7c72ca21879588d085272d3ef47799bc7fbd673a
[WebKit-https.git] / WebCore / ChangeLog
1 2010-07-12  Eric Carlson  <eric.carlson@apple.com>
2
3         Reviewed by Darin Adler.
4
5         Update media element's handling of empty 'src' attribute
6         https://bugs.webkit.org/show_bug.cgi?id=42001
7
8         * html/HTMLMediaElement.cpp:
9         (WebCore::HTMLMediaElement::src): Return getNonEmptyURLAttribute().
10         (WebCore::HTMLMediaElement::selectMediaResource): Call noneSupported() for empty 'src'.
11         (WebCore::HTMLMediaElement::selectNextSourceChild): Use getNonEmptyURLAttribute() to convert 
12         'src' to URL instead of document()->completeURL(). Don't consider a <source> with an empty 'src'. 
13
14         * html/HTMLMediaElement.idl: Add 'NonEmpty' option to 'src' attribute.
15
16         * html/HTMLSourceElement.cpp:
17         (WebCore::HTMLSourceElement::src): Return getNonEmptyURLAttribute().
18         (WebCore::HTMLSourceElement::isURLAttribute): New, 'src' is a URL attribute.
19         * html/HTMLSourceElement.h:
20
21         * html/HTMLSourceElement.idl: Add 'NonEmpty' option to 'src' attribute.
22
23         * html/HTMLVideoElement.cpp:
24         (WebCore::HTMLVideoElement::parseMappedAttribute): Use getNonEmptyURLAttribute() to convert 
25         'poster' to URL instead of document()->completeURL().
26
27         * html/HTMLVideoElement.idl:  Add 'NonEmpty' option to 'poster' attribute.
28
29 2010-07-12  Martin Robinson  <mrobinson@igalia.com>
30
31         Reviewed by Xan Lopez.
32
33         [GTK] make dist is broken because of missing headers and other miscellaneous reasons
34         https://bugs.webkit.org/show_bug.cgi?id=42107
35
36         * GNUmakefile.am: Remove InspectorFrontend2.idl from the list of IDL files, so
37         that it is not built into libWebCoreJS. Add missing header to the source list.
38         Make sure to distribute the new file: WebCore/inspector/CodeGeneratorInspector.pm.
39
40 2010-07-12  Eric Seidel  <eric@webkit.org>
41
42         Reviewed by Adam Barth.
43
44         HTMLTreeBuilder needs to support mixing SVG and MathML content
45         https://bugs.webkit.org/show_bug.cgi?id=42096
46
47         This is just a direct transcription of another paragraph of the
48         HTML5 spec.
49
50         This improved a couple results in html5lib/runner.html, but more
51         work to do yet to pass all the foreign content tests.
52
53         * html/HTMLTreeBuilder.cpp:
54         (WebCore::HTMLTreeBuilder::processStartTag):
55         (WebCore::HTMLTreeBuilder::processEndTag):
56         (WebCore::HTMLTreeBuilder::processUsingSecondaryInsertionModeAndAdjustInsertionMode):
57         * html/HTMLTreeBuilder.h:
58         * mathml/mathtags.in:
59
60 2010-07-12  Eric Seidel  <eric@webkit.org>
61
62         Reviewed by Adam Barth.
63
64         Update HTMLTreeBuilder now that MathMLNames is always generated
65         https://bugs.webkit.org/show_bug.cgi?id=42059
66
67         Fix the HTMLTreeBuilder MathML code path to compile and remove
68         the MathML and SVG guards now that SVGNames and MathMLNames are
69         always compiled into ever port after:
70         https://bugs.webkit.org/show_bug.cgi?id=42050
71
72         This fixed a whole bunch of libhtml5 tests now that we have the
73         mathml code paths enabled.
74
75         * html/HTMLTreeBuilder.cpp:
76         (WebCore::HTMLTreeBuilder::processStartTagForInBody):
77         (WebCore::HTMLTreeBuilder::processStartTag):
78         (WebCore::HTMLTreeBuilder::resetInsertionModeAppropriately):
79         (WebCore::HTMLTreeBuilder::processEndTag):
80         * page/Frame.cpp:
81         (WebCore::Frame::Frame):
82          - Always init SVGNames and MathML names.
83
84 2010-07-12  Simon Fraser  <simon.fraser@apple.com>
85
86         Reviewed by Dan Bernstein.
87
88         Need to do a layout when RenderLayers come and go because of compositing
89         https://bugs.webkit.org/show_bug.cgi?id=42108
90         
91         If we create or destroy RenderLayers for reasons other than style changes
92         (e.g. because of composited iframes or plugins), then we need to ensure
93         that we do a layout.
94
95         Test: compositing/iframes/layout-on-compositing-change.html
96
97         * rendering/RenderObject.cpp:
98         (WebCore::RenderObject::adjustStyleDifference):
99
100 2010-07-11  Eric Seidel  <eric@webkit.org>
101
102         Reviewed by Darin Adler.
103
104         Always generate SVGNames and MathMLNames for all ports (to support HTML5)
105         https://bugs.webkit.org/show_bug.cgi?id=42050
106
107         Historically, FOONames has only been generate when ENABLE(FOO) is defined.
108         However, for HTML5 parser support, we need to have access to the SVG
109         and MathML tag names regardless of whether we the engine is configured
110         to render SVG or MathML content.
111
112         This change enables generation of SVGNames and MathMLNames on all ports and
113         makes it so that ports can include FOOElementFactory.* regardless of whether
114         ENABLE(FOO) is defined (and have it do the right thing).
115
116         No functional change (yet) so no tests.
117
118         * DerivedSources.make:
119         * GNUmakefile.am:
120         * WebCore.gyp/WebCore.gyp:
121         * WebCore.pri:
122         * dom/make_names.pl:
123
124 2010-07-10  Eric Seidel  <eric@webkit.org>
125
126         Reviewed by Adam Barth.
127
128         make_names.pl should always generate all names in Names.* files
129         https://bugs.webkit.org/show_bug.cgi?id=42023
130
131         Only the *ElementFactory files need to have conditional contents
132         based on enabled features.  WebCore should always have all known
133         names for SVG, MathML, XML, XLink, HTML, etc. generated in the
134         various *Names files, even if features are disabled.
135
136         make_names.pl is kinda a big hack at this point.  I tried to clean
137         up a little as I went.  The way I made *Names include all names was to
138         read the .in files twice, once using the preprocessor and once without.
139
140         * dom/make_names.pl:
141
142 2010-07-12  Andreas Kling  <andreas.kling@nokia.com>
143
144         Reviewed by Oliver Hunt.
145
146         Canvas: Move fillRect() save/restore into GraphicsContext implementations
147         https://bugs.webkit.org/show_bug.cgi?id=42088
148
149         Saving the platform painter state is an expensive operation,
150         so don't do it in fillRect() for platforms that don't need it. (CG, Qt)
151
152         * html/canvas/CanvasRenderingContext2D.cpp:
153         (WebCore::CanvasRenderingContext2D::fillRect):
154         * platform/graphics/cairo/GraphicsContextCairo.cpp:
155         (WebCore::GraphicsContext::fillRect):
156         * platform/graphics/skia/GraphicsContextSkia.cpp:
157         (WebCore::GraphicsContext::fillRect):
158         * platform/graphics/wince/GraphicsContextWince.cpp:
159         (WebCore::GraphicsContext::fillRect):
160         * platform/graphics/wx/GraphicsContextWx.cpp:
161         (WebCore::GraphicsContext::fillRect):
162
163 2010-07-12  Nate Chapin  <japhet@chromium.org>
164
165         Reviewed by Darin Fisher.
166
167         Ensure that a cache policy that forces validation is cleared once
168         the load event is fired, rather than only doing so at the next
169         navigation. This leads to a lot of unnecessary load on AJAX-y
170         websites.
171
172         https://bugs.webkit.org/show_bug.cgi?id=41813
173
174         Test: http/tests/xmlhttprequest/cache-headers-after-reload.html
175
176         * loader/FrameLoader.cpp:
177         (WebCore::FrameLoader::handledOnloadEvents): Reset m_loadType to FrameLoadTypeStandard.
178         (WebCore::FrameLoader::addExtraFieldsToRequest): Only respect the original request's cache policy if the
179             DocumentLoader is still loading, and handle the other cache policy settings that were scattered around the loader.
180         (WebCore::FrameLoader::loadResourceSynchronously): Merge cachePolicy setting into FrameLoader::addExtraFieldsToRequest.
181         * loader/SubresourceLoader.cpp:
182         (WebCore::SubresourceLoader::create): Merge cachePolicy setting into FrameLoader::addExtraFieldsToRequest.
183
184 2010-07-12  Andreas Kling  <andreas.kling@nokia.com>
185
186         Reviewed by Oliver Hunt.
187
188         CSS color parsing optimizations
189         https://bugs.webkit.org/show_bug.cgi?id=42073
190
191         - Avoid instantiating a CSSParser in parseColor() unless necessary.
192         - Fixed hex color fast-path to support strings starting with '#'.
193         - Avoid allocating a new string for the value part of a '#' color.
