490         * xml/parser/CharacterReferenceParserInlineMethods.h:
1491         (WebCore::consumeCharacterReference): Changed to use currentChar() instead of
1492         SegmentedString '*' operator.
1493         * xml/parser/MarkupTokenizerBase.h:
1494         (WebCore::MarkupTokenizerBase::InputStreamPreprocessor::nextInputCharacter):
1495         (WebCore::MarkupTokenizerBase::InputStreamPreprocessor::peek): Changed to use
1496         currentChar() instead of SegmentedString '*' operator.
1497         (WebCore::MarkupTokenizerBase::InputStreamPreprocessor::advance):
1498
1499 2012-07-24  Kenneth Russell  <kbr@google.com>
1500
1501         [chromium] Hint garbage collector to run if page uses Canvas contexts
1502         https://bugs.webkit.org/show_bug.cgi?id=76225
1503
1504         Reviewed by Kentaro Hara.
1505
1506         Upon creating a canvas context, set a hint in the current isolate
1507         indicating that a full GC should be done upon the next page
1508         navigation.
1509
1510         This improves Chrome's robustness on some WebGL stress tests which
1511         simulate real-world behavior by repeatedly navigating among
1512         several samples. More general measures are being investigated, but
1513         this change makes V8 behave the same as JSC on these stress tests.
1514         JSC doesn't currently use generational garbage collection, so it
1515         has more opportunities to discover unreferenced canvas contexts.
1516
1517         Test: fast/canvas/webgl/context-creation-and-destruction.html
1518
1519         * bindings/v8/V8Binding.cpp:
1520         (WebCore::V8BindingPerIsolateData::V8BindingPerIsolateData):
1521             Initialize per-isolate low memory hint.
1522         * bindings/v8/V8Binding.h:
1523         (V8BindingPerIsolateData):
1524         (WebCore::V8BindingPerIsolateData::setLowMemoryNotificationHint):
1525             Set a per-isolate hint to signal a low memory condition upon the next page navigation.
1526         (WebCore::V8BindingPerIsolateData::clearLowMemoryNotificationHint):
1527             Clear the previously set hint.
1528         (WebCore::V8BindingPerIsolateData::isLowMemoryNotificationHint):
1529             Get the previously set hint.
1530         * bindings/v8/V8Proxy.cpp:
1531         (WebCore::V8Proxy::hintForGCIfNecessary):
1532             If necessary, send V8 a hint that it should GC.
1533         (WebCore):
1534         (WebCore::V8Proxy::clearForClose):
1535         (WebCore::V8Proxy::clearForNavigation):
1536             Call hintForGCIfNecessary.
1537         * bindings/v8/V8Proxy.h:
1538         (V8Proxy):
1539         * bindings/v8/custom/V8HTMLCanvasElementCustom.cpp:
1540         (WebCore::V8HTMLCanvasElement::getContextCallback):
1541             Set a hint that we should GC upon the next page navigation.
1542
1543 2012-07-24  Dave Tu  <dtu@chromium.org>
1544
1545         [chromium] Add time spent painting to GPU benchmarking renderingStats() API.
1546         https://bugs.webkit.org/show_bug.cgi?id=90019
1547
1548         Reviewed by Adrienne Walker.
1549
1550         * platform/graphics/chromium/BitmapCanvasLayerTextureUpdater.cpp:
1551         (WebCore::BitmapCanvasLayerTextureUpdater::prepareToUpdate):
1552         * platform/graphics/chromium/BitmapCanvasLayerTextureUpdater.h:
1553         (BitmapCanvasLayerTextureUpdater):
1554         * platform/graphics/chromium/BitmapSkPictureCanvasLayerTextureUpdater.cpp:
1555         (WebCore::BitmapSkPictureCanvasLayerTextureUpdater::Texture::prepareRect):
1556         (WebCore::BitmapSkPictureCanvasLayerTextureUpdater::prepareToUpdate):
1557         (WebCore::BitmapSkPictureCanvasLayerTextureUpdater::paintContentsRect):
1558         * platform/graphics/chromium/BitmapSkPictureCanvasLayerTextureUpdater.h:
1559         (Texture):
1560         (BitmapSkPictureCanvasLayerTextureUpdater):
1561         * platform/graphics/chromium/CanvasLayerTextureUpdater.cpp:
1562         (WebCore::CanvasLayerTextureUpdater::paintContents):
1563         * platform/graphics/chromium/CanvasLayerTextureUpdater.h:
1564         (CanvasLayerTextureUpdater):
1565         * platform/graphics/chromium/ContentLayerChromium.cpp:
1566         (WebCore::ContentLayerChromium::update):
1567         * platform/graphics/chromium/ContentLayerChromium.h:
1568         (ContentLayerChromium):
1569         * platform/graphics/chromium/ImageLayerChromium.cpp:
1570         (WebCore::ImageLayerChromium::update):
1571         * platform/graphics/chromium/ImageLayerChromium.h:
1572         (ImageLayerChromium):
1573         * platform/graphics/chromium/LayerChromium.h:
1574         (WebCore):
1575         (WebCore::LayerChromium::update):
1576         * platform/graphics/chromium/LayerTextureUpdater.h:
1577         (WebCore):
1578         (WebCore::LayerTextureUpdater::Texture::prepareRect):
1579         (WebCore::LayerTextureUpdater::prepareToUpdate):
1580         * platform/graphics/chromium/ScrollbarLayerChromium.cpp:
1581         (WebCore::ScrollbarLayerChromium::updatePart):
1582         (WebCore::ScrollbarLayerChromium::update):
1583         * platform/graphics/chromium/ScrollbarLayerChromium.h:
1584         (ScrollbarLayerChromium):
1585         * platform/graphics/chromium/SkPictureCanvasLayerTextureUpdater.cpp:
1586         (WebCore::SkPictureCanvasLayerTextureUpdater::prepareToUpdate):
1587         * platform/graphics/chromium/SkPictureCanvasLayerTextureUpdater.h:
1588         (SkPictureCanvasLayerTextureUpdater):
1589         * platform/graphics/chromium/TiledLayerChromium.cpp:
1590         (WebCore::TiledLayerChromium::updateTiles):
1591         (WebCore::TiledLayerChromium::updateContentRect):
1592         * platform/graphics/chromium/TiledLayerChromium.h:
1593         (TiledLayerChromium):
1594         * platform/graphics/chromium/cc/CCLayerTreeHost.cpp:
1595         (WebCore::CCLayerTreeHost::CCLayerTreeHost):
1596         (WebCore::CCLayerTreeHost::updateAnimations):
1597         (WebCore::CCLayerTreeHost::renderingStats):
1598         (WebCore::CCLayerTreeHost::paintMasksForRenderSurface):
1599         (WebCore::CCLayerTreeHost::paintLayerContents):
1600         * platform/graphics/chromium/cc/CCLayerTreeHost.h:
1601         (WebCore):
1602         (CCLayerTreeHost):
1603         * platform/graphics/chromium/cc/CCRenderingStats.h:
1604         (CCRenderingStats):
1605         (WebCore::CCRenderingStats::CCRenderingStats):
1606
1607 2012-07-24  Jian Li  <jianli@chromium.org>
1608
1609         [V8] Correct WebIDL type should be used in the testing IDL file
1610         https://bugs.webkit.org/show_bug.cgi?id=92168
1611
1612         Reviewed by Kentaro Hara.
1613
1614         Updated bindings test.
1615
1616         * bindings/scripts/test/GObject/WebKitDOMTestObj.cpp:
1617         (webkit_dom_test_obj_strict_function):
1618         * bindings/scripts/test/GObject/WebKitDOMTestObj.h:
1619         * bindings/scripts/test/JS/JSTestObj.cpp:
1620         (WebCore::jsTestObjPrototypeFunctionStrictFunction):
1621         * bindings/scripts/test/TestObj.idl:
1622         * bindings/scripts/test/V8/V8TestObj.cpp:
1623         (WebCore::TestObjV8Internal::enabledAtRuntimeMethod1Callback):
1624         (WebCore::TestObjV8Internal::enabledAtRuntimeMethod2Callback):
1625         (WebCore::TestObjV8Internal::enabledPerContextMethod1Callback):
1626         (WebCore::TestObjV8Internal::enabledPerContextMethod2Callback):
1627         (WebCore::TestObjV8Internal::strictFunctionCallback):
1628
1629 2012-07-24  Kent Tamura  <tkent@chromium.org>
1630
1631         Unreviewed, rolling out r123191.
1632         http://trac.webkit.org/changeset/123191
1633         https://bugs.webkit.org/show_bug.cgi?id=91804
1634
1635         It might cause another performance regression.
1636         http://code.google.com/p/chromium/issues/detail?id=138410
1637
1638         * html/FormController.cpp:
1639         (FormKeyGenerator):
1640         (WebCore::createKey):
1641         (WebCore::FormKeyGenerator::formKey):
1642         (WebCore::FormKeyGenerator::willDeleteForm):
1643
1644 2012-07-24  Vineet Chaudhary  <rgf748@motorola.com>
1645
1646         [JSBinding] Merging jsUnsignedLongArrayToVector() to toNativeArray() using traits.
1647         https://bugs.webkit.org/show_bug.cgi?id=90381
1648
1649         Reviewed by Kentaro Hara.
1650
1651         We can remove jsUnsignedLongArrayToVector() implementaion
1652         merging it to current and toNativeArray() traits.
1653
1654         Tests: TestObj.idl
1655         There should be no behavoiral changes as just refactoring.
1656
1657         * Modules/vibration/NavigatorVibration.idl: Use sequence<T> as method param than T[]
1658         * bindings/js/JSDOMBinding.cpp: Moved toJSSequence() to header file.
1659         * bindings/js/JSDOMBinding.h:
1660         (WebCore::toJSSequence): Moved toJSSequence() to header file.
1661         (WebCore::jsArray): Renamed Traits to JSValueTraits
1662         (WebCore::toNativeArray): Using traits.
1663         * bindings/js/JSDictionary.cpp:
1664         (WebCore::JSDictionary::convertValue):
1665         * bindings/js/JSMessagePortCustom.cpp:
1666         (WebCore::fillMessagePortArray):
1667         * bindings/scripts/CodeGeneratorJS.pm: Removed unsigned long[] specific binding code.
1668         (AddIncludesForType):
1669         (GenerateParametersCheckExpression):
1670         (JSValueToNative):
1671         (NativeToJSValue):
1672         * bindings/scripts/CodeGeneratorV8.pm:
1673         (GenerateParametersCheckExpression): Added check sequence<T> check for overloaded methods.
1674         * bindings/scripts/test/JS/JSTestObj.cpp: Rebased binding test.
1675         (WebCore::jsTestObjPrototypeFunctionOverloadedMethod10):
1676         (WebCore::jsTestObjPrototypeFunctionOverloadedMethod):
1677         (WebCore::jsTestObjPrototypeFunctionMethodWithUnsignedLongSequence):
1678         * bindings/scripts/test/JS/JSTestObj.h: Rebased binding test.
1679         * bindings/scripts/test/TestObj.idl: Added test for sequence<T> as method parameter.
1680         * bindings/scripts/test/V8/V8TestObj.cpp: Rebased binding test.
1681         (WebCore::TestObjV8Internal::overloadedMethod10Callback):
1682         (TestObjV8Internal):
1683         (WebCore::TestObjV8Internal::overloadedMethodCallback):
1684
1685 2012-07-24  Kwang Yul Seo  <skyul@company100.net>
1686
1687         Remove an invalid assertion introduced in r123536. Unreviewed.
1688
1689         * xml/parser/MarkupTokenBase.h:
1690         (WebCore::AtomicMarkupTokenBase::clearExternalCharacters):
1691
1692 2012-07-24  Sheriff Bot  <webkit.review.bot@gmail.com>
1693
1694         Unreviewed, rolling out r123504.
1695         http://trac.webkit.org/changeset/123504
1696         https://bugs.webkit.org/show_bug.cgi?id=92169
1697
1698         broke a bunch of text-related tests in chromium mac debug
1699         (Requested by zhenyao on #webkit).
1700
1701         * platform/text/cf/StringCF.cpp:
1702         (WTF::String::createCFString):
1703         * platform/text/cf/StringImplCF.cpp:
1704         (WTF::StringImpl::createCFString):
1705
1706 2012-07-24  Kwang Yul Seo  <skyul@company100.net>
1707
1708         Ensure Noah's ark without reading the DOM tree.
1709         https://bugs.webkit.org/show_bug.cgi?id=92065
1710
1711         Reviewed by Adam Barth.
1712
1713         Technically we shouldn't read attributes back from the DOM. If JavaScript changes
1714         the attributes values, we could get a slightly wrong output here.
1715
1716         Read attributes from tokens saved in the active formatting element list.
1717
1718         No new tests, covered by existing tests.
1719
1720         * html/parser/HTMLFormattingElementList.cpp:
1721         (WebCore::attributeCount):
1722         (WebCore::HTMLFormattingElementList::append):
1723         (WebCore::HTMLFormattingElementList::tryToEnsureNoahsArkConditionQuickly):
1724         (WebCore::HTMLFormattingElementList::ensureNoahsArkCondition):
1725         * html/parser/HTMLFormattingElementList.h:
1726         (HTMLFormattingElementList):
1727         * html/parser/HTMLStackItem.h:
1728         (WebCore::HTMLStackItem::localName):
1729         (HTMLStackItem):
1730
1731 2012-07-23  Kwang Yul Seo  <skyul@company100.net>
1732
1733         Clear the external characters pointer of an AtomicHTMLToken before the raw token is cleared.
1734         https://bugs.webkit.org/show_bug.cgi?id=92056
1735
1736         Reviewed by Adam Barth.
1737
1738         AtomicHTMLToken keeps a pointer to the HTMLToken's buffer instead of copying the
1739         characters for performance. Clear the external characters pointer before the raw token
1740         is cleared to make sure that we won't have a dangling pointer.
1741
1742         No new tests - no functional changes.
1743
1744         * html/parser/HTMLTreeBuilder.cpp:
1745         (WebCore::HTMLTreeBuilder::constructTreeFromToken):
1746         * xml/parser/MarkupTokenBase.h:
1747         (WebCore::AtomicMarkupTokenBase::clearExternalCharacters):
1748         (AtomicMarkupTokenBase):
1749
1750 2012-07-24  Jian Li  <jianli@chromium.org>
1751
1752         Add per-context setting for html notifications
1753         https://bugs.webkit.org/show_bug.cgi?id=91295
1754
1755         Reviewed by Adam Barth.
1756
1757         Added a new per-context setting to control the enabling of html notifications.
1758         For chromium port, we're going to disable html notifications for web
1759         pages, but still keep it enabled for extensions.
1760
1761         Also enhance V8 code generator to support V8EnabledPerContext attribute
1762         for methods.
1763
1764         Updated the binding tests.
1765
1766         * Modules/notifications/NotificationCenter.cpp:
1767         (WebCore::NotificationCenter::document): Expose document for being used by installPerContextProperties.
1768         (WebCore):
1769         * Modules/notifications/NotificationCenter.h:
1770         (NotificationCenter):
1771         * Modules/notifications/NotificationCenter.idl: Add V8EnabledPerContext attribute.
1772         * bindings/scripts/CodeGeneratorV8.pm: V8EnabledPerContext now supports methods.
1773         (GenerateHeader):
1774         (IsStandardFunction):
1775         (GenerateNonStandardFunction):
1776         (GenerateImplementation):
1777         (GenerateToV8Converters):
1778         (GetContextEnableFunction):
1779         * bindings/scripts/test/TestObj.idl: Add test case for V8EnabledPerContext methods.
1780         * bindings/scripts/test/V8/V8TestObj.cpp: Update test result.
1781         (WebCore::TestObjV8Internal::enabledAtContextMethod1Callback):
1782         (TestObjV8Internal):
1783         (WebCore::TestObjV8Internal::enabledAtContextMethod2Callback):
1784         (WebCore::V8TestObj::installPerContextProperties):
1785         (WebCore::V8TestObj::wrapSlow):
1786         * dom/ContextFeatures.cpp: Add a new per-context feature.
1787         (WebCore::ContextFeatures::htmlNotificationsEnabled):
1788         (WebCore):
1789         * dom/ContextFeatures.h: Add a new per-context feature.
