2010-07-11 Maciej Stachowiak <mjs@apple.com>
[WebKit.git] / WebCore / ChangeLog
1 2010-07-11  Maciej Stachowiak  <mjs@apple.com>
2
3         Reviewed by Dan Bernstein.
4
5         Implement animation-related methods for WebKitTestRunner
6         https://bugs.webkit.org/show_bug.cgi?id=42053
7
8         * WebCore.exp.in: Export Document::getElementById for WebKit2's benefit.
9
10 2010-07-11  Adam Barth  <abarth@webkit.org>
11
12         Rubber-stamped by Eric Seidel
13
14         Add a complete list of the HTML5 entities in JSON format.
15
16         * html/HTMLEntityNames.json: Added.
17
18 2010-07-11  Martin Robinson  <mrobinson@igalia.com>
19
20         Reviewed by Xan Lopez.
21
22         [GTK] WebKitWebView should support drops
23         https://bugs.webkit.org/show_bug.cgi?id=39843
24
25         Add support for dropping content onto GTK+ WebViews.
26
27         No new tests, as the DRT does not support simulating drops yet.
28
29         * platform/gtk/ClipboardUtilitiesGtk.cpp:
30         (WebCore::dragOperationToGdkDragAction): Added.
31         (WebCore::gdkDragActionToDragOperation): Properly detect DragOperationEvery.
32         * platform/gtk/ClipboardUtilitiesGtk.h: Add declaration for dragOperationToGdkDragAction.
33         * platform/gtk/PasteboardHelper.cpp:
34         Add new target atom and rename the markup target type to match the others.
35         Add a method which fills a data object from drop data.
36         (WebCore::PasteboardHelper::initializeTargetList): Add support for new atoms.
37         (WebCore::selectionDataToUTF8String): Added this helper.
38         (WebCore::PasteboardHelper::getClipboardContents): Use the selectionDataToUTF8String helper.
39         (WebCore::PasteboardHelper::targetListForDataObject): Change to reflect markup atom rename.
40         (WebCore::PasteboardHelper::fillDataObjectFromDropData): Added.
41         (WebCore::PasteboardHelper::dropAtoms): Added.
42         * platform/gtk/PasteboardHelper.h: Add declarations of new methods.
43
44 2010-07-10  Darin Adler  <darin@apple.com>
45
46         Reviewed by Anders Carlsson.
47
48         Enhance content attribute reflection for URL attributes, including adding a non-empty option
49         https://bugs.webkit.org/show_bug.cgi?id=42040
50
51         Test: fast/dom/URL-attribute-reflection.html
52
53         Changed syntax from [ReflectURL] to [Reflect,URL] and also added support for
54         a new option, NonEmpty, which implements the non-empty URL concept from the
55         HTML5 specification.
56
57         * bindings/scripts/CodeGenerator.pm: Changed code to expect the Reflect and URL
58         extended attributes to come in separately. The URL one simply means "the string
59         of this attribute is a URL", since we don't have a distinct type for URL. Also
60         added a new NonEmpty extended attribute.
61
62         * bindings/scripts/CodeGeneratorJS.pm: Removed now-unneeded code to handle ReflectURL.
63         * bindings/scripts/CodeGeneratorV8.pm: Ditto.
64
65         * bindings/scripts/test/CPP/WebDOMTestObj.cpp: Updated for new added test cases
66         and the fix I made to the reflectedCustomURLAttr test.
67         * bindings/scripts/test/CPP/WebDOMTestObj.h: Ditto.
68         * bindings/scripts/test/GObject/WebKitDOMTestObj.cpp: Ditto.
69         * bindings/scripts/test/GObject/WebKitDOMTestObj.h: Ditto.
70         * bindings/scripts/test/JS/JSTestObj.cpp: Ditto.
71         * bindings/scripts/test/JS/JSTestObj.h: Ditto.
72         * bindings/scripts/test/ObjC/DOMTestObj.h: Ditto.
73         * bindings/scripts/test/ObjC/DOMTestObj.mm: Ditto.
74         * bindings/scripts/test/V8/V8TestObj.cpp: Ditto. Also, for some reason the V8
75         bindig writes out the keywords into the generated file as comments, so the keyword
76         change had a direct efffect on the output file.
77
78         * bindings/scripts/test/TestObj.idl: Changed the test cases for ReflectURL to use
79         the new syntax. Added test cases for NonEmpty. Fixed the name of
80         reflectedNonEmptyURLAttr, which accidentally was repeating reflectedURLAttr
81         instead; never noticed because we never compile the test output.
82
83         * dom/Element.cpp:
84         (WebCore::Element::getNonEmptyURLAttribute): Added. For use by NonEmpty and also
85         by any code that wants to implement the non-empty URL content attribute semantic.
86         * dom/Element.h: Ditto.
87
88         * html/HTMLAnchorElement.idl: Use Reflect,URL instead of ReflectURL.
89         * html/HTMLAreaElement.idl: Ditto.
90         * html/HTMLFrameElement.idl: Ditto.
91         * html/HTMLImageElement.idl: Ditto.
92         * html/HTMLInputElement.idl: Ditto.
93         * html/HTMLLinkElement.idl: Ditto.
94         * html/HTMLMediaElement.idl: Ditto.
95         * html/HTMLObjectElement.idl: Ditto.
96         * html/HTMLScriptElement.idl: Ditto.
97         * html/HTMLVideoElement.idl: Ditto.
98
99 2010-07-10  Tony Gentilcore  <tonyg@chromium.org>
100
101         Reviewed by Darin Adler.
102
103         Add missing derrived sources to xcode project
104         https://bugs.webkit.org/show_bug.cgi?id=42034
105
106         This allows --web-timing to work with build-webkit on OSX with JSC.
107         This should have been done in bug 41442.
108
109         No new tests because no new functionality.
110
111         * WebCore.xcodeproj/project.pbxproj:
112
113 2010-07-10  Anders Carlsson  <andersca@apple.com>
114
115         Reviewed by Sam Weinig.
116
117         Don't initialize plug-ins until allowed by the page
118         https://bugs.webkit.org/show_bug.cgi?id=42033
119
120         * WebCore.exp.in:
121         Export Document::addMediaCanStartListener and Document::removeMediaCanStartListener.
122
123 2010-07-10  Sam Weinig  <sam@webkit.org>
124
125         Reviewed by Anders Carlsson.
126
127         Patch for https://bugs.webkit.org/show_bug.cgi?id=42021
128         isEqualNode should work for DocumentType nodes
129
130         Test: fast/dom/Node/isEqualNode.html
131
132         * dom/Node.cpp:
133         (WebCore::Node::isEqualNode): Add DocumentType logic from the DOM3 spec.
134
135 2010-07-10  Daniel Bates  <dbates@rim.com>
136
137         Reviewed by Adam Barth.
138
139         Move enum ReasonForCallingCanExecuteScripts to header ScriptControllerBase.h
140         https://bugs.webkit.org/show_bug.cgi?id=39339
141
142         Moved the enum ReasonForCallingCanExecuteScripts, which was
143         defined in both the JSC and V8 ScriptController.h file, into
144         a shared file called ScriptControllerBase.h.
145
146         No functionality was changed, so no new tests.
147
148         * GNUmakefile.am: Added file bindings/ScriptControllerBase.h.
149         * WebCore.gypi: Ditto.
150         * WebCore.pro: Ditto.
151         * WebCore.vcproj/WebCore.vcproj: Ditto.
152         * WebCore.vcproj/WebCoreCommon.vsprops: Ditto.
153         * WebCore.xcodeproj/project.pbxproj: Ditto.
154         * bindings/ScriptControllerBase.h: Added.
155         (WebCore::):
156         * bindings/js/ScriptController.h: #include ScriptControllerBase.h
157         * bindings/v8/ScriptController.h: Ditto.
158
159 2010-07-10  Dan Bernstein  <mitz@apple.com>
160
161         Reviewed by Anders Carlsson.
162
163         <rdar://problem/8153214> Continuation outlines in layers do not paint correctly
164
165         Test: fast/inline/continuation-outlines-with-layers-2.html
166
167         Continuation outlines are normally painted by the containing block. However, when the
168         block and the inline are not enclosed by the same self-painting layer, the inline has to
169         paint its own outlines. This was handled correctly only for the case where the inline had
170         its own self-painting layer, but now when an ancestor inline had the self-painting layer.
171
172         * rendering/InlineFlowBox.cpp:
173         (WebCore::InlineFlowBox::paint): Instead of testing for having a self-painting layer, test
174         whether any intermediate box between the inline and the containing block has a self-painting
175         layer.
176         * rendering/RenderBlock.cpp:
177         (WebCore::RenderBlock::paintObject): Ditto.
178         * rendering/RenderObject.cpp:
179         (WebCore::RenderObject::enclosingBoxModelObject): Added this utility method.
180         * rendering/RenderObject.h:
181
182 2010-07-10  François Sausset  <sausset@gmail.com>
183
184         Reviewed by Kenneth Rohde Christiansen.
185
186         Build fix: syntax and typo issues
187         https://bugs.webkit.org/show_bug.cgi?id=41859
188
189         * html/HTMLTreeBuilder.cpp:
190         (WebCore::HTMLTreeBuilder::processStartTagForInBody):
191         (WebCore::HTMLTreeBuilder::processStartTag):
192
193 2010-07-10  Eric Seidel  <eric@webkit.org>
194
195         Reviewed by Adam Barth.
196
197         HTMLTreeBuilder needs adjustForeignAttributes support
198         https://bugs.webkit.org/show_bug.cgi?id=42022
199
200         To add adjust foreign attributes support I had to add an
201         AtomicString (prefixed name) to QualifiedName hash.  Once I had
202         done that, I decided it would be best for the other "adjust" functions
203         to share the same hash logic, so I moved them to using the same
204         AtomicString -> QualifiedName hash as well.
205
206         Tested by html5lib/runner.html
207
208         * dom/Attribute.h:
209         (WebCore::Attribute::parserSetName):
210         * html/HTMLTreeBuilder.cpp:
211
212 2010-07-10  Rob Buis  <rwlbuis@gmail.com>
213
214         Reviewed by Darin Adler.
215
216         https://bugs.webkit.org/show_bug.cgi?id=41978
217         Remove namespace prefixes from idl files
218
219         No new tests, idl syntax fixes.
220
221         * css/CSSImportRule.idl:
222         * css/CSSMediaRule.idl:
223         * css/CSSStyleSheet.idl:
224         * css/CSSVariablesRule.idl:
225         * svg/SVGAnimatedString.idl:
226         * svg/SVGColor.idl:
227         * svg/SVGDocument.idl:
228         * svg/SVGLangSpace.idl:
229         * svg/SVGSVGElement.idl:
230         * svg/SVGStringList.idl:
231         * svg/SVGStylable.idl:
232         * svg/SVGStyleElement.idl:
233         * svg/SVGTests.idl:
234         * svg/SVGZoomEvent.idl:
235         * xml/XPathEvaluator.idl:
236
237 2010-07-10  Dumitru Daniliuc  <dumi@chromium.org>
238
239         Unreviewed, GTK build fix.
240
241         * bindings/gobject/GObjectEventListener.h:
242
243 2010-07-10  Dumitru Daniliuc  <dumi@chromium.org>
244
245         Reviewed by Darin Adler.
246
247         Remove unnecessary includes in header files in WebCore/dom.
248         https://bugs.webkit.org/show_bug.cgi?id=41941
249
250         * css/StyleMedia.h:
251         * dom/BeforeLoadEvent.h:
252         * dom/CustomEvent.h:
253         * dom/DOMStringMap.h:
254         * dom/DeviceOrientationEvent.h:
255         * dom/EventListener.h:
256         * dom/MessagePort.h:
257         * dom/Node.cpp:
258         * dom/NodeFilter.h:
259         * dom/NodeRareData.h:
260         * dom/RegisteredEventListener.h:
261         * dom/SelectorNodeList.cpp:
262         * dom/SelectorNodeList.h:
263         * dom/StyleElement.h:
264         * dom/Traversal.h:
265
266 2010-07-09  Tony Chang  <tony@chromium.org>
267
268         Reviewed by Ojan Vafai.
269
270         crash in WebCore::CompositeEditCommand::splitTreeToNode when indenting pre
271         https://bugs.webkit.org/show_bug.cgi?id=38231
272
273         Test: editing/execCommand/indent-pre.html
274
275         * editing/IndentOutdentCommand.cpp:
276         (WebCore::countParagraphs):
277         (WebCore::IndentOutdentCommand::indentRegion): Split text nodes into one node per paragraph
278                                                        so moveParagraph doesn't get confused.
279         (WebCore::IndentOutdentCommand::splitTextNodes):
280         * editing/IndentOutdentCommand.h:
281
282 2010-07-09  Erik Arvidsson  <arv@chromium.org>
283
284         Reviewed by Darin Adler.
285
286         Computed style is not implemented for padding-start, padding-end, margin-start, margin-end
287         https://bugs.webkit.org/show_bug.cgi?id=41496
288
289         * css/CSSComputedStyleDeclaration.cpp:
290         (WebCore::CSSComputedStyleDeclaration::getPropertyCSSValue):
291         * css/CSSProperty.cpp:
292         (WebCore::CSSProperty::resolveDirectionAwareProperty): This function resolves the property ID for a direction aware property.
293         * css/CSSProperty.h:
294         * css/CSSStyleSelector.cpp: Use helper function.
295         (WebCore::CSSStyleSelector::applyProperty):
296
297 2010-07-09  Adam Barth  <abarth@webkit.org>
298
299         Reviewed by Eric Seidel.
300
301         HTML5 tree builder should pass some LayoutTests
302         https://bugs.webkit.org/show_bug.cgi?id=41991
303
304         Before this patch, we weren't attaching text nodes to the render tree,
305         which turns out to be important.  :)
306
307         This patch fixes more than 10,000 LayoutTests.
308
309         * html/HTMLConstructionSite.cpp:
310         (WebCore::HTMLConstructionSite::attach):
311         (WebCore::HTMLConstructionSite::attachAtSite):
312
313 2010-07-09  Patrick Gansterer  <paroga@paroga.com>
314
315         Reviewed by Darin Adler.
316
317         [WINCE] Buildfix for TextEncodingRegistry
318         https://bugs.webkit.org/show_bug.cgi?id=41992
319
320         Renamed TextCodecWince to TextCodecWinCE.
321
322         * platform/text/TextEncodingRegistry.cpp:
323         (WebCore::buildBaseTextCodecMaps):
324         (WebCore::extendTextCodecMaps):
325
326 2010-07-09  Leon Clarke  <leonclarke@google.com>
327
328         Reviewed by Adam Barth.
329
330         add support for link prefetching
331         https://bugs.webkit.org/show_bug.cgi?id=3652
332
333         Test: fast/dom/HTMLLinkElement/prefetch.html
334
335         * Configurations/FeatureDefines.xcconfig:
336         * html/HTMLLinkElement.cpp:
337         (WebCore::HTMLLinkElement::tokenizeRelAttribute):
338         (WebCore::HTMLLinkElement::process):
339         * html/HTMLLinkElement.h:
340         (WebCore::HTMLLinkElement::RelAttribute::RelAttribute):
341         * loader/Cache.cpp:
342         (WebCore::createResource):
343         * loader/CachedResource.cpp:
344         (WebCore::CachedResource::data):
345         (WebCore::CachedResource::didAddClient):
346         * loader/CachedResource.h:
347         (WebCore::CachedResource::):
348         (WebCore::CachedResource::error):
349         (WebCore::CachedResource::isPrefetch):
350         (WebCore::CachedResource::schedule):
351         * loader/CachedScript.cpp:
352         * loader/CachedScript.h:
353         * loader/DocLoader.cpp:
354         (WebCore::DocLoader::requestLinkPrefetch):
355         (WebCore::DocLoader::canRequest):
356         (WebCore::DocLoader::incrementRequestCount):
357         (WebCore::DocLoader::decrementRequestCount):
358         * loader/DocLoader.h:
359         * loader/loader.cpp:
360         (WebCore::cachedResourceTypeToTargetType):
361         (WebCore::Loader::determinePriority):
362         (WebCore::Loader::load):
363         (WebCore::Loader::Host::servePendingRequests):
364         (WebCore::Loader::Host::didFinishLoading):
365         (WebCore::Loader::Host::didFail):
366         (WebCore::Loader::Host::didReceiveResponse):
367         (WebCore::Loader::Host::cancelPendingRequests):
368         * loader/loader.h:
369         (WebCore::Loader::):
370         * platform/network/ResourceRequestBase.h:
371         (WebCore::ResourceRequestBase::):
372
373 2010-07-09  James Robinson  <jamesr@chromium.org>
374
375         Unreviewed build fix.  Add #include to pick up ExceptionCode values for config (like Qt) where it's not picked up indirectly.
376
377         * html/HTMLCanvasElement.cpp:
378
379 2010-07-09  James Robinson  <jamesr@chromium.org>
380
381         Reviewed by Darin Adler.
382
383         Removes CanvasSurface and moves all of its functionality to HTMLCanvasElement.
384         https://bugs.webkit.org/show_bug.cgi?id=42005
385
386         http://trac.webkit.org/changeset/55201 introduced a new base class for HTMLCanvasElement called CanvasSurface.
387         The intention was that this would allow for code sharing with the then-proposed OffscreenCanvas.  However,
388         there is no OffscreenCanvas and there's unlikely to be one soon.  Additionally CanvasSurface breaks
389         encapsulation pretty badly by doing "static_cast<HTMLCanvasElement* const>(this)".  Until an abstraction is
390         really needed we should just use HTMLCanvasElement when we want to talk about a canvas.
391
392         This patch moves all of CanvasSurface's functionality back up to HTMLCanvasElement and reorders the header
393         to be a bit more logical.
394
395         Refactor with no behavior change, thus no new tests.
396
397         * CMakeLists.txt:
398         * GNUmakefile.am:
399         * WebCore.gypi:
400         * WebCore.pro:
401         * WebCore.vcproj/WebCore.vcproj:
402         * WebCore.xcodeproj/project.pbxproj:
403         * dom/CanvasSurface.cpp: Removed.
404         * dom/CanvasSurface.h: Removed.
405         * html/HTMLCanvasElement.cpp:
406         (WebCore::HTMLCanvasElement::HTMLCanvasElement):
407         (WebCore::HTMLCanvasElement::willDraw):
408         (WebCore::HTMLCanvasElement::setSurfaceSize):
409         (WebCore::HTMLCanvasElement::toDataURL):
410         (WebCore::HTMLCanvasElement::convertLogicalToDevice):
411         (WebCore::HTMLCanvasElement::securityOrigin):
412         (WebCore::HTMLCanvasElement::styleSelector):
413         (WebCore::HTMLCanvasElement::createImageBuffer):
414         (WebCore::HTMLCanvasElement::drawingContext):
415         (WebCore::HTMLCanvasElement::buffer):
416         (WebCore::HTMLCanvasElement::baseTransform):
417         * html/HTMLCanvasElement.h:
418         (WebCore::HTMLCanvasElement::setObserver):
419         (WebCore::HTMLCanvasElement::width):
420         (WebCore::HTMLCanvasElement::height):
421         (WebCore::HTMLCanvasElement::size):
422         (WebCore::HTMLCanvasElement::toDataURL):
423         (WebCore::HTMLCanvasElement::setOriginTainted):
424         (WebCore::HTMLCanvasElement::originClean):
425         (WebCore::HTMLCanvasElement::hasCreatedImageBuffer):
426
427 2010-07-09  Simon Fraser  <simon.fraser@apple.com>
428
429         Reviewed by Darin Adler.
430
431         Calling layoutTestController.layerTreeAsText() should update layout
432         https://bugs.webkit.org/show_bug.cgi?id=41818
433         
434         Need to call updateLayout() before we check for any layers, not after,
435         since layout will update compositing, and may create layers.
436
437         * page/Frame.cpp:
438         (WebCore::Frame::layerTreeAsText):
439
440 2010-07-09  Kenneth Russell  <kbr@google.com>
441
442         Reviewed by Nate Chapin.
443
444         bufferSubData causes crash in WebGLBuffer::associateBufferSubData
445         https://bugs.webkit.org/show_bug.cgi?id=42004
446
447         Test: fast/canvas/webgl/index-validation-crash-with-buffer-sub-data.html
448
449         * html/canvas/WebGLBuffer.cpp:
450         (WebCore::WebGLBuffer::associateBufferData):
451          - Allocate m_elementArrayBuffer for entry point taking only size.
452            Guard against allocation failures of m_elementArrayBuffer.
453         (WebCore::WebGLBuffer::associateBufferSubData):
454          - Guard against any possibility of crashes due to m_elementArrayBuffer being NULL.
455
456 2010-07-09  Dumitru Daniliuc  <dumi@chromium.org>
457
458         Unreviewed, but pre-approved by Eric Seidel.
459
460         Remove unnecessary includes in header files in WebCore/css.
461         https://bugs.webkit.org/show_bug.cgi?id=41941
462
463         * css/CSSComputedStyleDeclaration.h:
464         * css/CSSFontFaceSource.cpp:
465         (WebCore::CSSFontFaceSource::getFontData):
466         * css/CSSFontFaceSource.h:
467         * css/CSSPrimitiveValueMappings.h:
468         * css/StyleMedia.h:
469
470 2010-07-09  Eric Seidel  <eric@webkit.org>
471
472         Reviewed by Adam Barth.
473
474         Get rid of manual case maps in HTMLTreeBuilder
475         https://bugs.webkit.org/show_bug.cgi?id=42000
476
477         No functional change, thus no tests.
478
479         * html/HTMLTreeBuilder.cpp:
480
481 2010-07-09  Tony Gentilcore  <tonyg@chromium.org>
482
483         Reviewed by Dimitri Glazkov.
484
485         Implement performance.timing.navigationStart
486         https://bugs.webkit.org/show_bug.cgi?id=41815
487
488         See: http://dev.w3.org/2006/webapi/WebTiming/#nt-navigation-start
489
490         * loader/FrameLoader.cpp:
491         (WebCore::FrameLoader::loadWithDocumentLoader):
492         * loader/FrameLoaderTypes.h:
493         (WebCore::FrameLoadTimeline::FrameLoadTimeline):
494         * page/Timing.cpp:
495         (WebCore::Timing::navigationStart):
496
497 2010-07-09  Jesus Sanchez-Palencia  <jesus.palencia@openbossa.org>
498
499         Reviewed by Kenneth Rohde Christiansen.
500
501         [Qt] GraphicsLayerQt must have syncCompositingStateForThisLayerOnly() implemented
502         https://bugs.webkit.org/show_bug.cgi?id=41954
503
504         Add GraphicsLayerQt::syncCompositingStateForThisLayerOnly, a non-recursive
505         implementation of GraphicsLayerQt::syncCompositingState and which targets
506         the current layer only.