194
195         * css/CSSParser.cpp:
196         (WebCore::CSSParser::parseColor):
197         * platform/graphics/Color.cpp:
198         (WebCore::Color::parseHexColor):
199         (WebCore::Color::Color):
200         * platform/graphics/Color.h:
201
202 2010-07-09  Alexey Proskuryakov  <ap@apple.com>
203
204         Reviewed by Darin Adler.
205
206         https://bugs.webkit.org/show_bug.cgi?id=13075
207         XMLHttpRequest with failed authentication should set status to 401
208
209         https://bugs.webkit.org/show_bug.cgi?id=6871
210         <rdar://problem/3363403> 401 error page is never shown
211
212         * platform/network/mac/ResourceHandleMac.mm: (WebCore::ResourceHandle::receivedCredential):
213         Added a comment explaining why we handle empty credentials differently here.
214
215         * platform/network/cf/ResourceHandleCFNet.cpp: (WebCore::ResourceHandle::receivedCredential):
216         Bring this code in sync with Mac.
217
218 2010-07-12  Anders Carlsson  <andersca@apple.com>
219
220         Reviewed by Adam Roben.
221
222         Add a PluginController class, use it for invalidation and getting the user agent
223         https://bugs.webkit.org/show_bug.cgi?id=42084
224
225         * WebCore.exp.in:
226         Export Widget::convertToContainingWindow.
227
228 2010-07-12  Andreas Kling  <andreas.kling@nokia.com>
229
230         Reviewed by Simon Hausmann.
231
232         [Qt] Dragging within webkit with a drag created via Javascript ends up misinterpreting the data
233         https://bugs.webkit.org/show_bug.cgi?id=41457
234
235         Treat non-special-cased mime-types as Unicode strings in ClipboardQt's getData()
236         Fixes corruption when retrieving data that was set with anything other than text/plain
237
238         Also use QMimeData::setHtml() when applicable to be consistent with PasteboardQt.
239
240         * platform/qt/ClipboardQt.cpp:
241         (WebCore::isHtmlMimeType):
242         (WebCore::ClipboardQt::getData):
243         (WebCore::ClipboardQt::setData):
244
245 2010-07-12  Steve Block  <steveblock@google.com>
246
247         Reviewed by Jeremy Orlow.
248
249         add ANDROID to STORE_FONT_CUSTOM_PLATFORM_DATA
250         https://bugs.webkit.org/show_bug.cgi?id=32273
251
252         Tested by existing tests, just adding ANDROID to the list of platforms that use this feature.
253
254         * loader/CachedFont.cpp:
255
256 2010-07-12  Lucas De Marchi  <lucas.demarchi@profusion.mobi>
257
258         Unreviewed.
259
260         [EFL] Move ScriptConcotrollerEfl.cpp from CMakeLists.txt to
261         CMakeListsEfl.txt.
262
263         * CMakeLists.txt:
264         * CMakeListsEfl.txt:
265
266 2010-07-12  Lucas De Marchi  <lucas.demarchi@profusion.mobi>
267
268         Unreviewed build fix after r60050.
269
270         * CMakeLists.txt: Add WebCore/bindings to the include path.
271
272 2010-07-09  Gustavo Noronha Silva  <gustavo.noronha@collabora.co.uk>
273
274         Reviewed by Xan Lopez.
275
276         [GTK] Crashes when going back with page cache in unknown circunstances
277         https://bugs.webkit.org/show_bug.cgi?id=41710
278
279         Could not yet find a way to reproduce this in a layout test. The
280         issue is document being NULL, so this NULL-check should be enough
281         to get rid of the crash. We are working on trying to find a better
282         solution for these null cases, like attaching the document earlier
283         when openning a cached page.
284
285         * page/EventHandler.cpp:
286         (WebCore::EventHandler::sendScrollEvent):
287
288 2010-07-12  Alexander Pavlov  <apavlov@chromium.org>
289
290         Reviewed by Yury Semikhatsky.
291
292         [Chromium] Crash when stepping on a breakpoint while debugging Web Inspector
293         https://bugs.webkit.org/show_bug.cgi?id=41958
294
295         * page/PageGroupLoadDeferrer.cpp:
296         (WebCore::PageGroupLoadDeferrer::PageGroupLoadDeferrer):
297         * page/PageGroupLoadDeferrer.h:
298
299 2010-07-12  François Sausset  <sausset@gmail.com>
300
301         Reviewed by Kenneth Rohde Christiansen.
302
303         Make the mathsize MathML attribute handle values in em, px, pt,...
304         https://bugs.webkit.org/show_bug.cgi?id=42067
305
306         Test: mathml/presentation/attributes.xhtml
307
308         * css/mathml.css:
309         (math[mathsize="small"], mstyle[mathsize="small"], mo[mathsize="small"], mn[mathsize="small"], mi[mathsize="small"], mtext[mathsize="small"], mspace[mathsize="small"], ms[mathsize="small"]):
310         (math[mathsize="normal"], mstyle[mathsize="normal"], mo[mathsize="normal"], mn[mathsize="normal"], mi[mathsize="normal"], mtext[mathsize="normal"], mspace[mathsize="normal"], ms[mathsize="normal"]):
311         (math[mathsize="big"], mstyle[mathsize="big"], mo[mathsize="big"], mn[mathsize="big"], mi[mathsize="big"], mtext[mathsize="big"], mspace[mathsize="big"], ms[mathsize="big"]):
312         * mathml/MathMLElement.cpp:
313         (WebCore::MathMLElement::parseMappedAttribute):
314
315 2010-07-12  Xan Lopez  <xlopez@igalia.com>
316
317         Reviewed by Gustavo Noronha.
318
319         Fix compilation with sealed GTK+.
320
321         * platform/gtk/GtkVersioning.h:
322         * platform/gtk/PasteboardHelper.cpp:
323         (WebCore::PasteboardHelper::fillDataObjectFromDropData):
324
325 2010-07-12  François Sausset  <sausset@gmail.com>
326
327         Reviewed by Kenneth Rohde Christiansen.
328
329         Fix a bug preventing msqrt and mfrac to use style color to draw themselves.
330         In mfrac, the fraction bar is now using the color defined by the element style instead of black.
331         In msqrt, the radical was always drawn in black due to a colorSpace problem.
332         https://bugs.webkit.org/show_bug.cgi?id=41889
333
334         Test: mathml/presentation/roots.xhtml
335         Test: mathml/presentation/fractions.xhtml
336
337         * mathml/RenderMathMLFraction.cpp:
338         (WebCore::RenderMathMLFraction::paint):
339         * mathml/RenderMathMLSquareRoot.cpp:
340         (WebCore::RenderMathMLSquareRoot::paint):
341
342 2010-07-12  Andreas Kling  <andreas.kling@nokia.com>
343
344         Reviewed by Antti Koivisto.
345
346         Canvas: arc() with startAngle == endAngle shouldn't add to the path
347         https://bugs.webkit.org/show_bug.cgi?id=41420
348
349         Spec link:
350         http://www.whatwg.org/specs/web-apps/current-work/#dom-context-2d-stroke
351
352         * html/canvas/CanvasRenderingContext2D.cpp:
353         (WebCore::CanvasRenderingContext2D::arc):
354
355 2010-07-12  Shinichiro Hamaji  <hamaji@chromium.org>
356
357         Reviewed by Ojan Vafai.
358
359         Update padding on Windows?
360         https://bugs.webkit.org/show_bug.cgi?id=38016
361
362         Remove internal padding and add 1px vertical padding for Windows.
363
364         * css/themeWin.css:
365         (input[type="button"], input[type="submit"], input[type="reset"], input[type="file"]::-webkit-file-upload-button, button):
366         * rendering/RenderButton.cpp:
367         (WebCore::RenderButton::setupInnerStyle):
368         * rendering/RenderTheme.h:
369         * rendering/RenderThemeWin.cpp:
370         * rendering/RenderThemeWin.h:
371         * rendering/RenderThemeWince.cpp:
372         * rendering/RenderThemeWince.h:
373
374 2010-07-12  Pavel Feldman  <pfeldman@chromium.org>
375
376         Not reviewed. Chromium tests fix.
377
378         [Chromium, V8] r63057 regressed url tests.
379
380         https://bugs.webkit.org/show_bug.cgi?id=42063
381
382         * bindings/scripts/CodeGeneratorV8.pm:
383
384 2010-07-12  Steve Block  <steveblock@google.com>
385
386         Reviewed by Alexey Proskuryakov.
387
388         XPath substring function does not correctly handle non-positive values for the position argument
389         https://bugs.webkit.org/show_bug.cgi?id=41913
390
391         This patch changes the behavior of the XPath evaluate function when a non-positive
392         position argument is supplied and no length argument is supplied. In this case,
393         we reset the position to 1. This follows the spec and matches the current behaviour
394         when a length argument is supplied.
395
396         Test: fast/xpath/substring-non-positive-postion.html
397
398         * xml/XPathFunctions.cpp:
399         (WebCore::XPath::FunSubstring::evaluate):
400
401 2010-06-27  Jeremy Orlow  <jorlow@chromium.org>
402
403         Reviewed by Dumitru Daniliuc.
404
405         Implement IDBObjectStore.get/set/remove
406         https://bugs.webkit.org/show_bug.cgi?id=41250
407
408         Implement these functions in IDBObjectStore,
409         add plumbing, teach IDBAny/Callbacks how to deal
410         with IDBKey, and a few small bits of cleanup.
411
412         Test: Modified existing test to provide basic coverage.
413               Will add much more extensive layout test coverage
414               in future patches.
415
416         * Android.derived.jscbindings.mk:
417         * Android.derived.v8bindings.mk:
418         * Android.jscbindings.mk:
419         * Android.mk:
420         * Android.v8bindings.mk:
421         * CMakeLists.txt:
422         * WebCore.gypi:
423         * WebCore.pri:
424         * WebCore.pro:
425         * WebCore.vcproj/WebCore.vcproj:
426         * WebCore.xcodeproj/project.pbxproj:
427         * bindings/js/JSIDBAnyCustom.cpp:
428         (WebCore::toJS):
429         * bindings/v8/custom/V8IDBAnyCustom.cpp:
430         (WebCore::toV8):
431         * storage/IDBAny.cpp:
432         (WebCore::IDBAny::idbKey):
433         (WebCore::IDBAny::set):
434         * storage/IDBAny.h:
435         (WebCore::IDBAny::):
436         * storage/IDBAny.idl:
437         * storage/IDBCallbacks.h:
438         * storage/IDBDatabaseRequest.h:
439         * storage/IDBDatabaseRequest.idl:
440         * storage/IDBKeyRange.h:
441         * storage/IDBObjectStore.h:
442         (WebCore::IDBObjectStore::):
443         * storage/IDBObjectStoreImpl.cpp:
444         (WebCore::IDBObjectStoreImpl::IDBObjectStoreImpl):
445         (WebCore::IDBObjectStoreImpl::get):
446         (WebCore::IDBObjectStoreImpl::set):
447         (WebCore::IDBObjectStoreImpl::remove):
448         * storage/IDBObjectStoreImpl.h:
449         * storage/IDBObjectStoreRequest.cpp:
450         (WebCore::IDBObjectStoreRequest::get):
451         (WebCore::IDBObjectStoreRequest::add):
452         (WebCore::IDBObjectStoreRequest::modify):
453         (WebCore::IDBObjectStoreRequest::addOrModify):
454         (WebCore::IDBObjectStoreRequest::remove):
455         * storage/IDBObjectStoreRequest.h:
456         * storage/IDBObjectStoreRequest.idl:
457         * storage/IDBRequest.cpp:
458         (WebCore::IDBRequest::onSuccess):
459         * storage/IDBRequest.h:
460
461 2010-07-11  Maciej Stachowiak  <mjs@apple.com>
462
463         Reviewed by Dan Bernstein.
464
465         Implement animation-related methods for WebKitTestRunner
466         https://bugs.webkit.org/show_bug.cgi?id=42053
467
468         * WebCore.exp.in: Export Document::getElementById for WebKit2's benefit.