1790
1791 2012-07-24  Sheriff Bot  <webkit.review.bot@gmail.com>
1792
1793         Unreviewed, rolling out r123499.
1794         http://trac.webkit.org/changeset/123499
1795         https://bugs.webkit.org/show_bug.cgi?id=92161
1796
1797         Android fails to compile (Requested by zhenyao on #webkit).
1798
1799         * GNUmakefile.list.am:
1800         * Modules/mediastream/MediaStream.cpp:
1801         (WebCore::MediaStream::MediaStream):
1802         (WebCore::MediaStream::addTrack):
1803         * Modules/mediastream/MediaStreamTrack.cpp:
1804         (WebCore::MediaStreamTrack::create):
1805         (WebCore::MediaStreamTrack::MediaStreamTrack):
1806         (WebCore::MediaStreamTrack::~MediaStreamTrack):
1807         (WebCore::MediaStreamTrack::setEnabled):
1808         * Modules/mediastream/MediaStreamTrack.h:
1809         (MediaStreamTrack):
1810         * Modules/mediastream/MediaStreamTrack.idl:
1811         * WebCore.gypi:
1812         * dom/EventNames.h:
1813         (WebCore):
1814         * dom/EventTargetFactory.in:
1815         * platform/chromium/support/WebMediaStreamDescriptor.cpp:
1816         (WebKit):
1817         (WebKit::WebMediaStreamDescriptor::sources):
1818         * platform/chromium/support/WebMediaStreamSource.cpp:
1819         * platform/mediastream/MediaStreamSource.cpp: Removed.
1820         * platform/mediastream/MediaStreamSource.h:
1821         (WebCore::MediaStreamSource::create):
1822         (WebCore::MediaStreamSource::muted):
1823         (WebCore::MediaStreamSource::setMuted):
1824         (WebCore::MediaStreamSource::MediaStreamSource):
1825         (MediaStreamSource):
1826         * platform/mediastream/chromium/MediaStreamCenterChromium.cpp:
1827         (WebCore::MediaStreamCenterChromium::queryMediaStreamSources):
1828         (WebCore::MediaStreamCenterChromium::didCreateMediaStream):
1829
1830 2012-07-24  Julien Chaffraix  <jchaffraix@webkit.org>
1831
1832         Crash in RenderTableSection::layoutRows
1833         https://bugs.webkit.org/show_bug.cgi?id=92145
1834
1835         Reviewed by Abhishek Arya.
1836
1837         The crash came from the insertChildNode logic being confused and choosing the wrong
1838         renderer for insertion. This is because after r123159, table anonymous wrappers can
1839         be created as 'inline-table' and wouldn't pass the isAnonymousBlock() check.
1840
1841         This change also aligns the ancestor iteration with RenderBlock::addChildToAnonymousColumnBlocks.
1842
1843         Test: fast/table/crash-table-section-layoutRows.html
1844
1845         * rendering/RenderObjectChildList.cpp:
1846         (WebCore::RenderObjectChildList::insertChildNode):
1847         Changed the code to always get the right child. There is no justification to not
1848         walking up to the |owner|'s direct child in all cases.
1849
1850 2012-07-24  Tony Chang  <tony@chromium.org>
1851
1852         Make RenderBlock::paintChildren virtual so sub classes can change the paint order
1853         https://bugs.webkit.org/show_bug.cgi?id=92042
1854
1855         Reviewed by Ojan Vafai.
1856
1857         The flexbox spec says that the order property should change the paint order of flex children.
1858         To make that possible, make paintChildren virtual so RenderFlexibleBox can override it.
1859
1860         No new tests, this is just a refactoring.
1861
1862         * rendering/RenderBlock.cpp:
1863         (WebCore::RenderBlock::paintContents): Move some code out of paintChildren into paintContents.
1864         (WebCore::RenderBlock::paintChildren): Make virtual and just have it be a simple for loop.
1865         (WebCore::RenderBlock::paintChild): Move code out of paintChildren so subclasses can reuse.
1866         * rendering/RenderBlock.h:
1867         (RenderBlock): Make paintChildren virtual and protected so sub classes can call paintChild.
1868
1869 2012-07-24  Anna Cavender  <annacc@chromium.org>
1870
1871         Create SourceBuffer and SourceBufferList objects
1872         https://bugs.webkit.org/show_bug.cgi?id=91771
1873
1874         Reviewed by Eric Carlson.
1875
1876         SourceBuffer and SourceBufferList objects are needed in order to
1877         implement the new object-oriented MediaSource API:
1878         http://dvcs.w3.org/hg/html-media/raw-file/tip/media-source/media-source.html
1879
1880         No new tests - will be able to test after landing BUGWK91775
1881
1882         Adding new files to build files:
1883         * CMakeLists.txt:
1884         * DerivedSources.cpp:
1885         * DerivedSources.make:
1886         * DerivedSources.pri:
1887         * GNUmakefile.am:
1888         * GNUmakefile.list.am:
1889         * Target.pri:
1890         * WebCore.gyp/WebCore.gyp:
1891         * WebCore.gypi:
1892         * WebCore.vcproj/WebCore.vcproj:
1893         * WebCore.vcproj/WebCoreCommon.vsprops:
1894         * WebCore.vcproj/copyForwardingHeaders.cmd:
1895         * WebCore.xcodeproj/project.pbxproj:
1896
1897         New objects:
1898         * Modules/mediasource/SourceBuffer.cpp: Added.
1899         * Modules/mediasource/SourceBuffer.h: Added.
1900         * Modules/mediasource/SourceBuffer.idl: Added.
1901         * Modules/mediasource/SourceBufferList.cpp: Added.
1902         * Modules/mediasource/SourceBufferList.h: Added.
1903         * Modules/mediasource/SourceBufferList.idl: Added.
1904
1905         New events:
1906         * dom/EventNames.h:
1907         (WebCore):
1908         * dom/EventTarget.h:
1909         (WebCore):
1910         * dom/EventTargetFactory.in:
1911
1912         Object constructors:
1913         * page/DOMWindow.idl:
1914
1915 2012-07-24  Patrick Gansterer  <paroga@webkit.org>
1916
1917         Store the full year in GregorianDateTime
1918         https://bugs.webkit.org/show_bug.cgi?id=92067
1919
1920         Reviewed by Geoffrey Garen.
1921
1922         Use the full year instead of the offset from year 1900
1923         for the year member variable of GregorianDateTime.
1924
1925         * bridge/qt/qt_runtime.cpp:
1926         (JSC::Bindings::convertValueToQVariant):
1927         (JSC::Bindings::convertQVariantToValue):
1928         * bridge/qt/qt_runtime_qt4.cpp:
1929         (JSC::Bindings::convertValueToQVariant):
1930         (JSC::Bindings::convertQVariantToValue):
1931
1932 2012-07-24  Benjamin Poulain  <bpoulain@apple.com>
1933
1934         Do not perform 8 to 16bits characters conversion when converting a WTFString to NSString/CFString
1935         https://bugs.webkit.org/show_bug.cgi?id=90720
1936
1937         Reviewed by Geoffrey Garen.
1938
1939         In most String to CFString conversion, we should be able to use the "NoCopy" constructor and have
1940         a relatively cheap conversion from WTF::String to CFString.
1941
1942         When the String is 8 bits, it was converted to 16 bits by getData16SlowCase() because of the call
1943         to String::characters().
1944
1945         This patch adds a path for creating a CFString from a 8bits string using CFStringCreateWithBytes.
1946
1947         This is covered by existing tests.
1948
1949         * platform/text/cf/StringCF.cpp:
1950         (WTF::String::createCFString): CFSTR() create static CFString, it is unecessary to retain it.
1951         * platform/text/cf/StringImplCF.cpp:
1952         (WTF::StringImpl::createCFString): The logic to avoid the StringWrapperCFAllocator has also been simplified.
1953         The allocator creation is now closer to where it is useful.
1954
1955 2012-07-24  Kentaro Hara  <haraken@chromium.org>
1956
1957         [V8] String wrappers should be marked Independent
1958         https://bugs.webkit.org/show_bug.cgi?id=91251
1959
1960         Reviewed by Adam Barth.
1961
1962         Currently V8 String wrappers are not marked Independent.
1963         By marking them Independent, they can be reclaimed by the scavenger GC.
1964         Although I couldn't find a case where this change reduces memory usage,
1965         this change would be important for upcoming changes in string conversion
1966         between V8 and WebKit (https://bugs.webkit.org/show_bug.cgi?id=91850).
1967
1968         'm_lastStringImpl = 0' in StringCache::remove() is important.
1969         Look at the following code:
1970
1971             static void cachedStringCallback(v8::Persistent<v8::Value> wrapper, void* parameter)
1972             {
1973                 ...;
1974                 stringCache()->remove(stringImpl);
1975                 wrapper.Dispose();
1976             }
1977
1978             void StringCache::remove(StringImpl* stringImpl)
1979             {
1980                 ...
1981                 if (m_lastStringImpl.get() == stringImpl)
1982                     m_lastStringImpl = 0;
1983             }
1984
1985             v8::Local<v8::String> v8ExternalString(StringImpl* stringImpl, v8::Isolate* isolate)
1986             {
1987                 if (m_lastStringImpl.get() == stringImpl) {
1988                     return v8::Local<v8::String>::New(m_lastV8String); // m_lastV8String points to a wrapper object that was accessed most recently.
1989                 }
1990                 return v8ExternalStringSlow(stringImpl, isolate);
1991             }
1992
1993         Without 'm_lastStringImpl = 0', already disposed m_lastV8String can be used
1994         in v8ExternalString(). This was a cause of the crashes of r122614.
1995
1996         Tests: At the initial commit of this patch (r122614),
1997                the following tests had been broken due to missing 'm_lastStringImpl = 0'.
1998                fast/workers/worker-location.html
1999                Dromaeo/cssquery-jquery.html
2000                Dromaeo/jslib-event-jquery.html
2001                Dromaeo/jslib-style-jquery.html
2002                Dromaeo/jslib-style-prototype.html
2003
2004         * bindings/v8/V8Binding.cpp:
2005         (WebCore::StringCache::remove):
2006         (WebCore::StringCache::v8ExternalStringSlow):
2007
2008 2012-07-24  Tommy Widenflycht  <tommyw@google.com>
2009
2010         MediaStream API: Update MediaStreamTrack to match the specification
2011         https://bugs.webkit.org/show_bug.cgi?id=90180
2012
2013         Reviewed by Adam Barth.
2014
2015         MediaStreamTracks are now required to show the status of the underlying source,
2016         and trigger events when that status changes.
2017
2018         Test: fast/mediastream/MediaStreamTrack.html
2019
2020         * GNUmakefile.list.am:
2021         * Modules/mediastream/MediaStream.cpp:
2022         (WebCore::MediaStream::MediaStream):
2023         (WebCore::MediaStream::addTrack):
2024         * Modules/mediastream/MediaStreamTrack.cpp:
2025         (WebCore::MediaStreamTrack::create):
2026         (WebCore::MediaStreamTrack::MediaStreamTrack):
2027         (WebCore::MediaStreamTrack::~MediaStreamTrack):
2028         (WebCore::MediaStreamTrack::setEnabled):
2029         (WebCore::MediaStreamTrack::readyState):
2030         (WebCore):
2031         (WebCore::MediaStreamTrack::sourceChangedState):
2032         (WebCore::MediaStreamTrack::stop):
2033         (WebCore::MediaStreamTrack::interfaceName):
2034         (WebCore::MediaStreamTrack::scriptExecutionContext):
2035         (WebCore::MediaStreamTrack::eventTargetData):
2036         (WebCore::MediaStreamTrack::ensureEventTargetData):
2037         * Modules/mediastream/MediaStreamTrack.h:
2038         * Modules/mediastream/MediaStreamTrack.idl:
2039         * WebCore.gypi:
2040         * dom/EventNames.h:
2041         (WebCore):
2042         * dom/EventTargetFactory.in:
2043         * platform/chromium/support/WebMediaStreamDescriptor.cpp:
2044         * platform/chromium/support/WebMediaStreamSource.cpp:
2045         (WebKit::WebMediaStreamSource::setReadyState):
2046         (WebKit):
2047         (WebKit::WebMediaStreamSource::readyState):
2048         * platform/mediastream/MediaStreamSource.cpp: Copied from Source/WebCore/platform/mediastream/MediaStreamSource.h.
2049         (WebCore):
2050         (WebCore::MediaStreamSource::create):
2051         (WebCore::MediaStreamSource::MediaStreamSource):
2052         (WebCore::MediaStreamSource::setReadyState):
2053         (WebCore::MediaStreamSource::addObserver):
2054         (WebCore::MediaStreamSource::removeObserver):
2055         * platform/mediastream/MediaStreamSource.h:
2056         (Observer):
2057         (WebCore::MediaStreamSource::Observer::~Observer):
2058         (MediaStreamSource):
2059         (WebCore::MediaStreamSource::readyState):
2060         * platform/mediastream/chromium/MediaStreamCenterChromium.cpp:
2061         (WebCore::MediaStreamCenterChromium::queryMediaStreamSources):
2062         (WebCore::MediaStreamCenterChromium::didCreateMediaStream):
2063
2064 2012-07-24  Scott Graham  <scottmg@chromium.org>
2065
2066         Use native (non-cygwin) binaries for perl, gperf, and bison in Chromium
2067         https://bugs.webkit.org/show_bug.cgi?id=91667
2068
2069         Reviewed by Tony Chang.
2070
2071         Using native tools instead of cygwin version improves build time
2072         performance by roughly 50% (on top of previous cl-instead-of-gcc
2073         change).
2074
2075         Also, use - instead of / for cl flags, otherwise preprocessing fails
2076         very slowly (due to retry loop). And, replace \ with / in gperf
2077         command line. Without this, gperf emits the filename literally in
2078         #line directives which causes VS to error with "unescaped \ in
2079         string". Bizarrely, building ColorData.cpp with those \ in place
2080         causes the IDE to exit with no error message, which was the cause of
2081         the previous canary failures.
2082
2083         No new tests.
2084
2085         * WebCore.gyp/WebCore.gyp:
2086         * WebCore.gyp/scripts/rule_bison.py:
2087         * bindings/scripts/preprocessor.pm:
2088         (applyPreprocessor):
2089         * make-hash-tools.pl:
2090
2091 2012-07-24  Daniel Cheng  <dcheng@chromium.org>
2092
2093         Files from drag and file <input> should use getMIMETypeForExtension to determine content type.
2094         https://bugs.webkit.org/show_bug.cgi?id=91702
2095
2096         Reviewed by Jian Li.
2097
2098         Awhile back, we changed File to only use getWellKnownMIMETypeForExtension to prevent web
2099         pages from being able to detect what applications a user has installed indirectly through
2100         the MIME types. However, some sites like YouTube's drag and drop uploader use MIME types
2101         that aren't in WebKit's internal list, so we relax the restriction for Files that originate
2102         from an user action.