507
508         * platform/graphics/qt/GraphicsLayerQt.cpp:
509         (WebCore::GraphicsLayerQt::syncCompositingStateForThisLayerOnly):
510         * platform/graphics/qt/GraphicsLayerQt.h:
511
512 2010-07-09  David Kozub  <zub@linux.fjfi.cvut.cz>
513
514         Reviewed by Darin Adler.
515
516         Fix build by adding missing html/TimeRanges.idl to CMakeLists.txt.
517
518         https://bugs.webkit.org/show_bug.cgi?id=41945
519
520         No functional changes, thus no tests.
521
522         * CMakeLists.txt:
523
524 2010-07-09  Eric Seidel  <eric@webkit.org>
525
526         Reviewed by Adam Barth.
527
528         Fix SVG tag name casing for HTMLTreeBuilder
529         https://bugs.webkit.org/show_bug.cgi?id=41998
530
531         Tested by html5lib/runner.html
532
533         * html/HTMLTreeBuilder.cpp:
534         (WebCore::):
535
536 2010-07-09  Eric Seidel  <eric@webkit.org>
537
538         Reviewed by Adam Barth.
539
540         Implement SVG attribute case mapping for HTMLTreeBuilder
541         https://bugs.webkit.org/show_bug.cgi?id=41949
542
543         Tested by html5lib/runner.html.
544
545         * html/HTMLTreeBuilder.cpp:
546
547 2010-07-09  Andy Estes  <aestes@apple.com>
548
549         Reviewed by Adele Peterson.
550
551         Remove the workaround for a Core Animation bug on platforms where the
552         bug has been fixed.
553         https://bugs.webkit.org/show_bug.cgi?id=41927
554         <rdar://problem/7920153>
555
556         * platform/graphics/mac/GraphicsLayerCA.mm:
557         (WebCore::GraphicsLayerCA::createTransformAnimationsFromKeyframes): Add
558         a compile-time check for platforms that have a Core Animation bug that
559         needs working around.
560
561 2010-07-08  Tony Gentilcore  <tonyg@chromium.org>
562
563         Reviewed by Nate Chapin.
564
565         Implement unloadEventEnd, loadEventStart, and loadEventEnd for Web Timing
566         https://bugs.webkit.org/show_bug.cgi?id=41332
567
568         Test: fast/dom/webtiming.html
569
570         * loader/FrameLoader.cpp:
571         (WebCore::FrameLoader::stopLoading):
572         (WebCore::FrameLoader::loadWithDocumentLoader):
573         * loader/FrameLoader.h:
574         (WebCore::FrameLoader::frameLoadTimeline):
575         * loader/FrameLoaderTypes.h:
576         (WebCore::FrameLoadTimeline::FrameLoadTimeline):
577         * page/DOMWindow.cpp:
578         (WebCore::DOMWindow::dispatchLoadEvent):
579         * page/Timing.cpp:
580         (WebCore::Timing::navigationStart):
581         (WebCore::Timing::unloadEventEnd):
582         (WebCore::Timing::loadEventStart):
583         (WebCore::Timing::loadEventEnd):
584         * page/Timing.h:
585         * page/Timing.idl:
586
587 2010-07-09  Dumitru Daniliuc  <dumi@chromium.org>
588
589         Unreviewed, but pre-approved by Eric Seidel.
590
591         Remove all unnecessary includes from the header files in WebCore/accessibility/.
592         https://bugs.webkit.org/show_bug.cgi?id=41941
593
594         * accessibility/AXObjectCache.h:
595         * accessibility/AccessibilityListBox.cpp:
596         * accessibility/AccessibilityListBox.h:
597         * accessibility/AccessibilityMenuList.h:
598         * accessibility/mac/AccessibilityObjectWrapper.h:
599
600 2010-07-09  Patrick Gansterer  <paroga@paroga.com>
601
602         Reviewed by Kent Tamura.
603
604         [WINCE] Buildfix for EventHandlerWin
605         https://bugs.webkit.org/show_bug.cgi?id=41909
606
607         Use Clipboard.h instead of ClipboardWin.h on WinCE.
608
609         * page/win/EventHandlerWin.cpp:
610
611 2010-07-09  Sheriff Bot  <webkit.review.bot@gmail.com>
612
613         Unreviewed, rolling out r62946.
614         http://trac.webkit.org/changeset/62946
615         https://bugs.webkit.org/show_bug.cgi?id=41965
616
617         AppCache inspector support should be enabled in WebKit ToT.
618         (Requested by pfeldman_ on #webkit).
619
620         * inspector/front-end/StoragePanel.js:
621         (WebInspector.StoragePanel):
622         (WebInspector.StoragePanel.prototype.reset):
623         (WebInspector.StoragePanel.prototype.addApplicationCache):
624
625 2010-07-09  François Sausset  <sausset@gmail.com>
626
627         Reviewed by Kenneth Rohde Christiansen.
628
629         Implement MathML mathcolor & mathbackground attributes
630         https://bugs.webkit.org/show_bug.cgi?id=41895
631
632         Test: mathml/presentation/attributes.xhtml
633
634         * dom/MappedAttributeEntry.h:
635         (WebCore::):
636         * mathml/MathMLElement.cpp:
637         (WebCore::MathMLElement::MathMLElement):
638         (WebCore::MathMLElement::mapToEntry):
639         (WebCore::MathMLElement::parseMappedAttribute):
640         * mathml/MathMLElement.h:
641         * mathml/mathattrs.in:
642
643 2010-07-09  Xiaomei Ji  <xji@chromium.org>
644
645         Reviewed by David Levin.
646
647         Fix characters with unicode-bidi-mirror property are not correctly
648         mirrored in Linux.
649         https://bugs.webkit.org/show_bug.cgi?id=41305
650
651         Since harfbuzz does not do mirroring, chromium should iterate each
652         character in the string and mirror it if needed before passing the
653         string to harfbuzz for shaping.
654
655         Test: fast/text/international/bidi-mirror-he-ar.html
656
657         * platform/graphics/chromium/FontLinux.cpp:
658         (WebCore::TextRunWalker::TextRunWalker):
659         (WebCore::TextRunWalker::~TextRunWalker):
660         (WebCore::TextRunWalker::mirrorCharacters):
661
662
663 2010-07-09  Anders Carlsson  <andersca@apple.com>
664
665         Reviewed by Simon Fraser.
666
667         Instantiate Netscape plug-ins, pass geometry information to Plugin
668         https://bugs.webkit.org/show_bug.cgi?id=41960
669
670         * WebCore.exp.in:
671         Export ScrollView::contentsToWindow.
672
673 2010-07-09  Patrick Gansterer  <paroga@paroga.com>
674
675         Reviewed by Kent Tamura.
676
677         [WINCE] Buildfix for FrameWince after r47440
678         https://bugs.webkit.org/show_bug.cgi?id=41904
679
680         * page/wince/FrameWince.cpp: Added property svn:eol-style.
681         (WebCore::computePageRectsForFrame):
682         (WebCore::imageFromSelection):
683
684 2010-07-09  Vitaly Repeshko  <vitalyr@chromium.org>
685
686         Reviewed by Pavel Feldman.
687
688         [v8] Call JS gc in a fresh context to avoid retaining the current one.
689         https://bugs.webkit.org/show_bug.cgi?id=41963
690         http://crbug.com/46571
691
692         * bindings/v8/ScriptController.cpp:
693         (WebCore::ScriptController::collectGarbage):
694
695 2010-07-09  Kenneth Russell  <kbr@google.com>
696
697         Reviewed by Dimitri Glazkov.
698
699         Crash with uniform array test
700         https://bugs.webkit.org/show_bug.cgi?id=36028
701
702         * bindings/v8/custom/V8WebGLRenderingContextCustom.cpp:
703         (WebCore::vertexAttribAndUniformHelperf):
704         (WebCore::uniformHelperi):
705         (WebCore::uniformMatrixHelper):
706          - Fixed type tests and casting of incoming arrays.
707         * html/canvas/WebGLRenderingContext.cpp:
708         (WebCore::WebGLRenderingContext::getUniform):
709          - Fixed crash when null WebGLUniform is passed to getUniform.
710
711 2010-07-09  Chris Fleizach  <cfleizach@apple.com>
712
713         Reviewed by Darin Adler.
714
715         AX:  text editing not spoken by VO on web view contenteditable textbox
716         https://bugs.webkit.org/show_bug.cgi?id=41912
717
718         Test: platform/mac/accessibility/selection-value-changes-for-aria-textbox.html
719
720         * accessibility/AccessibilityRenderObject.cpp:
721         (WebCore::AccessibilityRenderObject::renderObjectIsObservable):
722
723 2010-07-09  Michael Nordman  <michaeln@google.com>
724
725         Reviewed by Dumitru Daniliuc.
726
727         Use class ProgressEvent when raising appcache related progress events and
728         set the 'total', 'loaded', and 'lengthComputable' attributes.
729         Also raise the final progress event with the 'total' and 'loaded' attribute
730         values are equal to one another to keep pace with the spec for this feature.
731         https://bugs.webkit.org/show_bug.cgi?id=37602
732
733         Test: http/tests/appcache/progress-counter.html
734
735         * loader/appcache/ApplicationCacheGroup.cpp:
736         (WebCore::ApplicationCacheGroup::ApplicationCacheGroup):
737         (WebCore::ApplicationCacheGroup::didFinishLoadingManifest):
738         (WebCore::ApplicationCacheGroup::checkIfLoadIsComplete):
739         (WebCore::ApplicationCacheGroup::startLoadingEntry):
740         (WebCore::CallCacheListenerTask::create):
741         (WebCore::CallCacheListenerTask::performTask):
742         (WebCore::CallCacheListenerTask::CallCacheListenerTask):
743         (WebCore::ApplicationCacheGroup::postListenerTask):
744         * loader/appcache/ApplicationCacheGroup.h:
745         (WebCore::ApplicationCacheGroup::postListenerTask):
746         * loader/appcache/ApplicationCacheHost.cpp:
747         (WebCore::ApplicationCacheHost::notifyDOMApplicationCache):
748         (WebCore::ApplicationCacheHost::stopDeferringEvents):
749         (WebCore::ApplicationCacheHost::dispatchDOMEvent):
750         * loader/appcache/ApplicationCacheHost.h:
751         (WebCore::ApplicationCacheHost::DeferredEvent::DeferredEvent):
752
753 2010-07-09  Sebastian Dröge  <sebastian.droege@collabora.co.uk>
754
755         Reviewed by Gustavo Noronha Silva.
756
757         Bug 41340 - [GStreamer] Subtle race condition during seeks
758         https://bugs.webkit.org/show_bug.cgi?id=41340
759
760         * platform/graphics/gstreamer/WebKitWebSourceGStreamer.cpp:
761         (webKitWebSrcStop):
762         (webKitWebSrcNeedDataMainCb):
763         (webKitWebSrcNeedDataCb):
764         (webKitWebSrcEnoughDataMainCb):
765         (webKitWebSrcEnoughDataCb):
766         (webKitWebSrcSeekDataCb):
767         (StreamingClient::didFinishLoading):
768         Fix two subtle race conditions that can happen during seeks:
769         - The timeout callback could be called before the callback ID is
770           assigned to the instance private data. This causes the ID to
771           be set after the callback has finished and breaks all future
772           processing.
773         - The source must not go EOS while a seek is pending because
774           this confuses appsrc due to the different threads involved here.
775
776 2010-07-09  Simon Hausmann  <simon.hausmann@nokia.com>
777
778         Reviewed by Tor Arne Vestbø.
779
780         [Qt] Removed hard-coded enabled default of video support
781
782         Properly detect video/audio instead of defaulting to true.Properly detect video/audio instead of defaulting to true.Properly detect video/audio instead of defaulting to true.Properly detect video/audio instead of defaulting to true.
783
784         * WebCore.pri:
785
786 2010-07-09  Ben Murdoch  <benm@google.com>
787
788         Reviewed by Steve Block.
789
790         [Arm] Missing NaN check in XPath substring function
791         https://bugs.webkit.org/show_bug.cgi?id=41862
792
793         Test: fast/xpath/substring-nan-position.html
794
795         * xml/XPathFunctions.cpp:
796         (WebCore::XPath::FunSubstring::evaluate): Add an isnan()
797             to the value returned from evaluating the position
798             argument and early out and return the empty string.
799
800 2010-07-09  Simon Hausmann  <simon.hausmann@nokia.com>
801
802         Reviewed by Tor Arne Vestbø.
803
804         [Qt] Prospective build fix.
805
806         Moved the media element detection from WebCore.pro into WebCore.pri, where
807         all the features are detected. This is also used by build-webkit to determine
808         the defaults, which may be the reason for the build breakage.
809
810         * WebCore.pri:
811         * WebCore.pro:
812
813 2010-07-09  Simon Hausmann  <simon.hausmann@nokia.com>
814
815         Reviewed by Tor Arne Vestbø.
816
817         [Qt] Re-enable support for QtMultimediaKit as backend for the media elements
818
819         Experimental support is re-enabled if QtMobility is available and the Qt
820         version is 4.7 or above.
821
822         * WebCore.pro:
823         * platform/graphics/MediaPlayer.cpp:
824         * platform/graphics/qt/MediaPlayerPrivateQt.cpp:
825         (WebCore::MediaPlayerPrivate::supportsType): Adapt to latest QtMultimediaKit API changes.
826         (WebCore::MediaPlayerPrivate::MediaPlayerPrivate):
827         (WebCore::MediaPlayerPrivate::totalBytes):
828
829 2010-07-09  Pavel Feldman  <pfeldman@chromium.org>
830
831         Reviewed by Yury Semikhatsky.
832
833         Web Inspector: do not show default tooltip when detailed network info is available.
834
835         https://bugs.webkit.org/show_bug.cgi?id=41957
836
837         * English.lproj/localizedStrings.js:
838         * inspector/front-end/ResourcesPanel.js:
839         (WebInspector.ResourcesPanel):
840         (WebInspector.ResourcesPanel.prototype.reset):
841         (WebInspector.ResourcesPanel.prototype.showResource):
842         (WebInspector.ResourcesPanel.prototype._getPopoverAnchor):
843         (WebInspector.ResourcesPanel.prototype._showPopover):
844         (WebInspector.ResourceTimeCalculator.prototype.computeBarGraphLabels):
845
846 2010-07-09  François Sausset  <sausset@gmail.com>
847
848         Reviewed by Beth Dakin.
849
850         Correct the default font variant of mtext (regular instead of italic).
851         https://bugs.webkit.org/show_bug.cgi?id=41626
852
853         Add test to be sure that only <mi> elements use italic fonts by default.
854         Update test with fractions to take into account the correct behaviour of mtext (regular instead of italic).
855
856         * css/mathml.css:
857         (mi):
858
859 2010-07-09  Tony Gentilcore  <tonyg@chromium.org>
860
861         Reviewed by Dimitri Glazkov.
862
863         Implement performance.navigation.type
864         https://bugs.webkit.org/show_bug.cgi?id=41564
865
866         Tests: fast/dom/navigation-type-back-forward.html
867                fast/dom/navigation-type-navigate.html
868                fast/dom/navigation-type-reload.html
869
870         * page/Navigation.cpp:
871         (WebCore::Navigation::type):
872         * page/Navigation.h:
873         (WebCore::Navigation::):
874         * page/Navigation.idl:
875
876 2010-07-09  Pavel Feldman  <pfeldman@chromium.org>
877
878         Reviewed by Yury Semikhatsky.
879
880         Web Inspector: hide AppCache until implemented.
881
882         https://bugs.webkit.org/show_bug.cgi?id=41858
883
884         * inspector/front-end/StoragePanel.js:
885         (WebInspector.StoragePanel):
886         (WebInspector.StoragePanel.prototype.reset):
887         (WebInspector.StoragePanel.prototype.addApplicationCache):
888
889 2010-07-09  François Sausset  <sausset@gmail.com>
890
891         Reviewed by Kenneth Rohde Christiansen.
892
893         Set the good default behaviour for the columalign attribute on MathML mtable element
894         https://bugs.webkit.org/show_bug.cgi?id=41631
895
896         Test update: mathml/tables.xhtml
897
898         * css/mathml.css:
899         (math):
900
901 2010-07-09  Sheriff Bot  <webkit.review.bot@gmail.com>
902
903         Unreviewed, rolling out r62937.
904         http://trac.webkit.org/changeset/62937
905         https://bugs.webkit.org/show_bug.cgi?id=41955
906
907         Crashes SnowLeopard leaks and Windows debug bot in fast/xsl
908         /xslt-relative-path.xml, with assertion in
909         XSLTProcessorLibxslt.cpp:264 (Requested by WildFox on
910         #webkit).
911
912         * xml/XSLTProcessor.h:
913         (WebCore::XSLTProcessor::XSLTProcessor):
914         * xml/XSLTProcessorLibxslt.cpp:
915         (WebCore::docLoaderFunc):
916         (WebCore::setXSLTLoadCallBack):
917         (WebCore::xsltStylesheetPointer):
918         (WebCore::XSLTProcessor::transformToString):
919
920 2010-07-09  Yael Aharon  <yael.aharon@nokia.com>
921
922         Reviewed by Laszlo Gombos.
923
924         NotificationPresenter needs a cancelRequestPermission API
925         https://bugs.webkit.org/show_bug.cgi?id=41783
926
927         Updated NotificationPresenter API, to use ScriptExecutionContext instead of origin.
928         Added new API NotificationPresenter::cancelRequestsForPermision
929         The new API will be implemented and a test will be added in a followup patch.
930
931         * notifications/Notification.cpp:
932         (WebCore::Notification::Notification):
933         * notifications/NotificationCenter.cpp:
934         (WebCore::NotificationCenter::checkPermission):
935         (WebCore::NotificationCenter::requestPermission):
936         (WebCore::NotificationCenter::disconnectFrame):
937         * notifications/NotificationCenter.h:
938         * notifications/NotificationPresenter.h:
939
940 2010-07-09  Andreas Wictor  <andreas.wictor@xcerion.com>
941
942         Reviewed by Alexey Proskuryakov.
943
944         Remove global variables from XSLTProcessorLibxslt.cpp
945         https://bugs.webkit.org/show_bug.cgi?id=41348
946
947         Remove the globalProcessor and globalDocLoader global variables
948         by using the _private field that exists on most libxml structs.
949
950         No new tests, existing tests covers this.
951
952         * xml/XSLTProcessor.h:
953         (WebCore::XSLTProcessor::sourceNode):
954         (WebCore::XSLTProcessor::XSLTProcessor):
955         * xml/XSLTProcessorLibxslt.cpp:
956         (WebCore::registeredXSLTProcessors):
957         (WebCore::registeredXSLStyleSheets):
958         (WebCore::docLoaderFunc):
959         (WebCore::clearSavedStyleSheetPointers):
960         (WebCore::xsltStylesheetPointer):
961         (WebCore::XSLTProcessor::transformToString):
962
963 2010-07-09  Adam Barth  <abarth@webkit.org>
964
965         Unreviewed build fix.
966
967         We recently taught the HTMLTokenizer to recognize self-closing tags,
968         <http://trac.webkit.org/changeset/62926>, but that confused the
969         LegacyHTMLTreeBuilder.  It turns out that it's much happier if we never
970         say a tag is self-closing.