469
470 2010-07-11  Adam Barth  <abarth@webkit.org>
471
472         Rubber-stamped by Eric Seidel
473
474         Add a complete list of the HTML5 entities in JSON format.
475
476         * html/HTMLEntityNames.json: Added.
477
478 2010-07-11  Martin Robinson  <mrobinson@igalia.com>
479
480         Reviewed by Xan Lopez.
481
482         [GTK] WebKitWebView should support drops
483         https://bugs.webkit.org/show_bug.cgi?id=39843
484
485         Add support for dropping content onto GTK+ WebViews.
486
487         No new tests, as the DRT does not support simulating drops yet.
488
489         * platform/gtk/ClipboardUtilitiesGtk.cpp:
490         (WebCore::dragOperationToGdkDragAction): Added.
491         (WebCore::gdkDragActionToDragOperation): Properly detect DragOperationEvery.
492         * platform/gtk/ClipboardUtilitiesGtk.h: Add declaration for dragOperationToGdkDragAction.
493         * platform/gtk/PasteboardHelper.cpp:
494         Add new target atom and rename the markup target type to match the others.
495         Add a method which fills a data object from drop data.
496         (WebCore::PasteboardHelper::initializeTargetList): Add support for new atoms.
497         (WebCore::selectionDataToUTF8String): Added this helper.
498         (WebCore::PasteboardHelper::getClipboardContents): Use the selectionDataToUTF8String helper.
499         (WebCore::PasteboardHelper::targetListForDataObject): Change to reflect markup atom rename.
500         (WebCore::PasteboardHelper::fillDataObjectFromDropData): Added.
501         (WebCore::PasteboardHelper::dropAtoms): Added.
502         * platform/gtk/PasteboardHelper.h: Add declarations of new methods.
503
504 2010-07-10  Darin Adler  <darin@apple.com>
505
506         Reviewed by Anders Carlsson.
507
508         Enhance content attribute reflection for URL attributes, including adding a non-empty option
509         https://bugs.webkit.org/show_bug.cgi?id=42040
510
511         Test: fast/dom/URL-attribute-reflection.html
512
513         Changed syntax from [ReflectURL] to [Reflect,URL] and also added support for
514         a new option, NonEmpty, which implements the non-empty URL concept from the
515         HTML5 specification.
516
517         * bindings/scripts/CodeGenerator.pm: Changed code to expect the Reflect and URL
518         extended attributes to come in separately. The URL one simply means "the string
519         of this attribute is a URL", since we don't have a distinct type for URL. Also
520         added a new NonEmpty extended attribute.
521
522         * bindings/scripts/CodeGeneratorJS.pm: Removed now-unneeded code to handle ReflectURL.
523         * bindings/scripts/CodeGeneratorV8.pm: Ditto.
524
525         * bindings/scripts/test/CPP/WebDOMTestObj.cpp: Updated for new added test cases
526         and the fix I made to the reflectedCustomURLAttr test.
527         * bindings/scripts/test/CPP/WebDOMTestObj.h: Ditto.
528         * bindings/scripts/test/GObject/WebKitDOMTestObj.cpp: Ditto.
529         * bindings/scripts/test/GObject/WebKitDOMTestObj.h: Ditto.
530         * bindings/scripts/test/JS/JSTestObj.cpp: Ditto.
531         * bindings/scripts/test/JS/JSTestObj.h: Ditto.
532         * bindings/scripts/test/ObjC/DOMTestObj.h: Ditto.
533         * bindings/scripts/test/ObjC/DOMTestObj.mm: Ditto.
534         * bindings/scripts/test/V8/V8TestObj.cpp: Ditto. Also, for some reason the V8
535         bindig writes out the keywords into the generated file as comments, so the keyword
536         change had a direct efffect on the output file.
537
538         * bindings/scripts/test/TestObj.idl: Changed the test cases for ReflectURL to use
539         the new syntax. Added test cases for NonEmpty. Fixed the name of
540         reflectedNonEmptyURLAttr, which accidentally was repeating reflectedURLAttr
541         instead; never noticed because we never compile the test output.
542
543         * dom/Element.cpp:
544         (WebCore::Element::getNonEmptyURLAttribute): Added. For use by NonEmpty and also
545         by any code that wants to implement the non-empty URL content attribute semantic.
546         * dom/Element.h: Ditto.
547
548         * html/HTMLAnchorElement.idl: Use Reflect,URL instead of ReflectURL.
549         * html/HTMLAreaElement.idl: Ditto.
550         * html/HTMLFrameElement.idl: Ditto.
551         * html/HTMLImageElement.idl: Ditto.
552         * html/HTMLInputElement.idl: Ditto.
553         * html/HTMLLinkElement.idl: Ditto.
554         * html/HTMLMediaElement.idl: Ditto.
555         * html/HTMLObjectElement.idl: Ditto.
556         * html/HTMLScriptElement.idl: Ditto.
557         * html/HTMLVideoElement.idl: Ditto.
558
559 2010-07-10  Tony Gentilcore  <tonyg@chromium.org>
560
561         Reviewed by Darin Adler.
562
563         Add missing derrived sources to xcode project
564         https://bugs.webkit.org/show_bug.cgi?id=42034
565
566         This allows --web-timing to work with build-webkit on OSX with JSC.
567         This should have been done in bug 41442.
568
569         No new tests because no new functionality.
570
571         * WebCore.xcodeproj/project.pbxproj:
572
573 2010-07-10  Anders Carlsson  <andersca@apple.com>
574
575         Reviewed by Sam Weinig.
576
577         Don't initialize plug-ins until allowed by the page
578         https://bugs.webkit.org/show_bug.cgi?id=42033
579
580         * WebCore.exp.in:
581         Export Document::addMediaCanStartListener and Document::removeMediaCanStartListener.
582
583 2010-07-10  Sam Weinig  <sam@webkit.org>
584
585         Reviewed by Anders Carlsson.
586
587         Patch for https://bugs.webkit.org/show_bug.cgi?id=42021
588         isEqualNode should work for DocumentType nodes
589
590         Test: fast/dom/Node/isEqualNode.html
591
592         * dom/Node.cpp:
593         (WebCore::Node::isEqualNode): Add DocumentType logic from the DOM3 spec.
594
595 2010-07-10  Daniel Bates  <dbates@rim.com>
596
597         Reviewed by Adam Barth.
598
599         Move enum ReasonForCallingCanExecuteScripts to header ScriptControllerBase.h
600         https://bugs.webkit.org/show_bug.cgi?id=39339
601
602         Moved the enum ReasonForCallingCanExecuteScripts, which was
603         defined in both the JSC and V8 ScriptController.h file, into
604         a shared file called ScriptControllerBase.h.
605
606         No functionality was changed, so no new tests.
607
608         * GNUmakefile.am: Added file bindings/ScriptControllerBase.h.
609         * WebCore.gypi: Ditto.
610         * WebCore.pro: Ditto.
611         * WebCore.vcproj/WebCore.vcproj: Ditto.
612         * WebCore.vcproj/WebCoreCommon.vsprops: Ditto.
613         * WebCore.xcodeproj/project.pbxproj: Ditto.
614         * bindings/ScriptControllerBase.h: Added.
615         (WebCore::):
616         * bindings/js/ScriptController.h: #include ScriptControllerBase.h
617         * bindings/v8/ScriptController.h: Ditto.
618
619 2010-07-10  Dan Bernstein  <mitz@apple.com>
620
621         Reviewed by Anders Carlsson.
622
623         <rdar://problem/8153214> Continuation outlines in layers do not paint correctly
624
625         Test: fast/inline/continuation-outlines-with-layers-2.html
626
627         Continuation outlines are normally painted by the containing block. However, when the
628         block and the inline are not enclosed by the same self-painting layer, the inline has to
629         paint its own outlines. This was handled correctly only for the case where the inline had
630         its own self-painting layer, but now when an ancestor inline had the self-painting layer.
631
632         * rendering/InlineFlowBox.cpp:
633         (WebCore::InlineFlowBox::paint): Instead of testing for having a self-painting layer, test
634         whether any intermediate box between the inline and the containing block has a self-painting
635         layer.
636         * rendering/RenderBlock.cpp:
637         (WebCore::RenderBlock::paintObject): Ditto.
638         * rendering/RenderObject.cpp:
639         (WebCore::RenderObject::enclosingBoxModelObject): Added this utility method.