2103
2104         * fileapi/File.cpp:
2105         (WebCore::getContentTypeFromFileName):
2106         (WebCore::createBlobDataForFile):
2107         (WebCore::createBlobDataForFileWithName):
2108         (WebCore::createBlobDataForFileWithMetadata):
2109         (WebCore::File::createWithRelativePath):
2110         (WebCore::File::File):
2111         * fileapi/File.h:
2112         (File):
2113         (WebCore::File::create):
2114         (WebCore::File::createWithName):
2115         * html/FileInputType.cpp:
2116         (WebCore::FileInputType::createFileList):
2117         * platform/chromium/ChromiumDataObject.cpp:
2118         (WebCore::ChromiumDataObject::addFilename):
2119         * platform/gtk/ClipboardGtk.cpp:
2120         (WebCore::ClipboardGtk::files):
2121         * platform/mac/ClipboardMac.mm:
2122         (WebCore::ClipboardMac::files):
2123         * platform/qt/ClipboardQt.cpp:
2124         (WebCore::ClipboardQt::files):
2125         * platform/win/ClipboardWin.cpp:
2126         (WebCore::ClipboardWin::files):
2127
2128 2012-07-24  Vsevolod Vlasov  <vsevik@chromium.org>
2129
2130         Web Inspector: Snippets UI polish
2131         https://bugs.webkit.org/show_bug.cgi?id=92142
2132
2133         Reviewed by Pavel Feldman.
2134
2135         Changed snippets context menu items.
2136         Added run snippet status bar button.
2137         Removed unused edit source button.
2138
2139         * English.lproj/localizedStrings.js:
2140         * inspector/front-end/Images/statusbarButtonGlyphs.png:
2141         * inspector/front-end/ScriptSnippetModel.js:
2142         (WebInspector.SnippetJavaScriptSourceFrame):
2143         (WebInspector.SnippetJavaScriptSourceFrame.prototype.statusBarItems):
2144         (WebInspector.SnippetJavaScriptSourceFrame.prototype._runButtonClicked):
2145         * inspector/front-end/ScriptsNavigator.js:
2146         (WebInspector.SnippetsNavigatorView.prototype.getOrCreateFolderTreeElement):
2147         (WebInspector.SnippetsNavigatorView.prototype.handleContextMenu):
2148         * inspector/front-end/ScriptsPanel.js:
2149         (WebInspector.ScriptsPanel.prototype._updateScriptViewStatusBarItems):
2150         (WebInspector.ScriptsPanel.prototype._createSourceFrame):
2151         * inspector/front-end/SourceFrame.js:
2152         (WebInspector.SourceFrame.prototype.statusBarItems):
2153         * inspector/front-end/inspector.css:
2154         * inspector/front-end/scriptsPanel.css:
2155         (.evaluate-snippet-status-bar-item .glyph):
2156         (.evaluate-snippet-status-bar-item.toggled .glyph):
2157
2158 2012-07-24  Nico Weber  <thakis@chromium.org>
2159
2160         [chromium/mac] Remove webcore_resource_files from WebCore.gypi
2161         https://bugs.webkit.org/show_bug.cgi?id=92129
2162
2163         Reviewed by Adam Barth.
2164
2165         It was added in http://trac.webkit.org/changeset/81706/trunk/Source/WebCore/WebCore.gypi .
2166         The changes in that revision were mostly reverted in http://trac.webkit.org/changeset/119921 ,
2167         but this block was forgotten. Since it just confused me for a bit, remove it.
2168
2169         * WebCore.gyp/WebCore.gyp:
2170
2171 2012-07-24  Julien Chaffraix  <jchaffraix@webkit.org>
2172
2173         inline-table wrapper should be generated for display: inline element only
2174         https://bugs.webkit.org/show_bug.cgi?id=92054
2175
2176         Reviewed by Abhishek Arya.
2177
2178         CSS 2.1 only takes into account the display when determining if we should generate
2179         an inline-table. My misreading of the spec would make us generate an inline-table
2180         wrapper for any inline formatting context.
2181
2182         Test: fast/table/inline-block-generates-table-wrapper.html
2183
2184         * rendering/RenderTable.cpp:
2185         (WebCore::RenderTable::createAnonymousWithParentRenderer):
2186         Changed to use display instead of isInline to properly match the spec.
2187
2188 2012-07-24  Alec Flett  <alecflett@chromium.org>
2189
2190         IndexedDB: Temporarily continue generating backend keys for empty key lists
2191         https://bugs.webkit.org/show_bug.cgi?id=92012
2192
2193         Reviewed by Tony Chang.
2194
2195         Temporarily change key behavior to allow chromium in_process_webkit
2196         to call putWithIndexKeys without inadvertently preventing index
2197         key generation. Note this is temporary until frontend key generation
2198         patch lands.
2199
2200         No new tests, this is part of a refactor that doesn't change behavior.
2201
2202         * Modules/indexeddb/IDBObjectStoreBackendImpl.cpp:
2203         (WebCore::IDBObjectStoreBackendImpl::putInternal):
2204
2205 2012-07-24  Pavel Feldman  <pfeldman@chromium.org>
2206
2207         Web Inspector: introduce UISourceCodeFrame.
2208         https://bugs.webkit.org/show_bug.cgi?id=92124
2209
2210         Reviewed by Vsevolod Vlasov.
2211
2212         Generic implementation of SourceFrame over UISourceCode added.
2213
2214         * WebCore.gypi:
2215         * WebCore.vcproj/WebCore.vcproj:
2216         * inspector/compile-front-end.py:
2217         * inspector/front-end/ScriptsPanel.js:
2218         (WebInspector.ScriptsPanel.prototype._createSourceFrame):
2219         * inspector/front-end/StylesPanel.js:
2220         * inspector/front-end/UISourceCodeFrame.js: Added.
2221         (WebInspector.UISourceCodeFrame):
2222         (WebInspector.UISourceCodeFrame.prototype.canEditSource):
2223         (WebInspector.UISourceCodeFrame.prototype.commitEditing):
2224         (WebInspector.UISourceCodeFrame.prototype.afterTextChanged):
2225         (WebInspector.UISourceCodeFrame.prototype._didEditContent):
2226         (WebInspector.UISourceCodeFrame.prototype._onContentChanged):
2227         (WebInspector.UISourceCodeFrame.prototype.populateTextAreaContextMenu):
2228         * inspector/front-end/WebKit.qrc:
2229         * inspector/front-end/inspector.html:
2230
2231 2012-07-24  Dan Bernstein  <mitz@apple.com>
2232
2233         <rdar://problem/11945102> REGRESSION (r109451): Overlay scrollbars always use the default style, regardless of background color
2234         https://bugs.webkit.org/show_bug.cgi?id=92115
2235
2236         Reviewed by Mark Rowe.
2237
2238         * platform/Scrollbar.cpp:
2239         (WebCore::Scrollbar::scrollbarOverlayStyle): Reversed an incorrect null check.
2240
2241 2012-07-24  Andrew Wilson  <atwilson@chromium.org>
2242
2243         Another prospective build fix for Chromium. Unreviewed.
2244
2245         * platform/graphics/harfbuzz/ng/HarfBuzzNGFace.h:
2246
2247 2012-07-24  Andrew Wilson  <atwilson@chromium.org>
2248
2249         Unreviewed prospective build fix for Chromium. Changed include of hb.h in HarfBuzzNGFace.h
2250
2251         * platform/graphics/harfbuzz/ng/HarfBuzzNGFace.h:
2252
2253 2012-07-24  Dominik Röttsches  <dominik.rottsches@intel.com>
2254
2255         HarfBuzzFaceNG.h forward declarations conflict with newer versions of Harfbuzz
2256         https://bugs.webkit.org/show_bug.cgi?id=92103
2257
2258         Reviewed by Philippe Normand.
2259
2260         As preparation for enabling harfbuzz-ng for EFL, I would like to include the header
2261         here since the forward declarations are conflicting with the harfbuzz version
2262         we will use for EFL.
2263
2264         No new tests, no change in behavior.
2265
2266         * platform/graphics/harfbuzz/ng/HarfBuzzNGFace.h:
2267
2268 2012-07-24  Allan Sandfeld Jensen  <allan.jensen@nokia.com>
2269
2270         Fix blend filter for autovectorizing
2271         https://bugs.webkit.org/show_bug.cgi?id=91398
2272
2273         Reviewed by Nikolas Zimmermann.
2274
2275         To support auto-vectorizing, the loop had to be unswitched, and regular arrays used that
2276         did not do boundary-checks in the inner loop. Finally the integer division by 255
2277         was optimized not use integer division intrinsics.
2278
2279         On a X86-64 architechture using GCC -O3 on the file, this gives a speed-up of 2.9x.
2280
2281         * platform/graphics/filters/FEBlend.cpp:
2282         (WebCore::fastDivideBy255):
2283         (BlendNormal):
2284         (WebCore::BlendNormal::apply):
2285         (BlendMultiply):
2286         (WebCore::BlendMultiply::apply):
2287         (BlendScreen):
2288         (WebCore::BlendScreen::apply):
2289         (BlendDarken):
2290         (WebCore::BlendDarken::apply):
2291         (BlendLighten):
2292         (WebCore::BlendLighten::apply):
2293         (BlendUnknown):
2294         (WebCore::BlendUnknown::apply):
2295         (WebCore::platformApply):
2296         (WebCore::FEBlend::platformApplyGeneric):
2297         (WebCore::FEBlend::platformApplySoftware):
2298         * platform/graphics/filters/FEBlend.h:
2299         (FEBlend):
2300
2301 2012-07-23  Hans Wennborg  <hans@chromium.org>
2302
2303         Speech JavaScript API: Add the SpeechRecognitionResult.emma attribute
2304         https://bugs.webkit.org/show_bug.cgi?id=91743
2305
2306         Reviewed by Adam Barth.
2307
2308         Implement the 'emma' attribute as described in the spec draft at
2309         http://dvcs.w3.org/hg/speech-api/raw-file/tip/speechapi.html
2310
2311         Test: fast/speech/scripted/emma.html
2312
2313         * Modules/speech/SpeechRecognitionResult.cpp:
2314         (WebCore::SpeechRecognitionResult::~SpeechRecognitionResult):
2315         (WebCore):
2316         (WebCore::SpeechRecognitionResult::emma):
2317         * Modules/speech/SpeechRecognitionResult.h:
2318         (WebCore):
2319         (SpeechRecognitionResult):
2320         * Modules/speech/SpeechRecognitionResult.idl:
2321         * WebCore.gypi:
2322         * bindings/v8/custom/V8SpeechRecognitionResultCustom.cpp:
2323         (WebCore):
2324         (WebCore::V8SpeechRecognitionResult::visitDOMWrapper):
2325           We need to add an implicit reference from the result object to the
2326           emma object to keep the latters DOM tree alive across
2327           garbage-collection.
2328
2329 2012-07-24  Andrei Poenaru  <poenaru@adobe.com>
2330
2331         Web Inspector: Protocol Extension: add getFlowByName command
2332         https://bugs.webkit.org/show_bug.cgi?id=91855
2333
2334         Reviewed by Pavel Feldman.
2335
2336         The "getFlowByName" command should return a NamedFlow for a given document and name.
2337
2338         Test: inspector/styles/protocol-css-regions-commands.html
2339
2340         * inspector/Inspector.json:
2341         * inspector/InspectorCSSAgent.cpp:
2342         (WebCore::InspectorCSSAgent::getFlowByName):
2343         (WebCore):
2344         * inspector/InspectorCSSAgent.h:
2345         (InspectorCSSAgent):
2346         * inspector/front-end/CSSStyleModel.js:
2347         (WebInspector.CSSStyleModel.prototype.getNamedFlowCollectionAsync):
2348         (WebInspector.CSSStyleModel.prototype.getFlowByNameAsync):
2349         (WebInspector.NamedFlow):
2350         (WebInspector.NamedFlow.parsePayload):
2351
2352 2012-07-23  Oswald Buddenhagen  <oswald.buddenhagen@nokia.com>
2353
2354         [Qt] Fix compilation against namespaced Qt
2355
2356         Reviewed by Simon Hausmann.
2357
2358         * platform/graphics/Font.h:
2359         * platform/network/ResourceHandleInternal.h:
2360         * platform/qt/ThirdPartyCookiesQt.h:
2361
2362 2012-07-24  Anthony Scian  <ascian@rim.com>
2363
2364         Web Inspector [JSC]: Enable initiator column in network panel.
2365         https://bugs.webkit.org/show_bug.cgi?id=65533
2366
2367         Reviewed by Vsevolod Vlasov.
2368
2369         With 40118 fixed for webkit, Preferences.displayInitiator should be set to true.
2370         Removed references to it since the optionality is no longer required.
2371
2372         Test: http/tests/inspector/network/network-initiator.html
2373
2374         * inspector/front-end/NetworkPanel.js:
2375         (WebInspector.NetworkLogView.prototype._createTable):
2376         (WebInspector.NetworkLogView.prototype.switchToDetailedView):
2377         (WebInspector.NetworkLogView.prototype.switchToBriefView):
2378         (WebInspector.NetworkDataGridNode.prototype.createCells):
2379         (WebInspector.NetworkDataGridNode.prototype.refreshRequest):
2380         * inspector/front-end/Settings.js:
2381
2382 2012-07-24  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
2383
2384         Missing *explicit* keyword in dom and page.
2385         https://bugs.webkit.org/show_bug.cgi?id=92074
2386
2387         Reviewed by Kentaro Hara.
2388
2389         Some constructors missed to use *explicit* keyword. They need to be added
2390         *explicit* keyword to contructor which has a parameter in order to avoid
2391         implicit type conversion.
2392
2393         Some files in dom directory can't use explicit keyword because of build breaks.
2394
2395         No new tests. Covered by existing tests.
2396
2397         * dom/BeforeLoadEvent.h:
2398         (WebCore::BeforeLoadEvent::BeforeLoadEvent):
2399         * dom/BeforeTextInsertedEvent.h:
2400         (BeforeTextInsertedEvent):
2401         * dom/ChildListMutationScope.h:
2402         (WebCore::ChildListMutationScope::ChildListMutationScope):
2403         * dom/ChildNodeList.h:
2404         (ChildNodeList):
2405         * dom/ClientRect.h:
2406         (ClientRect):
2407         * dom/ClientRectList.h:
2408         (ClientRectList):
2409         * dom/ComposedShadowTreeWalker.h:
2410         (ComposedShadowTreeParentWalker):
2411         * dom/DOMCoreException.h:
2412         (WebCore::DOMCoreException::DOMCoreException):
2413         * dom/DOMImplementation.h:
2414         (DOMImplementation):
2415         * dom/DatasetDOMStringMap.h:
2416         (WebCore::DatasetDOMStringMap::DatasetDOMStringMap):
2417         * dom/DeviceMotionController.h:
2418         (DeviceMotionController):
2419         * dom/DeviceOrientationController.h:
2420         (DeviceOrientationController):
2421         * dom/DocumentParser.h:
2422         (DocumentParser):
2423         * dom/EventException.h:
2424         (WebCore::EventException::EventException):
2425         * dom/EventListener.h:
2426         (WebCore::EventListener::EventListener):
2427         * dom/ExceptionBase.h:
2428         (ExceptionBase):
2429         * dom/GenericEventQueue.h:
2430         (GenericEventQueue):
2431         * dom/KeyboardEvent.h:
2432         (WebCore::KeypressCommand::KeypressCommand):
2433         * dom/MemoryInstrumentation.h:
2434         (WebCore::MemoryObjectInfo::MemoryObjectInfo):
2435         * dom/MessageChannel.h:
2436         (MessageChannel):
2437         * dom/MessagePort.h:
2438         (MessagePort):
2439         * dom/MessagePortChannel.h:
2440         (MessagePortChannel):
2441         * dom/MutationObserver.h:
2442         * dom/NamedNodeMap.h:
2443         (WebCore::NamedNodeMap::NamedNodeMap):
2444         * dom/NodeFilter.h:
2445         (WebCore::NodeFilter::NodeFilter):
2446         * dom/NodeWithIndex.h:
2447         (WebCore::NodeWithIndex::NodeWithIndex):
2448         * dom/Range.h:
2449         * dom/RangeException.h:
2450         (WebCore::RangeException::RangeException):
2451         * dom/RawDataDocumentParser.h:
2452         (WebCore::RawDataDocumentParser::RawDataDocumentParser):
2453         * dom/ScriptRunner.h:
2454         (ScriptRunner):
2455         * dom/SelectorQuery.h:
2456         (SelectorQuery):
2457         * dom/ShadowRoot.h:
2458         (ShadowRoot):
2459         * dom/SpaceSplitString.h:
2460         (SpaceSplitStringData):
2461         * dom/StaticHashSetNodeList.h:
2462         (StaticHashSetNodeList):
2463         * dom/StaticNodeList.h:
2464         (WebCore::StaticNodeList::StaticNodeList):
2465         * dom/TransformSource.h:
2466         (TransformSource):
2467         * dom/TreeScope.h:
2468         (TreeScope):
2469         * dom/WebKitNamedFlowCollection.h:
2470         (WebKitNamedFlowCollection):
2471         * page/EventHandler.h:
2472         (EventHandler):
2473         * page/FocusController.h:
2474         (FocusController):
2475         * page/FrameView.h:
2476         (FrameView):
2477         * page/MemoryInfo.h:
2478         (MemoryInfo):
2479         * page/Page.h:
2480         (Page):
2481         * page/PageGroup.h:
2482         (PageGroup):
2483         * page/PrintContext.h:
2484         (PrintContext):
2485         * page/Settings.h:
2486         (Settings):
2487         * page/SpeechInput.h:
2488         (SpeechInput):
2489         * page/SpeechInputResultList.h:
2490         (SpeechInputResultList):
2491         * page/WebKitAnimation.h:
2492         * page/WindowFeatures.h:
2493         (WindowFeatures):
2494         * page/WorkerNavigator.h:
2495         (WorkerNavigator):
2496
2497 2012-07-24  Kent Tamura  <tkent@chromium.org>
2498
2499         Replace some callsites of shadowAncestorNode() with shadowHost()
2500         https://bugs.webkit.org/show_bug.cgi?id=92060
2501
2502         Reviewed by Hajime Morita.
2503
2504         shadowAncestorNode() is deprecated.