971
972         * html/HTMLTreeBuilder.cpp:
973         (WebCore::convertToOldStyle):
974
975 2010-07-09  Nikolas Zimmermann  <nzimmermann@rim.com>
976
977         Reviewed by Rob Buis.
978
979         svg/custom/use-instanceRoot-event-bubbling.xhtml test crashes
980         https://bugs.webkit.org/show_bug.cgi?id=41931
981
982         Be careful to not mutate (marking it for recreation) the shadow tree, while building it.
983         The recent change that cloneNode() properly synchronizes the style/SVG attributes caused this problem.
984
985         Fixes crash seen on the buildbots in svg/custom/use-instanceRoot-event-bubbling.xhtml.
986
987         * svg/SVGUseElement.cpp:
988         (WebCore::SVGUseElement::SVGUseElement):
989         (WebCore::SVGUseElement::recalcStyle):
990         (WebCore::SVGUseElement::buildShadowAndInstanceTree):
991         (WebCore::SVGUseElement::invalidateShadowTree):
992         * svg/SVGUseElement.h:
993         (WebCore::SVGUseElement::setUpdatesBlocked):
994
995 2010-07-09  Adam Barth  <abarth@webkit.org>
996
997         Reviewed by Eric Seidel.
998
999         Skip leading newlines in <textarea>
1000         https://bugs.webkit.org/show_bug.cgi?id=41947
1001
1002         We would have caught this earlier with the ASSERT.
1003
1004         * html/HTMLTokenizer.cpp:
1005         (WebCore::HTMLTokenizer::nextToken):
1006
1007 2010-07-08  Pavel Feldman  <pfeldman@chromium.org>
1008
1009         Reviewed by Yury Semikhatsky.
1010
1011         Web Inspector: Provide detailed network info in the resources panel.
1012
1013         https://bugs.webkit.org/show_bug.cgi?id=40227
1014
1015         * English.lproj/localizedStrings.js:
1016         * inspector/InspectorController.cpp:
1017         (WebCore::InspectorController::willSendRequest):
1018         (WebCore::InspectorController::didReceiveResponse):
1019         * inspector/InspectorController.h:
1020         * inspector/InspectorResource.cpp:
1021         (WebCore::InspectorResource::InspectorResource):
1022         (WebCore::InspectorResource::updateResponse):
1023         (WebCore::InspectorResource::updateScriptObject):
1024         (WebCore::InspectorResource::buildObjectForTiming):
1025         * inspector/InspectorResource.h:
1026         * inspector/front-end/ResourcesPanel.js:
1027         (WebInspector.ResourcesPanel):
1028         (WebInspector.ResourcesPanel.prototype.elementsToRestoreScrollPositionsFor):
1029         (WebInspector.ResourcesPanel.prototype._getPopoverAnchor):
1030         (WebInspector.ResourcesPanel.prototype._showPopover):
1031         (WebInspector.ResourcesPanel.prototype.hide):
1032         (WebInspector.ResourceTimeCalculator.prototype.computeBarGraphLabels):
1033         (WebInspector.ResourceGraph):
1034         * inspector/front-end/inspector.js:
1035         (WebInspector.updateResource):
1036         * platform/network/ResourceLoadTiming.h:
1037         (WebCore::ResourceLoadTiming::create):
1038         (WebCore::ResourceLoadTiming::deepCopy):
1039         (WebCore::ResourceLoadTiming::operator==):
1040         (WebCore::ResourceLoadTiming::ResourceLoadTiming):
1041         * platform/network/ResourceRequestBase.h:
1042         (WebCore::ResourceRequestBase::reportLoadTiming):
1043         (WebCore::ResourceRequestBase::setReportLoadTiming):
1044         (WebCore::ResourceRequestBase::ResourceRequestBase):
1045         * platform/network/ResourceResponseBase.cpp:
1046         (WebCore::ResourceResponseBase::connectionID):
1047         (WebCore::ResourceResponseBase::setConnectionID):
1048         * platform/network/ResourceResponseBase.h:
1049
1050 2010-07-09  Eric Seidel  <eric@webkit.org>
1051
1052         Reviewed by Adam Barth.
1053
1054         Implement self closing start tag state in the tokenizer
1055         https://bugs.webkit.org/show_bug.cgi?id=41946
1056
1057         Amazingly we got this far w/o needing self closing.
1058         The LegacyHTMLTreeBuilder clearly uses the self-closing
1059         state, but the fact that we never exposed it seems to
1060         not have caused any test failures.  Sad.
1061
1062         The new HTMLTreeBuilder only needs the self closing state
1063         for foreign content mode (and a few parse error cases).
1064
1065         * html/HTMLToken.h:
1066         (WebCore::HTMLToken::setSelfClosing):
1067         * html/HTMLTokenizer.cpp:
1068         (WebCore::HTMLTokenizer::nextToken):
1069         * html/HTMLTreeBuilder.cpp:
1070         (WebCore::):
1071
1072 2010-07-09  Eric Seidel  <eric@webkit.org>
1073
1074         Reviewed by Adam Barth.
1075
1076         Add basic "in foreign content" support to the TreeBuilder
1077         https://bugs.webkit.org/show_bug.cgi?id=41943
1078
1079         This is covered by numerous tests in html5lib/runner.html.
1080
1081         "in foreign content" mode requires us to be able to process
1082         tokens using the "secondary insertion mode".  We have to set
1083         a fake insertion mode to do that, so much of this code is enabling
1084         setting of fake insertion modes, and then later restoration of the
1085         insertion mode after execution.
1086
1087         There is a lot more of foreign content mode to implement, but this is
1088         a good start resulting in huge test progressions.
1089
1090         * html/HTMLConstructionSite.cpp:
1091         (WebCore::HTMLConstructionSite::insertSelfClosingHTMLElement):
1092          - Use createHTMLElementAndAttachToCurrent instead of copy/paste code.
1093          - No need to include HTMLElementFactory in this file.
1094         * html/HTMLConstructionSite.h:
1095          - RedirectToFosterParentGuard does not need to be a friend class.
1096         * html/HTMLElementStack.cpp:
1097         (WebCore::HTMLElementStack::isOnlyHTMLElementsInScope):
1098         * html/HTMLElementStack.h:
1099         * html/HTMLTreeBuilder.cpp:
1100         (WebCore::HTMLTreeBuilder::processStartTagForInBody):
1101         (WebCore::):
1102         * html/HTMLTreeBuilder.h:
1103         (WebCore::HTMLTreeBuilder::setInsertionMode):
1104         (WebCore::HTMLTreeBuilder::isFakeInsertionMode):
1105         (WebCore::HTMLTreeBuilder::setFakeInsertionMode):
1106
1107 2010-07-09  Adam Barth  <abarth@webkit.org>
1108
1109         Reviewed by Eric Seidel.
1110
1111         Coalesce text nodes when foster parenting
1112         https://bugs.webkit.org/show_bug.cgi?id=41921
1113
1114         Introduces the notion of an AttachmentSite to the overall
1115         HTMLConstructionSite.  Maybe we should rename HTMLConstructionSite to
1116         HTMLConstructionArea since we construct things all over the tree?  :)
1117
1118         There's something wrong in the internal layering in this class, but I
1119         can't quite see what it is.  I added a FIXME for the some of the
1120         symptoms.
1121
1122         * html/HTMLConstructionSite.cpp:
1123         (WebCore::HTMLConstructionSite::attach):
1124         (WebCore::HTMLConstructionSite::attachAtSite):
1125         (WebCore::HTMLConstructionSite::insertTextNode):
1126         (WebCore::HTMLConstructionSite::findFosterSite):
1127         (WebCore::HTMLConstructionSite::fosterParent):
1128         * html/HTMLConstructionSite.h:
1129
1130 2010-07-08  Nikolas Zimmermann  <nzimmermann@rim.com>
1131
1132         Reviewed by Dirk Schulze.
1133
1134         RenderSVGRoot does not include border/padding while repainting
1135         https://bugs.webkit.org/show_bug.cgi?id=41854
1136
1137         RenderSVGRoot does not include border/padding in the repaint rect.
1138         clippedOverflowRectForRepaint() was missing. Affects all DRT results,
1139         as <svg> now gets properly sized.
1140
1141         Test: svg/custom/repaint-moving-svg-and-div.xhtml
1142
1143         * rendering/RenderSVGRoot.cpp:
1144         (WebCore::RenderSVGRoot::repaintRectInLocalCoordinates):
1145         (WebCore::RenderSVGRoot::clippedOverflowRectForRepaint):
1146         * rendering/RenderSVGRoot.h:
1147
1148 2010-07-09  Albert J. Wong  <ajwong@chromium.org>
1149
1150         Reviewed by Nikolas Zimmermann.
1151
1152         Add RuntimeEnabledFeatures::timeRangesEnabled() required by r62880
1153
1154         https://bugs.webkit.org/show_bug.cgi?id=41935
1155
1156         * bindings/generic/RuntimeEnabledFeatures.cpp:
1157         (WebCore::RuntimeEnabledFeatures::timeRangesEnabled):
1158                 Add in timeRangesEnabled() implementation.
1159         * bindings/generic/RuntimeEnabledFeatures.h:
1160                 Add in timeRangesEnabled() declaration.
1161
1162
1163 2010-07-09  Eric Seidel  <eric@webkit.org>
1164
1165         Reviewed by Adam Barth.
1166
1167         Add setInsertionMode setter in preparation for "in foreign content" mode
1168         https://bugs.webkit.org/show_bug.cgi?id=41942
1169
1170         "in foreign content" mode needs to be able to use a fake
1171         insertion mode for processing.  We need to be able to save the
1172         original insertion mode, set a fake one, and then restore the original
1173         if it wasn't changed.  To detect changes, we need all callsites to
1174         use a setInsertionMode accessor instead of m_insertionMode =
1175
1176         No functional changes, thus no tests.
1177
1178         * html/HTMLTreeBuilder.cpp:
1179         (WebCore::HTMLTreeBuilder::processStartTagForInBody):
1180         (WebCore::HTMLTreeBuilder::processColgroupEndTagForInColumnGroup):
1181         (WebCore::HTMLTreeBuilder::processStartTagForInTable):
1182         (WebCore::HTMLTreeBuilder::processStartTag):
1183         (WebCore::HTMLTreeBuilder::processBodyEndTagForInBody):
1184         (WebCore::HTMLTreeBuilder::setInsertionModeAndEnd):
1185         (WebCore::HTMLTreeBuilder::processEndTagForInTableBody):
1186         (WebCore::HTMLTreeBuilder::processEndTagForInCell):
1187         (WebCore::HTMLTreeBuilder::processCaptionEndTagForInCaption):
1188         (WebCore::HTMLTreeBuilder::processTrEndTagForInRow):
1189         (WebCore::HTMLTreeBuilder::processEndTag):
1190         (WebCore::HTMLTreeBuilder::processCharacter):
1191         (WebCore::HTMLTreeBuilder::processDefaultForInTableTextMode):
1192         (WebCore::HTMLTreeBuilder::processGenericRCDATAStartTag):
1193         (WebCore::HTMLTreeBuilder::processGenericRawTextStartTag):
1194         (WebCore::HTMLTreeBuilder::processScriptStartTag):
1195
1196 2010-07-09  Eric Seidel  <eric@webkit.org>
1197
1198         Reviewed by Adam Barth.
1199
1200         Add insertForeignElement in preparation for adding "in foreign content" support
1201         https://bugs.webkit.org/show_bug.cgi?id=41940
1202
1203         No functional changes, thus no tests.
1204
1205         * html/HTMLConstructionSite.cpp:
1206         (WebCore::HTMLConstructionSite::insertForeignElement):
1207         (WebCore::HTMLConstructionSite::createElement):
1208         (WebCore::HTMLConstructionSite::createHTMLElement):
1209         * html/HTMLConstructionSite.h:
1210
1211 2010-07-09  Adam Barth  <abarth@webkit.org>
1212
1213         Reviewed by Eric Seidel.
1214
1215         Implementing pending table characters
1216         https://bugs.webkit.org/show_bug.cgi?id=41916
1217
1218         This turned out to not be as scary as I thought it would be.
1219
1220         * html/HTMLTreeBuilder.cpp:
1221         (WebCore::HTMLTreeBuilder::processDoctypeToken):
1222         (WebCore::HTMLTreeBuilder::processStartTag):
1223         (WebCore::HTMLTreeBuilder::processEndTag):
1224         (WebCore::HTMLTreeBuilder::processComment):
1225         (WebCore::HTMLTreeBuilder::processCharacter):
1226         (WebCore::HTMLTreeBuilder::processEndOfFile):
1227         (WebCore::HTMLTreeBuilder::processDefaultForInTableTextMode):
1228         * html/HTMLTreeBuilder.h:
1229
1230 2010-07-08  Adam Barth  <abarth@webkit.org>
1231
1232         Reviewed by Eric Seidel.
1233
1234         Handle whitespace correctly
1235         https://bugs.webkit.org/show_bug.cgi?id=41907
1236
1237         This patch introduces an extra memcpy in the character token pipeline.
1238         I'll remove the memcpy in a future patch.
1239
1240         * html/HTMLConstructionSite.cpp:
1241         (WebCore::HTMLConstructionSite::insertTextNode):
1242         * html/HTMLConstructionSite.h:
1243         * html/HTMLTreeBuilder.cpp:
1244         (WebCore::HTMLTreeBuilder::processCharacter):
1245         * html/HTMLTreeBuilder.h:
1246
1247 2010-07-09  Eric Seidel  <eric@webkit.org>
1248
1249         Reviewed by Adam Barth.
1250
1251         Move more mode handling into functions for later re-use
1252         https://bugs.webkit.org/show_bug.cgi?id=41939
1253
1254         No functional changes, thus no tests.
1255
1256         * html/HTMLTreeBuilder.cpp:
1257         (WebCore::HTMLTreeBuilder::processEndTagForInTableBody):
1258         (WebCore::HTMLTreeBuilder::processEndTagForInRow):
1259         (WebCore::HTMLTreeBuilder::processEndTagForInCell):
1260         (WebCore::HTMLTreeBuilder::processEndTag):
1261         * html/HTMLTreeBuilder.h:
1262
1263 2010-07-09  Eric Seidel  <eric@webkit.org>
1264
1265         Reviewed by Adam Barth.
1266
1267         Add new popUntilPopped functions to clean up code
1268         https://bugs.webkit.org/show_bug.cgi?id=41936
1269
1270         Pretty self explanatory.  Much awesome code reduction.
1271
1272         No functional changes, thus no tests.
1273
1274         * html/HTMLElementStack.cpp:
1275         (WebCore::HTMLElementStack::popUntilPopped):
1276         * html/HTMLElementStack.h:
1277         * html/HTMLTreeBuilder.cpp:
1278         (WebCore::HTMLTreeBuilder::processAnyOtherEndTagForInBody):
1279         (WebCore::HTMLTreeBuilder::callTheAdoptionAgency):
1280         (WebCore::HTMLTreeBuilder::processEndTagForInBody):
1281         (WebCore::HTMLTreeBuilder::processCaptionEndTagForInCaption):
1282         (WebCore::HTMLTreeBuilder::processTableEndTagForInTable):
1283         (WebCore::HTMLTreeBuilder::processEndTag):
1284
1285 2010-07-09  Eric Seidel  <eric@webkit.org>
1286
1287         Unreviewed.  Just renaming (discussed with Adam Barth).
1288
1289         Rename createElement* to createHTMLElement* to better reflect
1290         behavior.  This is in preparation for adding foreign content support.
1291
1292         * html/HTMLConstructionSite.cpp:
1293         (WebCore::HTMLConstructionSite::createHTMLElementAndAttachToCurrent):
1294         (WebCore::HTMLConstructionSite::insertHTMLHtmlElement):
1295         (WebCore::HTMLConstructionSite::insertHTMLHeadElement):
1296         (WebCore::HTMLConstructionSite::insertHTMLBodyElement):
1297         (WebCore::HTMLConstructionSite::insertHTMLElement):
1298         (WebCore::HTMLConstructionSite::insertSelfClosingHTMLElement):
1299         (WebCore::HTMLConstructionSite::createHTMLElement):
1300         * html/HTMLConstructionSite.h:
1301         * html/HTMLTreeBuilder.cpp:
1302         (WebCore::HTMLTreeBuilder::callTheAdoptionAgency):
1303
1304 2010-07-09  Eric Seidel  <eric@webkit.org>
1305
1306         Unreviewed.  Just renaming (discussed with Adam Barth).
1307
1308         Rename insertElement to insertHTMLElement and
1309         insertSelfClosingElement to insertSelfClosingHTMLElement
1310         to better reflect what they actually do.  This is in preparation
1311         for adding foreign content support.
1312
1313         * html/HTMLConstructionSite.cpp:
1314         (WebCore::HTMLConstructionSite::insertHTMLElement):
1315         (WebCore::HTMLConstructionSite::insertSelfClosingHTMLElement):
1316         (WebCore::HTMLConstructionSite::insertFormattingElement):
1317         (WebCore::HTMLConstructionSite::reconstructTheActiveFormattingElements):
1318         (WebCore::HTMLConstructionSite::fosterParent):
1319         * html/HTMLConstructionSite.h:
1320         * html/HTMLTreeBuilder.cpp:
1321         (WebCore::HTMLTreeBuilder::processCloseWhenNestedTag):
1322         (WebCore::HTMLTreeBuilder::processStartTagForInBody):
1323         (WebCore::HTMLTreeBuilder::processStartTagForInTable):
1324         (WebCore::HTMLTreeBuilder::processStartTag):
1325         (WebCore::HTMLTreeBuilder::callTheAdoptionAgency):
1326         (WebCore::HTMLTreeBuilder::processStartTagForInHead):
1327         (WebCore::HTMLTreeBuilder::processGenericRCDATAStartTag):
1328         (WebCore::HTMLTreeBuilder::processGenericRawTextStartTag):
1329
1330 2010-07-08  Rob Buis  <rwlbuis@gmail.com>
1331
1332         Reviewed by Eric Seidel.
1333
1334         Implement SVGSVGElement.getElementById
1335         https://bugs.webkit.org/show_bug.cgi?id=41655
1336
1337         Implement getElementById for SVGSVGElement by trying to
1338         reuse Document.getElementById. If that fails to find an
1339         SVG element in the document fragent do a subtree search.
1340
1341         Test: svg/custom/svg-getelementid.xhtml
1342
1343         * svg/SVGSVGElement.cpp:
1344         (WebCore::SVGSVGElement::getElementById):
1345         * svg/SVGSVGElement.h:
1346         * svg/SVGSVGElement.idl:
1347
1348 2010-07-08  Eric Seidel  <eric@webkit.org>
1349
1350         Reviewed by Adam Barth.
1351
1352         Unwrap a few || blocks for easier readability
1353         https://bugs.webkit.org/show_bug.cgi?id=41838
1354
1355         No functional change, thus no tests.
1356
1357         * html/HTMLTreeBuilder.cpp:
1358         (WebCore::HTMLTreeBuilder::adjustedLexerState):
1359         (WebCore::HTMLTreeBuilder::processStartTagForInBody):
1360         (WebCore::HTMLTreeBuilder::processStartTagForInTable):
1361         (WebCore::HTMLTreeBuilder::processStartTag):
1362         (WebCore::HTMLTreeBuilder::processEndTagForInBody):
1363         (WebCore::HTMLTreeBuilder::processEndTag):
1364         (WebCore::HTMLTreeBuilder::processComment):
1365         (WebCore::HTMLTreeBuilder::processStartTagForInHead):
1366
1367 2010-07-08  Sam Magnuson  <smagnuson@netflix.com>
1368
1369         Reviewed by Simon Hausmann.
1370
1371         [Qt] for debugging purposes nice I'm contributing back my FPS
1372         counter in the AnimationQtBase
1373         https://bugs.webkit.org/show_bug.cgi?id=40381
1374
1375         Simply start a timer and count frames in the AnimationQtBase and
1376         spit out the FPS count at the end of a single animation.
1377
1378         No new tests: this is a simple debugging aid.
1379
1380         * platform/graphics/qt/GraphicsLayerQt.cpp:
1381         (WebCore::AnimationQt::updateState):
1382         (WebCore::AnimationQt::updateCurrentTime):
1383         (WebCore::TransformAnimationQt::updateState):
1384         (WebCore::OpacityAnimationQt::updateState):
1385
1386 2010-07-08  Sam Magnuson  <smagnuson@netflix.com>
1387
1388         Reviewed by Simon Hausmann.
1389
1390         [Qt] instance objects created for QObjects are somtimes GC'd
1391         https://bugs.webkit.org/show_bug.cgi?id=40352
1392
1393         In markAggregate loop over all the current fields and for any that
1394         still have a dynamic property or a child, mark them as
1395         well. Otherwise the proxy instance will go away and the JS that
1396         was bound to it will be lost.
1397
1398         * bridge/qt/qt_instance.cpp:
1399         (JSC::Bindings::QtInstance::getQtInstance):
1400         (JSC::Bindings::QtInstance::removeCachedMethod):
1401         (JSC::Bindings::QtInstance::markAggregate):
1402         (JSC::Bindings::QtInstance::getPropertyNames):
1403
1404 2010-07-08  Ilya Tikhonovsky  <loislo@chromium.org>
1405
1406         Reviewed by Pavel Feldman.
1407
1408         WebInspector: migrate InspectorDOMAgent to the generated version of
1409         InspectorFrontend interface. This is the next step on the way to
1410         Remote Debugging.
1411
1412         * bindings/js/ScriptEventListener.cpp:
1413         (WebCore::eventListenerHandlerBody):
1414         (WebCore::eventListenerHandlerLocation):
1415         * bindings/js/ScriptEventListener.h:
1416         * bindings/v8/ScriptEventListener.cpp:
1417         (WebCore::eventListenerHandlerBody):
1418         (WebCore::eventListenerHandlerLocation):
1419         * bindings/v8/ScriptEventListener.h:
1420         * inspector/CodeGeneratorInspector.pm:
1421         * inspector/InspectorCSSStore.cpp:
1422         (WebCore::InspectorCSSStore::inspectorStyleSheet):
1423         * inspector/InspectorController.cpp:
1424         (WebCore::InspectorController::connectFrontend):
1425         * inspector/InspectorController.h:
1426         (WebCore::InspectorController::inspectorFrontend2):
1427         * inspector/InspectorDOMAgent.cpp:
1428         (WebCore::InspectorDOMAgent::InspectorDOMAgent):
1429         (WebCore::InspectorDOMAgent::setDocument):
1430         (WebCore::InspectorDOMAgent::handleEvent):
1431         (WebCore::InspectorDOMAgent::pushChildNodesToFrontend):
1432         (WebCore::InspectorDOMAgent::getEventListenersForNode):
1433         (WebCore::InspectorDOMAgent::buildObjectForNode):
1434         (WebCore::InspectorDOMAgent::buildArrayForElementAttributes):
1435         (WebCore::InspectorDOMAgent::buildArrayForContainerChildren):
1436         (WebCore::InspectorDOMAgent::buildObjectForEventListener):
1437         (WebCore::InspectorDOMAgent::didInsertDOMNode):
1438         (WebCore::InspectorDOMAgent::getStyles):
1439         (WebCore::InspectorDOMAgent::getAllStyles):
1440         (WebCore::InspectorDOMAgent::getStyleSheet):
1441         (WebCore::InspectorDOMAgent::getRuleRangesForStyleSheetId):
1442         (WebCore::InspectorDOMAgent::getInlineStyle):
1443         (WebCore::InspectorDOMAgent::getComputedStyle):
1444         (WebCore::InspectorDOMAgent::buildObjectForAttributeStyles):
1445         (WebCore::InspectorDOMAgent::buildArrayForCSSRules):
1446         (WebCore::InspectorDOMAgent::buildArrayForPseudoElements):
1447         (WebCore::InspectorDOMAgent::applyStyleText):
1448         (WebCore::InspectorDOMAgent::toggleStyleEnabled):
1449         (WebCore::InspectorDOMAgent::setRuleSelector):
1450         (WebCore::InspectorDOMAgent::addRule):
1451         (WebCore::InspectorDOMAgent::buildObjectForStyle):
1452         (WebCore::InspectorDOMAgent::populateObjectWithStyleProperties):
1453         (WebCore::InspectorDOMAgent::buildArrayForDisabledStyleProperties):
1454         (WebCore::InspectorDOMAgent::buildObjectForStyleSheet):
1455         (WebCore::InspectorDOMAgent::buildObjectForRule):
1456         (WebCore::InspectorDOMAgent::toArray):
1457         (WebCore::InspectorDOMAgent::reportNodesAsSearchResults):
1458         * inspector/InspectorDOMAgent.h:
1459         (WebCore::InspectorDOMAgent::create):
1460         * inspector/InspectorFrontend.cpp:
1461         * inspector/InspectorFrontend.h:
1462         * inspector/InspectorFrontend2.idl:
1463         * inspector/front-end/inspector.js:
1464         (WebInspector.dispatchMessageFromBackend):
1465
1466 2010-07-08  Simon Fraser  <simon.fraser@apple.com>
1467
1468         Reviewed by Darin Adler.
1469
1470         compositing/iframes/iframe-resize.html displays incorrectly after the resize
1471         https://bugs.webkit.org/show_bug.cgi?id=41794
1472         
1473         The clip and scroll layers of a composited iframe's RenderLayerCompositor are updated from
1474         from updateGraphicsLayerGeometry(), but this is too early to get the correct layoutWidth and
1475         layoutHeight from the FrameView which happen later in layout. So when a widget size changes,
1476         call updateAfterWidgetResize() directly on the RenderLayerBacking (if any).