640         * rendering/RenderObject.h:
641
642 2010-07-10  François Sausset  <sausset@gmail.com>
643
644         Reviewed by Kenneth Rohde Christiansen.
645
646         Build fix: syntax and typo issues
647         https://bugs.webkit.org/show_bug.cgi?id=41859
648
649         * html/HTMLTreeBuilder.cpp:
650         (WebCore::HTMLTreeBuilder::processStartTagForInBody):
651         (WebCore::HTMLTreeBuilder::processStartTag):
652
653 2010-07-10  Eric Seidel  <eric@webkit.org>
654
655         Reviewed by Adam Barth.
656
657         HTMLTreeBuilder needs adjustForeignAttributes support
658         https://bugs.webkit.org/show_bug.cgi?id=42022
659
660         To add adjust foreign attributes support I had to add an
661         AtomicString (prefixed name) to QualifiedName hash.  Once I had
662         done that, I decided it would be best for the other "adjust" functions
663         to share the same hash logic, so I moved them to using the same
664         AtomicString -> QualifiedName hash as well.
665
666         Tested by html5lib/runner.html
667
668         * dom/Attribute.h:
669         (WebCore::Attribute::parserSetName):
670         * html/HTMLTreeBuilder.cpp:
671
672 2010-07-10  Rob Buis  <rwlbuis@gmail.com>
673
674         Reviewed by Darin Adler.
675
676         https://bugs.webkit.org/show_bug.cgi?id=41978
677         Remove namespace prefixes from idl files
678
679         No new tests, idl syntax fixes.
680
681         * css/CSSImportRule.idl:
682         * css/CSSMediaRule.idl:
683         * css/CSSStyleSheet.idl:
684         * css/CSSVariablesRule.idl:
685         * svg/SVGAnimatedString.idl:
686         * svg/SVGColor.idl:
687         * svg/SVGDocument.idl:
688         * svg/SVGLangSpace.idl:
689         * svg/SVGSVGElement.idl:
690         * svg/SVGStringList.idl:
691         * svg/SVGStylable.idl:
692         * svg/SVGStyleElement.idl:
693         * svg/SVGTests.idl:
694         * svg/SVGZoomEvent.idl:
695         * xml/XPathEvaluator.idl:
696
697 2010-07-10  Dumitru Daniliuc  <dumi@chromium.org>
698
699         Unreviewed, GTK build fix.
700
701         * bindings/gobject/GObjectEventListener.h:
702
703 2010-07-10  Dumitru Daniliuc  <dumi@chromium.org>
704
705         Reviewed by Darin Adler.
706
707         Remove unnecessary includes in header files in WebCore/dom.
708         https://bugs.webkit.org/show_bug.cgi?id=41941
709
710         * css/StyleMedia.h:
711         * dom/BeforeLoadEvent.h:
712         * dom/CustomEvent.h:
713         * dom/DOMStringMap.h:
714         * dom/DeviceOrientationEvent.h:
715         * dom/EventListener.h:
716         * dom/MessagePort.h:
717         * dom/Node.cpp:
718         * dom/NodeFilter.h:
719         * dom/NodeRareData.h:
720         * dom/RegisteredEventListener.h:
721         * dom/SelectorNodeList.cpp:
722         * dom/SelectorNodeList.h:
723         * dom/StyleElement.h:
724         * dom/Traversal.h:
725
726 2010-07-09  Tony Chang  <tony@chromium.org>
727
728         Reviewed by Ojan Vafai.
729
730         crash in WebCore::CompositeEditCommand::splitTreeToNode when indenting pre
731         https://bugs.webkit.org/show_bug.cgi?id=38231
732
733         Test: editing/execCommand/indent-pre.html
734
735         * editing/IndentOutdentCommand.cpp:
736         (WebCore::countParagraphs):
737         (WebCore::IndentOutdentCommand::indentRegion): Split text nodes into one node per paragraph
738                                                        so moveParagraph doesn't get confused.
739         (WebCore::IndentOutdentCommand::splitTextNodes):
740         * editing/IndentOutdentCommand.h:
741
742 2010-07-09  Erik Arvidsson  <arv@chromium.org>
743
744         Reviewed by Darin Adler.
745
746         Computed style is not implemented for padding-start, padding-end, margin-start, margin-end
747         https://bugs.webkit.org/show_bug.cgi?id=41496
748
749         * css/CSSComputedStyleDeclaration.cpp:
750         (WebCore::CSSComputedStyleDeclaration::getPropertyCSSValue):
751         * css/CSSProperty.cpp:
752         (WebCore::CSSProperty::resolveDirectionAwareProperty): This function resolves the property ID for a direction aware property.
753         * css/CSSProperty.h:
754         * css/CSSStyleSelector.cpp: Use helper function.
755         (WebCore::CSSStyleSelector::applyProperty):
756
757 2010-07-09  Adam Barth  <abarth@webkit.org>
758
759         Reviewed by Eric Seidel.
760
761         HTML5 tree builder should pass some LayoutTests
762         https://bugs.webkit.org/show_bug.cgi?id=41991
763
764         Before this patch, we weren't attaching text nodes to the render tree,
765         which turns out to be important.  :)
766
767         This patch fixes more than 10,000 LayoutTests.
768
769         * html/HTMLConstructionSite.cpp:
770         (WebCore::HTMLConstructionSite::attach):
771         (WebCore::HTMLConstructionSite::attachAtSite):
772
773 2010-07-09  Patrick Gansterer  <paroga@paroga.com>
774
775         Reviewed by Darin Adler.
776
777         [WINCE] Buildfix for TextEncodingRegistry
778         https://bugs.webkit.org/show_bug.cgi?id=41992
779
780         Renamed TextCodecWince to TextCodecWinCE.
781
782         * platform/text/TextEncodingRegistry.cpp:
783         (WebCore::buildBaseTextCodecMaps):
784         (WebCore::extendTextCodecMaps):
785
786 2010-07-09  Leon Clarke  <leonclarke@google.com>
787
788         Reviewed by Adam Barth.
789
790         add support for link prefetching
791         https://bugs.webkit.org/show_bug.cgi?id=3652
792
793         Test: fast/dom/HTMLLinkElement/prefetch.html
794
795         * Configurations/FeatureDefines.xcconfig:
796         * html/HTMLLinkElement.cpp:
797         (WebCore::HTMLLinkElement::tokenizeRelAttribute):
798         (WebCore::HTMLLinkElement::process):
799         * html/HTMLLinkElement.h:
800         (WebCore::HTMLLinkElement::RelAttribute::RelAttribute):
801         * loader/Cache.cpp:
802         (WebCore::createResource):
803         * loader/CachedResource.cpp:
804         (WebCore::CachedResource::data):
805         (WebCore::CachedResource::didAddClient):
806         * loader/CachedResource.h:
807         (WebCore::CachedResource::):
808         (WebCore::CachedResource::error):
809         (WebCore::CachedResource::isPrefetch):
810         (WebCore::CachedResource::schedule):
811         * loader/CachedScript.cpp:
812         * loader/CachedScript.h:
813         * loader/DocLoader.cpp:
814         (WebCore::DocLoader::requestLinkPrefetch):
815         (WebCore::DocLoader::canRequest):
816         (WebCore::DocLoader::incrementRequestCount):
817         (WebCore::DocLoader::decrementRequestCount):
818         * loader/DocLoader.h:
819         * loader/loader.cpp:
820         (WebCore::cachedResourceTypeToTargetType):
821         (WebCore::Loader::determinePriority):
822         (WebCore::Loader::load):
823         (WebCore::Loader::Host::servePendingRequests):
824         (WebCore::Loader::Host::didFinishLoading):
825         (WebCore::Loader::Host::didFail):
826         (WebCore::Loader::Host::didReceiveResponse):
827         (WebCore::Loader::Host::cancelPendingRequests):
828         * loader/loader.h:
829         (WebCore::Loader::):
830         * platform/network/ResourceRequestBase.h:
831         (WebCore::ResourceRequestBase::):
832
833 2010-07-09  James Robinson  <jamesr@chromium.org>
834
835         Unreviewed build fix.  Add #include to pick up ExceptionCode values for config (like Qt) where it's not picked up indirectly.
836
837         * html/HTMLCanvasElement.cpp:
838
839 2010-07-09  James Robinson  <jamesr@chromium.org>
840
841         Reviewed by Darin Adler.
842
843         Removes CanvasSurface and moves all of its functionality to HTMLCanvasElement.
844         https://bugs.webkit.org/show_bug.cgi?id=42005
845
846         http://trac.webkit.org/changeset/55201 introduced a new base class for HTMLCanvasElement called CanvasSurface.
847         The intention was that this would allow for code sharing with the then-proposed OffscreenCanvas.  However,
848         there is no OffscreenCanvas and there's unlikely to be one soon.  Additionally CanvasSurface breaks
849         encapsulation pretty badly by doing "static_cast<HTMLCanvasElement* const>(this)".  Until an abstraction is
850         really needed we should just use HTMLCanvasElement when we want to talk about a canvas.
851
852         This patch moves all of CanvasSurface's functionality back up to HTMLCanvasElement and reorders the header
853         to be a bit more logical.
854
855         Refactor with no behavior change, thus no new tests.
856
857         * CMakeLists.txt:
858         * GNUmakefile.am:
859         * WebCore.gypi:
860         * WebCore.pro:
861         * WebCore.vcproj/WebCore.vcproj:
862         * WebCore.xcodeproj/project.pbxproj:
863         * dom/CanvasSurface.cpp: Removed.
864         * dom/CanvasSurface.h: Removed.
865         * html/HTMLCanvasElement.cpp:
866         (WebCore::HTMLCanvasElement::HTMLCanvasElement):
867         (WebCore::HTMLCanvasElement::willDraw):
868         (WebCore::HTMLCanvasElement::setSurfaceSize):
869         (WebCore::HTMLCanvasElement::toDataURL):
870         (WebCore::HTMLCanvasElement::convertLogicalToDevice):
871         (WebCore::HTMLCanvasElement::securityOrigin):
872         (WebCore::HTMLCanvasElement::styleSelector):
873         (WebCore::HTMLCanvasElement::createImageBuffer):
874         (WebCore::HTMLCanvasElement::drawingContext):
875         (WebCore::HTMLCanvasElement::buffer):
876         (WebCore::HTMLCanvasElement::baseTransform):
877         * html/HTMLCanvasElement.h:
878         (WebCore::HTMLCanvasElement::setObserver):
879         (WebCore::HTMLCanvasElement::width):
880         (WebCore::HTMLCanvasElement::height):
881         (WebCore::HTMLCanvasElement::size):
882         (WebCore::HTMLCanvasElement::toDataURL):
883         (WebCore::HTMLCanvasElement::setOriginTainted):
884         (WebCore::HTMLCanvasElement::originClean):
885         (WebCore::HTMLCanvasElement::hasCreatedImageBuffer):
886
887 2010-07-09  Simon Fraser  <simon.fraser@apple.com>
888
889         Reviewed by Darin Adler.
890
891         Calling layoutTestController.layerTreeAsText() should update layout
892         https://bugs.webkit.org/show_bug.cgi?id=41818
893         
894         Need to call updateLayout() before we check for any layers, not after,
895         since layout will update compositing, and may create layers.