2505
2506         No new tests. This shouldn't change behaviors.
2507
2508         * html/shadow/MediaControlElements.cpp:
2509         (WebCore::toParentMediaElement):
2510         * html/shadow/TextControlInnerElements.h:
2511         (WebCore::SpinButtonElement::isEnabledFormControl):
2512         (WebCore::SpinButtonElement::isReadOnlyFormControl):
2513         * platform/efl/RenderThemeEfl.cpp:
2514         (WebCore::RenderThemeEfl::paintMediaFullscreenButton):
2515         (WebCore::RenderThemeEfl::paintMediaMuteButton):
2516         * platform/gtk/RenderThemeGtk.cpp:
2517         (WebCore::getMediaElementFromRenderObject):
2518         (WebCore::centerRectVerticallyInParentInputElement):
2519         * platform/qt/RenderThemeQt.cpp:
2520         (WebCore::RenderThemeQt::paintSearchFieldCancelButton):
2521         (WebCore::RenderThemeQt::paintMediaSliderThumb):
2522         Removed an ASSERT() because nullness of o->node() is already asserted.
2523         * rendering/RenderInputSpeech.cpp:
2524         (WebCore::RenderInputSpeech::paintInputFieldSpeechButton):
2525         * rendering/RenderMediaControlsChromium.cpp:
2526         (WebCore::paintMediaSliderThumb):
2527         Passing shadowHost() to toParentMediaElement(), which accepts 0 and will return 0.
2528         (WebCore::paintMediaVolumeSliderThumb): ditto.
2529         * rendering/RenderThemeChromiumSkia.cpp:
2530         (WebCore::RenderThemeChromiumSkia::paintSearchFieldCancelButton):
2531         (WebCore::RenderThemeChromiumSkia::paintSearchFieldResultsDecoration):
2532         (WebCore::RenderThemeChromiumSkia::paintSearchFieldResultsButton):
2533         * rendering/RenderThemeMac.mm:
2534         (WebCore::RenderThemeMac::paintSearchFieldCancelButton):
2535         Remove an ASSERT(input) because input never be null unless o->node() is
2536         null. If o->node() is null, this function crashes.
2537         (WebCore::RenderThemeMac::paintSearchFieldResultsDecoration):
2538         (WebCore::RenderThemeMac::paintSearchFieldResultsButton):
2539         (WebCore::RenderThemeMac::paintMediaMuteButton):
2540         It's ok to ignore "node is a media element" case because this function
2541         works only if node->isMediaControlElement().
2542         (WebCore::RenderThemeMac::paintMediaPlayButton): ditto.
2543         (WebCore::RenderThemeMac::paintMediaSliderTrack): ditto.
2544         * rendering/RenderThemeSafari.cpp:
2545         (WebCore::RenderThemeSafari::paintSearchFieldCancelButton):
2546         Remove an ASSERT(input) because input never be null unless o->node() is
2547         null. If o->node() is null, this function crashes.
2548         (WebCore::RenderThemeSafari::paintSearchFieldResultsDecoration): ditto.
2549         (WebCore::RenderThemeSafari::paintSearchFieldResultsButton): ditto.
2550         * rendering/RenderThemeWinCE.cpp:
2551         (WebCore::mediaElementParent):
2552
2553 2012-07-24  Peter Beverloo  <peter@chromium.org>
2554
2555         [Chromium] Build fix for Android after r123424
2556         https://bugs.webkit.org/show_bug.cgi?id=92082
2557
2558         Unreviewed build fix.
2559
2560         Mark removed the PlatformSupport.h header from ScrollbarThemeChromiumAndroid.cpp
2561         in r123424, while there still are two references left to another method and a
2562         constant. Re-add the include directive.
2563
2564         * platform/chromium/ScrollbarThemeChromiumAndroid.cpp:
2565
2566 2012-07-24  Kentaro Hara  <haraken@chromium.org>
2567
2568         [JSC] REGRESSION(r122912): CodeGeneratorJS.pm generates wrong code
2569         for static attributes in supplemental IDL files
2570         https://bugs.webkit.org/show_bug.cgi?id=91925
2571
2572         Reviewed by Adam Barth.
2573
2574         r122912 implemented static attributes in CodeGeneratorJS.pm.
2575         However, it generates wrong code for static attributes in supplemental
2576         IDL files. This patch fixes it as shown in JSTestInterface.cpp.
2577         This fix aligns with the fix in CodeGeneratorV8.pm (r123308).
2578
2579         Test: bindings/scripts/test/TestSupplemental.idl
2580
2581         * bindings/scripts/CodeGeneratorJS.pm:
2582         (GenerateImplementation):
2583         * bindings/scripts/test/JS/JSTestInterface.cpp:
2584         (WebCore::jsTestInterfaceConstructorSupplementalStaticReadOnlyAttr):
2585         (WebCore::jsTestInterfaceConstructorSupplementalStaticAttr):
2586         (WebCore::setJSTestInterfaceConstructorSupplementalStaticAttr):
2587
2588 2012-07-24  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
2589
2590         Add *explicit* keyword to constructor which has a parameter in bridge, history, loader and plugins.
2591         https://bugs.webkit.org/show_bug.cgi?id=92064
2592
2593         Reviewed by Kentaro Hara.
2594
2595         Some constructors missed to use *explicit* keyword. They need to be added
2596         *explicit* keyword to contructor which has a parameter in order to avoid
2597         implicit type conversion.
2598
2599         No new tests. Covered by existing tests.
2600
2601         * bridge/IdentifierRep.h:
2602         (WebCore::IdentifierRep::IdentifierRep):
2603         * history/BackForwardListImpl.h:
2604         (BackForwardListImpl):
2605         * history/CachedFrame.h:
2606         (CachedFrame):
2607         * history/HistoryItem.h:
2608         (HistoryItem):
2609         * loader/DocumentWriter.h:
2610         (DocumentWriter):
2611         * loader/FrameNetworkingContext.h:
2612         (WebCore::FrameNetworkingContext::FrameNetworkingContext):
2613         * loader/HistoryController.h:
2614         * loader/ImageLoader.h:
2615         (ImageLoader):
2616         * loader/LinkLoader.h:
2617         (LinkLoader):
2618         * loader/MainResourceLoader.h:
2619         (MainResourceLoader):
2620         * loader/NavigationAction.h:
2621         (NavigationAction):
2622         * loader/NavigationScheduler.h:
2623         (NavigationScheduler):
2624         * loader/PolicyChecker.h:
2625         (PolicyChecker):
2626         * loader/SubframeLoader.h:
2627         (SubframeLoader):
2628         * loader/ThreadableLoaderClientWrapper.h:
2629         (WebCore::ThreadableLoaderClientWrapper::ThreadableLoaderClientWrapper):
2630         * plugins/PluginData.h:
2631         (PluginData):
2632         * plugins/PluginViewBase.h:
2633         (WebCore::PluginViewBase::PluginViewBase):
2634
2635 2012-07-24  Patrick Gansterer  <paroga@webkit.org>
2636
2637         [Qt] Fix conversion from GregorianDateTime to QTime
2638         https://bugs.webkit.org/show_bug.cgi?id=92068
2639
2640         Reviewed by Simon Hausmann.
2641
2642         * bridge/qt/qt_runtime.cpp:
2643         (JSC::Bindings::convertValueToQVariant):
2644         * bridge/qt/qt_runtime_qt4.cpp:
2645         (JSC::Bindings::convertValueToQVariant):
2646
2647 2012-07-24  Arko Saha  <arko@motorola.com>
2648
2649         Microdata: Implement PropertyNodeList interface.
2650         https://bugs.webkit.org/show_bug.cgi?id=80269
2651
2652         Reviewed by Ryosuke Niwa.
2653
2654         Added support for PropertyNodeList interface. Currently HTMLPropertiesCollection.namedItem(name)
2655         returns NodeList object. Made changes so that it will return PropertyNodeList object.
2656
2657         Implemented propertyNodeList.getValues() method. It returns an array of various values. Its
2658         values are obtained from the "itemValue" DOM property of each of the elements represented by
2659         the object, in tree order.
2660
2661         Spec: http://www.whatwg.org/specs/web-apps/current-work/multipage/common-dom-interfaces.html#propertynodelist
2662
2663         Tests: fast/dom/MicroData/nameditem-returns-propertynodelist.html
2664                fast/dom/MicroData/propertynodelist-getvalues-array-values-obtained-from-itemvalue-of-each-element.html
2665                fast/dom/MicroData/propertynodelist-getvalues-test.html
2666                fast/dom/MicroData/propertynodelist-test-add-remove-element.html
2667
2668         * CMakeLists.txt:
2669         * DerivedSources.cpp:
2670         * DerivedSources.make:
2671         * DerivedSources.pri:
2672         * GNUmakefile.list.am:
2673         * Target.pri:
2674         * UseJSC.cmake:
2675         * UseV8.cmake:
2676         * WebCore.gypi:
2677         * WebCore.vcproj/WebCore.vcproj:
2678         * WebCore.xcodeproj/project.pbxproj:
2679         * bindings/gobject/GNUmakefile.am:
2680         * bindings/js/JSBindingsAllInOne.cpp:
2681         * bindings/js/JSMicroDataItemValueCustom.cpp: Added.
2682         (WebCore):
2683         (WebCore::toJS): Added toJS() custom method. Returns Node if itemValue is node else returns String.
2684         * bindings/v8/custom/V8MicroDataItemValueCustom.cpp: Added.
2685         (WebCore):
2686         (WebCore::toV8): Added toV8 custom code.
2687         * dom/DynamicNodeList.cpp:
2688         (WebCore::DynamicNodeListCacheBase::rootNode): Return document if owner node has itemref attribute specified.
2689         (WebCore):
2690         (WebCore::DynamicNodeListCacheBase::ownerNodeHasItemRefAttribute): Returns true if owner node has itemref attribute
2691         specified with root type NodeListRootedAtDocumentIfOwnerHasItemrefAttr.
2692         (WebCore::DynamicNodeListCacheBase::invalidateCache):
2693         (WebCore::DynamicNodeList::itemWithName): Update itemrefElements cache if root type is
2694         NodeListIsRootedAtDocumentIfOwnerHasItemrefAttr.
2695         * dom/DynamicNodeList.h: Added m_hasItemRefElements in DynamicNodeListCacheBase to check if itemRefElmenets cache
2696         is valid.
2697         (WebCore::DynamicNodeListCacheBase::DynamicNodeListCacheBase):
2698         (WebCore::DynamicNodeListCacheBase::isRootedAtDocument):
2699         (DynamicNodeListCacheBase):
2700         (WebCore::DynamicNodeListCacheBase::isItemRefElementsCacheValid):
2701         (WebCore::DynamicNodeListCacheBase::setItemRefElementsCacheValid):
2702         (WebCore::DynamicNodeListCacheBase::rootType): Returns cached root type.
2703         (WebCore::DynamicNodeListCacheBase::shouldInvalidateTypeOnAttributeChange): Invalidate NodeListCache on itemref
2704         attribute change.
2705         * dom/Node.cpp:
2706         (WebCore::Node::propertyNodeList):
2707         * dom/Node.h:
2708         (WebCore):
2709         (Node):
2710         * dom/PropertyNodeList.cpp: Added.
2711         (WebCore):
2712         (WebCore::PropertyNodeList::PropertyNodeList):
2713         (WebCore::PropertyNodeList::~PropertyNodeList):
2714         (WebCore::PropertyNodeList::elementIsPropertyOfRefElement): Checks if the testElement is a property of Microdata
2715         item.
2716         (WebCore::PropertyNodeList::updateRefElements):
2717         (WebCore::PropertyNodeList::nodeMatches):
2718         (WebCore::PropertyNodeList::getValues): Returns PropertyValueArray whose values are obtained from the
2719         itemValue DOM property of each of the elements represented by the object, in tree order.
2720         * dom/PropertyNodeList.h: Added.
2721         (WebCore):
2722         (PropertyNodeList):
2723         (WebCore::PropertyNodeList::create):
2724         * dom/PropertyNodeList.idl: Added.
2725         * html/HTMLCollection.cpp:
2726         (WebCore::DynamicNodeListCacheBase::itemCommon): Update itemrefElements cache if its a NodeListCollectionType with
2727         root type NodeListIsRootedAtDocumentIfOwnerHasItemrefAttr.
2728         * html/HTMLElement.cpp:
2729         (WebCore::HTMLElement::properties):
2730         (WebCore):
2731         (WebCore::HTMLElement::getItemRefElements): Returns the ItemRefElements list of given element. Added this
2732         method to share the code between HTMLPropertiesCollection and PropertyNodeList to update itemRefElements.
2733         * html/HTMLElement.h:
2734         (HTMLElement):
2735         * html/HTMLPropertiesCollection.cpp:
2736         (WebCore::HTMLPropertiesCollection::HTMLPropertiesCollection):
2737         (WebCore::HTMLPropertiesCollection::updateRefElements):
2738         (WebCore::HTMLPropertiesCollection::updateNameCache):
2739         (WebCore::HTMLPropertiesCollection::namedItem): Return PropertyNodeList object instead of NodeList.
2740         (WebCore::HTMLPropertiesCollection::hasNamedItem): Return true if it finds an item with given
2741         name in the properties collection.
2742         * html/HTMLPropertiesCollection.h: Removed m_hasItemRefElements member variable, now DynamicNodeListCacheBase
2743         has isItemRefElementsCacheValid. Also removed m_hasPropertyNameCache, as DynamicNodeListCacheBase already has
2744         hasNameCache() and setHasNameCache().
2745         (WebCore):
2746         (HTMLPropertiesCollection):
2747         (WebCore::HTMLPropertiesCollection::invalidateCache):
2748         (WebCore::HTMLPropertiesCollection::updatePropertyCache):
2749         * html/HTMLPropertiesCollection.idl:
2750         * html/MicroDataItemValue.idl: Added.
2751         * page/DOMWindow.idl:
2752
2753 2012-07-24  Kentaro Hara  <haraken@chromium.org>
2754
2755         [JSC] REGRESSION(r122912): CodeGeneratorJS.pm should not
2756         implicitly assume ScriptExecutionContext for static attributes
2757         https://bugs.webkit.org/show_bug.cgi?id=91924
2758
2759         Reviewed by Adam Barth.
2760
2761         r122912 implemented static attributes in CodeGeneratorJS.pm.
2762         However, the generated code assumes that static attributes
2763         always require ScriptExecutionContext, which is wrong.
2764         If we need a ScriptExecutionContext, we should specify
2765         [CallWith=ScriptExecutionContext].
2766
2767         This patch fixes CodeGeneratorJS.pm so that static attributes
2768         do not assume ScriptExecutionContext. This fix aligns with
2769         the fix in CodeGeneratorV8.pm in r123308.