1477
1478         * rendering/RenderLayerBacking.cpp:
1479         (WebCore::RenderLayerBacking::updateAfterWidgetResize): New method that updates the clip
1480         and scroll layers of the iframe's content RenderLayerCompositor.
1481         (WebCore::RenderLayerBacking::updateGraphicsLayerGeometry): Call updateAfterWidgetResize()
1482         * rendering/RenderLayerBacking.h: Add updateAfterWidgetResize().
1483         * rendering/RenderWidget.cpp:
1484         (WebCore::RenderWidget::setWidgetGeometry): Call updateAfterWidgetResize().
1485
1486 2010-07-08  Simon Fraser  <simon.fraser@apple.com>
1487
1488         Reviewed by Dan Bernstein.
1489
1490         Calling layoutTestController.layerTreeAsText() should update layout
1491         https://bugs.webkit.org/show_bug.cgi?id=41818
1492
1493         Call updateLayout() on the document before obtaining the layer tree.
1494         This will update compositing layers.
1495         
1496         * page/Frame.cpp:
1497         (WebCore::Frame::layerTreeAsText):
1498
1499 2010-07-08  Adam Barth  <abarth@webkit.org>
1500
1501         Reviewed by Eric Seidel.
1502
1503         Get my head wrapped around processCharacter
1504         https://bugs.webkit.org/show_bug.cgi?id=41812
1505
1506         The bulk of the patch is just stubbing out functions to document what
1507         the spec tells us to do.  I'll implement these functions in subsequent
1508         patches.
1509
1510         * html/HTMLConstructionSite.h:
1511         (WebCore::HTMLConstructionSite::insertLeadingWhitespace):
1512         (WebCore::HTMLConstructionSite::insertLeadingWhitespaceWithActiveFormattingElements):
1513         * html/HTMLTreeBuilder.cpp:
1514         (WebCore::HTMLTreeBuilder::processCharacter):
1515         * html/HTMLTreeBuilder.h:
1516         (WebCore::HTMLTreeBuilder::skipLeadingWhitespace):
1517
1518 2010-07-08  Tony Chang  <tony@chromium.org>
1519
1520         Reviewed by Ojan Vafai.
1521
1522         WebCore::InsertLineBreakCommand::shouldUseBreakElement ReadAV@NULL
1523         https://bugs.webkit.org/show_bug.cgi?id=30116
1524         Fixing the crash causes text insertions on hidden elements to get ignored
1525         (not a new bug).  This is https://bugs.webkit.org/show_bug.cgi?id=40342
1526
1527         Test: editing/inserting/return-key-in-hidden-textarea.html
1528
1529         * editing/InsertLineBreakCommand.cpp:
1530         (WebCore::InsertLineBreakCommand::doApply):
1531         * editing/InsertParagraphSeparatorCommand.cpp:
1532         (WebCore::InsertParagraphSeparatorCommand::doApply):
1533
1534 2010-07-08  Sam Weinig  <sam@webkit.org>
1535
1536         Reviewed by Oliver Hunt.
1537
1538         Fix for https://bugs.webkit.org/show_bug.cgi?id=41923
1539         TimeRanges should expose a JS constructor
1540
1541         * html/TimeRanges.idl: Remove [OmitConstructor].
1542         * page/DOMWindow.idl: Add constructor.
1543
1544 2010-07-08  Erik Arvidsson  <arv@chromium.org>
1545
1546         Reviewed by Ojan Vafai.
1547
1548         Add missing padding-end and margin-end CSS properties.
1549         https://bugs.webkit.org/show_bug.cgi?id=25761
1550
1551         Tests: fast/css/margin-start-end.html
1552                fast/css/padding-start-end.html
1553
1554         * css/CSSComputedStyleDeclaration.cpp:
1555         (WebCore::CSSComputedStyleDeclaration::getPropertyCSSValue):
1556         * css/CSSParser.cpp:
1557         (WebCore::CSSParser::parseValue):
1558         * css/CSSPropertyNames.in:
1559         * css/CSSStyleSelector.cpp:
1560         (WebCore::CSSStyleSelector::applyProperty):
1561         * inspector/front-end/SourceCSSTokenizer.js:
1562         (WebInspector.SourceCSSTokenizer):
1563
1564 2010-07-08  Aaron Boodman  <aa@chromium.org>
1565
1566         Reviewed by Timothy Hatcher.
1567
1568         Add the ability for user scripts and user styles to affect just the top frame.
1569
1570         https://bugs.webkit.org/show_bug.cgi?id=41529
1571
1572         Tests: userscripts/user-script-all-frames.html
1573                userscripts/user-script-top-frame-only.html
1574                userscripts/user-style-all-frames.html
1575                userscripts/user-style-top-frame-only.html
1576
1577         * WebCore.base.exp: Update PageGroup method signatures.
1578         * dom/Document.cpp:
1579         (WebCore::Document::pageGroupUserSheets): Check allFrames before injecting.
1580         * page/Frame.cpp:
1581         (WebCore::Frame::injectUserScriptsForWorld): Ditto.
1582         * page/PageGroup.cpp:
1583         (WebCore::PageGroup::addUserScriptToWorld):
1584         (WebCore::PageGroup::addUserStyleSheetToWorld):
1585         * page/PageGroup.h:
1586         * page/UserScript.h:
1587         (WebCore::UserScript::UserScript):
1588         (WebCore::UserScript::injectedFrames):
1589         * page/UserScriptTypes.h:
1590         (WebCore::):
1591         * page/UserStyleSheet.h:
1592         (WebCore::UserStyleSheet::UserStyleSheet):
1593         (WebCore::UserStyleSheet::injectedFrames):
1594
1595 2010-07-08  Adele Peterson  <adele@apple.com>
1596
1597         Reviewed by Jon Honeycutt, Adam Roben, and Darin Adler.
1598
1599         Fix for https://bugs.webkit.org/show_bug.cgi?id=41721
1600         <rdar://problem/8158561> Missing plug-in indicator should have a pressed state
1601
1602         Test: plugins/clicking-missing-plugin-fires-delegate.html
1603
1604         * html/HTMLPlugInElement.cpp:
1605         (WebCore::HTMLPlugInElement::HTMLPlugInElement): Initialize m_isCapturingMouseEvents.
1606         (WebCore::HTMLPlugInElement::detach): If we're still capturing when getting detached, clear the capturing node on the EventHandler.
1607         (WebCore::HTMLPlugInElement::defaultEventHandler): Call handleMissingPluginIndicatorEvent when the missing plugin indicator is showing.
1608         * html/HTMLPlugInElement.h:
1609         (WebCore::HTMLPlugInElement::isCapturingMouseEvents):
1610         (WebCore::HTMLPlugInElement::setIsCapturingMouseEvents):
1611         * page/ChromeClient.h:
1612         (WebCore::ChromeClient::shouldMissingPluginMessageBeButton): Added default implementation.
1613         * rendering/RenderEmbeddedObject.cpp:
1614         (WebCore::replacementTextRoundedRectPressedColor):
1615         (WebCore::RenderEmbeddedObject::RenderEmbeddedObject):
1616         (WebCore::RenderEmbeddedObject::setMissingPluginIndicatorIsPressed): Added.  Causes a repaint when the state changes.
1617         (WebCore::RenderEmbeddedObject::paintReplaced): Call getReplacementTextGeometry.
1618         (WebCore::RenderEmbeddedObject::getReplacementTextGeometry): Factored this out so it can be used in paintReplaced and in isInMissingPluginIndicator.
1619         (WebCore::RenderEmbeddedObject::isInMissingPluginIndicator): Hit test to see if the mouse event is in the missing plugin indicator.
1620         (WebCore::RenderEmbeddedObject::handleMissingPluginIndicatorEvent): Capture mouse events as needed and track the pressed appearance.
1621         * rendering/RenderEmbeddedObject.h:
1622
1623 2010-07-08  James Robinson  <jamesr@google.com>
1624
1625         Reviewed by Darin Fisher.
1626
1627         Allow resizing and getting the texture id from an offscreen GLES2Context
1628         https://bugs.webkit.org/show_bug.cgi?id=41828
1629
1630         When using an offscreen GLES2Context the caller needs to be able to resize the backing store
1631         managed by the embedder and get access to a texture id to pass to the compositor.  WebGL
1632         does these actions in an indirect way, it will be refactored to use this path.
1633
1634         * platform/chromium/GLES2Context.h:
1635
1636 2010-07-02  Ojan Vafai  <ojan@chromium.org>
1637
1638         Reviewed by Adam Barth.
1639
1640         Crash in RenderObject::containingBlock when clearing selection in a display:none node.
1641         https://bugs.webkit.org/show_bug.cgi?id=41523
1642
1643         updateStyleIfNeeded before clearing the selection in the RenderView. Otherwise,
1644         m_selectionStart and m_selectionEnd in RenderView point to garbage object.
1645         This fixes the crash because updateStyleIfNeeded clears the selection before
1646         clobbering nodes that contain the selection.
1647
1648         Test: editing/selection/crash-on-clear-selection.html
1649
1650         * editing/SelectionController.cpp:
1651         (WebCore::SelectionController::updateAppearance):
1652
1653 2010-07-08  Brent Fulgham  <bfulgham@webkit.org>
1654
1655         Reviewed by Xan Lopez.
1656
1657         Correct a mistake in calculating squiggle extents.  The entire
1658         quantity was being divided by two, rather than just the unitWidth.
1659         This error caused the squiggle to be about half the expected length.
1660
1661         * platform/graphics/cairo/DrawErrorUnderline.h:
1662         (drawErrorUnderline):
1663
1664 2010-07-08  John Abd-El-Malek  <jam@chromium.org>
1665
1666         Reviewed by Darin Fisher.
1667
1668         [V8] Navigation policy doesn't play nicely with pepper plugins
1669         https://bugs.webkit.org/show_bug.cgi?id=41864
1670
1671         * bindings/v8/V8Utilities.cpp:
1672         (WebCore::callingOrEnteredFrame):
1673         (WebCore::shouldAllowNavigation):
1674         (WebCore::navigateIfAllowed):
1675
1676 2010-07-08  Jon Honeycutt  <jhoneycutt@apple.com>
1677
1678         Missing plug-ins may cause an assertion failure.
1679         https://bugs.webkit.org/show_bug.cgi?id=41900
1680
1681         Reviewed by Adele Peterson.
1682
1683         Test: plugins/missing-plugin.html
1684
1685         * plugins/PluginView.cpp:
1686         (WebCore::PluginView::~PluginView):
1687         Check whether m_instance is null before trying to remove it from the
1688         map. Trying to remove a null instance results in an assertion failure.
1689         (WebCore::PluginView::PluginView):
1690         Initialize m_instance to 0, because there is an early return that may
1691         skip its being assigned its proper value.
1692
1693 2010-07-08  Darin Adler  <darin@apple.com>
1694
1695         Reviewed by Oliver Hunt.
1696
1697         Fix a test failure seen only on the Leopard bot.
1698
1699         * bindings/js/JSSharedWorkerCustom.cpp:
1700         (WebCore::JSSharedWorkerConstructor::constructJSSharedWorker):
1701         If SharedWorker::create returns an exception, don't try to create a wrapper
1702         for its return value, which can be 0.
1703
1704 2010-07-08  Albert J. Wong  <ajwong@chromium.org>
1705
1706         Not reviewed, build break fix for chromium.
1707
1708         A new MediaControlElementType enum was added which caused a warning in
1709         the Chromium build.
1710
1711         * rendering/RenderMediaControlsChromium.cpp:
1712         (WebCore::RenderMediaControlsChromium::paintMediaControlsPart):
1713             Add MediaVolumeSliderMuteButton enumartion entry.
1714
1715 2010-07-08  Dmitry Titov  <dimich@chromium.org>
1716
1717         Reviewed by David Levin.
1718
1719         Remove IDL declarations and stubs for navigator.registerProtocolHandler and navigator.registerContentHandler.
1720         https://bugs.webkit.org/show_bug.cgi?id=41878
1721
1722         This is practically a rollback of http://trac.webkit.org/changeset/50477.
1723
1724         * page/Chrome.cpp:
1725         * page/Chrome.h:
1726         * page/ChromeClient.h:
1727         * page/Navigator.cpp:
1728         * page/Navigator.h:
1729         * page/Navigator.idl:
1730
1731 2010-07-08  Xan Lopez  <xlopez@igalia.com>
1732
1733         Reviewed by Darin Adler.
1734
1735         Fix adoptRef assertion failures caused by stack-allocated ResourceHandle objects
1736         https://bugs.webkit.org/show_bug.cgi?id=41823
1737
1738         Create the ResourceHandle manually for now instead of calling
1739         ::create, since that methods does a few extra checks that make us
1740         fail a couple of HTTP tests.
1741
1742         * platform/network/soup/ResourceHandleSoup.cpp:
1743         (WebCore::):
1744
1745 2010-07-08  Alexey Proskuryakov  <ap@apple.com>
1746
1747         Reviewed by David Levin.
1748
1749         https://bugs.webkit.org/show_bug.cgi?id=41886
1750         Shorten access control failure explanation messages
1751
1752         * loader/CrossOriginAccessControl.cpp:
1753         (WebCore::passesAccessControlCheck):
1754         * loader/CrossOriginPreflightResultCache.cpp:
1755         (WebCore::CrossOriginPreflightResultCacheItem::allowsCrossOriginMethod):
1756         (WebCore::CrossOriginPreflightResultCacheItem::allowsCrossOriginHeaders):
1757         Removed "response header field" parts from error text - it's long, but doesn't add much
1758         to header field name.
1759
1760         * platform/network/ResourceErrorBase.cpp:
1761         * platform/network/ResourceErrorBase.h:
1762         * xml/XMLHttpRequest.cpp:
1763         (WebCore::XMLHttpRequest::didFail):
1764         * loader/DocumentThreadableLoader.cpp:
1765         (WebCore::DocumentThreadableLoader::DocumentThreadableLoader):
1766         (WebCore::DocumentThreadableLoader::makeSimpleCrossOriginAccessRequest):
1767         (WebCore::DocumentThreadableLoader::didReceiveResponse):
1768         (WebCore::DocumentThreadableLoader::preflightFailure):
1769         Renamed the constant for domain. WebCore shouldn't be creating errors observable by WebKit
1770         clients, because it can't create a platform error, and cannot load a localized string.
1771         We don't treat cross origin access check failures as true loading failures, so we don't tell
1772         clients about these.
1773
1774 2010-07-08  Eric Carlson  <eric.carlson@apple.com>
1775
1776         Reviewed by Dan Bernstein.
1777
1778         Mac OS X media controls should have a way to adjust volume incrementally
1779         https://bugs.webkit.org/show_bug.cgi?id=41718
1780         <rdar://problem/5679472>
1781
1782         * css/CSSPrimitiveValueMappings.h:
1783         (WebCore::CSSPrimitiveValue::CSSPrimitiveValue): Handle MediaVolumeSliderMuteButtonPart.
1784
1785         * css/CSSSelector.cpp: 
1786         (WebCore::CSSSelector::pseudoId): Handle PseudoMediaControlsVolumeSliderMuteButton.
1787         (WebCore::nameToPseudoTypeMap): Define mediaControlsVolumeSliderMuteButton.
1788         (WebCore::CSSSelector::extractPseudoType): Handle PseudoMediaControlsVolumeSliderMuteButton.
1789
1790         * css/CSSSelector.h:
1791         (WebCore::CSSSelector::): Define PseudoMediaControlsVolumeSliderMuteButton.
1792
1793         * css/CSSValueKeywords.in: Define media-volume-slider-mute-button.
1794
1795         * css/mediaControls.css: 
1796         (audio::-webkit-media-controls-volume-slider-mute-button, video::-webkit-media-controls-volume-slider-mute-button): New.
1797
1798         * css/mediaControlsQuickTime.css:
1799         (audio::-webkit-media-controls-panel, video::-webkit-media-controls-panel): Add "overflow: visible" 
1800         so volume slider pop-up will be visible.
1801
1802         (audio::-webkit-media-controls-mute-button, video::-webkit-media-controls-mute-button): Make 
1803         mute button 14x12 instead of 15x14 so it matches the size in the volume slider.
1804
1805         (audio::-webkit-media-controls-volume-slider-container, video::-webkit-media-controls-volume-slider-container):
1806         (audio::-webkit-media-controls-volume-slider, video::-webkit-media-controls-volume-slider):
1807         (audio::-webkit-media-controls-volume-slider-mute-button, video::-webkit-media-controls-volume-slider-mute-button): 
1808         New.
1809
1810         * platform/ThemeTypes.h: Define MediaVolumeSliderMuteButtonPart.
1811
1812         * rendering/MediaControlElements.cpp:
1813         (WebCore::MediaControlInputElement::MediaControlInputElement): Handle MEDIA_CONTROLS_VOLUME_SLIDER_MUTE_BUTTON.
1814         (WebCore::MediaControlMuteButtonElement::MediaControlMuteButtonElement): Add ButtonLocation 
1815         parameter both mute buttons can be created.
1816         (WebCore::MediaControlMuteButtonElement::create): Ditto.
1817
1818         * rendering/MediaControlElements.h: Define MediaVolumeSliderMuteButton and ButtonLocation.
1819
1820         * rendering/RenderMedia.cpp: 
1821         (WebCore::RenderMedia::styleDidChange): Update volume slider mute button.
1822         (WebCore::RenderMedia::createMuteButton): Pass ButtonLocation parameter to MediaControlMuteButtonElement::create.
1823         (WebCore::RenderMedia::createVolumeSliderMuteButton): New.
1824         (WebCore::RenderMedia::updateControls): Deal with volume slider mute button.
1825         (WebCore::RenderMedia::updateVolumeSliderContainer): Don't assume all ports want to position
1826         the volume slider in the same place, call new volumeSliderOffsetFromMuteButton theme function.
1827         (WebCore::RenderMedia::forwardEvent): Handle volume slider mute button.
1828         * rendering/RenderMedia.h:
1829
1830         * rendering/RenderTheme.cpp: 
1831         (WebCore::RenderTheme::paint): Handle MediaVolumeSliderMuteButtonPart.
1832         (WebCore::RenderTheme::volumeSliderOffsetFromMuteButton): New, return location previously hard
1833         coded in RenderMedia::updateVolumeSliderContainer.
1834         * rendering/RenderTheme.h:
1835
1836         * rendering/RenderThemeMac.h:
1837         * rendering/RenderThemeMac.mm:
1838         (WebCore::RenderThemeMac::adjustMediaSliderThumbSize): Deal with the volume slider thumb.
1839         (WebCore::RenderThemeMac::paintMediaVolumeSliderContainer): New, call the WKSI paint function 
1840         for this button.
1841         (WebCore::RenderThemeMac::paintMediaVolumeSliderTrack): Ditto.
1842         (WebCore::RenderThemeMac::paintMediaVolumeSliderThumb): Ditto.
1843         (WebCore::RenderThemeMac::shouldRenderMediaControlPart): Draw the volume slider parts when
1844         using the new controller UI for a movie that has audio.
1845         (WebCore::RenderThemeMac::volumeSliderOffsetFromMuteButton): New, position the volume slider 
1846         directly above the mute button.
1847
1848         * rendering/style/RenderStyleConstants.h: Define MEDIA_CONTROLS_VOLUME_SLIDER_MUTE_BUTTON.