896
897         * page/Frame.cpp:
898         (WebCore::Frame::layerTreeAsText):
899
900 2010-07-09  Kenneth Russell  <kbr@google.com>
901
902         Reviewed by Nate Chapin.
903
904         bufferSubData causes crash in WebGLBuffer::associateBufferSubData
905         https://bugs.webkit.org/show_bug.cgi?id=42004
906
907         Test: fast/canvas/webgl/index-validation-crash-with-buffer-sub-data.html
908
909         * html/canvas/WebGLBuffer.cpp:
910         (WebCore::WebGLBuffer::associateBufferData):
911          - Allocate m_elementArrayBuffer for entry point taking only size.
912            Guard against allocation failures of m_elementArrayBuffer.
913         (WebCore::WebGLBuffer::associateBufferSubData):
914          - Guard against any possibility of crashes due to m_elementArrayBuffer being NULL.
915
916 2010-07-09  Dumitru Daniliuc  <dumi@chromium.org>
917
918         Unreviewed, but pre-approved by Eric Seidel.
919
920         Remove unnecessary includes in header files in WebCore/css.
921         https://bugs.webkit.org/show_bug.cgi?id=41941
922
923         * css/CSSComputedStyleDeclaration.h:
924         * css/CSSFontFaceSource.cpp:
925         (WebCore::CSSFontFaceSource::getFontData):
926         * css/CSSFontFaceSource.h:
927         * css/CSSPrimitiveValueMappings.h:
928         * css/StyleMedia.h:
929
930 2010-07-09  Eric Seidel  <eric@webkit.org>
931
932         Reviewed by Adam Barth.
933
934         Get rid of manual case maps in HTMLTreeBuilder
935         https://bugs.webkit.org/show_bug.cgi?id=42000
936
937         No functional change, thus no tests.
938
939         * html/HTMLTreeBuilder.cpp:
940
941 2010-07-09  Tony Gentilcore  <tonyg@chromium.org>
942
943         Reviewed by Dimitri Glazkov.
944
945         Implement performance.timing.navigationStart
946         https://bugs.webkit.org/show_bug.cgi?id=41815
947
948         See: http://dev.w3.org/2006/webapi/WebTiming/#nt-navigation-start
949
950         * loader/FrameLoader.cpp:
951         (WebCore::FrameLoader::loadWithDocumentLoader):
952         * loader/FrameLoaderTypes.h:
953         (WebCore::FrameLoadTimeline::FrameLoadTimeline):
954         * page/Timing.cpp:
955         (WebCore::Timing::navigationStart):
956
957 2010-07-09  Jesus Sanchez-Palencia  <jesus.palencia@openbossa.org>
958
959         Reviewed by Kenneth Rohde Christiansen.
960
961         [Qt] GraphicsLayerQt must have syncCompositingStateForThisLayerOnly() implemented
962         https://bugs.webkit.org/show_bug.cgi?id=41954
963
964         Add GraphicsLayerQt::syncCompositingStateForThisLayerOnly, a non-recursive
965         implementation of GraphicsLayerQt::syncCompositingState and which targets
966         the current layer only.
967
968         * platform/graphics/qt/GraphicsLayerQt.cpp:
969         (WebCore::GraphicsLayerQt::syncCompositingStateForThisLayerOnly):
970         * platform/graphics/qt/GraphicsLayerQt.h:
971
972 2010-07-09  David Kozub  <zub@linux.fjfi.cvut.cz>
973
974         Reviewed by Darin Adler.
975
976         Fix build by adding missing html/TimeRanges.idl to CMakeLists.txt.
977
978         https://bugs.webkit.org/show_bug.cgi?id=41945
979
980         No functional changes, thus no tests.
981
982         * CMakeLists.txt:
983
984 2010-07-09  Eric Seidel  <eric@webkit.org>
985
986         Reviewed by Adam Barth.
987
988         Fix SVG tag name casing for HTMLTreeBuilder
989         https://bugs.webkit.org/show_bug.cgi?id=41998
990
991         Tested by html5lib/runner.html
992
993         * html/HTMLTreeBuilder.cpp:
994         (WebCore::):
995
996 2010-07-09  Eric Seidel  <eric@webkit.org>
997
998         Reviewed by Adam Barth.
999
1000         Implement SVG attribute case mapping for HTMLTreeBuilder
1001         https://bugs.webkit.org/show_bug.cgi?id=41949
1002
1003         Tested by html5lib/runner.html.
1004
1005         * html/HTMLTreeBuilder.cpp:
1006
1007 2010-07-09  Andy Estes  <aestes@apple.com>
1008
1009         Reviewed by Adele Peterson.
1010
1011         Remove the workaround for a Core Animation bug on platforms where the
1012         bug has been fixed.
1013         https://bugs.webkit.org/show_bug.cgi?id=41927
1014         <rdar://problem/7920153>
1015
1016         * platform/graphics/mac/GraphicsLayerCA.mm:
1017         (WebCore::GraphicsLayerCA::createTransformAnimationsFromKeyframes): Add
1018         a compile-time check for platforms that have a Core Animation bug that
1019         needs working around.
1020
1021 2010-07-08  Tony Gentilcore  <tonyg@chromium.org>
1022
1023         Reviewed by Nate Chapin.
1024
1025         Implement unloadEventEnd, loadEventStart, and loadEventEnd for Web Timing
1026         https://bugs.webkit.org/show_bug.cgi?id=41332
1027
1028         Test: fast/dom/webtiming.html
1029
1030         * loader/FrameLoader.cpp:
1031         (WebCore::FrameLoader::stopLoading):
1032         (WebCore::FrameLoader::loadWithDocumentLoader):
1033         * loader/FrameLoader.h:
1034         (WebCore::FrameLoader::frameLoadTimeline):
1035         * loader/FrameLoaderTypes.h:
1036         (WebCore::FrameLoadTimeline::FrameLoadTimeline):
1037         * page/DOMWindow.cpp:
1038         (WebCore::DOMWindow::dispatchLoadEvent):
1039         * page/Timing.cpp:
1040         (WebCore::Timing::navigationStart):
1041         (WebCore::Timing::unloadEventEnd):
1042         (WebCore::Timing::loadEventStart):
1043         (WebCore::Timing::loadEventEnd):
1044         * page/Timing.h:
1045         * page/Timing.idl:
1046
1047 2010-07-09  Dumitru Daniliuc  <dumi@chromium.org>
1048
1049         Unreviewed, but pre-approved by Eric Seidel.
1050
1051         Remove all unnecessary includes from the header files in WebCore/accessibility/.
1052         https://bugs.webkit.org/show_bug.cgi?id=41941
1053
1054         * accessibility/AXObjectCache.h:
1055         * accessibility/AccessibilityListBox.cpp:
1056         * accessibility/AccessibilityListBox.h:
1057         * accessibility/AccessibilityMenuList.h:
1058         * accessibility/mac/AccessibilityObjectWrapper.h:
1059
1060 2010-07-09  Patrick Gansterer  <paroga@paroga.com>
1061
1062         Reviewed by Kent Tamura.
1063
1064         [WINCE] Buildfix for EventHandlerWin
1065         https://bugs.webkit.org/show_bug.cgi?id=41909
1066
1067         Use Clipboard.h instead of ClipboardWin.h on WinCE.
1068
1069         * page/win/EventHandlerWin.cpp:
1070
1071 2010-07-09  Sheriff Bot  <webkit.review.bot@gmail.com>
1072
1073         Unreviewed, rolling out r62946.
1074         http://trac.webkit.org/changeset/62946
1075         https://bugs.webkit.org/show_bug.cgi?id=41965
1076
1077         AppCache inspector support should be enabled in WebKit ToT.
1078         (Requested by pfeldman_ on #webkit).
1079
1080         * inspector/front-end/StoragePanel.js:
1081         (WebInspector.StoragePanel):
1082         (WebInspector.StoragePanel.prototype.reset):
1083         (WebInspector.StoragePanel.prototype.addApplicationCache):
1084
1085 2010-07-09  François Sausset  <sausset@gmail.com>
1086
1087         Reviewed by Kenneth Rohde Christiansen.
1088
1089         Implement MathML mathcolor & mathbackground attributes
1090         https://bugs.webkit.org/show_bug.cgi?id=41895
1091
1092         Test: mathml/presentation/attributes.xhtml
1093
1094         * dom/MappedAttributeEntry.h:
1095         (WebCore::):
1096         * mathml/MathMLElement.cpp:
1097         (WebCore::MathMLElement::MathMLElement):
1098         (WebCore::MathMLElement::mapToEntry):
1099         (WebCore::MathMLElement::parseMappedAttribute):
1100         * mathml/MathMLElement.h:
1101         * mathml/mathattrs.in:
1102
1103 2010-07-09  Xiaomei Ji  <xji@chromium.org>
1104
1105         Reviewed by David Levin.
1106
1107         Fix characters with unicode-bidi-mirror property are not correctly
1108         mirrored in Linux.
1109         https://bugs.webkit.org/show_bug.cgi?id=41305
1110
1111         Since harfbuzz does not do mirroring, chromium should iterate each
1112         character in the string and mirror it if needed before passing the
1113         string to harfbuzz for shaping.