2770
2771         Test: bindings/scripts/test/TestObj.idl
2772
2773         * bindings/scripts/CodeGeneratorJS.pm:
2774         (GenerateImplementation):
2775         * bindings/scripts/test/JS/JSTestInterface.cpp:
2776         (WebCore::jsTestInterfaceConstructorSupplementalStaticReadOnlyAttr):
2777         (WebCore::jsTestInterfaceConstructorSupplementalStaticAttr):
2778         (WebCore::setJSTestInterfaceConstructorSupplementalStaticAttr):
2779         * bindings/scripts/test/JS/JSTestObj.cpp:
2780         (WebCore::jsTestObjConstructorStaticReadOnlyIntAttr):
2781         (WebCore::jsTestObjConstructorStaticStringAttr):
2782         (WebCore::setJSTestObjConstructorStaticStringAttr):
2783
2784 2012-07-23  Gyuyoung Kim  <gyuyoung.kim@samsung.com>
2785
2786         Missing *explicit* keyword in editing.
2787         https://bugs.webkit.org/show_bug.cgi?id=92055
2788
2789         Reviewed by Ryosuke Niwa.
2790
2791         Some constructors missed to use *explicit* keyword. They need to be added
2792         *explicit* keyword to contructor which has a parameter in order to avoid
2793         implicit type conversion.
2794
2795         No new tests. Covered by existing tests.
2796
2797         * editing/AlternativeTextController.h:
2798         (WebCore::AlternativeTextController::UNLESS_ENABLED):
2799         * editing/BreakBlockquoteCommand.h:
2800         (BreakBlockquoteCommand):
2801         * editing/DeleteButton.h:
2802         (DeleteButton):
2803         * editing/DeleteButtonController.h:
2804         (DeleteButtonController):
2805         * editing/EditCommand.h:
2806         (EditCommand):
2807         (WebCore::SimpleEditCommand::SimpleEditCommand):
2808         * editing/EditingBehavior.h:
2809         (WebCore::EditingBehavior::EditingBehavior):
2810         * editing/EditingStyle.h:
2811         (EditingStyle):
2812         * editing/Editor.h:
2813         (Editor):
2814         * editing/FrameSelection.h:
2815         (FrameSelection):
2816         * editing/InsertLineBreakCommand.h:
2817         (InsertLineBreakCommand):
2818         * editing/ModifySelectionListLevel.h:
2819         (ModifySelectionListLevelCommand):
2820         (DecreaseSelectionListLevelCommand):
2821         * editing/RemoveFormatCommand.h:
2822         (RemoveFormatCommand):
2823         * editing/RemoveNodeCommand.h:
2824         (RemoveNodeCommand):
2825         * editing/RemoveNodePreservingChildrenCommand.h:
2826         (RemoveNodePreservingChildrenCommand):
2827         * editing/ReplaceNodeWithSpanCommand.h:
2828         (ReplaceNodeWithSpanCommand):
2829         * editing/TextInsertionBaseCommand.h:
2830         (TextInsertionBaseCommand):
2831         * editing/UndoManager.h:
2832         (UndoManager):
2833         * editing/UnlinkCommand.h:
2834         (UnlinkCommand):
2835         * editing/WrapContentsInDummySpanCommand.h:
2836         (WrapContentsInDummySpanCommand):
2837
2838 2012-07-23  Dominic Mazzoni  <dmazzoni@google.com>
2839
2840         AX: Need AccessibilityObjects for nodes without renderers in canvas subtree
2841         https://bugs.webkit.org/show_bug.cgi?id=87899
2842
2843         Reviewed by Chris Fleizach.
2844
2845         Refactors AccessibilityRenderObject so that it inherits from a new class,
2846         AccessibilityNodeObject, that can be constructed from a Node without a
2847         renderer. Modifies AXObjectCache so that it automatically creates an
2848         AccessibilityNodeObject for elements in a canvas subtree but not otherwise.
2849         A new layout test verifies that this correctly exposes an accessibility
2850         tree with appropriate roles for elements in a canvas subtree.
2851
2852         This patch does not try to complete the implementation of
2853         AccessibilityNodeObject. Most AX methods are still unimplemented and need
2854         to be migrated from AccessibilityRenderObject to AccessibilityNodeObject
2855         in a future patch.
2856
2857         This patch also doesn't change anything outside of WebCore/accessibility, so
2858         the rest of WebCore only calls AXObjectCache when there are changes to a
2859         RenderObject, not to a Node. Accessible notifications on nodes without
2860         renderers need to be implemented in a future patch.
2861
2862         Test: accessibility/canvas-accessibilitynodeobject.html
2863
2864         * CMakeLists.txt:
2865         * GNUmakefile.list.am:
2866         * Target.pri:
2867         * WebCore.gypi:
2868         * accessibility/AXObjectCache.cpp:
2869         (WebCore::AXObjectCache::get):
2870         (WebCore):
2871         (WebCore::createFromNode):
2872         (WebCore::AXObjectCache::getOrCreate):
2873         (WebCore::AXObjectCache::remove):
2874         * accessibility/AXObjectCache.h:
2875         (AXObjectCache):
2876         (WebCore::AXObjectCache::get):
2877         (WebCore::AXObjectCache::getOrCreate):
2878         (WebCore::AXObjectCache::remove):
2879         * accessibility/AccessibilityARIAGrid.cpp:
2880         (WebCore::AccessibilityARIAGrid::AccessibilityARIAGrid):
2881         (WebCore):
2882         (WebCore::AccessibilityARIAGrid::~AccessibilityARIAGrid):
2883         (WebCore::AccessibilityARIAGrid::init):
2884         (WebCore::AccessibilityARIAGrid::create):
2885         * accessibility/AccessibilityARIAGrid.h:
2886         (AccessibilityARIAGrid):
2887         * accessibility/AccessibilityARIAGridCell.cpp:
2888         (WebCore::AccessibilityARIAGridCell::create):
2889         * accessibility/AccessibilityARIAGridRow.cpp:
2890         (WebCore::AccessibilityARIAGridRow::create):
2891         * accessibility/AccessibilityAllInOne.cpp:
2892         * accessibility/AccessibilityList.cpp:
2893         (WebCore::AccessibilityList::create):
2894         * accessibility/AccessibilityListBox.cpp:
2895         (WebCore::AccessibilityListBox::create):
2896         * accessibility/AccessibilityMediaControls.cpp:
2897         (WebCore::AccessibilityMediaControl::create):
2898         (WebCore::AccessibilityMediaControlsContainer::create):
2899         (WebCore::AccessibilityMediaTimeline::create):
2900         (WebCore::AccessibilityMediaTimeDisplay::create):
2901         * accessibility/AccessibilityMenuList.cpp:
2902         (WebCore::AccessibilityMenuList::create):
2903         (WebCore):
2904         * accessibility/AccessibilityMenuList.h:
2905         (AccessibilityMenuList):
2906         * accessibility/AccessibilityNodeObject.cpp: Added.
2907         (WebCore):
2908         (WebCore::AccessibilityNodeObject::AccessibilityNodeObject):
2909         (WebCore::AccessibilityNodeObject::~AccessibilityNodeObject):
2910         (WebCore::AccessibilityNodeObject::init):
2911         (WebCore::AccessibilityNodeObject::create):
2912         (WebCore::AccessibilityNodeObject::detach):
2913         (WebCore::AccessibilityNodeObject::childrenChanged):
2914         (WebCore::AccessibilityNodeObject::updateAccessibilityRole):
2915         (WebCore::AccessibilityNodeObject::firstChild):
2916         (WebCore::AccessibilityNodeObject::lastChild):
2917         (WebCore::AccessibilityNodeObject::previousSibling):
2918         (WebCore::AccessibilityNodeObject::nextSibling):
2919         (WebCore::AccessibilityNodeObject::parentObjectIfExists):
2920         (WebCore::AccessibilityNodeObject::parentObject):
2921         (WebCore::AccessibilityNodeObject::elementRect):
2922         (WebCore::AccessibilityNodeObject::setNode):
2923         (WebCore::AccessibilityNodeObject::document):
2924         (WebCore::AccessibilityNodeObject::determineAccessibilityRole):
2925         (WebCore::AccessibilityNodeObject::addChildren):
2926         (WebCore::AccessibilityNodeObject::accessibilityIsIgnored):
2927         (WebCore::AccessibilityNodeObject::canSetFocusAttribute):
2928         (WebCore::AccessibilityNodeObject::determineAriaRoleAttribute):
2929         (WebCore::AccessibilityNodeObject::ariaRoleAttribute):
2930         (WebCore::AccessibilityNodeObject::remapAriaRoleDueToParent):
2931         * accessibility/AccessibilityNodeObject.h: Added.
2932         (WebCore):
2933         (AccessibilityNodeObject):
2934         (WebCore::AccessibilityNodeObject::isAccessibilityNodeObject):
2935         (WebCore::AccessibilityNodeObject::node):
2936         (WebCore::AccessibilityNodeObject::isDetached):
2937         (WebCore::toAccessibilityNodeObject):
2938         * accessibility/AccessibilityObject.h:
2939         (WebCore::AccessibilityObject::isAccessibilityNodeObject):
2940         * accessibility/AccessibilityProgressIndicator.cpp:
2941         (WebCore::AccessibilityProgressIndicator::create):
2942         * accessibility/AccessibilityRenderObject.cpp:
2943         (WebCore::AccessibilityRenderObject::AccessibilityRenderObject):
2944         (WebCore::AccessibilityRenderObject::init):
2945         (WebCore):
2946         (WebCore::AccessibilityRenderObject::create):
2947         (WebCore::AccessibilityRenderObject::detach):
2948         (WebCore::AccessibilityRenderObject::setRenderer):
2949         (WebCore::AccessibilityRenderObject::canHaveChildren):
2950         (WebCore::AccessibilityRenderObject::addCanvasChildren):
2951         (WebCore::AccessibilityRenderObject::addChildren):
2952         * accessibility/AccessibilityRenderObject.h:
2953         (AccessibilityRenderObject):
2954         * accessibility/AccessibilitySlider.cpp:
2955         (WebCore::AccessibilitySlider::create):
2956         * accessibility/AccessibilityTable.cpp:
2957         (WebCore::AccessibilityTable::AccessibilityTable):
2958         (WebCore):
2959         (WebCore::AccessibilityTable::~AccessibilityTable):
2960         (WebCore::AccessibilityTable::init):
2961         (WebCore::AccessibilityTable::create):
2962         * accessibility/AccessibilityTable.h:
2963         (AccessibilityTable):
2964         * accessibility/AccessibilityTableCell.cpp:
2965         (WebCore::AccessibilityTableCell::create):
2966         * accessibility/AccessibilityTableRow.cpp:
2967         (WebCore::AccessibilityTableRow::create):
2968
2969 2012-07-23  David Barr  <davidbarr@chromium.org>
2970
2971         Fix ENABLE_CSS_IMAGE_ORIENTATION compile flag
2972         https://bugs.webkit.org/show_bug.cgi?id=91961
2973
2974         Reviewed by Tony Chang.
2975
2976         https://bugs.webkit.org/show_bug.cgi?id=89055
2977         A rule was missed implementing the ENABLE_CSS_IMAGE_ORIENTATION compile flag.
2978
2979         https://bugs.webkit.org/show_bug.cgi?id=90046
2980         Half of the missing rule was inadvertently added together with ENABLE_CSS_STICKY_POSITION.
2981
2982         No new tests; build configuration fix.
2983
2984         * Configurations/FeatureDefines.xcconfig:
2985
2986 2012-07-23  Douglas Stockwell  <dstockwell@chromium.org>
2987
2988         Null-pointer crash when a derived color like -webkit-activelink is set in a gradient stop
2989         https://bugs.webkit.org/show_bug.cgi?id=89148
2990
2991         Reviewed by Simon Fraser.
2992
2993         CSSGradientValue attempted to resolve colors during paint, this crashed
2994         when a derived color like -webkit-activelink was encountered because the
2995         corresponding element was no longer available in the StyleResolver.
2996         Instead, by adding a field to CSSGradientColorStop we can resolve and
2997         then cache the resolved colors at the correct time. To avoid sharing
2998         cached derived colors between elements we clone the gradient values when
2999         needed.
3000
3001         Test: fast/css/crash-on-gradient-with-derived-color.html
3002
3003         * css/CSSGradientValue.cpp:
3004         (WebCore::CSSGradientValue::gradientWithStylesResolved): Added. Resolve and cache
3005         resolved colors, clone if colors are derived from the element.
3006         (WebCore):
3007         (WebCore::CSSGradientValue::addStops):
3008         (WebCore::CSSGradientValue::isCacheable): Defer to new logic in
3009         StyleResolver.
3010         * css/CSSGradientValue.h:
3011         (CSSGradientColorStop): Added cache of resolved color.
3012         (CSSGradientValue):
3013         (CSSLinearGradientValue):
3014         (CSSRadialGradientValue):
3015         * css/CSSImageGeneratorValue.h:
3016         (WebCore):
3017         * css/CSSValue.h:
3018         (WebCore::CSSValue::isGradientValue):
3019         * css/StyleResolver.cpp:
3020         (WebCore::StyleResolver::collectMatchingRulesForList):
3021         * css/StyleResolver.h:
3022         (StyleResolver):
3023         * rendering/style/StyleGeneratedImage.cpp:
3024         (WebCore::StyleGeneratedImage::image): Revert change from r96449. This
3025         is no longer necessary as the gradient colors are now resolved at a
3026         time when the style is set on StyleResolver.
3027
3028 2012-07-23  Patrick Gansterer  <paroga@webkit.org>
3029
3030         Move GregorianDateTime from JSC to WTF namespace
3031         https://bugs.webkit.org/show_bug.cgi?id=91948
3032
3033         Reviewed by Geoffrey Garen.
3034
3035         Moving GregorianDateTime into the WTF namespace allows us to us to
3036         use it in WebCore too. The new class has the same behaviour as the
3037         old struct. Only the unused timeZone member has been removed.
3038
3039         * bridge/qt/qt_runtime.cpp:
3040         (JSC::Bindings::convertValueToQVariant):
3041         (JSC::Bindings::convertQVariantToValue):
3042         * bridge/qt/qt_runtime_qt4.cpp:
3043         (JSC::Bindings::convertValueToQVariant):
3044         (JSC::Bindings::convertQVariantToValue):
3045
3046 2012-07-23  Mark Pilgrim  <pilgrim@chromium.org>
3047
3048         [Chromium] Move layoutTestMode to WebCore
3049         https://bugs.webkit.org/show_bug.cgi?id=92010
3050
3051         Reviewed by Adam Barth.
3052
3053         Part of a refactoring series. See tracking bug 82948.
3054
3055         * CMakeLists.txt:
3056         * GNUmakefile.list.am:
3057         * Target.pri:
3058         * WebCore.gypi:
3059         * WebCore.vcproj/WebCore.vcproj:
3060         * WebCore.xcodeproj/project.pbxproj:
3061         * bindings/generic/RuntimeEnabledFeatures.h:
3062         (RuntimeEnabledFeatures):
3063         * platform/LayoutTestSupport.cpp: Added.
3064         (WebCore):
3065         (WebCore::isRunningLayoutTest):
3066         (WebCore::setIsRunningLayoutTest):
3067         * platform/LayoutTestSupport.h: Added.
3068         (WebCore):
3069         * platform/chromium/PlatformSupport.h:
3070         (PlatformSupport):
3071         * platform/chromium/ScrollbarThemeChromiumAndroid.cpp:
3072         (WebCore::ScrollbarThemeChromiumAndroid::scrollbarThickness):
3073         (WebCore::ScrollbarThemeChromiumAndroid::paintScrollbarBackground):
3074         * platform/chromium/ScrollbarThemeChromiumWin.cpp:
3075         (WebCore::ScrollbarThemeChromiumWin::scrollbarThickness):
3076         (WebCore::ScrollbarThemeChromiumWin::buttonSize):
3077         * platform/graphics/chromium/FontCacheChromiumWin.cpp:
3078         (WebCore):
3079         (WebCore::LookupAltName):
3080         (WebCore::fontContainsCharacter):
3081         (WebCore::FillLogFont):
3082         (WebCore::FontCache::getFontDataForCharacters):
3083         (WebCore::FontCache::createFontPlatformData):
3084         * platform/graphics/skia/FontCustomPlatformData.cpp:
3085         (WebCore::FontCustomPlatformData::fontPlatformData):
3086         * platform/graphics/skia/FontSkia.cpp:
3087         (WebCore::Font::drawGlyphs):
3088         * rendering/RenderThemeChromiumAndroid.cpp:
3089         (WebCore::RenderThemeChromiumAndroid::systemColor):
3090         (WebCore::RenderThemeChromiumAndroid::adjustInnerSpinButtonStyle):
3091         * rendering/RenderThemeChromiumMac.mm:
3092         (WebCore::RenderThemeChromiumMac::usesTestModeFocusRingColor):
3093         * rendering/RenderThemeChromiumSkia.cpp:
3094         (WebCore::RenderThemeChromiumSkia::caretBlinkInterval):
3095         * rendering/RenderThemeChromiumWin.cpp:
3096         (WebCore):
3097         (WebCore::RenderThemeChromiumWin::platformActiveSelectionBackgroundColor):
3098         (WebCore::RenderThemeChromiumWin::platformInactiveSelectionBackgroundColor):
3099         (WebCore::RenderThemeChromiumWin::platformActiveSelectionForegroundColor):
3100         (WebCore::RenderThemeChromiumWin::systemColor):
3101         (WebCore::menuListButtonWidth):
3102         (WebCore::RenderThemeChromiumWin::determineClassicState):
3103
3104 2012-07-23  Dave Tu  <dtu@chromium.org>
3105
3106         [chromium] Add droppedFrameCount to renderingStats.