1849
1850 2010-07-08  Antonio Gomes  <tonikitoo@webkit.org>
1851
1852         Unreviewed complementary fix for r62815
1853
1854         One last time change made me blind to a "!" in a if statment.
1855
1856         * editing/EditorCommand.cpp:
1857         (WebCore::executeToggleStyle):
1858
1859 2010-07-07  Ojan Vafai  <ojan@chromium.org>
1860
1861         Reviewed by Darin Adler.
1862
1863         Regression: Selection anchor + focus swap when arrow keys after setBaseAndExtent
1864         https://bugs.webkit.org/show_bug.cgi?id=32605
1865
1866         Only have selections be non-directional when they are set via the mouse.
1867         Otherwise, all selections are directional.
1868
1869         Test: editing/selection/extend-forward-after-set-base-and-extent.html
1870
1871         * WebCore.base.exp:
1872         * editing/SelectionController.cpp:
1873         (WebCore::SelectionController::setSelection):
1874         * editing/SelectionController.h:
1875         (WebCore::SelectionController::setSelection):
1876         * page/EventHandler.cpp:
1877         (WebCore::EventHandler::selectClosestWordFromMouseEvent):
1878         (WebCore::EventHandler::selectClosestWordOrLinkFromMouseEvent):
1879         (WebCore::EventHandler::handleMousePressEventTripleClick):
1880         (WebCore::EventHandler::handleMousePressEventSingleClick):
1881         (WebCore::EventHandler::updateSelectionForMouseDrag):
1882
1883 2010-07-08  Antonio Gomes  <tonikitoo@webkit.org>
1884
1885         Reviewed by Ojan Vafai.
1886
1887         Refactor platform dependent editing behavior code out of Settings (part II)
1888         https://bugs.webkit.org/show_bug.cgi?id=39854
1889
1890         Mac port is currently the only port relying on a style to be present on the start
1891         of a selection to consider the style as applied or not. All other ports
1892         have to have the style present in all text node of the selection for such.
1893
1894         Patch makes situations that depend on this check like this to be controlled by
1895         the EditingBehavior class.
1896
1897         Refactoring, so no new tests needed.
1898
1899         * editing/EditingBehavior.h:
1900         (WebCore::EditingBehavior::shouldToggleStyleBasedOnStartOfSelection):
1901         * editing/EditorCommand.cpp:
1902         (WebCore::executeToggleStyle):
1903
1904 2010-07-08  Anders Carlsson  <andersca@apple.com>
1905
1906         Reviewed by Sam Weinig.
1907
1908         Add stubbed out PluginView class
1909         https://bugs.webkit.org/show_bug.cgi?id=41879
1910
1911         Export functions needed by WebKit2.
1912
1913         * WebCore.exp.in:
1914
1915 2010-07-08  Ben Murdoch  <benm@google.com>
1916
1917         Reviewed by Pavel Feldman.
1918
1919         Fix build break in V8ConsoleCustom.cpp
1920         https://bugs.webkit.org/show_bug.cgi?id=40825
1921
1922         No new tests as just fixing a build break.
1923
1924         * bindings/v8/custom/V8ConsoleCustom.cpp: Guard profilesAccessorGetter
1925             appropriately.
1926
1927 2010-07-08  Sam Weinig  <sam@webkit.org>
1928
1929         Reviewed by Anders Carlsson.
1930
1931         Pass Page to BackForwardControllerClient::createBackForwardList since it
1932         may be called before implementations of BackForwardControllerClient have
1933         access to a Page.
1934
1935         * history/BackForwardController.cpp:
1936         (WebCore::BackForwardController::BackForwardController):
1937         * history/BackForwardControllerClient.h:
1938
1939 2010-07-08  Sheriff Bot  <webkit.review.bot@gmail.com>
1940
1941         Unreviewed, rolling out r62778.
1942         http://trac.webkit.org/changeset/62778
1943         https://bugs.webkit.org/show_bug.cgi?id=41866
1944
1945         Broke svg/W3C-SVG-1.1/text-intro-05-t.svg in debug (Requested
1946         by pfeldman on #webkit).
1947
1948         * platform/graphics/chromium/FontLinux.cpp:
1949         (WebCore::TextRunWalker::TextRunWalker):
1950         (WebCore::TextRunWalker::~TextRunWalker):
1951
1952 2010-07-08  Vitaly Repeshko  <vitalyr@chromium.org>
1953
1954         Reviewed by Pavel Feldman.
1955
1956         Fix adoptRef usage violation in IDBObjectStoreRequest
1957         https://bugs.webkit.org/show_bug.cgi?id=41869
1958
1959         * storage/IDBObjectStoreRequest.cpp:
1960         (WebCore::IDBObjectStoreRequest::IDBObjectStoreRequest):
1961
1962 2010-07-08  Sam Weinig  <sam@webkit.org>
1963
1964         Another chromium build fix.
1965
1966         * history/BackForwardListChromium.cpp:
1967         (WebCore::BackForwardListImpl::goBack):
1968         (WebCore::BackForwardListImpl::goForward):
1969         (WebCore::BackForwardListImpl::backListWithLimit):
1970         (WebCore::BackForwardListImpl::forwardListWithLimit):
1971         (WebCore::BackForwardListImpl::containsItem):
1972         (WebCore::BackForwardListImpl::removeItem):
1973
1974 2010-07-08  Sam Weinig  <sam@webkit.org>
1975
1976         Fix Chromium build.
1977
1978         * WebCore.gyp/WebCore.gyp:
1979
1980 2010-07-08  Sam Weinig  <sam@webkit.org>
1981
1982         Reviewed by Anders Carlsson.
1983
1984         Patch for https://bugs.webkit.org/show_bug.cgi?id=41826
1985         Convert BackForwardList to an abstract base class and add BackForwardListImpl
1986         as the concrete implementation of it.
1987
1988         * CMakeLists.txt:
1989         * GNUmakefile.am:
1990         * WebCore.exp.in:
1991         * WebCore.gypi:
1992         * WebCore.pro:
1993         * WebCore.vcproj/WebCore.vcproj:
1994         * WebCore.xcodeproj/project.pbxproj:
1995         * history/BackForwardController.cpp:
1996         (WebCore::BackForwardController::BackForwardController):
1997         * history/BackForwardController.h:
1998         * history/BackForwardList.cpp: Removed.
1999         * history/BackForwardList.h:
2000         (WebCore::BackForwardList::~BackForwardList):
2001         (WebCore::BackForwardList::isBackForwardListImpl):
2002         (WebCore::BackForwardList::BackForwardList):
2003         * history/BackForwardListChromium.cpp:
2004         * history/BackForwardListImpl.cpp: Copied from WebCore/history/BackForwardList.cpp.
2005         * history/BackForwardListImpl.h: Copied from WebCore/history/BackForwardList.h.
2006         (WebCore::BackForwardListImpl::isBackForwardListImpl):
2007
2008 2010-07-08  Vitaly Repeshko  <vitalyr@chromium.org>
2009
2010         Reviewed by Pavel Feldman.
2011
2012         Fix adoptRef usage violations (mostly in chromium)
2013         https://bugs.webkit.org/show_bug.cgi?id=41863
2014
2015         * bindings/v8/V8DOMWrapper.cpp:
2016         (WebCore::V8DOMWrapper::wrapNativeNodeFilter):
2017         * bindings/v8/V8NodeFilterCondition.h:
2018         (WebCore::V8NodeFilterCondition::create):
2019         * bindings/v8/custom/V8HTMLAllCollectionCustom.cpp:
2020         (WebCore::getNamedItems):
2021         * bindings/v8/custom/V8HTMLCollectionCustom.cpp:
2022         (WebCore::getNamedItems):
2023         * bindings/v8/custom/V8HTMLFormElementCustom.cpp:
2024         (WebCore::V8HTMLFormElement::namedPropertyGetter):
2025         * bindings/v8/custom/V8HTMLSelectElementCustom.cpp:
2026         (WebCore::V8HTMLSelectElement::namedPropertyGetter):
2027         * bindings/v8/custom/V8NamedNodesCollection.h:
2028         (WebCore::V8NamedNodesCollection::create):
2029         (WebCore::V8NamedNodesCollection::V8NamedNodesCollection):
2030         * storage/IDBDatabaseRequest.cpp:
2031         (WebCore::IDBDatabaseRequest::IDBDatabaseRequest):
2032         * storage/IndexedDatabaseRequest.cpp:
2033         (WebCore::IndexedDatabaseRequest::IndexedDatabaseRequest):
2034
2035 2010-07-08  Xan Lopez  <xlopez@igalia.com>
2036
2037         Reviewed by Gustavo Noronha.
2038
2039         Silence a few noisy build rules.
2040
2041         * GNUmakefile.am:
2042
2043 2010-07-08  Andreas Kling  <andreas.kling@nokia.com>
2044
2045         Reviewed by Simon Hausmann.
2046
2047         [Qt] Canvas putImageData() resets painter state
2048         https://bugs.webkit.org/show_bug.cgi?id=41827
2049
2050         Use drawImage() to copy pixels in putImageData() instead of QPixmap::operator=
2051
2052         Test: fast/canvas/canvas-state-intact-after-putImageData.html
2053
2054         * platform/graphics/qt/ImageBufferQt.cpp:
2055         (WebCore::putImageData):
2056
2057 2010-07-08  Andrey Kosyakov  <caseq@chromium.org>
2058
2059         Reviewed by Pavel Feldman.
2060
2061         Web Inspector: Do not invoke shortcuts popup upon bare '?' if it's typed
2062         into an input field.
2063         https://bugs.webkit.org/show_bug.cgi?id=41760
2064
2065         * inspector/front-end/inspector.js:
2066         (WebInspector.documentKeyDown):
2067
2068 2010-07-08  Patrick Gansterer  <paroga@paroga.com>
2069
2070         Reviewed by Kent Tamura.
2071
2072         [WINCE] Buildfix for EventHandler
2073         https://bugs.webkit.org/show_bug.cgi?id=41829
2074
2075         SM_MENUDROPALIGNMENT isn't supported on WinCE.
2076
2077         * page/EventHandler.cpp:
2078         (WebCore::EventHandler::sendContextMenuEventForKey):
2079
2080 2010-07-08  Xiaomei Ji  <xji@chromium.org>
2081
2082         Reviewed by David Levin.
2083
2084         Fix characters with unicode-bidi-mirror property are not correctly
2085         mirrored in Linux.
2086         https://bugs.webkit.org/show_bug.cgi?id=41305
2087
2088         Since harfbuzz does not do mirroring, chromium should iterate each
2089         character in the string and mirror it if needed before passing the
2090         string to harfbuzz for shaping.
2091
2092         Test: fast/text/international/bidi-mirror-he-ar.html
2093
2094         * platform/graphics/chromium/FontLinux.cpp:
2095         (WebCore::TextRunWalker::TextRunWalker):
2096         (WebCore::TextRunWalker::~TextRunWalker):
2097
2098 2010-07-08  Nikolas Zimmermann  <nzimmermann@rim.com>
2099
2100         Reviewed by Rob Buis & Dirk Schulze.
2101
2102         SVG text transformed incorrectly when a transform and gradient applied
2103         https://bugs.webkit.org/show_bug.cgi?id=41563
2104
2105         When rendering filled/stroked text with a gradient, RenderSVGResourceGradient creates a mask image,
2106         renders the text into the image, and then clips the current context against that image buffer,
2107         and filling a rect afterwards. This happened in the wrong coordinate space so far. Don't actually
2108         try to compute the right transformation matrix (which failed), but extract it directly from the
2109         GraphicsContext and apply this transformation to the mask image context.
2110
2111         Fixes pixellation in svg/batik/text/textEffect3.svg and makes the new test svg/custom/text-rotated-gradient.svg pass.
2112         Only affects CoreGraphics platforms.
2113
2114         Test: svg/custom/text-rotated-gradient.svg
2115
2116         * rendering/RenderSVGResourceGradient.cpp:
2117         (WebCore::absoluteTransformFromContext):
2118         (WebCore::createMaskAndSwapContextForTextGradient):
2119         (WebCore::clipToTextMask):
2120
2121 2010-07-07  Alexander Pavlov  <apavlov@chromium.org>
2122
2123         Reviewed by Darin Fisher.
2124
2125         [Chromium] Crash when re-entering message loop
2126         https://bugs.webkit.org/show_bug.cgi?id=41697
2127
2128         A Chromium-specific test case will be submitted into Chromium shortly.
2129
2130         * page/PageGroupLoadDeferrer.cpp:
2131         (WebCore::PageGroupLoadDeferrer::PageGroupLoadDeferrer):
2132         * page/PageGroupLoadDeferrer.h:
2133         (WebCore::PageGroupLoadDeferrer::nextDeferrer):
2134
2135 2010-07-07  Yury Semikhatsky  <yurys@chromium.org>
2136
2137         Reviewed by Pavel Feldman.
2138
2139         Web Inspector: skip breakpoints when script is already paused
2140         https://bugs.webkit.org/show_bug.cgi?id=41768
2141
2142         Test: inspector/debugger-no-nested-pause.html
2143
2144         * bindings/v8/ScriptDebugServer.cpp:
2145         (WebCore::ScriptDebugServer::handleV8DebugEvent):
2146
2147 2010-07-08  Eric Seidel  <eric@webkit.org>
2148
2149         Reviewed by Adam Barth.
2150
2151         WebCore/benchmarks/parser/html-parser.html spends a lot of time in deprecatedParseURL
2152         https://bugs.webkit.org/show_bug.cgi?id=41807
2153
2154         Wow.  This was an awful bug.  We were always taking the slow case
2155         every time we parsed a URL.  This is about a 10% speedup on our
2156         parsing benchmark, and might cause as much as a 1% speedup for Apple's
2157         PLT (even though I can't run that).
2158
2159         We still spend a lot of time in deprecatedParseURL.  We might consider
2160         inlining it if its being kept around much longer.
2161
2162         No behavioral change, just fixing a broken optimization.
2163
2164         * css/CSSHelper.cpp:
2165         (WebCore::deprecatedParseURL):
2166          - We only need to strip characters <= '\r', not >.
2167
2168 2010-07-08  Yury Semikhatsky  <yurys@chromium.org>
2169
2170         Reviewed by Pavel Feldman.
2171
2172         Web Inspector: factor out common part of debugger tests
2173         https://bugs.webkit.org/show_bug.cgi?id=41836
2174
2175         * bindings/js/ScriptDebugServer.cpp:
2176         (WebCore::ScriptDebugServer::recompileAllJSFunctions): postpone script recompilation if JS stack is not empty.
2177
2178 2010-07-08  Patrick Gansterer  <paroga@paroga.com>
2179
2180         Reviewed by Kent Tamura.
2181
2182         Buildfix for !ENABLE(SVG_ANIMATION) after r51567.
2183         https://bugs.webkit.org/show_bug.cgi?id=41803
2184
2185         * svg/SVGDocumentExtensions.cpp:
2186         (WebCore::SVGDocumentExtensions::sampleAnimationAtTime):
2187
2188 2010-07-07  Pavel Podivilov  <podivilov@chromium.org>
2189
2190         Reviewed by Nate Chapin.
2191
2192         [V8] Fix document wrapper memory leak in bindings.
2193         https://bugs.webkit.org/show_bug.cgi?id=41771
2194
2195         * bindings/v8/custom/V8HTMLDocumentCustom.cpp:
2196         (WebCore::V8HTMLDocument::WrapInShadowObject): Do not create static persistent handle
2197         to shadowConstructor because it keeps the first context alive forever.
2198
2199 2010-07-07  Andreas Kling  <andreas.kling@nokia.com>
2200
2201         Reviewed by Tor Arne Vestbø.
2202
2203         [Qt] Enable smooth pixmap transforms by default
2204         https://bugs.webkit.org/show_bug.cgi?id=41774
2205
2206         * platform/graphics/qt/GraphicsContextQt.cpp:
2207         (WebCore::GraphicsContextPlatformPrivate::GraphicsContextPlatformPrivate):
2208         (WebCore::GraphicsContext::setImageInterpolationQuality):
2209
2210 2010-07-07  Jesus Sanchez-Palencia  <jesus@webkit.org>
2211
2212         Reviewed by Antti Koivisto.
2213
2214         [Qt] Missing include to build QtWebKit with WebKit2 in MediaPlayerPrivatePhonon.cpp
2215         https://bugs.webkit.org/show_bug.cgi?id=41767
2216
2217         * platform/graphics/qt/MediaPlayerPrivatePhonon.cpp: Add Logging.h include
2218
2219 2010-07-07  Pavel Podivilov  <podivilov@chromium.org>
2220
2221         Reviewed by Yury Semikhatsky.
2222
2223         Web Inspector: Move v8-related script offset conversion to DebuggerScript.js
2224         https://bugs.webkit.org/show_bug.cgi?id=41755
2225
2226         * bindings/v8/ScriptDebugServer.cpp:
2227         (WebCore::ScriptDebugServer::dispatchDidParseSource):
2228
2229 2010-07-07  Nicolas Weber  <thakis@chromium.org>
2230
2231         Reviewed by Dimitri Glazkov.
2232         https://bugs.webkit.org/show_bug.cgi?id=41580
2233
2234         Fix rendering of radial gradients in skia if both points of the
2235         gradient are the same and r0 > 0.
2236
2237         Test: fast/gradients/radial-centered.html
2238
2239         * platform/graphics/skia/GradientSkia.cpp:
2240         (WebCore::Gradient::platformGradient):
2241
2242 2010-06-18  MORITA Hajime  <morrita@google.com>
2243
2244         Reviewed by Ojan Vafai.
2245
2246         https://bugs.webkit.org/show_bug.cgi?id=26526
2247         Add support for input events (oninput) to contentEditable elements
2248         
2249         Made a default event handler on the Node to dispatch an input event when
2250         webkitEditableContentChanged arrived.
2251         
2252         Test: fast/events/event-input-contentEditable.html
2253
2254         * dom/Node.cpp:
2255         (WebCore::Node::defaultEventHandler):
2256
2257 2010-07-07  Darin Adler  <darin@apple.com>
2258
2259         Reviewed by Anders Carlsson.
2260
2261         Fix adoptRef assertion failures caused by stack-allocated ResourceHandle objects
2262         https://bugs.webkit.org/show_bug.cgi?id=41823
2263
2264         * platform/network/android/ResourceHandleAndroid.cpp:
2265         (WebCore::ResourceHandle::loadResourceSynchronously): Use adoptRef and new instead
2266         of allocating an object on the stack.
2267         * platform/network/curl/ResourceHandleCurl.cpp:
2268         (WebCore::ResourceHandle::loadResourceSynchronously): Ditto.
2269         * platform/network/qt/ResourceHandleQt.cpp:
2270         (WebCore::ResourceHandle::loadResourceSynchronously): Ditto.
2271
2272         * platform/network/soup/ResourceHandleSoup.cpp:
2273         (WebCore::ResourceHandle::loadResourceSynchronously): Use create instead of
2274         allocating an object on the stack.
2275
2276 2010-07-07  Joseph Pecoraro  <joepeck@webkit.org>
2277
2278         Rubber-stamped by Pavel Feldman.
2279
2280         Web Inspector: All RefPtr arguments and return values should be changed to PassRefPtr according to common practice.
2281         https://bugs.webkit.org/show_bug.cgi?id=41759
2282
2283         * inspector/CodeGeneratorInspector.pm:
2284
2285 2010-07-07  Chris Fleizach  <cfleizach@apple.com>
2286
2287         Reviewed by Darin Adler.
2288
2289         AX: TextArea should return AXSelectedTextRange of 0,0 if the cursor is not in the text area
2290         https://bugs.webkit.org/show_bug.cgi?id=41810
2291
2292         Test: platform/mac/accessibility/selected-text-range-for-empty-textarea.html
2293
2294         * accessibility/mac/AccessibilityObjectWrapper.mm:
2295         (-[AccessibilityObjectWrapper accessibilityAttributeValue:]):
2296
2297 2010-07-07  Chris Fleizach  <cfleizach@apple.com>
2298
2299         Reviewed by Beth Dakin.
2300
2301         AX: Data table heuristics: assume data table if at least one row or column of TH cells exist
2302         https://bugs.webkit.org/show_bug.cgi?id=41806
2303
2304         Test: platform/mac/accessibility/table-with-row-col-of-headers.html
2305
2306         * accessibility/AccessibilityTable.cpp:
2307         (WebCore::AccessibilityTable::isTableExposableThroughAccessibility):
2308
2309 2010-07-07  Mark Rowe  <mrowe@apple.com>
2310
2311         Fix the build.
2312
2313         Xcode decided to change the reference type from "Relative to Build Product" to "Relative to Group"
2314         when I moved the file between folders. That's incredibly confusing so I'll file a bug against Xcode.
2315
2316         * WebCore.xcodeproj/project.pbxproj:
2317
2318 2010-07-07  Mark Rowe  <mrowe@apple.com>
2319
2320         Fix the MathML build.
2321
2322         * html/HTMLTreeBuilder.cpp:
2323         (WebCore::HTMLTreeBuilder::resetInsertionModeAppropriately): Fix a typo in a variable name.
2324
2325 2010-07-07  James Hawkins  <jhawkins@chromium.org>
2326
2327         Reviewed by Dimitri Glazkov.
2328
2329         Move setSuggestedValue() and suggestedValue() back to public as it's
2330         now used by WebInputElement.
2331         https://bugs.webkit.org/show_bug.cgi?id=41785
2332
2333         No new tests, as the method is only being used by Chromium's WebKit
2334         API.
2335
2336         * html/HTMLInputElement.h:
2337
2338 2010-07-07  Mark Rowe  <mrowe@apple.com>
2339
2340         Rubber-stamped by Sam Weinig.
2341
2342         Clean up the project file slightly.
2343
2344         Remove obsolete .exp files, add the new .exp.in file, and move ExportFileGenerator.cpp in to the Exports collection.