1114
1115         Test: fast/text/international/bidi-mirror-he-ar.html
1116
1117         * platform/graphics/chromium/FontLinux.cpp:
1118         (WebCore::TextRunWalker::TextRunWalker):
1119         (WebCore::TextRunWalker::~TextRunWalker):
1120         (WebCore::TextRunWalker::mirrorCharacters):
1121
1122
1123 2010-07-09  Anders Carlsson  <andersca@apple.com>
1124
1125         Reviewed by Simon Fraser.
1126
1127         Instantiate Netscape plug-ins, pass geometry information to Plugin
1128         https://bugs.webkit.org/show_bug.cgi?id=41960
1129
1130         * WebCore.exp.in:
1131         Export ScrollView::contentsToWindow.
1132
1133 2010-07-09  Patrick Gansterer  <paroga@paroga.com>
1134
1135         Reviewed by Kent Tamura.
1136
1137         [WINCE] Buildfix for FrameWince after r47440
1138         https://bugs.webkit.org/show_bug.cgi?id=41904
1139
1140         * page/wince/FrameWince.cpp: Added property svn:eol-style.
1141         (WebCore::computePageRectsForFrame):
1142         (WebCore::imageFromSelection):
1143
1144 2010-07-09  Vitaly Repeshko  <vitalyr@chromium.org>
1145
1146         Reviewed by Pavel Feldman.
1147
1148         [v8] Call JS gc in a fresh context to avoid retaining the current one.
1149         https://bugs.webkit.org/show_bug.cgi?id=41963
1150         http://crbug.com/46571
1151
1152         * bindings/v8/ScriptController.cpp:
1153         (WebCore::ScriptController::collectGarbage):
1154
1155 2010-07-09  Kenneth Russell  <kbr@google.com>
1156
1157         Reviewed by Dimitri Glazkov.
1158
1159         Crash with uniform array test
1160         https://bugs.webkit.org/show_bug.cgi?id=36028
1161
1162         * bindings/v8/custom/V8WebGLRenderingContextCustom.cpp:
1163         (WebCore::vertexAttribAndUniformHelperf):
1164         (WebCore::uniformHelperi):
1165         (WebCore::uniformMatrixHelper):
1166          - Fixed type tests and casting of incoming arrays.
1167         * html/canvas/WebGLRenderingContext.cpp:
1168         (WebCore::WebGLRenderingContext::getUniform):
1169          - Fixed crash when null WebGLUniform is passed to getUniform.
1170
1171 2010-07-09  Chris Fleizach  <cfleizach@apple.com>
1172
1173         Reviewed by Darin Adler.
1174
1175         AX:  text editing not spoken by VO on web view contenteditable textbox
1176         https://bugs.webkit.org/show_bug.cgi?id=41912
1177
1178         Test: platform/mac/accessibility/selection-value-changes-for-aria-textbox.html
1179
1180         * accessibility/AccessibilityRenderObject.cpp:
1181         (WebCore::AccessibilityRenderObject::renderObjectIsObservable):
1182
1183 2010-07-09  Michael Nordman  <michaeln@google.com>
1184
1185         Reviewed by Dumitru Daniliuc.
1186
1187         Use class ProgressEvent when raising appcache related progress events and
1188         set the 'total', 'loaded', and 'lengthComputable' attributes.
1189         Also raise the final progress event with the 'total' and 'loaded' attribute
1190         values are equal to one another to keep pace with the spec for this feature.
1191         https://bugs.webkit.org/show_bug.cgi?id=37602
1192
1193         Test: http/tests/appcache/progress-counter.html
1194
1195         * loader/appcache/ApplicationCacheGroup.cpp:
1196         (WebCore::ApplicationCacheGroup::ApplicationCacheGroup):
1197         (WebCore::ApplicationCacheGroup::didFinishLoadingManifest):
1198         (WebCore::ApplicationCacheGroup::checkIfLoadIsComplete):
1199         (WebCore::ApplicationCacheGroup::startLoadingEntry):
1200         (WebCore::CallCacheListenerTask::create):
1201         (WebCore::CallCacheListenerTask::performTask):
1202         (WebCore::CallCacheListenerTask::CallCacheListenerTask):
1203         (WebCore::ApplicationCacheGroup::postListenerTask):
1204         * loader/appcache/ApplicationCacheGroup.h:
1205         (WebCore::ApplicationCacheGroup::postListenerTask):
1206         * loader/appcache/ApplicationCacheHost.cpp:
1207         (WebCore::ApplicationCacheHost::notifyDOMApplicationCache):
1208         (WebCore::ApplicationCacheHost::stopDeferringEvents):
1209         (WebCore::ApplicationCacheHost::dispatchDOMEvent):
1210         * loader/appcache/ApplicationCacheHost.h:
1211         (WebCore::ApplicationCacheHost::DeferredEvent::DeferredEvent):
1212
1213 2010-07-09  Sebastian Dröge  <sebastian.droege@collabora.co.uk>
1214
1215         Reviewed by Gustavo Noronha Silva.
1216
1217         Bug 41340 - [GStreamer] Subtle race condition during seeks
1218         https://bugs.webkit.org/show_bug.cgi?id=41340
1219
1220         * platform/graphics/gstreamer/WebKitWebSourceGStreamer.cpp:
1221         (webKitWebSrcStop):
1222         (webKitWebSrcNeedDataMainCb):
1223         (webKitWebSrcNeedDataCb):
1224         (webKitWebSrcEnoughDataMainCb):
1225         (webKitWebSrcEnoughDataCb):
1226         (webKitWebSrcSeekDataCb):
1227         (StreamingClient::didFinishLoading):
1228         Fix two subtle race conditions that can happen during seeks:
1229         - The timeout callback could be called before the callback ID is
1230           assigned to the instance private data. This causes the ID to
1231           be set after the callback has finished and breaks all future
1232           processing.
1233         - The source must not go EOS while a seek is pending because
1234           this confuses appsrc due to the different threads involved here.
1235
1236 2010-07-09  Simon Hausmann  <simon.hausmann@nokia.com>
1237
1238         Reviewed by Tor Arne Vestbø.
1239
1240         [Qt] Removed hard-coded enabled default of video support
1241
1242         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.
1243
1244         * WebCore.pri:
1245
1246 2010-07-09  Ben Murdoch  <benm@google.com>
1247
1248         Reviewed by Steve Block.
1249
1250         [Arm] Missing NaN check in XPath substring function
1251         https://bugs.webkit.org/show_bug.cgi?id=41862
1252
1253         Test: fast/xpath/substring-nan-position.html
1254
1255         * xml/XPathFunctions.cpp:
1256         (WebCore::XPath::FunSubstring::evaluate): Add an isnan()
1257             to the value returned from evaluating the position
1258             argument and early out and return the empty string.
1259
1260 2010-07-09  Simon Hausmann  <simon.hausmann@nokia.com>
1261
1262         Reviewed by Tor Arne Vestbø.
1263
1264         [Qt] Prospective build fix.
1265
1266         Moved the media element detection from WebCore.pro into WebCore.pri, where
1267         all the features are detected. This is also used by build-webkit to determine
1268         the defaults, which may be the reason for the build breakage.
1269
1270         * WebCore.pri:
1271         * WebCore.pro:
1272
1273 2010-07-09  Simon Hausmann  <simon.hausmann@nokia.com>
1274
1275         Reviewed by Tor Arne Vestbø.
1276
1277         [Qt] Re-enable support for QtMultimediaKit as backend for the media elements
1278
1279         Experimental support is re-enabled if QtMobility is available and the Qt
1280         version is 4.7 or above.
1281
1282         * WebCore.pro:
1283         * platform/graphics/MediaPlayer.cpp:
1284         * platform/graphics/qt/MediaPlayerPrivateQt.cpp:
1285         (WebCore::MediaPlayerPrivate::supportsType): Adapt to latest QtMultimediaKit API changes.
1286         (WebCore::MediaPlayerPrivate::MediaPlayerPrivate):
1287         (WebCore::MediaPlayerPrivate::totalBytes):
1288
1289 2010-07-09  Pavel Feldman  <pfeldman@chromium.org>
1290
1291         Reviewed by Yury Semikhatsky.
1292
1293         Web Inspector: do not show default tooltip when detailed network info is available.
1294
1295         https://bugs.webkit.org/show_bug.cgi?id=41957
1296
1297         * English.lproj/localizedStrings.js:
1298         * inspector/front-end/ResourcesPanel.js:
1299         (WebInspector.ResourcesPanel):
1300         (WebInspector.ResourcesPanel.prototype.reset):
1301         (WebInspector.ResourcesPanel.prototype.showResource):
1302         (WebInspector.ResourcesPanel.prototype._getPopoverAnchor):
1303         (WebInspector.ResourcesPanel.prototype._showPopover):
1304         (WebInspector.ResourceTimeCalculator.prototype.computeBarGraphLabels):
1305
1306 2010-07-09  François Sausset  <sausset@gmail.com>
1307
1308         Reviewed by Beth Dakin.
1309
1310         Correct the default font variant of mtext (regular instead of italic).
1311         https://bugs.webkit.org/show_bug.cgi?id=41626
1312
1313         Add test to be sure that only <mi> elements use italic fonts by default.
1314         Update test with fractions to take into account the correct behaviour of mtext (regular instead of italic).
1315
1316         * css/mathml.css:
1317         (mi):
1318
1319 2010-07-09  Tony Gentilcore  <tonyg@chromium.org>
1320
1321         Reviewed by Dimitri Glazkov.
1322
1323         Implement performance.navigation.type
1324         https://bugs.webkit.org/show_bug.cgi?id=41564
1325
1326         Tests: fast/dom/navigation-type-back-forward.html
1327                fast/dom/navigation-type-navigate.html
1328                fast/dom/navigation-type-reload.html
1329
1330         * page/Navigation.cpp:
1331         (WebCore::Navigation::type):
1332         * page/Navigation.h:
1333         (WebCore::Navigation::):
1334         * page/Navigation.idl:
1335
1336 2010-07-09  Pavel Feldman  <pfeldman@chromium.org>
1337
1338         Reviewed by Yury Semikhatsky.
1339
1340         Web Inspector: hide AppCache until implemented.
1341
1342         https://bugs.webkit.org/show_bug.cgi?id=41858
1343
1344         * inspector/front-end/StoragePanel.js:
1345         (WebInspector.StoragePanel):
1346         (WebInspector.StoragePanel.prototype.reset):
1347         (WebInspector.StoragePanel.prototype.addApplicationCache):
1348
1349 2010-07-09  François Sausset  <sausset@gmail.com>
1350
1351         Reviewed by Kenneth Rohde Christiansen.
1352
1353         Set the good default behaviour for the columalign attribute on MathML mtable element
1354         https://bugs.webkit.org/show_bug.cgi?id=41631
1355
1356         Test update: mathml/tables.xhtml
1357
1358         * css/mathml.css:
1359         (math):
1360
1361 2010-07-09  Sheriff Bot  <webkit.review.bot@gmail.com>
1362
1363         Unreviewed, rolling out r62937.
1364         http://trac.webkit.org/changeset/62937
1365         https://bugs.webkit.org/show_bug.cgi?id=41955
1366
1367         Crashes SnowLeopard leaks and Windows debug bot in fast/xsl
1368         /xslt-relative-path.xml, with assertion in
1369         XSLTProcessorLibxslt.cpp:264 (Requested by WildFox on
1370         #webkit).