3107         https://bugs.webkit.org/show_bug.cgi?id=91694
3108
3109         Reviewed by Adrienne Walker.
3110
3111         * platform/graphics/chromium/cc/CCFrameRateCounter.cpp:
3112         (WebCore):
3113         (WebCore::CCFrameRateCounter::frameInterval):
3114         (WebCore::CCFrameRateCounter::frameIndex):
3115         (WebCore::CCFrameRateCounter::CCFrameRateCounter):
3116         (WebCore::CCFrameRateCounter::markBeginningOfFrame):
3117         (WebCore::CCFrameRateCounter::isBadFrame):
3118         * platform/graphics/chromium/cc/CCFrameRateCounter.h:
3119         (WebCore::CCFrameRateCounter::droppedFrameCount):
3120         (CCFrameRateCounter):
3121         * platform/graphics/chromium/cc/CCLayerTreeHost.cpp:
3122         (WebCore::CCLayerTreeHost::renderingStats):
3123         * platform/graphics/chromium/cc/CCLayerTreeHostImpl.cpp:
3124         (WebCore::CCLayerTreeHostImpl::CCLayerTreeHostImpl):
3125         (WebCore::CCLayerTreeHostImpl::drawLayers):
3126         * platform/graphics/chromium/cc/CCLayerTreeHostImpl.h:
3127         (CCLayerTreeHostImpl):
3128         * platform/graphics/chromium/cc/CCRenderingStats.h:
3129         (CCRenderingStats):
3130         (WebCore::CCRenderingStats::CCRenderingStats):
3131         * platform/graphics/chromium/cc/CCSingleThreadProxy.cpp:
3132         (WebCore::CCSingleThreadProxy::implSideRenderingStats):
3133         * platform/graphics/chromium/cc/CCThreadProxy.cpp:
3134         (WebCore::CCThreadProxy::implSideRenderingStatsOnImplThread):
3135
3136 2012-07-23  Dan Bernstein  <mitz@apple.com>
3137
3138         Fixed the build after r123418.
3139
3140         * WebCore.exp.in: Added DocumentMarkerController::addTextMatchMarker.
3141
3142 2012-07-23  Mike Lawther  <mikelawther@chromium.org>
3143
3144         CSS3 calc: optimise blending expression
3145         https://bugs.webkit.org/show_bug.cgi?id=90037
3146
3147         Reviewed by Tony Chang.
3148
3149         Introduce a new calc expression node - a BlendLength node. Given two Lengths and a progress
3150         this blends them together when evaluated.
3151
3152         An alternative and more general approach of being able to blend two CalcExpressionNodes was
3153         tested, but involves more memory allocations, was measurably slower, and nothing currently
3154         would make use of the generality.
3155
3156         No functional change. Covered by existing tests.
3157
3158         * platform/CalculationValue.h:
3159         (CalcExpressionBlendLength):
3160         (WebCore::CalcExpressionBlendLength::CalcExpressionBlendLength):
3161         (WebCore::CalcExpressionBlendLength::operator==):
3162         (WebCore::CalcExpressionBlendLength::evaluate):
3163         (WebCore):
3164         * platform/Length.cpp:
3165         (WebCore::Length::blendCalculation):
3166
3167 2012-07-23  Nico Weber  <thakis@chromium.org>
3168
3169         [chromium] Show search tickmarks on css-styled scrollbars
3170         https://bugs.webkit.org/show_bug.cgi?id=91949
3171
3172         Reviewed by Adrienne Walker.
3173
3174         This is done by letting RenderScrollbarTheme::paintTickmarks()
3175         delegate to the native ScrollbarTheme. To make this possible,
3176         move paintTickmarks() from ScrollbarThemeComposite to the
3177         superclass ScrollbarTheme.
3178
3179         For testing, add internals.addTextMatchMarker() and add a pixel test.
3180
3181         * platform/ScrollbarTheme.h:
3182         (WebCore::ScrollbarTheme::paintTickmarks):
3183         Moved paintTickmarks() from ScrollbarThemeComposite to here.
3184         * platform/ScrollbarThemeComposite.h:
3185         Remove paintTickmarks().
3186         * platform/chromium/ScrollbarThemeChromium.cpp:
3187         (WebCore::ScrollbarThemeChromium::paintTickmarks):
3188         Switch to drawing tickmarks as vectors, so they can be arbitrarily wide.
3189         * platform/chromium/ScrollbarThemeChromiumMac.h:
3190         (ScrollbarThemeChromiumMac):
3191         Implement paintTickmarks(), so that css-styled scrollbars get tickmarks on mac.
3192         * platform/chromium/ScrollbarThemeChromiumMac.mm:
3193         (WebCore::ScrollbarThemeChromiumMac::paint):
3194         (WebCore::ScrollbarThemeChromiumMac::paintTickmarks):
3195         Implement paintTickmarks(), so that css-styled scrollbars get tickmarks on mac.
3196         * rendering/RenderScrollbarTheme.cpp:
3197         (WebCore::RenderScrollbarTheme::paintTickmarks):
3198         Delegate to the native ScrollbarTheme for tickmark drawing.
3199         (WebCore):
3200         * rendering/RenderScrollbarTheme.h:
3201         (RenderScrollbarTheme):
3202         Override paintTickmarks().
3203
3204 2012-07-23  Brian Anderson  <brianderson@chromium.org>
3205
3206         [chromium] Use shallow flushes that don't glFlush
3207         https://bugs.webkit.org/show_bug.cgi?id=90325
3208
3209         Reviewed by Kenneth Russell.
3210
3211         Adds plumbing for the shallowFlushCHROMIUM extension and uses
3212         shallow flushes instead of normal flushes when uploading textures.
3213         Shallow flushes allow us to initiate command buffer flushes to
3214         the GPU thread/process without the overhead of a true glFlush.
3215
3216         CCTextureUpdaterTest updated to recognize shallow flushes.
3217
3218         * platform/chromium/support/Extensions3DChromium.cpp:
3219         (WebCore::Extensions3DChromium::shallowFlushCHROMIUM):
3220         (WebCore):
3221         * platform/graphics/chromium/Extensions3DChromium.h:
3222         (Extensions3DChromium):
3223         * platform/graphics/chromium/cc/CCGraphicsContext.h:
3224         * platform/graphics/chromium/cc/CCResourceProvider.cpp:
3225         (WebCore::CCResourceProvider::shallowFlushIfSupported):
3226         (WebCore):
3227         (WebCore::CCResourceProvider::CCResourceProvider):
3228         (WebCore::CCResourceProvider::initialize):
3229         * platform/graphics/chromium/cc/CCResourceProvider.h:
3230         (CCResourceProvider):
3231         * platform/graphics/chromium/cc/CCTextureUpdater.cpp:
3232         (WebCore::CCTextureUpdater::update):
3233
3234 2012-07-23  Hanyee Kim  <choco@company100.net>
3235
3236         When using TextureMapper, WebKit does not retain final value of opacity in animations, unless it is specified on the last key frame.
3237         https://bugs.webkit.org/show_bug.cgi?id=91322
3238
3239         Reviewed by Noam Rosenthal.
3240
3241         When -webkit-fill-mode is forwards, GraphicsLayerAnimation::apply does not apply
3242         the property values defined in the last executing keyframe after the final iteration of animation.
3243
3244         This patch does not return right after the end of animation, but it keeps applying
3245         the property values defined in the last executing keyframe.
3246         Add normalizedAnimationValueForFillsForwards function that returns a normalized value of the
3247         last executing keyframe.
3248
3249         Link to CSS Animations spec: http://www.w3.org/TR/css3-animations/#animation-fill-mode-property
3250
3251         Test: animations/fill-mode-forwards2.html
3252
3253         * platform/graphics/GraphicsLayerAnimation.cpp:
3254         (WebCore::normalizedAnimationValueForFillsForwards):
3255         (WebCore):
3256         (WebCore::GraphicsLayerAnimation::apply):
3257
3258 2012-07-23  Sheriff Bot  <webkit.review.bot@gmail.com>
3259
3260         Unreviewed, rolling out r123184, r123195, and r123197.
3261         http://trac.webkit.org/changeset/123184
3262         http://trac.webkit.org/changeset/123195
3263         http://trac.webkit.org/changeset/123197
3264         https://bugs.webkit.org/show_bug.cgi?id=92049
3265
3266         pagecycler regression (Requested by morrita on #webkit).
3267
3268         * WebCore.exp.in:
3269         * dom/Document.cpp:
3270         (WebCore::Document::Document):
3271         (WebCore::Document::~Document):
3272         (WebCore::Document::suggestedMIMEType):
3273         * dom/Document.h:
3274         (WebCore):
3275         (WebCore::Node::isDocumentNode):
3276         (WebCore::Node::Node):
3277         * dom/Node.cpp:
3278         (WebCore::Node::~Node):
3279         (WebCore::Node::setDocument):
3280         (WebCore):
3281         (WebCore::Node::setTreeScope):
3282         (WebCore::Node::treeScope):
3283         (WebCore::Node::isInShadowTree):
3284         (WebCore::Node::reportMemoryUsage):
3285         * dom/Node.h:
3286         (Node):
3287         (WebCore::Node::document):
3288         (WebCore::Node::inDocument):
3289         * dom/NodeRareData.h:
3290         (WebCore::NodeRareData::NodeRareData):
3291         (WebCore::NodeRareData::treeScope):
3292         (WebCore::NodeRareData::setTreeScope):
3293         (NodeRareData):
3294         * dom/ShadowRoot.cpp:
3295         (WebCore::ShadowRoot::ShadowRoot):
3296         * dom/TreeScope.cpp:
3297         (WebCore::TreeScope::TreeScope):
3298         (WebCore::TreeScope::setParentTreeScope):
3299         * dom/TreeScope.h:
3300         (WebCore):
3301         (WebCore::TreeScope::idTargetObserverRegistry):
3302         (TreeScope):
3303         * dom/TreeScopeAdopter.cpp:
3304         (WebCore::TreeScopeAdopter::moveTreeToNewScope):
3305         (WebCore::TreeScopeAdopter::moveNodeToNewDocument):
3306         * editing/MoveSelectionCommand.cpp:
3307         * editing/RemoveNodeCommand.cpp:
3308         * editing/RemoveNodePreservingChildrenCommand.cpp:
3309         * inspector/PageConsoleAgent.cpp:
3310
3311 2012-07-23  Roger Fong  <roger_fong@apple.com>
3312
3313         On Windows, if select element is off screen horizontally,
3314         menu is either inappropriately resized or positioned offscreen.
3315         https://bugs.webkit.org/show_bug.cgi?id=91913
3316         <rdar://problem/7611229>
3317
3318         Reviewed by Tim Horton.
3319
3320         If the select element is positioned off the edge of the screen to the left,
3321         the menu is resized. It should not be resized, just shifted to remain on the screen.
3322         If the select element is positioned off the edge of the screen to the right,
3323         the menu goes off screen instead of being shifted over to appear on screen.
3324         This problem only occurs on Windows.
3325
3326         Test: ManualTests/win/select-menu-off-screen.html
3327
3328         * platform/win/PopupMenuWin.cpp:
3329         (WebCore::PopupMenuWin::calculatePositionAndSize):
3330         Modified final horizontal position calculation code to position
3331         popup menu on screen if it would otherwise go off.
3332
3333 2012-07-23  David Dorwin  <ddorwin@chromium.org>
3334
3335         Fixed crash in webkitAddKey() when key parameter is null.
3336         https://bugs.webkit.org/show_bug.cgi?id=85444
3337
3338         Reviewed by Kentaro Hara.
3339
3340         Reference: Step 1 of http://dvcs.w3.org/hg/html-media/raw-file/5f76a0b43836/encrypted-media/encrypted-media.html#dom-addkey
3341
3342         Tests: media/encrypted-media/encrypted-media-syntax.html
3343
3344         * html/HTMLMediaElement.cpp:
3345         (WebCore::HTMLMediaElement::webkitAddKey):
3346
3347 2012-07-23  Hayato Ito  <hayato@chromium.org>
3348
3349         A FocusScope for a distributed node should not be its TreeScope.
3350         https://bugs.webkit.org/show_bug.cgi?id=91829
3351
3352         Reviewed by Dimitri Glazkov.
3353
3354         Current implementation of FocusScope::focusScopeOf(Node*) returns
3355         the given node's treeScope().  That does not apply if the node is
3356         a distributed node.  We should calculate a FocusScope for a
3357         distributed node by traversing ancestor nodes in Composed Shadow
3358         Tree.
3359
3360         Test: fast/dom/shadow/focus-navigation-with-distributed-nodes.html
3361
3362         * page/FocusController.cpp:
3363         (WebCore::FocusScope::focusScopeOf):
3364
3365 2012-07-23  Douglas Stockwell  <dstockwell@google.com>
3366
3367         Null-pointer crash when parsing border-image
3368         https://bugs.webkit.org/show_bug.cgi?id=91963
3369
3370         Reviewed by Darin Adler.
3371
3372         Test: fast/css/border-image-fill-crash.html
3373
3374         * css/CSSParser.cpp:
3375         (WebCore::CSSParser::parseBorderImageRepeat): Handle the case where
3376         parseBorderImageRepeat is called when the value list is empty.
3377
3378 2012-07-23  Bem Jones-Bey  <bjonesbe@adobe.com>
3379
3380         [CSS Exclusions] SVG shape errors should invalidate exclusion shapes
3381         https://bugs.webkit.org/show_bug.cgi?id=91761
3382
3383         Reviewed by Andreas Kling.
3384
3385         CSS exclusion shapes that are specified with negative radiuses or
3386         height/width are now considered invalid and ignored.
3387
3388         Tests added to
3389         LayoutTests/fast/exclusions/parsing-wrap-shape-lengths.html
3390
3391         * css/CSSParser.cpp:
3392         (WebCore::CSSParser::parseExclusionShapeRectangle):
3393         (WebCore::CSSParser::parseExclusionShapeCircle):
3394         (WebCore::CSSParser::parseExclusionShapeEllipse):
3395
3396 2012-07-23  Tien-Ren Chen  <trchen@chromium.org>
3397
3398         [chromium] Implement scrollbar theme for Android
3399         https://bugs.webkit.org/show_bug.cgi?id=91674
3400
3401         Reviewed by Adam Barth.
3402
3403         Previously the scrollbar for Android was implemented as an extra
3404         drawing pass in the compositor. Now we switch to use the standard
3405         ScrollbarTheme mechanism.
3406
3407         No new tests. We use mock scrollbars during layout test to share
3408         pixel results with Linux.