2345
2346         * WebCore.xcodeproj/project.pbxproj:
2347
2348 2010-07-07  James Robinson  <jamesr@chromium.org> and Vincent Scheib <schieb@chromium.org>
2349
2350         Reviewed by Dimitri Glazkov.
2351
2352         WebCore::GLES2Context should allow creating onscreen and offscreen contexts
2353         https://bugs.webkit.org/show_bug.cgi?id=41492
2354
2355         You should be able to create a GLES2Context for rendering both on screen
2356         and off screen (to a texture).  Chromium's layer renderer (used for compositing)
2357         does the former.  Currently WebGL does the latter by side stepping the
2358         WebCore::GLES2Context API completely.  Longer term it should move over to using this.
2359
2360         * platform/chromium/GLES2Context.h:
2361         * platform/graphics/chromium/LayerRendererChromium.cpp:
2362         (WebCore::LayerRendererChromium::initGL):
2363
2364 2010-07-07  Simon Fraser  <simon.fraser@apple.com>
2365
2366         Reviewed by Dan Bernstein.
2367
2368         Document::ownerElement() should return an HTMLFrameOwnerElement
2369         https://bugs.webkit.org/show_bug.cgi?id=41789
2370
2371         Change return type of RenderLayerCompositor::enclosingIFrameElement() to
2372         HTMLFrameOwnerElement*. No behavioral changes.
2373
2374         * rendering/RenderLayerCompositor.cpp:
2375         (WebCore::RenderLayerCompositor::shouldPropagateCompositingToEnclosingIFrame):
2376         (WebCore::RenderLayerCompositor::enclosingIFrameElement):
2377         * rendering/RenderLayerCompositor.h:
2378
2379 2010-07-07  Adam Barth  <abarth@webkit.org>
2380
2381         Reviewed by Sam Weinig.
2382
2383         Add reverseFind to Vector and deploy in HTML5 parser
2384         https://bugs.webkit.org/show_bug.cgi?id=41778
2385
2386         This patch moves reverseFind from begin an
2387         HTMLFormattingElementList-specific concept to begin a general Vector
2388         concept.  Also, instead of using Entry as the type for operator==, we
2389         now use elements directly.  The old code compiled because the Entry
2390         constructor wasn't explicit, which means we were churning refs on every
2391         comparison!
2392
2393         * html/HTMLFormattingElementList.cpp:
2394         (WebCore::HTMLFormattingElementList::find):
2395         (WebCore::HTMLFormattingElementList::bookmarkFor):
2396         (WebCore::HTMLFormattingElementList::insertAt):
2397         (WebCore::HTMLFormattingElementList::remove):
2398         * html/HTMLFormattingElementList.h:
2399         (WebCore::HTMLFormattingElementList::Entry::operator==):
2400         (WebCore::HTMLFormattingElementList::Entry::operator!=):
2401
2402 2010-07-07  Darin Adler  <darin@apple.com>
2403
2404         Reviewed by Adam Barth.
2405
2406         Turn on adoptRef assertion for RefCounted
2407         https://bugs.webkit.org/show_bug.cgi?id=41547
2408
2409         The WebCore part of this fixes all the assertions I saw in testing.
2410
2411         * html/FileReader.cpp:
2412         (WebCore::FileReader::readAsBinaryString): Added null checks.
2413         Callers from JavaScript can pass the wrong type, which becomes null.
2414         (WebCore::FileReader::readAsText): Ditto.
2415         (WebCore::FileReader::readAsDataURL): Ditto.
2416
2417         * html/FileStreamClient.h: Removed unneeded include.
2418
2419         * html/FileStreamProxy.cpp:
2420         (WebCore::FileStreamProxy::FileStreamProxy): Made inline and moved
2421         some of the code, including the ref, into the create function.
2422         (WebCore::FileStreamProxy::create): Moved some of the code from
2423         the constructor here. It's safe to ref once the object has been
2424         created and adopted.
2425
2426         * html/FileStreamProxy.h: Changed create function to no longer be
2427         inlined. Also removed an unneeded include.
2428
2429         * page/EventSource.cpp:
2430         (WebCore::EventSource::EventSource): Made inline, changed arguments
2431         and moved code that involves the need to ref this object into the
2432         create function. Also moved failure handling out there since it's
2433         cleaner to have a function that fails than a constructor. For
2434         example, the function can return 0.
2435         (WebCore::EventSource::create): Moved some of the code from the
2436         constructor here.
2437
2438         * page/EventSource.h: Removed unneeded includes. Made the
2439         creation function non-inline. Changed the arguments to the constructor.
2440
2441         * storage/StorageAreaSync.cpp:
2442         (WebCore::StorageAreaSync::StorageAreaSync): Made inline. Moved
2443         code that requires ref'ing this object out to the create function.
2444         (WebCore::StorageAreaSync::create): Moved some of the code from the
2445         constructor here.
2446
2447         * storage/StorageAreaSync.h: Removed unneeded includes. Changed
2448         the type of one of the constructor arguments from String to
2449         const String&.
2450
2451         * workers/SharedWorker.cpp:
2452         (WebCore::SharedWorker::SharedWorker): Made inline. Moved most of
2453         the setup code out of here into the create function.
2454         (WebCore::SharedWorker::create): Moved the code here.
2455
2456         * workers/SharedWorker.h: Removed unneeded includes. Made the
2457         create function non-inline. Marked the toSharedWorker override private
2458         to catch people doing an unnecessary virtual function call if they
2459         already have a SharedWorker*.
2460
2461         * workers/Worker.cpp:
2462         (WebCore::Worker::Worker): Made inline. Moved most of the setup code
2463         out of here into the create function.
2464         (WebCore::Worker::create): Moved the code here.
2465
2466         * workers/Worker.h: Made the create function non-inline. Changed
2467         the arguments to the constructor.
2468
2469 2010-07-07  Chris Fleizach  <cfleizach@apple.com>
2470
2471         Reviewed by Beth Dakin.
2472
2473         AX: when a node's role changes, the AX tree might need to be updated
2474         https://bugs.webkit.org/show_bug.cgi?id=41784
2475
2476         Test: platform/mac/accessibility/update-children-when-aria-role-changes.html
2477
2478         * accessibility/AccessibilityRenderObject.cpp:
2479         (WebCore::AccessibilityRenderObject::AccessibilityRenderObject):
2480         (WebCore::AccessibilityRenderObject::updateAccessibilityRole):
2481
2482 2010-07-07  Chris Fleizach  <cfleizach@apple.com>
2483
2484         Reviewed by Beth Dakin.
2485
2486         AX: when an element uses role="text" but no aria-label, it should default to textUnderElement()
2487         https://bugs.webkit.org/show_bug.cgi?id=41780
2488
2489         Test: platform/mac/accessibility/static-text-role-uses-text-under-element.html
2490
2491         * accessibility/AccessibilityRenderObject.cpp:
2492         (WebCore::AccessibilityRenderObject::stringValue):
2493            Fallback to textUnderElement() if text() does not return any explicitly set text.
2494         (WebCore::AccessibilityRenderObject::accessibilityIsIgnored): 
2495            Remove erroneous check to ignore text elements that didn't have a explicit text value.
2496
2497 2010-07-07  Simon Fraser  <simon.fraser@apple.com>
2498
2499         Reviewed by Darin Adler.
2500
2501         Document::ownerElement() should return an HTMLFrameOwnerElement
2502         https://bugs.webkit.org/show_bug.cgi?id=41789
2503
2504         Change the return type of Document::ownerElement() to an HTMLFrameOwnerElement, for better
2505         type checking. No behavioral changes.
2506
2507         * accessibility/mac/AccessibilityObjectWrapper.mm:
2508         * dom/Document.cpp:
2509         (WebCore::Document::ownerElement):
2510         * dom/Document.h:
2511         * editing/SelectionController.cpp:
2512         * rendering/RenderBoxModelObject.cpp:
2513         * rendering/RenderLayer.cpp:
2514         * rendering/RenderView.cpp:
2515
2516 2010-07-07  Simon Fraser  <simon.fraser@apple.com>
2517
2518         Reviewed by Dan Bernstein.
2519
2520         REGRESSION: GMail becomes blank after closing other tabs
2521         https://bugs.webkit.org/show_bug.cgi?id=40421
2522
2523         RenderLayerCompositor's attach/detachRootPlatformLayer methods use
2524         setNeedsStyleRecalc() in order to trigger acclerated compositing layers
2525         to be hooked together across iframe boundaries. However, it was possible
2526         for these to get called while inside of Document::recalcStyle(), which
2527         is bad because it can cause the recalc to fail to get processed.
2528         
2529         Fix this by using the existing queuePostAttachCallback() functionality
2530         to delay the call to setNeedsStyleRecalc() if post-attach callbacks 
2531         are suspended (indicating that we're inside recalcStyle()).
2532         
2533         No new tests because I wasn't able to make a test that shows the problem.
2534
2535         * dom/ContainerNode.h: Make queuePostAttachCallback() public.
2536         * dom/ContainerNode.cpp:
2537         (WebCore::ContainerNode::postAttachCallbacksAreSuspended): Added; returns
2538         whether s_attachDepth is non-zero.
2539
2540         * rendering/RenderLayerCompositor.cpp:
2541         (WebCore::RenderLayerCompositor::attachRootPlatformLayer): Call scheduleNeedsStyleRecalc()
2542         instead of setNeedsStyleRecalc().
2543         (WebCore::RenderLayerCompositor::detachRootPlatformLayer): Ditto.
2544         (WebCore::needsStyleRecalcCallback): Here we call setNeedsStyleRecalc().
2545         (WebCore::RenderLayerCompositor::scheduleNeedsStyleRecalc):
2546         (WebCore::RenderLayerCompositor::notifyIFramesOfCompositingChange):
2547         * rendering/RenderLayerCompositor.h: Add scheduleNeedsStyleRecalc().
2548
2549 2010-07-06  Simon Fraser  <simon.fraser@apple.com>
2550
2551         Reviewed by Dan Bernstein.
2552
2553         Incorrect compositing order with negative z-index
2554         https://bugs.webkit.org/show_bug.cgi?id=38959
2555         
2556         When painting the contents of compositing layers whose phase is "GraphicsLayerPaintBackground"
2557         (indicating that they are used for the background of elements with negative z-index children),
2558         we need to paint the non-composited negative-z-order descendants at the end of the background phase, so
2559         they appear behind composited negative-z-order descendants.
2560
2561         Test: compositing/z-order/negative-z-index.html
2562
2563         * rendering/RenderLayerBacking.cpp:
2564         (WebCore::RenderLayerBacking::paintIntoLayer):
2565
2566 2010-07-06  Simon Fraser  <simon.fraser@apple.com>
2567
2568         Reviewed by Dan Bernstein.
2569
2570         Incorrect compositing order with negative z-index
2571         https://bugs.webkit.org/show_bug.cgi?id=38959
2572
2573         Prerequisite: adopt new paintList() method that was added when painting compositing layer contents.
2574         No behavioral changes.
2575
2576         * rendering/RenderLayerBacking.cpp:
2577         (WebCore::RenderLayerBacking::paintIntoLayer):
2578
2579 2010-07-07  Adam Barth  <abarth@webkit.org>
2580
2581         Unreviewed.
2582
2583         Fix ASSERT in Eric's previous patch.  I think he tested his patch in
2584         Release and not Debug.
2585
2586         * html/HTMLFormattingElementList.h:
2587         (WebCore::HTMLFormattingElementList::Entry::Entry):
2588         (WebCore::HTMLFormattingElementList::findIndex):
2589
2590 2010-07-07  Eric Seidel  <eric@webkit.org>
2591
2592         Reviewed by Adam Barth.
2593
2594         Grease the TreeBuilder's lightning
2595         https://bugs.webkit.org/show_bug.cgi?id=41756
2596
2597         Brings the new TreeBuilder from 7s to 3s on the parser benchmark.
2598         This makes performance comparable to the old parser.
2599
2600         We have not begun to fight!  There is so much fat left on these bones.
2601
2602         * html/HTMLFormattingElementList.cpp:
2603         (WebCore::HTMLFormattingElementList::find):
2604         (WebCore::HTMLFormattingElementList::bookmarkFor):
2605         (WebCore::HTMLFormattingElementList::insertAt):
2606         (WebCore::HTMLFormattingElementList::remove):
2607         * html/HTMLFormattingElementList.h:
2608         (WebCore::HTMLFormattingElementList::findIndex):
2609
2610 2010-07-07  Eric Seidel  <eric@webkit.org>
2611
2612         Reviewed by Adam Barth.
2613
2614         HTMLTreeBuilder is way too slow
2615         https://bugs.webkit.org/show_bug.cgi?id=41754
2616
2617         This takes us from 14s to 7s on our parsing benchmark.
2618         That's still much slower than the old tree builder, but there
2619         is a huge amount of fat left to trim.
2620
2621         Vector<T> wasn't able to inline all the Entry functions when
2622         they were buried in the cpp.  Turns out the active formatting elements
2623         list is very hot.
2624
2625         I'm not sure Vector<T> is going to be the right data structure for us
2626         in the end, but it has done alright for bring-up.
2627
2628         * html/HTMLFormattingElementList.cpp:
2629         * html/HTMLFormattingElementList.h:
2630         (WebCore::HTMLFormattingElementList::Entry::Entry):
2631         (WebCore::HTMLFormattingElementList::Entry::~Entry):
2632         (WebCore::HTMLFormattingElementList::Entry::isMarker):
2633         (WebCore::HTMLFormattingElementList::Entry::element):
2634         (WebCore::HTMLFormattingElementList::Entry::replaceElement):
2635         (WebCore::HTMLFormattingElementList::Entry::operator==):
2636         (WebCore::HTMLFormattingElementList::Entry::operator!=):
2637
2638 2010-07-06  Darin Adler  <darin@apple.com>
2639
2640         Reviewed by Adam Barth.
2641
2642         More OwnPtr work, including making clear set the pointer to 0 before deletion
2643         https://bugs.webkit.org/show_bug.cgi?id=41727
2644
2645         * WebCore.exp.in: Updated.
2646
2647         * css/CSSSelector.h:
2648         (WebCore::CSSSelector::RareData::RareData): Use adoptPtr.
2649         (WebCore::CSSSelector::createRareData): Ditto.
2650         * dom/SpaceSplitString.h:
2651         (WebCore::SpaceSplitString::SpaceSplitString): Ditto.
2652         (WebCore::SpaceSplitString::set): Ditto.
2653         * history/CachedFrame.cpp:
2654         (WebCore::CachedFrame::CachedFrame): Ditto.
2655         (WebCore::CachedFrame::setCachedFramePlatformData): Ditto.
2656
2657         * history/CachedFrame.h: Use PassOwnPtr.
2658
2659         * loader/appcache/ApplicationCacheGroup.cpp:
2660         (WebCore::CallCacheListenerTask::create): Use adoptPtr.
2661         * loader/appcache/ApplicationCacheStorage.cpp:
2662         (WebCore::ApplicationCacheStorage::storeCopyOfCache): Ditto.
2663         * platform/PurgeableBuffer.h:
2664         (WebCore::PurgeableBuffer::create): Ditto.
2665         * platform/graphics/GlyphMetricsMap.h:
2666         (WebCore::::locatePageSlowCase): Ditto.
2667         * platform/graphics/GraphicsLayer.h:
2668         (WebCore::AnimationValue::AnimationValue): Ditto.
2669         (WebCore::TransformAnimationValue::TransformAnimationValue): Ditto.
2670         * platform/graphics/MediaPlayer.h:
2671         (WebCore::MediaPlayer::create): Ditto.
2672         * platform/graphics/SimpleFontData.h:
2673         (WebCore::SimpleFontData::boundsForGlyph): Ditto.
2674         * platform/mac/PurgeableBufferMac.cpp:
2675         (WebCore::PurgeableBuffer::create): Ditto.
2676         * rendering/InlineFlowBox.h:
2677         (WebCore::InlineFlowBox::setHorizontalOverflowPositions): Ditto.
2678         (WebCore::InlineFlowBox::setVerticalOverflowPositions): Ditto.
2679         * rendering/RootInlineBox.h:
2680         (WebCore::RootInlineBox::floats): Ditto.
2681         * rendering/style/RenderStyle.h:
2682         (WebCore::InheritedFlags::inheritAnimations): Ditto.
2683         (WebCore::InheritedFlags::inheritTransitions): Ditto.
2684
2685         * rendering/style/SVGRenderStyleDefs.h: Use PassOwnPtr.
2686
2687 2010-07-07  Sam Weinig  <sam@webkit.org>
2688
2689         Reviewed by Anders Carlsson.
2690
2691         Patch for https://bugs.webkit.org/show_bug.cgi?id=41772
2692         Add basic piping for BackForwardControllerClient.
2693
2694         * WebCore.exp.in:
2695         * WebCore.xcodeproj/project.pbxproj:
2696         * history/BackForwardController.cpp:
2697         (WebCore::BackForwardController::BackForwardController):
2698         * history/BackForwardControllerClient.h:
2699         * page/Page.cpp:
2700         (WebCore::Page::Page):
2701         * page/Page.h:
2702         * svg/graphics/SVGImage.cpp:
2703         (WebCore::SVGImage::dataChanged):
2704
2705 2010-07-07  Lucas De Marchi  <lucas.demarchi@profusion.mobi>
2706
2707         Unreviewed build fix after r62577.
2708
2709         [EFL] build fix after r62577
2710         https://bugs.webkit.org/show_bug.cgi?id=41764
2711
2712         No new functionality so no new tests.
2713
2714 2010-07-07  Gustavo Noronha Silva  <gustavo.noronha@collabora.co.uk>
2715
2716         Reviewed by Xan Lopez.
2717
2718         [GTK] Scrollbars sometimes go dead and stop scrolling the view
2719         https://bugs.webkit.org/show_bug.cgi?id=41711
2720
2721         Rework the setGtkAdjustment function slightly, make it have an
2722         early return, and be more readable.
2723
2724         * platform/gtk/ScrollViewGtk.cpp:
2725         (WebCore::ScrollView::setGtkAdjustments):
2726
2727 2010-07-07  Kristian Monsen  <kristianm@google.com>
2728
2729         Reviewed by Steve Block.
2730
2731         Updating Android to use the UserGestureIndicator class instead of Android specific code.
2732         The rest of the changes are in parts of the Android port which are yet to be upstreamed.
2733
2734         No change in behavior, so no new test.
2735
2736         * platform/network/android/ResourceRequest.h:
2737         (WebCore::ResourceRequest::ResourceRequest):
2738         (WebCore::ResourceRequest::doUpdateResourceRequest):
2739
2740 2010-07-07  Andy Estes  <aestes@apple.com>
2741
2742         Reviewed by Adam Barth.
2743
2744         Allow a beforeload listener to prevent loading of images in <object> tags.
2745         https://bugs.webkit.org/show_bug.cgi?id=41027
2746         <rdar://problem/8120596>
2747
2748         Tests: fast/dom/beforeload/image-object-before-load-innerHTML.html
2749                fast/dom/beforeload/image-object-before-load.html
2750
2751         * html/HTMLObjectElement.cpp:
2752         (WebCore::HTMLObjectElement::attach): Do not call
2753         RenderImage::setCachedImage() at the end of attach(). Instead, allow
2754         this to happen conditionally after beforeload is dispatched.
2755         * loader/ImageLoader.cpp:
2756         (WebCore::ImageLoader::dispatchPendingBeforeLoadEvent): Render fallback
2757         content if an object's load was cancelled.
2758
2759 2010-07-06  Nikolas Zimmermann  <nzimmermann@rim.com>
2760
2761         Reviewed by Dirk Schulze.
2762
2763         <use> on <font-face> causes crashes, if SVGUseElement gets detached
2764         https://bugs.webkit.org/show_bug.cgi?id=41621
2765
2766         Do not call removeFromMappedElementSheet() from the SVGFontFaceElement destructor,
2767         as that can potentially cause the element to be reattached while destructing.
2768
2769         In order to fix the crash in the testcase, the order of calling the base-class detach
2770         method in SVGUseElement and the instance/shadow tree destruction has to be reversed,
2771         matching the order in removedFromDocument().
2772
2773         Test: svg/custom/use-font-face-crash.svg
2774
2775         * svg/SVGFontFaceElement.cpp:
2776         (WebCore::SVGFontFaceElement::~SVGFontFaceElement): Remove removeFromMappedElementSheet() call.
2777         * svg/SVGUseElement.cpp:
2778         (WebCore::SVGUseElement::detach): Reverse order of calling base-class detach method and instance/shadow tree destruction.
2779
2780 2010-07-07  Steve Block  <steveblock@google.com>
2781
2782         Reviewed by Adam Barth.
2783
2784         document.createEvent() should support DeviceOrientationEvent
2785         https://bugs.webkit.org/show_bug.cgi?id=41618
2786
2787         Tests: fast/dom/DeviceOrientation/create-event.html
2788
2789         * dom/Document.cpp:
2790         (WebCore::Document::createEvent):
2791
2792 2010-07-07  Yury Semikhatsky  <yurys@chromium.org>
2793
2794         Reviewed by Pavel Feldman.
2795
2796         Web Inspector: show functions from internal browser scripts on the call stack if they
2797         are there.
2798         https://bugs.webkit.org/show_bug.cgi?id=41762
2799
2800         * English.lproj/localizedStrings.js:
2801         * inspector/front-end/CallStackSidebarPane.js:
2802         (WebInspector.CallStackSidebarPane.prototype.update):
2803
2804 2010-07-07  Ilya Tikhonovsky  <loislo@chromium.org>
2805
2806         Reviewed by Pavel Feldman.
2807
2808         Web Inspector: All RefPtr arguments and return values should be
2809         changed to PassRefPtr whenever it is possible according to common practice.