1371
1372         * xml/XSLTProcessor.h:
1373         (WebCore::XSLTProcessor::XSLTProcessor):
1374         * xml/XSLTProcessorLibxslt.cpp:
1375         (WebCore::docLoaderFunc):
1376         (WebCore::setXSLTLoadCallBack):
1377         (WebCore::xsltStylesheetPointer):
1378         (WebCore::XSLTProcessor::transformToString):
1379
1380 2010-07-09  Yael Aharon  <yael.aharon@nokia.com>
1381
1382         Reviewed by Laszlo Gombos.
1383
1384         NotificationPresenter needs a cancelRequestPermission API
1385         https://bugs.webkit.org/show_bug.cgi?id=41783
1386
1387         Updated NotificationPresenter API, to use ScriptExecutionContext instead of origin.
1388         Added new API NotificationPresenter::cancelRequestsForPermision
1389         The new API will be implemented and a test will be added in a followup patch.
1390
1391         * notifications/Notification.cpp:
1392         (WebCore::Notification::Notification):
1393         * notifications/NotificationCenter.cpp:
1394         (WebCore::NotificationCenter::checkPermission):
1395         (WebCore::NotificationCenter::requestPermission):
1396         (WebCore::NotificationCenter::disconnectFrame):
1397         * notifications/NotificationCenter.h:
1398         * notifications/NotificationPresenter.h:
1399
1400 2010-07-09  Andreas Wictor  <andreas.wictor@xcerion.com>
1401
1402         Reviewed by Alexey Proskuryakov.
1403
1404         Remove global variables from XSLTProcessorLibxslt.cpp
1405         https://bugs.webkit.org/show_bug.cgi?id=41348
1406
1407         Remove the globalProcessor and globalDocLoader global variables
1408         by using the _private field that exists on most libxml structs.
1409
1410         No new tests, existing tests covers this.
1411
1412         * xml/XSLTProcessor.h:
1413         (WebCore::XSLTProcessor::sourceNode):
1414         (WebCore::XSLTProcessor::XSLTProcessor):
1415         * xml/XSLTProcessorLibxslt.cpp:
1416         (WebCore::registeredXSLTProcessors):
1417         (WebCore::registeredXSLStyleSheets):
1418         (WebCore::docLoaderFunc):
1419         (WebCore::clearSavedStyleSheetPointers):
1420         (WebCore::xsltStylesheetPointer):
1421         (WebCore::XSLTProcessor::transformToString):
1422
1423 2010-07-09  Adam Barth  <abarth@webkit.org>
1424
1425         Unreviewed build fix.
1426
1427         We recently taught the HTMLTokenizer to recognize self-closing tags,
1428         <http://trac.webkit.org/changeset/62926>, but that confused the
1429         LegacyHTMLTreeBuilder.  It turns out that it's much happier if we never
1430         say a tag is self-closing.
1431
1432         * html/HTMLTreeBuilder.cpp:
1433         (WebCore::convertToOldStyle):
1434
1435 2010-07-09  Nikolas Zimmermann  <nzimmermann@rim.com>
1436
1437         Reviewed by Rob Buis.
1438
1439         svg/custom/use-instanceRoot-event-bubbling.xhtml test crashes
1440         https://bugs.webkit.org/show_bug.cgi?id=41931
1441
1442         Be careful to not mutate (marking it for recreation) the shadow tree, while building it.
1443         The recent change that cloneNode() properly synchronizes the style/SVG attributes caused this problem.
1444
1445         Fixes crash seen on the buildbots in svg/custom/use-instanceRoot-event-bubbling.xhtml.
1446
1447         * svg/SVGUseElement.cpp:
1448         (WebCore::SVGUseElement::SVGUseElement):
1449         (WebCore::SVGUseElement::recalcStyle):
1450         (WebCore::SVGUseElement::buildShadowAndInstanceTree):
1451         (WebCore::SVGUseElement::invalidateShadowTree):
1452         * svg/SVGUseElement.h:
1453         (WebCore::SVGUseElement::setUpdatesBlocked):
1454
1455 2010-07-09  Adam Barth  <abarth@webkit.org>
1456
1457         Reviewed by Eric Seidel.
1458
1459         Skip leading newlines in <textarea>
1460         https://bugs.webkit.org/show_bug.cgi?id=41947
1461
1462         We would have caught this earlier with the ASSERT.
1463
1464         * html/HTMLTokenizer.cpp:
1465         (WebCore::HTMLTokenizer::nextToken):
1466
1467 2010-07-08  Pavel Feldman  <pfeldman@chromium.org>
1468
1469         Reviewed by Yury Semikhatsky.
1470
1471         Web Inspector: Provide detailed network info in the resources panel.
1472
1473         https://bugs.webkit.org/show_bug.cgi?id=40227
1474
1475         * English.lproj/localizedStrings.js:
1476         * inspector/InspectorController.cpp:
1477         (WebCore::InspectorController::willSendRequest):
1478         (WebCore::InspectorController::didReceiveResponse):
1479         * inspector/InspectorController.h:
1480         * inspector/InspectorResource.cpp:
1481         (WebCore::InspectorResource::InspectorResource):
1482         (WebCore::InspectorResource::updateResponse):
1483         (WebCore::InspectorResource::updateScriptObject):
1484         (WebCore::InspectorResource::buildObjectForTiming):
1485         * inspector/InspectorResource.h:
1486         * inspector/front-end/ResourcesPanel.js:
1487         (WebInspector.ResourcesPanel):
1488         (WebInspector.ResourcesPanel.prototype.elementsToRestoreScrollPositionsFor):
1489         (WebInspector.ResourcesPanel.prototype._getPopoverAnchor):
1490         (WebInspector.ResourcesPanel.prototype._showPopover):
1491         (WebInspector.ResourcesPanel.prototype.hide):
1492         (WebInspector.ResourceTimeCalculator.prototype.computeBarGraphLabels):
1493         (WebInspector.ResourceGraph):
1494         * inspector/front-end/inspector.js:
1495         (WebInspector.updateResource):
1496         * platform/network/ResourceLoadTiming.h:
1497         (WebCore::ResourceLoadTiming::create):
1498         (WebCore::ResourceLoadTiming::deepCopy):
1499         (WebCore::ResourceLoadTiming::operator==):
1500         (WebCore::ResourceLoadTiming::ResourceLoadTiming):
1501         * platform/network/ResourceRequestBase.h:
1502         (WebCore::ResourceRequestBase::reportLoadTiming):
1503         (WebCore::ResourceRequestBase::setReportLoadTiming):
1504         (WebCore::ResourceRequestBase::ResourceRequestBase):
1505         * platform/network/ResourceResponseBase.cpp:
1506         (WebCore::ResourceResponseBase::connectionID):
1507         (WebCore::ResourceResponseBase::setConnectionID):
1508         * platform/network/ResourceResponseBase.h:
1509
1510 2010-07-09  Eric Seidel  <eric@webkit.org>
1511
1512         Reviewed by Adam Barth.
1513
1514         Implement self closing start tag state in the tokenizer
1515         https://bugs.webkit.org/show_bug.cgi?id=41946
1516
1517         Amazingly we got this far w/o needing self closing.
1518         The LegacyHTMLTreeBuilder clearly uses the self-closing
1519         state, but the fact that we never exposed it seems to
1520         not have caused any test failures.  Sad.
1521
1522         The new HTMLTreeBuilder only needs the self closing state
1523         for foreign content mode (and a few parse error cases).
1524
1525         * html/HTMLToken.h:
1526         (WebCore::HTMLToken::setSelfClosing):
1527         * html/HTMLTokenizer.cpp:
1528         (WebCore::HTMLTokenizer::nextToken):
1529         * html/HTMLTreeBuilder.cpp:
1530         (WebCore::):
1531
1532 2010-07-09  Eric Seidel  <eric@webkit.org>
1533
1534         Reviewed by Adam Barth.
1535
1536         Add basic "in foreign content" support to the TreeBuilder
1537         https://bugs.webkit.org/show_bug.cgi?id=41943
1538
1539         This is covered by numerous tests in html5lib/runner.html.
1540
1541         "in foreign content" mode requires us to be able to process
1542         tokens using the "secondary insertion mode".  We have to set
1543         a fake insertion mode to do that, so much of this code is enabling
1544         setting of fake insertion modes, and then later restoration of the
1545         insertion mode after execution.
1546
1547         There is a lot more of foreign content mode to implement, but this is
1548         a good start resulting in huge test progressions.
1549
1550         * html/HTMLConstructionSite.cpp:
1551         (WebCore::HTMLConstructionSite::insertSelfClosingHTMLElement):
1552          - Use createHTMLElementAndAttachToCurrent instead of copy/paste code.
1553          - No need to include HTMLElementFactory in this file.
1554         * html/HTMLConstructionSite.h:
1555          - RedirectToFosterParentGuard does not need to be a friend class.