3409
3410         * platform/chromium/ScrollbarThemeChromiumAndroid.cpp:
3411         (WebCore::ScrollbarThemeChromiumAndroid::scrollbarThickness):
3412         (WebCore::ScrollbarThemeChromiumAndroid::thumbPosition):
3413         (WebCore::ScrollbarThemeChromiumAndroid::thumbLength):
3414         (WebCore::ScrollbarThemeChromiumAndroid::backButtonRect):
3415         (WebCore::ScrollbarThemeChromiumAndroid::forwardButtonRect):
3416         (WebCore::ScrollbarThemeChromiumAndroid::trackRect):
3417         (WebCore):
3418         (WebCore::ScrollbarThemeChromiumAndroid::paintThumb):
3419         (WebCore::ScrollbarThemeChromiumAndroid::paintScrollbarBackground):
3420         * platform/chromium/ScrollbarThemeChromiumAndroid.h:
3421         (WebCore::ScrollbarThemeChromiumAndroid::usesOverlayScrollbars):
3422         (ScrollbarThemeChromiumAndroid):
3423         (WebCore::ScrollbarThemeChromiumAndroid::hasButtons):
3424         (WebCore::ScrollbarThemeChromiumAndroid::hasThumb):
3425
3426 2012-07-23  Kwang Yul Seo  <skyul@company100.net>
3427
3428         Use the original token to create an element in "reconstruct the active formatting elements" and "call the adoption agency"
3429         https://bugs.webkit.org/show_bug.cgi?id=91703
3430
3431         Reviewed by Adam Barth.
3432
3433         The current WebKit HTML5 parser implementation does not hold the original token
3434         in the stack of open elements and the active formatting elements. This is
3435         problematic because the original token is used to create an element in
3436         "reconstruct the active formatting elements" and "call the adoption agency".
3437
3438         As a workaround, WebKit uses the saved element instead of the original token
3439         to create an element. But this causes us to fail examples like this:
3440         <b id="1"><p><script>document.getElementById("1").id = "2"</script></p>TEXT</b>
3441         reconstructTheActiveFormattingElements calls this method to open a second <b>
3442         tag to wrap TEXT, it will have id "2", even though the HTML5 spec implies it
3443         should be "1".
3444
3445         Created a ref-counted container class, HTMLStackItem to hold the original token
3446         and the namespace URI as well as the element. Changed HTMLElementStack and
3447         HTMLFormattingElementList to use HTMLStackItem.
3448         Changed HTMLConstructionSite::reconstructTheActiveFormattingElements and
3449         HTMLTreeBuilder::callTheAdoptionAgency to create an element from the saved token
3450         instead of the saved element.
3451
3452         Updated test expectation for html5lib/runner-expected.txt
3453         because now resources/scripted/adoption01.dat passes.
3454
3455         * html/parser/HTMLConstructionSite.cpp:
3456         (WebCore::HTMLConstructionSite::insertHTMLHtmlStartTagBeforeHTML):
3457         (WebCore::HTMLConstructionSite::insertHTMLHeadElement):
3458         (WebCore::HTMLConstructionSite::insertHTMLBodyElement):
3459         (WebCore::HTMLConstructionSite::insertHTMLFormElement):
3460         (WebCore::HTMLConstructionSite::insertHTMLElement):
3461         (WebCore::HTMLConstructionSite::insertFormattingElement):
3462         (WebCore::HTMLConstructionSite::insertScriptElement):
3463         (WebCore::HTMLConstructionSite::insertForeignElement):
3464         (WebCore::HTMLConstructionSite::createElementFromSavedToken):
3465         (WebCore::HTMLConstructionSite::reconstructTheActiveFormattingElements):
3466         * html/parser/HTMLConstructionSite.h:
3467         (HTMLConstructionSite):
3468         (WebCore::HTMLConstructionSite::currentElementRecord):
3469         * html/parser/HTMLElementStack.cpp:
3470         (WebCore::HTMLElementStack::ElementRecord::ElementRecord):
3471         (WebCore::HTMLElementStack::ElementRecord::replaceElement):
3472         (WebCore::HTMLElementStack::pushRootNode):
3473         (WebCore::HTMLElementStack::pushHTMLHtmlElement):
3474         (WebCore::HTMLElementStack::pushRootNodeCommon):
3475         (WebCore::HTMLElementStack::pushHTMLHeadElement):
3476         (WebCore::HTMLElementStack::pushHTMLBodyElement):
3477         (WebCore::HTMLElementStack::push):
3478         (WebCore::HTMLElementStack::insertAbove):
3479         (WebCore::HTMLElementStack::pushCommon):
3480         * html/parser/HTMLElementStack.h:
3481         (WebCore::HTMLElementStack::ElementRecord::element):
3482         (WebCore::HTMLElementStack::ElementRecord::node):
3483         (WebCore::HTMLElementStack::ElementRecord::stackItem):
3484         (ElementRecord):
3485         (HTMLElementStack):
3486         * html/parser/HTMLFormattingElementList.cpp:
3487         (WebCore::HTMLFormattingElementList::swapTo):
3488         (WebCore::HTMLFormattingElementList::append):
3489         * html/parser/HTMLFormattingElementList.h:
3490         (WebCore::HTMLFormattingElementList::Entry::Entry):
3491         (WebCore::HTMLFormattingElementList::Entry::isMarker):
3492         (WebCore::HTMLFormattingElementList::Entry::stackItem):
3493         (WebCore::HTMLFormattingElementList::Entry::element):
3494         (WebCore::HTMLFormattingElementList::Entry::replaceElement):
3495         (WebCore::HTMLFormattingElementList::Entry::operator==):
3496         (WebCore::HTMLFormattingElementList::Entry::operator!=):
3497         (Entry):
3498         (HTMLFormattingElementList):
3499         * html/parser/HTMLStackItem.h: Added.
3500         (WebCore):
3501         (HTMLStackItem):
3502         (WebCore::HTMLStackItem::create):
3503         (WebCore::HTMLStackItem::element):
3504         (WebCore::HTMLStackItem::node):
3505         (WebCore::HTMLStackItem::token):
3506         (WebCore::HTMLStackItem::namespaceURI):
3507         (WebCore::HTMLStackItem::HTMLStackItem):
3508         * html/parser/HTMLTreeBuilder.cpp:
3509         (WebCore::HTMLTreeBuilder::HTMLTreeBuilder):
3510         (WebCore::HTMLTreeBuilder::processStartTag):
3511         (WebCore::HTMLTreeBuilder::callTheAdoptionAgency):
3512
3513 2012-07-23  Andreas Kling  <kling@webkit.org>
3514
3515         Report the extra memory used by immutable StylePropertySet objects.
3516         <http://webkit.org/b/92032>
3517
3518         Reviewed by Anders Carlsson.
3519
3520         Add an optional parameter to the MemoryClassInfo constructor for passing in additional base object
3521         size on top of sizeof(T).
3522
3523         Use this in StylePropertySet::reportMemoryUsage() to properly account for the CSSProperty array
3524         tacked onto the end of the object when m_isMutable == false.
3525
3526         * css/StylePropertySet.h:
3527         (WebCore::StylePropertySet::reportMemoryUsage):
3528         * dom/MemoryInstrumentation.h:
3529         (WebCore::MemoryObjectInfo::reportObjectInfo):
3530         (WebCore::MemoryClassInfo::MemoryClassInfo):
3531
3532 2012-07-23  Gregg Tavares  <gman@google.com>
3533
3534         Fix WebGL texSubImage2D for cube maps
3535         https://bugs.webkit.org/show_bug.cgi?id=91927
3536
3537         Reviewed by Kenneth Russell.
3538
3539         Fixes texSubImage2D so any size rectangle can be passed in
3540         for cube maps.
3541
3542         Test: fast/canvas/webgl/tex-sub-image-cube-maps.html
3543
3544         * html/canvas/WebGLRenderingContext.cpp:
3545         (WebCore):
3546         (WebCore::WebGLRenderingContext::copyTexImage2D):
3547         (WebCore::WebGLRenderingContext::texImage2DBase):
3548         (WebCore::WebGLRenderingContext::texSubImage2DBase):
3549         (WebCore::WebGLRenderingContext::validateTexFuncParameters):
3550         * html/canvas/WebGLRenderingContext.h:
3551         (WebGLRenderingContext):
3552
3553 2012-07-23  Sheriff Bot  <webkit.review.bot@gmail.com>
3554
3555         Unreviewed, rolling out r123387.
3556         http://trac.webkit.org/changeset/123387
3557         https://bugs.webkit.org/show_bug.cgi?id=92036
3558
3559         Broke chromium win build (Requested by tony^work on #webkit).
3560
3561         * WebCore.gyp/WebCore.gyp:
3562         * WebCore.gyp/scripts/rule_bison.py:
3563         * bindings/scripts/preprocessor.pm:
3564         (applyPreprocessor):
3565         * make-hash-tools.pl:
3566
3567 2012-07-23  Kwang Yul Seo  <skyul@company100.net>
3568
3569         Ref-count AtomicHTMLToken
3570         https://bugs.webkit.org/show_bug.cgi?id=91981
3571
3572         Reviewed by Adam Barth.
3573
3574         Ref-count AtomicHTMLToken to avoid copying AtomicHTMLToken in Bug 91703.
3575
3576         No new tests - no functional changes.
3577
3578         * html/parser/HTMLConstructionSite.cpp:
3579         (WebCore::HTMLConstructionSite::insertHTMLHtmlStartTagBeforeHTML):
3580         (WebCore::HTMLConstructionSite::mergeAttributesFromTokenIntoElement):
3581         (WebCore::HTMLConstructionSite::insertHTMLHtmlStartTagInBody):
3582         (WebCore::HTMLConstructionSite::insertHTMLBodyStartTagInBody):
3583         (WebCore::HTMLConstructionSite::insertDoctype):
3584         (WebCore::HTMLConstructionSite::insertComment):
3585         (WebCore::HTMLConstructionSite::insertCommentOnDocument):
3586         (WebCore::HTMLConstructionSite::insertCommentOnHTMLHtmlElement):
3587         (WebCore::HTMLConstructionSite::insertHTMLHeadElement):
3588         (WebCore::HTMLConstructionSite::insertHTMLBodyElement):
3589         (WebCore::HTMLConstructionSite::insertHTMLFormElement):
3590         (WebCore::HTMLConstructionSite::insertHTMLElement):
3591         (WebCore::HTMLConstructionSite::insertSelfClosingHTMLElement):
3592         (WebCore::HTMLConstructionSite::insertFormattingElement):
3593         (WebCore::HTMLConstructionSite::insertScriptElement):
3594         (WebCore::HTMLConstructionSite::insertForeignElement):
3595         (WebCore::HTMLConstructionSite::createElement):
3596         (WebCore::HTMLConstructionSite::createHTMLElement):
3597         (WebCore::HTMLConstructionSite::createHTMLElementFromSavedElement):
3598         * html/parser/HTMLConstructionSite.h:
3599         (HTMLConstructionSite):
3600         * html/parser/HTMLToken.h:
3601         (WebCore::AtomicHTMLToken::create):
3602         (AtomicHTMLToken):
3603         (WebCore::AtomicHTMLToken::AtomicHTMLToken):
3604         * html/parser/HTMLTreeBuilder.cpp:
3605         (WebCore::HTMLTreeBuilder::ExternalCharacterTokenBuffer::ExternalCharacterTokenBuffer):
3606         (WebCore::HTMLTreeBuilder::constructTreeFromToken):
3607         (WebCore::HTMLTreeBuilder::constructTreeFromAtomicToken):
3608         (WebCore::HTMLTreeBuilder::processToken):
3609         (WebCore::HTMLTreeBuilder::processDoctypeToken):
3610         (WebCore::HTMLTreeBuilder::processFakeStartTag):
3611         (WebCore::HTMLTreeBuilder::processFakeEndTag):
3612         (WebCore::HTMLTreeBuilder::processFakePEndTagIfPInButtonScope):
3613         (WebCore::HTMLTreeBuilder::attributesForIsindexInput):
3614         (WebCore::HTMLTreeBuilder::processIsindexStartTagForInBody):
3615         (WebCore):
3616         (WebCore::HTMLTreeBuilder::processCloseWhenNestedTag):
3617         (WebCore::HTMLTreeBuilder::processStartTagForInBody):
3618         (WebCore::HTMLTreeBuilder::processStartTagForInTable):
3619         (WebCore::HTMLTreeBuilder::processStartTag):
3620         (WebCore::HTMLTreeBuilder::processHtmlStartTagForInBody):
3621         (WebCore::HTMLTreeBuilder::processBodyEndTagForInBody):
3622         (WebCore::HTMLTreeBuilder::processAnyOtherEndTagForInBody):
3623         (WebCore::HTMLTreeBuilder::callTheAdoptionAgency):
3624         (WebCore::HTMLTreeBuilder::processEndTagForInTableBody):
3625         (WebCore::HTMLTreeBuilder::processEndTagForInRow):
3626         (WebCore::HTMLTreeBuilder::processEndTagForInCell):
3627         (WebCore::HTMLTreeBuilder::processEndTagForInBody):
3628         (WebCore::HTMLTreeBuilder::processEndTagForInTable):
3629         (WebCore::HTMLTreeBuilder::processEndTag):
3630         (WebCore::HTMLTreeBuilder::processComment):
3631         (WebCore::HTMLTreeBuilder::processCharacter):
3632         (WebCore::HTMLTreeBuilder::processEndOfFile):
3633         (WebCore::HTMLTreeBuilder::defaultForBeforeHTML):
3634         (WebCore::HTMLTreeBuilder::defaultForBeforeHead):
3635         (WebCore::HTMLTreeBuilder::defaultForInHead):
3636         (WebCore::HTMLTreeBuilder::defaultForInHeadNoscript):
3637         (WebCore::HTMLTreeBuilder::defaultForAfterHead):
3638         (WebCore::HTMLTreeBuilder::processStartTagForInHead):
3639         (WebCore::HTMLTreeBuilder::processGenericRCDATAStartTag):
3640         (WebCore::HTMLTreeBuilder::processGenericRawTextStartTag):
3641         (WebCore::HTMLTreeBuilder::processScriptStartTag):
3642         (WebCore::HTMLTreeBuilder::shouldProcessTokenInForeignContent):
3643         (WebCore::HTMLTreeBuilder::processTokenInForeignContent):
3644         (WebCore::HTMLTreeBuilder::parseError):
3645         * html/parser/HTMLTreeBuilder.h:
3646         (HTMLTreeBuilder):
3647         * html/parser/TextDocumentParser.cpp:
3648         (WebCore::TextDocumentParser::insertFakePreElement):
3649
3650 2012-07-23  Scott Graham  <scottmg@google.com>
3651
3652         Use native (non-cygwin) binaries for perl, gperf, and bison in Chromium
3653         https://bugs.webkit.org/show_bug.cgi?id=91667
3654
3655         Reviewed by Tony Chang.
3656
3657         Using native tools instead of cygwin version improves build time
3658         performance by roughly 50% (on top of previous cl-instead-of-gcc
3659         change).
3660
3661         Also, use - instead of / for cl flags because a layer of project
3662         generator converts them to \ otherwise, which causes the preprocessing
3663         to fail (very slowly because of the cygwin-loop with a sleep 1).
3664
3665         No new tests.
3666
3667         * WebCore.gyp/WebCore.gyp:
3668         * WebCore.gyp/scripts/rule_bison.py:
3669         * bindings/scripts/preprocessor.pm:
3670         (applyPreprocessor):
3671         * make-hash-tools.pl:
3672
3673 2012-06-12  Jer Noble  <jer.noble@apple.com>
3674
3675         MediaController.currentTime should be kept stable during script execution.
3676         https://bugs.webkit.org/show_bug.cgi?id=88555
3677
3678         Reviewed by Eric Carlson.
3679
3680         Test: media/media-controller-time-constant.html
3681
3682         To keep MediaController.currentTime stable, add a new m_position variable and
3683         a new m_clearPositionTimer timer.  Both must be mutable variables as they will
3684         be updated from within const functions.  Calls to currentTime() will result in
3685         stable values until the next runloop iteration.
3686
3687         * html/MediaController.cpp:
3688         (MediaController::MediaController):
3689         (MediaController::currentTime):
3690         (MediaController::setCurrentTime):
3691         (MediaController::clearPositionTimerFired):
3692         * html/MediaController.h:
3693
3694 2012-07-23  Huang Dongsung  <luxtella@company100.net>
3695
3696         Destroy CSS decoded data more eagerly once they become dead caches.
3697         https://bugs.webkit.org/show_bug.cgi?id=91733
3698
3699         Reviewed by Geoffrey Garen.
3700
3701         Internal review by Kwang Yul Seo.
3702
3703         There are three CachedResources with decoded data: CachedImage, CachedScript
3704         and CachedCSSStyleSheet. In the cases of CachedImage and CachedScript, we
3705         eagerly destroy the decoded data using Timer in CacehdResource::allClientsRemoved().