2810         https://bugs.webkit.org/show_bug.cgi?id=41759
2811
2812         * inspector/CodeGeneratorInspector.pm:
2813         * inspector/InspectorTimelineAgent.cpp:
2814         (WebCore::InspectorTimelineAgent::pushGCEventRecords):
2815         (WebCore::InspectorTimelineAgent::didInstallTimer):
2816         (WebCore::InspectorTimelineAgent::didRemoveTimer):
2817         (WebCore::InspectorTimelineAgent::didScheduleResourceRequest):
2818         (WebCore::InspectorTimelineAgent::willSendResourceRequest):
2819         (WebCore::InspectorTimelineAgent::didFinishLoadingResource):
2820         (WebCore::InspectorTimelineAgent::didMarkTimeline):
2821         (WebCore::InspectorTimelineAgent::didMarkDOMContentEvent):
2822         (WebCore::InspectorTimelineAgent::didMarkLoadEvent):
2823         (WebCore::InspectorTimelineAgent::addRecordToTimeline):
2824         (WebCore::InspectorTimelineAgent::setHeapSizeStatistic):
2825         (WebCore::InspectorTimelineAgent::pushCurrentRecord):
2826         * inspector/InspectorTimelineAgent.h:
2827         (WebCore::InspectorTimelineAgent::TimelineRecordEntry::TimelineRecordEntry):
2828         * inspector/TimelineRecordFactory.cpp:
2829         (WebCore::TimelineRecordFactory::createGenericRecord):
2830         (WebCore::TimelineRecordFactory::createGCEventData):
2831         (WebCore::TimelineRecordFactory::createFunctionCallData):
2832         (WebCore::TimelineRecordFactory::createEventDispatchData):
2833         (WebCore::TimelineRecordFactory::createGenericTimerData):
2834         (WebCore::TimelineRecordFactory::createTimerInstallData):
2835         (WebCore::TimelineRecordFactory::createXHRReadyStateChangeData):
2836         (WebCore::TimelineRecordFactory::createXHRLoadData):
2837         (WebCore::TimelineRecordFactory::createEvaluateScriptData):
2838         (WebCore::TimelineRecordFactory::createMarkTimelineData):
2839         (WebCore::TimelineRecordFactory::createScheduleResourceRequestData):
2840         (WebCore::TimelineRecordFactory::createResourceSendRequestData):
2841         (WebCore::TimelineRecordFactory::createResourceReceiveResponseData):
2842         (WebCore::TimelineRecordFactory::createResourceFinishData):
2843         (WebCore::TimelineRecordFactory::createReceiveResourceData):
2844         (WebCore::TimelineRecordFactory::createPaintData):
2845         (WebCore::TimelineRecordFactory::createParseHTMLData):
2846         * inspector/TimelineRecordFactory.h:
2847
2848 2010-07-07  Antti Koivisto  <koivisto@iki.fi>
2849
2850         Reviewed by Kenneth Rohde Christiansen.
2851         
2852         [Qt] Initial WebKit2 implementation
2853         https://bugs.webkit.org/show_bug.cgi?id=40233
2854
2855         Export some event conversion functions needed for WebKit2.
2856
2857         * platform/PlatformKeyboardEvent.h:
2858         * platform/qt/PlatformKeyboardEventQt.cpp:
2859         (WebCore::keyIdentifierForQtKeyCode):
2860         (WebCore::windowsKeyCodeForKeyEvent):
2861
2862 2010-07-06  Steve Block  <steveblock@google.com>
2863
2864         Reviewed by Adam Barth.
2865
2866         DeviceOrientationEventConstructor should be exposed at window.DeviceOrientationEvent
2867         https://bugs.webkit.org/show_bug.cgi?id=41685
2868
2869         Test: fast/dom/Window/window-properties-device-orientation.html
2870               fast/dom/DeviceOrientation/window-property.txt
2871
2872         * page/DOMWindow.idl:
2873
2874 2010-07-07  Sheriff Bot  <webkit.review.bot@gmail.com>
2875
2876         Unreviewed, rolling out r62645.
2877         http://trac.webkit.org/changeset/62645
2878         https://bugs.webkit.org/show_bug.cgi?id=41757
2879
2880         "Broken inspector tests on several platforms" (Requested by
2881         eseidel on #webkit).
2882
2883         * inspector/front-end/ScriptView.js:
2884         (WebInspector.ScriptView.prototype._addBreakpoint):
2885         * inspector/front-end/ScriptsPanel.js:
2886         (WebInspector.ScriptsPanel):
2887         (WebInspector.ScriptsPanel.prototype._resourceLoadingFinished):
2888         (WebInspector.ScriptsPanel.prototype._breakpointAdded):
2889         (WebInspector.ScriptsPanel.prototype._scriptOrResourceForURLAndLine):
2890         (WebInspector.ScriptsPanel.prototype._showScriptOrResource):
2891         (WebInspector.ScriptsPanel.prototype._toggleBreakpointsClicked):
2892         * inspector/front-end/SourceView.js:
2893         (WebInspector.SourceView.prototype._addBreakpoint):
2894
2895 2010-07-07  Steve Block  <steveblock@google.com>
2896
2897         Reviewed by Adam Barth.
2898
2899         JavaInstanceV8 needs to push a local reference frame to avoid table overflow.
2900         https://bugs.webkit.org/show_bug.cgi?id=41516
2901
2902         This change is required to help prevent the Java virtual machine from
2903         running out of local references. The Java virtual machine supports only
2904         a limited number of local references. Normally, local references are
2905         cleared when the native method returns.
2906
2907         This change adds calls to PushLocalFrame() and PopLocalFrame() around
2908         each invocation of a method on JavaInstance. These calls instruct the
2909         virtual machine to clear local references when the frame is popped.
2910         This means that in the case where many calls to JavaInstance methods
2911         are made within the same native call, local references are cleared as
2912         soon as they are no longer needed, thus conserving references.
2913
2914         This change mirrors exactly JavaInstanceJSC.
2915
2916         No new tests.
2917
2918         * bridge/jni/v8/JavaInstanceV8.cpp:
2919         (JavaInstance::virtualBegin):
2920         (JavaInstance::virtualEnd):
2921         * bridge/jni/v8/JavaInstanceV8.h:
2922
2923 2010-07-06  Yury Semikhatsky  <yurys@chromium.org>
2924
2925         Reviewed by Pavel Feldman.
2926
2927         Inspector should remember the size of sidebars set by the user
2928         https://bugs.webkit.org/show_bug.cgi?id=19207
2929
2930         * inspector/front-end/AbstractTimelinePanel.js:
2931         (WebInspector.AbstractTimelinePanel):
2932         * inspector/front-end/AuditsPanel.js:
2933         (WebInspector.AuditsPanel):
2934         * inspector/front-end/ConsolePanel.js:
2935         (WebInspector.ConsolePanel):
2936         * inspector/front-end/ElementsPanel.js:
2937         (WebInspector.ElementsPanel.prototype.rightSidebarResizerDragEnd):
2938         * inspector/front-end/Panel.js:
2939         (WebInspector.Panel):
2940         (WebInspector.Panel.prototype.get toolbarItemClass):
2941         (WebInspector.Panel.prototype.show):
2942         (WebInspector.Panel.prototype.createSidebar):
2943         (WebInspector.Panel.prototype.get _sidebarWidthSettingName):
2944         (WebInspector.Panel.prototype._endSidebarDragging):
2945         (WebInspector.Panel.prototype.updateSidebarWidth):
2946         (WebInspector.Panel.prototype.restoreSidebarWidth):
2947         (WebInspector.Panel.prototype.saveSidebarWidth):
2948         * inspector/front-end/ProfilesPanel.js:
2949         (WebInspector.ProfilesPanel):
2950         * inspector/front-end/ResourcesPanel.js:
2951         (WebInspector.ResourcesPanel):
2952         (WebInspector.ResourcesPanel.prototype.resourceTrackingWasEnabled):
2953         * inspector/front-end/ScriptsPanel.js:
2954         (WebInspector.ScriptsPanel):
2955         (WebInspector.ScriptsPanel.prototype._endSidebarResizeDrag):
2956         (WebInspector.ScriptsPanel.prototype._sidebarResizeDrag):
2957         (WebInspector.ScriptsPanel.prototype.setSidebarWidth):
2958         * inspector/front-end/Settings.js:
2959         (WebInspector.populateApplicationSettings):
2960         * inspector/front-end/StoragePanel.js:
2961         (WebInspector.StoragePanel):
2962         * inspector/front-end/TimelinePanel.js:
2963         (WebInspector.TimelinePanel):
2964
2965 2010-07-05  Steve Block  <steveblock@google.com>
2966
2967         Reviewed by Adam Barth.
2968
2969         DeviceOrientation should be renamed to DeviceOrientationController
2970         https://bugs.webkit.org/show_bug.cgi?id=41608
2971
2972         No new tests, renaming only.
2973
2974         * Android.mk:
2975         * CMakeLists.txt:
2976         * GNUmakefile.am:
2977         * WebCore.gypi:
2978         * WebCore.pro:
2979         * WebCore.vcproj/WebCore.vcproj:
2980         * WebCore.xcodeproj/project.pbxproj:
2981         * dom/DeviceOrientation.cpp: Removed.
2982         * dom/DeviceOrientation.h: Removed.
2983         * dom/DeviceOrientationController.cpp: Copied from WebCore/dom/DeviceOrientation.cpp.
2984         (WebCore::DeviceOrientationController::DeviceOrientationController):
2985         (WebCore::DeviceOrientationController::onDeviceOrientationChange):
2986         * dom/DeviceOrientationController.h: Copied from WebCore/dom/DeviceOrientation.h.
2987         * page/Page.cpp:
2988         (WebCore::Page::Page):
2989         * page/Page.h:
2990         (WebCore::Page::deviceOrientationController):
2991
2992 2010-07-07  Pavel Podivilov  <podivilov@chromium.org>
2993
2994         Reviewed by Yury Semikhatsky.
2995
2996         Web Inspector: do not activate all breakpoints on page reload
2997         https://bugs.webkit.org/show_bug.cgi?id=41461
2998
2999         Test: inspector/debugger-breakpoints-not-activated-on-reload.html
3000
3001         * inspector/front-end/ScriptView.js:
3002         (WebInspector.ScriptView.prototype._addBreakpoint):
3003         * inspector/front-end/ScriptsPanel.js:
3004         (WebInspector.ScriptsPanel):
3005         (WebInspector.ScriptsPanel.prototype._breakpointAdded):
3006         (WebInspector.ScriptsPanel.prototype.toggleBreakpointsClicked):
3007         * inspector/front-end/SourceView.js:
3008         (WebInspector.SourceView.prototype._addBreakpoint):
3009
3010 2010-07-07  Nikolas Zimmermann  <nzimmermann@rim.com>
3011
3012         Not reviewed. Rollout r62633 - test crashes after some last-minute modifications.
3013
3014         * svg/SVGFontFaceElement.cpp:
3015         (WebCore::SVGFontFaceElement::~SVGFontFaceElement):
3016
3017 2010-07-07  Eric Seidel  <eric@webkit.org>
3018
3019         Reviewed by Adam Barth.
3020
3021         Fix </html> handling in before head mode
3022         https://bugs.webkit.org/show_bug.cgi?id=41752
3023
3024         Covered by html5lib/runner.html.
3025
3026         * html/HTMLTreeBuilder.cpp:
3027         (WebCore::HTMLTreeBuilder::processEndTag):
3028
3029 2010-07-07  Eric Seidel  <eric@webkit.org>
3030
3031         Reviewed by Adam Barth.
3032
3033         </div> pops too many <div> elements in body
3034         https://bugs.webkit.org/show_bug.cgi?id=41751
3035
3036         Another missing return!
3037
3038         While tracking this down I added some debugging code to both the
3039         open elements stack and the active formatting elements list.
3040
3041         I also unwrapped a very long || chain to make it more readable.
3042
3043         * html/HTMLElementStack.cpp:
3044         (WebCore::HTMLElementStack::show):
3045         * html/HTMLElementStack.h:
3046         * html/HTMLFormattingElementList.cpp:
3047         (WebCore::HTMLFormattingElementList::show):
3048         * html/HTMLFormattingElementList.h:
3049         * html/HTMLTreeBuilder.cpp:
3050         (WebCore::HTMLTreeBuilder::processEndTagForInBody):
3051
3052 2010-07-06  Eric Seidel  <eric@webkit.org>
3053
3054         Reviewed by Adam Barth.
3055
3056         HTMLTreeBuilder needs to handle <table><input>
3057         https://bugs.webkit.org/show_bug.cgi?id=41744
3058
3059         * html/HTMLTreeBuilder.cpp:
3060         (WebCore::HTMLTreeBuilder::processStartTagForInTable):
3061
3062 2010-07-06  Eric Seidel  <eric@webkit.org>
3063
3064         Reviewed by Adam Barth.
3065
3066         Teach HTMLTreeBuilder how to handle <button><button>
3067         https://bugs.webkit.org/show_bug.cgi?id=41743
3068
3069         * html/HTMLTreeBuilder.cpp:
3070         (WebCore::HTMLTreeBuilder::processStartTagForInBody):
3071
3072 2010-07-06  Eric Seidel  <eric@webkit.org>
3073
3074         Reviewed by Adam Barth.
3075
3076         Teach TreeBuilder how to handle <form><form>
3077         https://bugs.webkit.org/show_bug.cgi?id=41741
3078
3079         * html/HTMLTreeBuilder.cpp:
3080         (WebCore::HTMLTreeBuilder::processStartTagForInBody):
3081
3082 2010-07-06  Eric Seidel  <eric@webkit.org>
3083
3084         Reviewed by Adam Barth.
3085
3086         <table> should auto-close <p> when not in quirks mode
3087         https://bugs.webkit.org/show_bug.cgi?id=41740
3088
3089         This is already covered by multiple tests in html5lib/runner.html
3090
3091         * html/HTMLTreeBuilder.cpp:
3092         (WebCore::HTMLTreeBuilder::processStartTagForInBody):
3093
3094 2010-07-06  Eric Seidel  <eric@webkit.org>
3095
3096         Reviewed by Adam Barth.
3097
3098         Make "in body" character handling reconstruct active formatting elements
3099         https://bugs.webkit.org/show_bug.cgi?id=41739
3100
3101         This fixes a bunch of tests, and due to some other bug in our
3102         formatting code, regresses two.  I'll fix that bug in a separate
3103         patch.
3104
3105         * html/HTMLTreeBuilder.cpp:
3106         (WebCore::HTMLTreeBuilder::processCharacter):
3107
3108 2010-07-06  Nikolas Zimmermann  <nzimmermann@rim.com>
3109
3110         Reviewed by Darin Adler.
3111
3112         <use> on <font-face> causes crashes, if SVGUseElement gets detached
3113         https://bugs.webkit.org/show_bug.cgi?id=41621
3114
3115         Do not call removeFromMappedElementSheet() from the destructor, as the call to document()->updateStyleSelector() that can potentially
3116         cause the element to be reattached while destructing. It's not needed at all, because removedFromDocument() is called before destruction,
3117         which already calls removeFromMappedElementSheet() - at this point it's still safe to update the style selector.
3118
3119         The crash is reproducable when using <use> on <font-face>.
3120
3121         Test: svg/custom/use-font-face-crash.svg
3122
3123         * svg/SVGFontFaceElement.cpp:
3124         (WebCore::SVGFontFaceElement::~SVGFontFaceElement):
3125
3126 2010-07-07  Hayato Ito  <hayato@chromium.org>
3127
3128         Reviewed by Darin Adler.
3129
3130         Rolling out 'page-break-inside:avoid' part of the r54929.
3131         Rebased the related layout tests, which are now expected to fail, as well.
3132
3133         https://bugs.webkit.org/show_bug.cgi?id=41532
3134
3135         * rendering/RenderBlock.cpp:
3136         (WebCore::RenderBlock::paintChildren):
3137
3138 2010-07-07  Mark Rowe  <mrowe@apple.com>
3139
3140         Fix failures in a handful of Java-related tests.
3141
3142         * WebCore.exp.in: Remove an extraneous ENABLE_ from an #if so that the condition
3143         evalutes to true when we expect it to.
3144
3145 2010-07-06  Pavel Feldman  <pfeldman@chromium.org>
3146
3147         Reviewed by Joseph Pecoraro.
3148
3149         Web Inspector: provide line numbers for inline styles.
3150
3151         https://bugs.webkit.org/show_bug.cgi?id=41691
3152
3153         Test: inspector/styles-source-lines-inline.html
3154
3155         * css/CSSParser.cpp:
3156         (WebCore::CSSParser::parseSheet):
3157         * css/CSSParser.h:
3158         * css/CSSStyleSheet.cpp:
3159         (WebCore::CSSStyleSheet::parseString):
3160         (WebCore::CSSStyleSheet::parseStringAtLine):
3161         * css/CSSStyleSheet.h:
3162         * dom/StyleElement.cpp:
3163         (WebCore::StyleElement::process):
3164         (WebCore::StyleElement::createSheet):
3165         * dom/StyleElement.h:
3166         * html/HTMLStyleElement.cpp:
3167         (WebCore::HTMLStyleElement::HTMLStyleElement):
3168         (WebCore::HTMLStyleElement::finishParsingChildren):
3169         * html/HTMLStyleElement.h:
3170         * inspector/InspectorCSSStore.cpp:
3171         (WebCore::InspectorCSSStore::getRuleRangesForStyleSheet):
3172         * inspector/InspectorDOMAgent.cpp:
3173         (WebCore::InspectorDOMAgent::buildObjectForRule):
3174         * inspector/front-end/DOMAgent.js:
3175         (WebInspector.CSSStyleDeclaration.parseRule):
3176         * inspector/front-end/StylesSidebarPane.js:
3177         (WebInspector.StylePropertiesSection.else.linkifyUncopyable):
3178         (WebInspector.StylePropertiesSection):
3179
3180 2010-07-06  Dan Bernstein  <mitz@apple.com>
3181
3182         Reviewed by Jon Honeycutt.
3183
3184         <rdar://problem/8163651> Wrong value type being supplied for kCTTypesetterOptionForcedEmbeddingLevel key
3185
3186         No change in behavior, so no new test.
3187
3188         * platform/graphics/mac/ComplexTextControllerCoreText.cpp:
3189         (WebCore::ComplexTextController::collectComplexTextRunsForCharactersCoreText): Use CFNumber instances
3190         as the values of the kCTTypesetterOptionForcedEmbeddingLevel key, instead of CFBooleans.
3191
3192 2010-07-06  Mark Rowe  <mrowe@apple.com>
3193
3194         Fix production builds on Mac OS X.
3195
3196         * DerivedSources.make: Don't try and generate the export file during the installhdrs build phase.
3197         It's not useful until much later in the build process, and it's not possible to generate it until
3198         after the tool has been compiled anyway.
3199         * WebCore.xcodeproj/project.pbxproj: Tweak how the ICU and forwarding headers are copied so that it
3200         works in production builds once more.
3201
3202 2010-07-06  Steve Falkenburg  <sfalken@apple.com>
3203
3204         Reviewed by Simon Fraser.
3205
3206         Expose URL matching from WebUserContentURLPattern
3207         https://bugs.webkit.org/show_bug.cgi?id=41726
3208         <rdar://problem/7910144>
3209
3210         * WebCore.exp.in: Export UserContentURLPattern::matches for use in WebKit.
3211
3212 2010-07-06  Eric Seidel  <eric@webkit.org>
3213
3214         Reviewed by Adam Barth.
3215
3216         Fix <nobr><nobr> case in HTMLTreeBuilder
3217         https://bugs.webkit.org/show_bug.cgi?id=41735
3218
3219         We were both not handling <nobr> correctly, as well as
3220         never hitting the <nobr> case because our formatting
3221         elements check was overzealous.
3222
3223         * html/HTMLTreeBuilder.cpp:
3224         (WebCore::HTMLTreeBuilder::processStartTagForInBody):
3225
3226 2010-07-06  Adam Barth  <abarth@webkit.org>
3227
3228         Reviewed by Eric Seidel.
3229
3230         Implement start table tag in table mode
3231         https://bugs.webkit.org/show_bug.cgi?id=41736
3232
3233         * html/HTMLTreeBuilder.cpp:
3234         (WebCore::HTMLTreeBuilder::processStartTagForInTable):
3235         (WebCore::HTMLTreeBuilder::processTableEndTagForInTable):
3236         (WebCore::HTMLTreeBuilder::processEndTagForInTable):
3237         * html/HTMLTreeBuilder.h:
3238
3239 2010-07-06  Adam Barth  <abarth@webkit.org>
3240
3241         Reviewed by Eric Seidel.
3242
3243         Implement "act as if the banana has been peeled" (for lack of a better term)
3244         https://bugs.webkit.org/show_bug.cgi?id=41734
3245
3246         The spec has a tricky passage that says "whenever a node would be
3247         inserted into the current node, it must instead be foster parented."
3248         This patch attempts to implement that requirement.
3249
3250         * html/HTMLConstructionSite.cpp:
3251         (WebCore::HTMLConstructionSite::attach):
3252         (WebCore::HTMLConstructionSite::HTMLConstructionSite):
3253         (WebCore::HTMLConstructionSite::insertHTMLHtmlElement):
3254         (WebCore::HTMLConstructionSite::insertHTMLHeadElement):
3255         (WebCore::HTMLConstructionSite::insertHTMLBodyElement):
3256         (WebCore::HTMLConstructionSite::insertScriptElement):
3257         (WebCore::HTMLConstructionSite::fosterParent):
3258         * html/HTMLConstructionSite.h:
3259         (WebCore::HTMLConstructionSite::RedirectToFosterParentGuard::RedirectToFosterParentGuard):
3260         (WebCore::HTMLConstructionSite::RedirectToFosterParentGuard::~RedirectToFosterParentGuard):
3261         * html/HTMLTreeBuilder.cpp:
3262         (WebCore::HTMLTreeBuilder::processStartTagForInTable):
3263         (WebCore::HTMLTreeBuilder::processEndTagForInTable):
3264         * html/HTMLTreeBuilder.h:
3265
3266 2010-07-06  Eric Seidel  <eric@webkit.org>
3267
3268         Reviewed by Adam Barth.
3269
3270         Fix </optgroup> in "in select" mode
3271         https://bugs.webkit.org/show_bug.cgi?id=41733
3272
3273         I had to add a oneBelowTop() accessor.
3274         I added some ASSERTs after looking at the fragment case
3275         documentation.  I'm now convinced that top() and oneBeforeTop()
3276         should never be NULL, so we just ASSERT they aren't.
3277
3278         This is a very obscure corner of the HTML spec, but at least
3279         we have a test for it.  This makes one more html5lib test pass.
3280
3281         * html/HTMLConstructionSite.h:
3282         (WebCore::HTMLConstructionSite::oneBelowTop):
3283         * html/HTMLElementStack.cpp:
3284         (WebCore::HTMLElementStack::topRecord):
3285         (WebCore::HTMLElementStack::top):
3286         (WebCore::HTMLElementStack::oneBelowTop):
3287         * html/HTMLElementStack.h:
3288         * html/HTMLTreeBuilder.cpp:
3289         (WebCore::HTMLTreeBuilder::processEndTag):
3290
3291 2010-07-06  Eric Seidel  <eric@webkit.org>
3292
3293         Reviewed by Adam Barth.
3294
3295         Fix "in row" "Anything else" handling of end tags
3296         https://bugs.webkit.org/show_bug.cgi?id=41731
3297
3298         I also fixed </br> handling to follow the spec more closely
3299         since it implementation was old and didn't use all our
3300         new processFake* hotness.
3301
3302         This is already covered by 3 tests, however we don't
3303         pass them yet, due to lack of table foster parenting
3304         code (which Adam is working on as we speak).
3305
3306         * html/HTMLTreeBuilder.cpp:
3307         (WebCore::HTMLTreeBuilder::processEndTagForInBody):
3308         (WebCore::HTMLTreeBuilder::processEndTag):
3309
3310 2010-07-06  Mark Rowe  <mrowe@apple.com>
3311
3312         Fix clean builds on Mac OS X.
3313
3314         Since "Copy Forwarding and ICU Headers" was split in to a separate target
3315         from the WebCore framework it no longer had automatica access to the
3316         PRIVATE_HEADERS_FOLDER_PATH variable. This caused it to copy the header
3317         files in to the wrong location. Fix that by giving PRIVATE_HEADERS_FOLDER_PATH
3318         its correct value and ensuring that the path exists.
3319
3320         * WebCore.xcodeproj/project.pbxproj:
3321
3322 2010-07-06  Sam Weinig  <sam@webkit.org>
3323
3324         Fix Gtk build.
3325
3326         * GNUmakefile.am:
3327
3328 2010-07-06  Mark Rowe  <mrowe@apple.com>
3329
3330         Tiger build fix.
3331
3332         Don't use WebCore's LDFLAGS when building WebCoreExportFileGenerator.
3333
3334         * WebCore.xcodeproj/project.pbxproj:
3335
3336 2010-07-06  Eric Seidel  <eric@webkit.org>
3337
3338         Reviewed by Adam Barth.
3339
3340         "In cell" does not correctly handle <td><tr> or <td><td>
3341         https://bugs.webkit.org/show_bug.cgi?id=41729
3342
3343         This change is mostly cleanup to try and prevent forgetting
3344         tag name checks in the future by using inlines to reduce
3345         copy/paste code.
3346
3347         3 little bugs in InCellMode:
3348          - Missing trTag from the long or statement (reason for the cleanup)
3349          - Used || instead of &&
3350          - Forgot to reprocess the tag after closeTheCell()
3351
3352         * html/HTMLTreeBuilder.cpp:
3353         (WebCore::HTMLTreeBuilder::processStartTagForInBody):
3354         (WebCore::HTMLTreeBuilder::processStartTagForInTable):
3355         (WebCore::HTMLTreeBuilder::processStartTag):
3356         (WebCore::HTMLTreeBuilder::processEndTagForInTable):
3357         (WebCore::HTMLTreeBuilder::processEndTag):
3358
3359 2010-07-06  Adam Barth  <abarth@webkit.org>
3360
3361         Reviewed by Eric Seidel.
3362
3363         Move fosterParent to HTMLConstructionSite
3364         https://bugs.webkit.org/show_bug.cgi?id=41728
3365
3366         This paves the way for handling the default case of InTableMode
3367         properly.
3368
3369         * html/HTMLConstructionSite.cpp:
3370         (WebCore::HTMLConstructionSite::fosterParent):
3371         * html/HTMLConstructionSite.h:
3372         * html/HTMLTreeBuilder.cpp:
3373         (WebCore::HTMLTreeBuilder::callTheAdoptionAgency):
3374
3375 2010-07-06  Adam Barth  <abarth@webkit.org>
3376
3377         Rubber-stamped by Eric Seidel.
3378
3379         Move HTMLConstructionSite to its own file.  No behavior change.
3380
3381         * Android.mk:
3382         * CMakeLists.txt:
3383         * GNUmakefile.am:
3384         * WebCore.gypi:
3385         * WebCore.pro:
3386         * WebCore.vcproj/WebCore.vcproj:
3387         * WebCore.xcodeproj/project.pbxproj:
3388         * html/HTMLConstructionSite.cpp: Copied from WebCore/html/HTMLTreeBuilder.cpp.
3389         (WebCore::HTMLNames::hasImpliedEndTag):
3390         (WebCore::HTMLConstructionSite::attach):
3391         (WebCore::HTMLConstructionSite::HTMLConstructionSite):
3392         (WebCore::HTMLConstructionSite::~HTMLConstructionSite):
3393         (WebCore::HTMLConstructionSite::insertHTMLHtmlStartTagBeforeHTML):
3394         (WebCore::HTMLConstructionSite::mergeAttributesFromTokenIntoElement):
3395         (WebCore::HTMLConstructionSite::insertHTMLHtmlStartTagInBody):
3396         (WebCore::HTMLConstructionSite::insertHTMLBodyStartTagInBody):
3397         * html/HTMLConstructionSite.h: Copied from WebCore/html/HTMLTreeBuilder.h.
3398         * html/HTMLTreeBuilder.cpp:
3399         * html/HTMLTreeBuilder.h:
3400
3401 2010-07-06  Mark Rowe  <mrowe@apple.com>
3402
3403         Build fix.
3404
3405         The path to the generated file needs to be relative to the built product directory,
3406         not to the project file.
3407
3408         * WebCore.xcodeproj/project.pbxproj:
3409
3410 2010-07-06  Mark Rowe  <mrowe@apple.com>
3411
3412         Reviewed by Darin Adler.
3413
3414         <http://webkit.org/b/41725> REGRESSION(r62283): No longer possible to build for Tiger from SnowLeopard due to export file shenanigans
3415
3416         Change how the exports file is generated. Rather than having the logic for which symbols
3417         to include be encoded in a Makefile that concatenated separate files, the logic is now
3418         encoded as #if's in the base exports file. This ensures that the exact same preprocessor
3419         settings are used when determining which symbols to export as when compiling the code that
3420         defines the symbols.
3421
3422         * DerivedSources.make: Remove logic that determines which export files to concatenate.
3423         Replace with an invocation of WebCoreExportFileGenerator.
3424         * WebCore.3DRendering.exp: Removed.
3425         * WebCore.AcceleratedCompositing.exp: Removed.
3426         * WebCore.ClientBasedGeolocation.exp: Removed.
3427         * WebCore.ContextMenus.exp: Removed.
3428         * WebCore.DashboardSupport.exp: Removed.
3429         * WebCore.DragSupport.exp: Removed.
3430         * WebCore.Geolocation.exp: Removed.
3431         * WebCore.Inspector.exp: Removed.
3432         * WebCore.JNI.exp: Removed.
3433         * WebCore.NPAPI.exp: Removed.
3434         * WebCore.OrientationEvents.exp: Removed.
3435         * WebCore.PluginHostProcess.exp: Removed.
3436         * WebCore.ProtectionSpaceAuthCallback.exp: Removed.
3437         * WebCore.SVG.Animation.exp: Removed.
3438         * WebCore.SVG.Filters.exp: Removed.
3439         * WebCore.SVG.ForeignObject.exp: Removed.
3440         * WebCore.SVG.exp: Removed.
3441         * WebCore.Tiger.exp: Removed.
3442         * WebCore.Video.exp: Removed.
3443         * WebCore.VideoProxy.exp: Removed.
3444         * WebCore.exp.in: Renamed from WebCore/WebCore.base.exp. Merge the individual .exp
3445         files in, wrapping the appropriate parts with the necessary #if's.
3446         * WebCore.xcodeproj/project.pbxproj:
3447         * make-exports-file-generator: Added.
3448
3449 2010-07-06  Sam Weinig  <sam@webkit.org>
3450
3451         Reviewed by Anders Carlsson.
3452
3453         Patch for https://bugs.webkit.org/show_bug.cgi?id=41723
3454         Add BackForwardController class
3455
3456         - Add BackForwardController which currently only forwards to the existing
3457           BackForwardList implementation. This is a first step toward allowing
3458           for a proxying BackForwardList implementation.
3459
3460         * CMakeLists.txt:
3461         * GNUmakefile.am:
3462         * WebCore.base.exp:
3463         * WebCore.gypi:
3464         * WebCore.pro:
3465         * WebCore.vcproj/WebCore.vcproj:
3466         * WebCore.xcodeproj/project.pbxproj:
3467         * history/BackForwardController.cpp: Added.
3468         (WebCore::BackForwardController::BackForwardController):
3469         (WebCore::BackForwardController::~BackForwardController):
3470         * history/BackForwardController.h: Added.
3471         (WebCore::BackForwardController::client):
3472         (WebCore::BackForwardController::list):
3473         * history/BackForwardControllerClient.h: Added.
3474         (WebCore::BackForwardControllerClient::~BackForwardControllerClient):
3475         * page/Page.cpp:
3476         (WebCore::Page::Page):
3477         (WebCore::Page::~Page):
3478         (WebCore::Page::backForwardList):
3479         (WebCore::Page::goBack):
3480         (WebCore::Page::goForward):
3481         (WebCore::Page::canGoBackOrForward):
3482         (WebCore::Page::goBackOrForward):
3483         (WebCore::Page::getHistoryLength):
3484         * page/Page.h:
3485
3486 2010-07-06  Gustavo Noronha Silva  <gustavo.noronha@collabora.co.uk>
3487
3488         Unreviewed.
3489
3490         Fixes regression on API test by disabling the scrollbars before
3491         configuring the adjustments.
3492
3493         * platform/gtk/ScrollViewGtk.cpp:
3494         (WebCore::ScrollView::setGtkAdjustments):
3495
3496 2010-07-06  Eric Seidel  <eric@webkit.org>
3497
3498         Reviewed by Adam Barth.
3499
3500         Add support for <li>, <dd> and <dt> nested tag closing
3501         https://bugs.webkit.org/show_bug.cgi?id=41720
3502
3503         li, dd, and dt have nearly identical logic in HTML5, so I
3504         attempted to make them share as much code as possible.
3505
3506         This fixes a bunch of subtests for html5lib/runner.html
3507
3508         * html/HTMLTreeBuilder.cpp:
3509         (WebCore::HTMLTreeBuilder::processCloseWhenNestedTag):
3510         (WebCore::HTMLTreeBuilder::processStartTagForInBody):
3511         (WebCore::HTMLTreeBuilder::processAnyOtherEndTagForInBody):
3512         (WebCore::HTMLTreeBuilder::furthestBlockForFormattingElement):
3513         * html/HTMLTreeBuilder.h:
3514
3515 2010-07-06  Adam Barth  <abarth@webkit.org>
3516
3517         Reviewed by Eric Seidel.
3518
3519         Factor HTMLConstructionSite out of HTMLTreeBuilder
3520         https://bugs.webkit.org/show_bug.cgi?id=41716
3521
3522         The HTMLContructionSite is the model object on which the
3523         HTMLTreeBuilder (a controller) acts.
3524
3525         No behavior change.  I'll move this class into its own file in a
3526         followup patch.
3527
3528         * html/HTMLFormattingElementList.h:
3529         (WebCore::HTMLFormattingElementList::at):
3530         * html/HTMLTreeBuilder.cpp:
3531         (WebCore::HTMLTreeBuilder::HTMLTreeBuilder):
3532         (WebCore::HTMLConstructionSite::HTMLConstructionSite):
3533         (WebCore::HTMLTreeBuilder::passTokenToLegacyParser):
3534         (WebCore::HTMLTreeBuilder::processDoctypeToken):
3535         (WebCore::HTMLConstructionSite::insertHTMLHtmlStartTagBeforeHTML):
3536         (WebCore::HTMLConstructionSite::mergeAttributesFromTokenIntoElement):
3537         (WebCore::HTMLConstructionSite::insertHTMLHtmlStartTagInBody):
3538         (WebCore::HTMLConstructionSite::insertHTMLBodyStartTagInBody):
3539         (WebCore::HTMLTreeBuilder::processFakePEndTagIfPInScope):
3540         (WebCore::HTMLTreeBuilder::processIsindexStartTagForInBody):
3541         (WebCore::HTMLTreeBuilder::processStartTagForInBody):
3542         (WebCore::HTMLTreeBuilder::processColgroupEndTagForInColumnGroup):
3543         (WebCore::HTMLTreeBuilder::closeTheCell):
3544         (WebCore::HTMLTreeBuilder::processStartTagForInTable):
3545         (WebCore::HTMLTreeBuilder::processStartTag):
3546         (WebCore::HTMLTreeBuilder::processBodyEndTagForInBody):
3547         (WebCore::HTMLTreeBuilder::processAnyOtherEndTagForInBody):
3548         (WebCore::HTMLTreeBuilder::furthestBlockForFormattingElement):
3549         (WebCore::HTMLTreeBuilder::findFosterParentFor):
3550         (WebCore::HTMLTreeBuilder::callTheAdoptionAgency):
3551         (WebCore::HTMLTreeBuilder::resetInsertionModeAppropriately):
3552         (WebCore::HTMLTreeBuilder::processEndTagForInBody):
3553         (WebCore::HTMLTreeBuilder::processCaptionEndTagForInCaption):
3554         (WebCore::HTMLTreeBuilder::processTrEndTagForInRow):
3555         (WebCore::HTMLTreeBuilder::processEndTagForInTable):
3556         (WebCore::HTMLTreeBuilder::processEndTag):
3557         (WebCore::HTMLTreeBuilder::processComment):
3558         (WebCore::HTMLTreeBuilder::processCharacter):
3559         (WebCore::HTMLTreeBuilder::processEndOfFile):
3560         (WebCore::HTMLTreeBuilder::processDefaultForBeforeHTMLMode):
3561         (WebCore::HTMLTreeBuilder::processStartTagForInHead):
3562         (WebCore::HTMLConstructionSite::insertDoctype):
3563         (WebCore::HTMLConstructionSite::insertComment):
3564         (WebCore::HTMLConstructionSite::insertCommentOnDocument):
3565         (WebCore::HTMLConstructionSite::insertCommentOnHTMLHtmlElement):
3566         (WebCore::HTMLConstructionSite::createElementAndAttachToCurrent):
3567         (WebCore::HTMLConstructionSite::insertHTMLHtmlElement):
3568         (WebCore::HTMLConstructionSite::insertHTMLHeadElement):
3569         (WebCore::HTMLConstructionSite::insertHTMLBodyElement):
3570         (WebCore::HTMLConstructionSite::insertElement):
3571         (WebCore::HTMLConstructionSite::insertSelfClosingElement):
3572         (WebCore::HTMLConstructionSite::insertFormattingElement):
3573         (WebCore::HTMLTreeBuilder::processGenericRCDATAStartTag):
3574         (WebCore::HTMLTreeBuilder::processGenericRawTextStartTag):
3575         (WebCore::HTMLConstructionSite::insertScriptElement):
3576         (WebCore::HTMLTreeBuilder::processScriptStartTag):
3577         (WebCore::HTMLConstructionSite::insertTextNode):
3578         (WebCore::HTMLConstructionSite::createElement):
3579         (WebCore::HTMLTreeBuilder::indexOfFirstUnopenFormattingElement):
3580         (WebCore::HTMLTreeBuilder::reconstructTheActiveFormattingElements):
3581         (WebCore::HTMLTreeBuilder::generateImpliedEndTagsWithExclusion):
3582         (WebCore::HTMLTreeBuilder::generateImpliedEndTags):
3583         * html/HTMLTreeBuilder.h:
3584         (WebCore::HTMLConstructionSite::currentElement):
3585         (WebCore::HTMLConstructionSite::openElements):
3586         (WebCore::HTMLConstructionSite::activeFormattingElements):
3587         (WebCore::HTMLConstructionSite::head):
3588         (WebCore::HTMLConstructionSite::form):
3589         (WebCore::HTMLConstructionSite::releaseForm):
3590         (WebCore::HTMLConstructionSite::setForm):
3591         (WebCore::HTMLConstructionSite::fragmentScriptingPermission):
3592         (WebCore::HTMLConstructionSite::attach):
3593
3594 2010-07-06  Gustavo Noronha Silva  <gustavo.noronha@collabora.co.uk>
3595
3596         Reviewed by Xan Lopez.
3597
3598         [GTK] Scrollbars sometimes go dead and stop scrolling the view
3599         https://bugs.webkit.org/show_bug.cgi?id=41711
3600
3601         Fixes GtkScrolledWindow scrollbars not actually scrolling the page
3602         in certain conditions. No tests because it is hard to reproduce,
3603         and depends on interaction with a widget that is outside of the
3604         WebView, which is tricky.
3605
3606         * platform/gtk/ScrollViewGtk.cpp:
3607         (WebCore::ScrollView::setGtkAdjustments):
3608
3609 2010-07-06  Darin Adler  <darin@apple.com>
3610
3611         Reviewed by Geoffrey Garen.
3612
3613         Simplify access to element attribute map, removing unneeded functions
3614         https://bugs.webkit.org/show_bug.cgi?id=41703
3615
3616         * css/CSSStyleSelector.cpp:
3617         (WebCore::CSSStyleSelector::canShareStyleWithElement): Use the
3618         attributeMap function instead of the mappedAttributes function.
3619         (WebCore::CSSStyleSelector::styleForElement): Ditto.
3620
3621         * dom/Element.cpp:
3622         (WebCore::Element::~Element): Renamed namedAttrMap to m_attributeMap.
3623         (WebCore::Element::removeAttribute): Ditto.
3624         (WebCore::Element::getAttribute): Ditto.
3625         (WebCore::Element::setAttribute): Ditto.
3626         (WebCore::Element::setAttributeMap): Ditto.
3627         (WebCore::Element::hasAttributes): Ditto.
3628         (WebCore::Element::createAttributeMap): Ditto.
3629         (WebCore::Element::insertedIntoDocument): Ditto.
3630         (WebCore::Element::removedFromDocument): Ditto.
3631         (WebCore::Element::getURLAttribute): Ditto.
3632
3633         * dom/Element.h: Removed unneeded virtual attributes function.
3634         Made read-only argument to attributes function default to false.
3635         (Later, we'll eliminate this.) Renamed namedAttrMap to m_attributeMap
3636         and made it private instead of protected.
3637
3638         * dom/StyledElement.cpp:
3639         (WebCore::StyledElement::attributeChanged): Use attributeMap function
3640         instead of namedAttrMap data member or mappedAttributes function.
3641         (WebCore::StyledElement::classAttributeChanged): Ditto.
3642         (WebCore::StyledElement::parseMappedAttribute): Ditto.
3643
3644         * dom/StyledElement.h: Removed mappedAttributes function and changed
3645         callers to use attributeMap function instead.
3646
3647         * html/HTMLInputElement.cpp:
3648         (WebCore::HTMLInputElement::setInputType): Use attributeMap instead of
3649         mappedAttributes.
3650         * svg/SVGStyledElement.cpp:
3651         (WebCore::SVGStyledElement::getPresentationAttribute): Ditto.
3652
3653 2010-07-06  Anders Carlsson  <andersca@apple.com>
3654
3655         Reviewed by Sam Weinig.
3656
3657         Add a build rule that removes the old generated Obj-C bindings for PluginArray and MimeTypeArray.
3658
3659         * DerivedSources.make:
3660
3661 2010-07-06  Adam Barth  <abarth@webkit.org>
3662
3663         Unreviewed.
3664
3665         Make the remaining insertion modes explicit.
3666
3667         * html/HTMLTreeBuilder.cpp:
3668         (WebCore::HTMLTreeBuilder::processStartTag):
3669         (WebCore::HTMLTreeBuilder::processEndTag):
3670         (WebCore::HTMLTreeBuilder::processCharacter):
3671         (WebCore::HTMLTreeBuilder::processEndOfFile):
3672
3673 2010-07-06  Alexey Proskuryakov  <ap@apple.com>
3674
3675         Not reviewed, fixing a simple copy/paste mistake.
3676
3677         https://bugs.webkit.org/show_bug.cgi?id=41156
3678         Cross origin XMLHttpRequest should log the reason why connection failed
3679
3680         Covered by xmlhttprequest/access-control-basic-non-simple-deny-cached.html
3681
3682         * loader/CrossOriginPreflightResultCache.cpp:
3683         (WebCore::CrossOriginPreflightResultCacheItem::allowsCrossOriginHeaders): Fix incorrectly
3684         copy/pasted header field name.
3685
3686 2010-07-06  Xan Lopez  <xlopez@igalia.com>
3687
3688         GTK+ fixes for the DOM plug-in renaming Apocalypse.
3689
3690         * GNUmakefile.am:
3691
3692 2010-07-06  Anders Carlsson  <andersca@apple.com>
3693
3694         Yet another Windows build fix.
3695
3696         * DerivedSources.cpp:
3697
3698 2010-07-06  Anders Carlsson  <andersca@apple.com>
3699
3700         Fix Windows build.
3701
3702         * WebCore.vcproj/WebCore.vcproj:
3703
3704 2010-07-06  Anders Carlsson  <andersca@apple.com>
3705
3706         More build fixes.
3707
3708         * WebCore.gypi:
3709         * WebCore.pro:
3710
3711 2010-07-06  Anders Carlsson  <andersca@apple.com>
3712
3713         Try to fix Qt build.
3714
3715         * WebCore.pro:
3716
3717 2010-07-06  Anders Carlsson  <andersca@apple.com>
3718
3719         Attempt to fix GTK+ build.
3720
3721         * GNUmakefile.am:
3722
3723 2010-07-06  Adam Barth  <abarth@webkit.org>