3706         We must apply the same policy here in CachedCSSStyleSheet because priority
3707         inversion can occur. For example, we can't destroy the decoded data of CachedImages
3708         when they are referenced by CachedCSSStyleSheet as background, mask or border
3709         images.
3710
3711         No new tests - no new testable functionality.
3712
3713         * loader/cache/CachedCSSStyleSheet.cpp:
3714         (WebCore::CachedCSSStyleSheet::CachedCSSStyleSheet):
3715         (WebCore::CachedCSSStyleSheet::didAddClient):
3716         (WebCore::CachedCSSStyleSheet::allClientsRemoved):
3717         (WebCore::CachedCSSStyleSheet::destroyDecodedData):
3718         (WebCore):
3719         (WebCore::CachedCSSStyleSheet::decodedDataDeletionTimerFired):
3720         * loader/cache/CachedCSSStyleSheet.h:
3721         (CachedCSSStyleSheet):
3722
3723 2012-07-23  Simon Fraser  <simon.fraser@apple.com>
3724
3725         Part 2 of: Implement sticky positioning
3726         https://bugs.webkit.org/show_bug.cgi?id=90046
3727
3728         Reviewed by Ojan Vafai.
3729
3730         Turn on ENABLE_CSS_STICKY_POSITION. Add support for parsing the new '-webkit-sticky'
3731         value for position, returning it from getComputedStyle(), and storing it in RenderStyle.
3732
3733         Test: fast/css/sticky/parsing-position-sticky.html
3734
3735         * Configurations/FeatureDefines.xcconfig:
3736         * css/CSSComputedStyleDeclaration.cpp:
3737         (WebCore::getPositionOffsetValue):
3738         * css/CSSParser.cpp:
3739         (WebCore::isValidKeywordPropertyAndValue):
3740         * css/CSSPrimitiveValueMappings.h:
3741         (WebCore::CSSPrimitiveValue::CSSPrimitiveValue):
3742         (WebCore::CSSPrimitiveValue::operator EPosition):
3743         * css/CSSValueKeywords.in:
3744         * rendering/style/RenderStyle.h:
3745         * rendering/style/RenderStyleConstants.h:
3746
3747 2012-07-23  Stephen Chenney  <schenney@chromium.org>
3748
3749         Crash when setting empty class name on a new element
3750         https://bugs.webkit.org/show_bug.cgi?id=92024
3751
3752         Reviewed by Andreas Kling.
3753
3754         Add a check for null attributeData() when setting the className to an
3755         empty string on a newly created element. New SVG elements have null
3756         attributeData() on baseVal upon creation.
3757
3758         Test: svg/custom/empty-className-baseVal-crash.html
3759
3760         * dom/StyledElement.cpp:
3761         (WebCore::StyledElement::classAttributeChanged): Add check for null attributeData()
3762
3763 2012-07-23  Shawn Singh  <shawnsingh@chromium.org>
3764
3765         [chromium] Refactor CCLayerTreeHostCommon: clean up clipRect and drawableContentRect design
3766         https://bugs.webkit.org/show_bug.cgi?id=80622
3767
3768         Reviewed by Adrienne Walker.
3769
3770         clipRect(), usesLayerClipping(), and drawableContentRect() have been
3771         very confusing in CCLayerTreeHostCommon for a while. This patch
3772         refactors it so that (1) clipping is only done locally in
3773         calcDrawTransforms, and (2) the layer's drawableContentRect value
3774         is now meaningful value outside of calcDrawTransforms.
3775         Additionally, the layer is now always clipped to the root
3776         surface's contentBounds (which are set to the viewport bounds).
3777         This refactor not only makes calcDrawTransforms far more readable and intuitive, but
3778         this patch enables more upcoming beneficial refactors, including
3779         the pending refactor in https://bugs.webkit.org/show_bug.cgi?id=88953.
3780
3781         Tests are also significantly updated to keep up with this refactoring change.
3782
3783         * platform/graphics/chromium/LayerChromium.cpp:
3784         (WebCore::LayerChromium::LayerChromium):
3785         * platform/graphics/chromium/LayerChromium.h:
3786         (LayerChromium):
3787         Removed m_usesLayerClipping and m_clipRect and associated accessors.
3788
3789         * platform/graphics/chromium/RenderSurfaceChromium.h:
3790         (RenderSurfaceChromium):
3791         Updated comment
3792
3793         * platform/graphics/chromium/cc/CCLayerImpl.cpp:
3794         (WebCore::CCLayerImpl::CCLayerImpl):
3795         * platform/graphics/chromium/cc/CCLayerImpl.h:
3796         (CCLayerImpl):
3797         Removed m_usesLayerClipping and m_clipRect and associated accessors.
3798
3799         * platform/graphics/chromium/cc/CCLayerTreeHost.cpp:
3800         (WebCore::CCLayerTreeHost::updateLayers):
3801         removed setClipRect code that no longer applies
3802
3803         * platform/graphics/chromium/cc/CCLayerTreeHostCommon.cpp:
3804         (WebCore::calculateLayerScissorRect):
3805         scissor rect is now a little bit tighter, the intersection between damage and layer's new drawableContentRect.
3806
3807         (WebCore::calculateSurfaceScissorRect):
3808         scissor rect is now a little bit tighter, except when filters are involved.
3809
3810         (WebCore::layerClipsSubtree):
3811         new helper function
3812
3813         (WebCore):
3814         (WebCore::calculateVisibleContentRect):
3815         (WebCore::subtreeShouldRenderToSeparateSurface):
3816         (WebCore::calculateDrawTransformsInternal):
3817            - added drawableContentRectOfSubtree to the function args, it is valid only after recursion returns,
3818            - added clipRectFromAncestor and bool ancestorClipsSubtree to function args, this replaces the layer's clipRect and usesLayerClipping.
3819            - removed the boolean return value, which was redundant with drawableContentRectOfSubtree.
3820            - replaced all the "setDrawableContentRect" logic with more intuitive, clear logic.
3821            - now, layer's drawableContentRect represents the clipped bounds of the layer expressed in the target surface space.
3822
3823         (WebCore::CCLayerTreeHostCommon::calculateDrawTransforms):
3824         (WebCore::pointIsClippedBySurfaceOrClipRect):
3825         * platform/graphics/chromium/cc/CCLayerTreeHostImpl.cpp:
3826         (WebCore::CCLayerTreeHostImpl::calculateRenderSurfaceLayerList):
3827         * platform/graphics/chromium/cc/CCOcclusionTracker.cpp:
3828         (WebCore::::layerScissorRectInTargetSurface):
3829         Updated this accessor.  It could be removed in the future, but not appropriate for this patch.
3830
3831         * platform/graphics/chromium/cc/CCRenderSurface.h:
3832
3833 2012-07-23  Patrick Gansterer  <paroga@webkit.org>
3834
3835         [CMAKE] Added missing RenderLayerFilterInfo.cpp to build system.
3836
3837         * CMakeLists.txt:
3838
3839 2012-07-23  Ryosuke Niwa  <rniwa@webkit.org>
3840
3841         REGRESSION(r123281): childNodes sometimes returns wrong nodes
3842         https://bugs.webkit.org/show_bug.cgi?id=92014
3843
3844         Reviewed by Anders Carlsson.
3845
3846         The bug was caused by a typo in itemBeforeOrAfter. Namely, it should have been calling firstNode as
3847         firstNode(forward, rootNode(), shouldOnlyIncludeDirectChildren()),
3848         NOT firstNode(forward, rootNode(), previous)
3849         as evident from the argument list of the function.
3850
3851         Test: fast/dom/NodeList/childNodes-reverse-iteration.html
3852
3853         * html/HTMLCollection.cpp:
3854         (WebCore::DynamicNodeListCacheBase::itemBeforeOrAfter):
3855
3856 2012-07-23  Steve VanDeBogart  <vandebo@chromium.org>
3857
3858         Chrome/Skia: PDF print output does not have clickable links.
3859         https://bugs.webkit.org/show_bug.cgi?id=91171
3860
3861         Reviewed by Stephen White.
3862
3863         Connect GraphicsContext::setURLForRect to Skia's new API for annotations.
3864
3865         Printing is not generally testable.
3866
3867         * platform/graphics/skia/GraphicsContextSkia.cpp:
3868         (WebCore::GraphicsContext::setURLForRect):
3869
3870 2012-07-23  Pierre Rossi  <pierre.rossi@gmail.com>
3871
3872         Unify numTouchEventHandlersChanged and needTouchEvents in the chrome client
3873         https://bugs.webkit.org/show_bug.cgi?id=91006
3874
3875         Reviewed by Ryosuke Niwa.
3876
3877         The rationale here is that the client doesn't need to know about the touch
3878         event handler count. needTouchEvents was already used for that purpose.
3879
3880         Test: fast/events/touch/touch-handler-count.html
3881
3882         * dom/Document.cpp:
3883         (WebCore::Document::Document):
3884         (WebCore::Document::didAddTouchEventHandler): Only notify the client if needed.
3885         (WebCore::Document::didRemoveTouchEventHandler): Ditto. Also unset the TOUCH_LISTENER
3886         flag for the document if we reach a count of zero. The rationale being that
3887         hasListenerType() is relied upon in other places in combination with TOUCH_LISTENER for
3888         the same purpose.
3889         * dom/Document.h:
3890         (Document):
3891         (WebCore::Document::touchEventHandlerCount):
3892         * loader/EmptyClients.h:
3893         * page/ChromeClient.h:
3894         (ChromeClient):
3895         * page/Frame.cpp: Removed notifyChromeClientTouchEventHandlerCountChanged.
3896         (WebCore::Frame::setDocument): call needsTouchEvents directly.
3897         * page/Frame.h: Ditto.
3898         (Frame):
3899         * testing/Internals.cpp:
3900         (WebCore::Internals::hasTouchEventListener): expose Document::hasListenerType indirectly
3901         for additional testing.
3902         (WebCore):
3903         * testing/Internals.h:
3904         (Internals):
3905         * testing/Internals.idl:
3906
3907 2012-07-23  Caio Marcelo de Oliveira Filho  <caio.oliveira@openbossa.org>
3908
3909         [Qt] RenderThemeQtMobile highlight colors are not being used
3910         https://bugs.webkit.org/show_bug.cgi?id=92004
3911
3912         Reviewed by Noam Rosenthal.
3913
3914         The issue here is that setPaletteFromPageClientIfExists() is being used as a
3915         virtual function, but it isn't, so when platformActiveSelectionBackgroundColor()
3916         runs, it doesn't pick the right palette.
3917
3918         Besides fixing this virtual behavior, the patch changes the structure a bit,
3919         because setPaletteFromPageClientIfExists() was being "overriden" in mobile theme
3920         to set the palette, which isn't exactly what the function name says.
3921
3922         * platform/qt/RenderThemeQt.cpp:
3923         (WebCore::RenderThemeQt::platformActiveSelectionBackgroundColor):
3924         (WebCore::RenderThemeQt::platformInactiveSelectionBackgroundColor):
3925         (WebCore::RenderThemeQt::platformActiveSelectionForegroundColor):
3926         (WebCore::RenderThemeQt::platformInactiveSelectionForegroundColor):
3927         (WebCore::RenderThemeQt::platformFocusRingColor):
3928         (WebCore::RenderThemeQt::systemColor):
3929         (WebCore::RenderThemeQt::getMediaControlForegroundColor):
3930         (WebCore::RenderThemeQt::paintMediaVolumeSliderTrack):
3931         Use the virtual colorPalette() to get the palette.
3932
3933         (WebCore::RenderThemeQt::colorPalette):
3934         (WebCore): Removed the code for getting the page client from here since it is
3935         used only by the QStyle variant.
3936
3937         * platform/qt/RenderThemeQt.h:
3938         (RenderThemeQt):
3939         * platform/qt/RenderThemeQtMobile.cpp:
3940         (WebCore::RenderThemeQtMobile::colorPalette):
3941         (WebCore):
3942         * platform/qt/RenderThemeQtMobile.h:
3943         (RenderThemeQtMobile):
3944
3945 2012-07-23  Simon Fraser  <simon.fraser@apple.com>
3946
3947         Part 1 of: Implement sticky positioning
3948         https://bugs.webkit.org/show_bug.cgi?id=90046
3949
3950         Reviewed by Ojan Vafai.
3951
3952         Add ENABLE_CSS_STICKY_POSITION, defaulting to off initially.
3953
3954         Sort the ENABLE_CSS lines in the file. Make sure all the flags
3955         are in FEATURE_DEFINES.
3956
3957         * Configurations/FeatureDefines.xcconfig:
3958
3959 2012-07-23  Sheriff Bot  <webkit.review.bot@gmail.com>
3960
3961         Unreviewed, rolling out r123339.
3962         http://trac.webkit.org/changeset/123339
3963         https://bugs.webkit.org/show_bug.cgi?id=92006
3964
3965         massive media tests failure (Requested by philn on #webkit).
3966
3967         * platform/graphics/gstreamer/GRefPtrGStreamer.cpp:
3968         (WTF::adoptGRef):
3969         * platform/graphics/gstreamer/GStreamerVersioning.cpp:
3970         (setGstElementClassMetadata):
3971         * platform/graphics/gstreamer/GStreamerVersioning.h:
3972
3973 2012-07-21  Vincent Scheib  <scheib@chromium.org>
3974
3975         webkitFullscreenElement, webkitCurrentFullScreenElement, webkitPointerLockElement block cross origin access.
3976         https://bugs.webkit.org/show_bug.cgi?id=91892
3977
3978         Reviewed by Adam Barth.
3979
3980         PointerLockElement only returned when requested from the document that owns it.
3981
3982         Tests: http/tests/fullscreen/fullscreenelement-different-origin.html
3983                http/tests/fullscreen/fullscreenelement-same-origin.html
3984                http/tests/pointer-lock/pointerlockelement-different-origin.html
3985                http/tests/pointer-lock/pointerlockelement-same-origin.html
3986
3987         * dom/Document.cpp:
3988         (WebCore::Document::webkitPointerLockElement):
3989
3990 2012-07-23  Philippe Normand  <pnormand@igalia.com>
3991
3992         [GTK][jhbuild] Switch to GStreamer 0.11 build
3993         https://bugs.webkit.org/show_bug.cgi?id=91727
3994
3995         Reviewed by Gustavo Noronha Silva.
3996
3997         Add a new function to encapsulate the GStreamer API removal of
3998         GST_OBJECT_IS_FLOATING in the upcoming 1.0 release. Use of this
3999         macro can now be replaced by calling the g_object_is_floating
4000         function.
4001
4002         * platform/graphics/gstreamer/GRefPtrGStreamer.cpp:
4003         (WTF::adoptGRef):
4004         * platform/graphics/gstreamer/GStreamerVersioning.cpp:
4005         (gstObjectIsFloating):
4006         * platform/graphics/gstreamer/GStreamerVersioning.h:
4007
4008 2012-07-23  Christophe Dumez  <christophe.dumez@intel.com>
4009
4010         [EFL] media/controls-styling.html is failing
4011         https://bugs.webkit.org/show_bug.cgi?id=91984
4012
4013         Reviewed by Eric Carlson.
4014
4015         Make sure that the CSS properties letter-spacing, word-spacing,
4016         line-height, text-transform, text-indent, text-shadow,
4017         text-decoration and color do not affect the media element controls,
4018         that display text.
4019
4020         No new tests. Already tested by media/controls-styling.html and
4021         media/controls-styling-strict.html
4022
4023         * css/mediaControlsEfl.css:
4024         (audio::-webkit-media-controls-current-time-display, video::-webkit-media-controls-current-time-display):
4025
4026 2012-07-23  Carlos Garcia Campos  <cgarcia@igalia.com>
4027
4028         Unreviewed. Fix make distcheck.
4029
4030         * GNUmakefile.list.am: Add missing header files.
4031
4032 2012-07-23  Kent Tamura  <tkent@chromium.org>
4033
4034         Replace some instances of shadowAncestorNode() with shadowHost()
4035         https://bugs.webkit.org/show_bug.cgi?id=91966
4036
4037         Reviewed by Hajime Morita.
4038
4039         shadowAncestorNode() is deprecated. We should use shadowHost().