1556         * html/HTMLElementStack.cpp:
1557         (WebCore::HTMLElementStack::isOnlyHTMLElementsInScope):
1558         * html/HTMLElementStack.h:
1559         * html/HTMLTreeBuilder.cpp:
1560         (WebCore::HTMLTreeBuilder::processStartTagForInBody):
1561         (WebCore::):
1562         * html/HTMLTreeBuilder.h:
1563         (WebCore::HTMLTreeBuilder::setInsertionMode):
1564         (WebCore::HTMLTreeBuilder::isFakeInsertionMode):
1565         (WebCore::HTMLTreeBuilder::setFakeInsertionMode):
1566
1567 2010-07-09  Adam Barth  <abarth@webkit.org>
1568
1569         Reviewed by Eric Seidel.
1570
1571         Coalesce text nodes when foster parenting
1572         https://bugs.webkit.org/show_bug.cgi?id=41921
1573
1574         Introduces the notion of an AttachmentSite to the overall
1575         HTMLConstructionSite.  Maybe we should rename HTMLConstructionSite to
1576         HTMLConstructionArea since we construct things all over the tree?  :)
1577
1578         There's something wrong in the internal layering in this class, but I
1579         can't quite see what it is.  I added a FIXME for the some of the
1580         symptoms.
1581
1582         * html/HTMLConstructionSite.cpp:
1583         (WebCore::HTMLConstructionSite::attach):
1584         (WebCore::HTMLConstructionSite::attachAtSite):
1585         (WebCore::HTMLConstructionSite::insertTextNode):
1586         (WebCore::HTMLConstructionSite::findFosterSite):
1587         (WebCore::HTMLConstructionSite::fosterParent):
1588         * html/HTMLConstructionSite.h:
1589
1590 2010-07-08  Nikolas Zimmermann  <nzimmermann@rim.com>
1591
1592         Reviewed by Dirk Schulze.
1593
1594         RenderSVGRoot does not include border/padding while repainting
1595         https://bugs.webkit.org/show_bug.cgi?id=41854
1596
1597         RenderSVGRoot does not include border/padding in the repaint rect.
1598         clippedOverflowRectForRepaint() was missing. Affects all DRT results,
1599         as <svg> now gets properly sized.
1600
1601         Test: svg/custom/repaint-moving-svg-and-div.xhtml
1602
1603         * rendering/RenderSVGRoot.cpp:
1604         (WebCore::RenderSVGRoot::repaintRectInLocalCoordinates):
1605         (WebCore::RenderSVGRoot::clippedOverflowRectForRepaint):
1606         * rendering/RenderSVGRoot.h:
1607
1608 2010-07-09  Albert J. Wong  <ajwong@chromium.org>
1609
1610         Reviewed by Nikolas Zimmermann.
1611
1612         Add RuntimeEnabledFeatures::timeRangesEnabled() required by r62880
1613
1614         https://bugs.webkit.org/show_bug.cgi?id=41935
1615
1616         * bindings/generic/RuntimeEnabledFeatures.cpp:
1617         (WebCore::RuntimeEnabledFeatures::timeRangesEnabled):
1618                 Add in timeRangesEnabled() implementation.
1619         * bindings/generic/RuntimeEnabledFeatures.h:
1620                 Add in timeRangesEnabled() declaration.
1621
1622
1623 2010-07-09  Eric Seidel  <eric@webkit.org>
1624
1625         Reviewed by Adam Barth.
1626
1627         Add setInsertionMode setter in preparation for "in foreign content" mode
1628         https://bugs.webkit.org/show_bug.cgi?id=41942
1629
1630         "in foreign content" mode needs to be able to use a fake
1631         insertion mode for processing.  We need to be able to save the
1632         original insertion mode, set a fake one, and then restore the original
1633         if it wasn't changed.  To detect changes, we need all callsites to
1634         use a setInsertionMode accessor instead of m_insertionMode =
1635
1636         No functional changes, thus no tests.
1637
1638         * html/HTMLTreeBuilder.cpp:
1639         (WebCore::HTMLTreeBuilder::processStartTagForInBody):
1640         (WebCore::HTMLTreeBuilder::processColgroupEndTagForInColumnGroup):
1641         (WebCore::HTMLTreeBuilder::processStartTagForInTable):
1642         (WebCore::HTMLTreeBuilder::processStartTag):
1643         (WebCore::HTMLTreeBuilder::processBodyEndTagForInBody):
1644         (WebCore::HTMLTreeBuilder::setInsertionModeAndEnd):
1645         (WebCore::HTMLTreeBuilder::processEndTagForInTableBody):
1646         (WebCore::HTMLTreeBuilder::processEndTagForInCell):
1647         (WebCore::HTMLTreeBuilder::processCaptionEndTagForInCaption):
1648         (WebCore::HTMLTreeBuilder::processTrEndTagForInRow):
1649         (WebCore::HTMLTreeBuilder::processEndTag):
1650         (WebCore::HTMLTreeBuilder::processCharacter):
1651         (WebCore::HTMLTreeBuilder::processDefaultForInTableTextMode):
1652         (WebCore::HTMLTreeBuilder::processGenericRCDATAStartTag):
1653         (WebCore::HTMLTreeBuilder::processGenericRawTextStartTag):
1654         (WebCore::HTMLTreeBuilder::processScriptStartTag):
1655
1656 2010-07-09  Eric Seidel  <eric@webkit.org>
1657
1658         Reviewed by Adam Barth.
1659
1660         Add insertForeignElement in preparation for adding "in foreign content" support
1661         https://bugs.webkit.org/show_bug.cgi?id=41940
1662
1663         No functional changes, thus no tests.
1664
1665         * html/HTMLConstructionSite.cpp:
1666         (WebCore::HTMLConstructionSite::insertForeignElement):
1667         (WebCore::HTMLConstructionSite::createElement):
1668         (WebCore::HTMLConstructionSite::createHTMLElement):
1669         * html/HTMLConstructionSite.h:
1670
1671 2010-07-09  Adam Barth  <abarth@webkit.org>
1672
1673         Reviewed by Eric Seidel.
1674
1675         Implementing pending table characters
1676         https://bugs.webkit.org/show_bug.cgi?id=41916
1677
1678         This turned out to not be as scary as I thought it would be.
1679
1680         * html/HTMLTreeBuilder.cpp:
1681         (WebCore::HTMLTreeBuilder::processDoctypeToken):
1682         (WebCore::HTMLTreeBuilder::processStartTag):
1683         (WebCore::HTMLTreeBuilder::processEndTag):
1684         (WebCore::HTMLTreeBuilder::processComment):
1685         (WebCore::HTMLTreeBuilder::processCharacter):
1686         (WebCore::HTMLTreeBuilder::processEndOfFile):
1687         (WebCore::HTMLTreeBuilder::processDefaultForInTableTextMode):
1688         * html/HTMLTreeBuilder.h:
1689
1690 2010-07-08  Adam Barth  <abarth@webkit.org>
1691
1692         Reviewed by Eric Seidel.
1693
1694         Handle whitespace correctly
1695         https://bugs.webkit.org/show_bug.cgi?id=41907
1696
1697         This patch introduces an extra memcpy in the character token pipeline.
1698         I'll remove the memcpy in a future patch.
1699
1700         * html/HTMLConstructionSite.cpp:
1701         (WebCore::HTMLConstructionSite::insertTextNode):
1702         * html/HTMLConstructionSite.h:
1703         * html/HTMLTreeBuilder.cpp:
1704         (WebCore::HTMLTreeBuilder::processCharacter):
1705         * html/HTMLTreeBuilder.h:
1706
1707 2010-07-09  Eric Seidel  <eric@webkit.org>
1708
1709         Reviewed by Adam Barth.
1710
1711         Move more mode handling into functions for later re-use
1712         https://bugs.webkit.org/show_bug.cgi?id=41939
1713
1714         No functional changes, thus no tests.
1715
1716         * html/HTMLTreeBuilder.cpp:
1717         (WebCore::HTMLTreeBuilder::processEndTagForInTableBody):
1718         (WebCore::HTMLTreeBuilder::processEndTagForInRow):
1719         (WebCore::HTMLTreeBuilder::processEndTagForInCell):
1720         (WebCore::HTMLTreeBuilder::processEndTag):
1721         * html/HTMLTreeBuilder.h:
1722
1723 2010-07-09  Eric Seidel  <eric@webkit.org>
1724
1725         Reviewed by Adam Barth.
1726
1727         Add new popUntilPopped functions to clean up code
1728         https://bugs.webkit.org/show_bug.cgi?id=41936
1729
1730         Pretty self explanatory.  Much awesome code reduction.
1731
1732         No functional changes, thus no tests.
1733
1734         * html/HTMLElementStack.cpp:
1735         (WebCore::HTMLElementStack::popUntilPopped):
1736         * html/HTMLElementStack.h:
1737         * html/HTMLTreeBuilder.cpp:
1738         (WebCore::HTMLTreeBuilder::processAnyOtherEndTagForInBody):
1739         (WebCore::HTMLTreeBuilder::callTheAdoptionAgency):
1740         (WebCore::HTMLTreeBuilder::processEndTagForInBody):
1741         (WebCore::HTMLTreeBuilder::processCaptionEndTagForInCaption):
1742         (WebCore::HTMLTreeBuilder::processTableEndTagForInTable):
1743         (WebCore::HTMLTreeBuilder::processEndTag):
1744
1745 2010-07-09  Eric Seidel  <eric@webkit.org>
1746
1747         Unreviewed.  Just renaming (discussed with Adam Barth).
1748
1749         Rename createElement* to createHTMLElement* to better reflect
1750         behavior.  This is in preparation for adding foreign content support.
1751
1752         * html/HTMLConstructionSite.cpp:
1753         (WebCore::HTMLConstructionSite::createHTMLElementAndAttachToCurrent):
1754         (WebCore::HTMLConstructionSite::insertHTMLHtmlElement):
1755         (WebCore::HTMLConstructionSite::insertHTMLHeadElement):
1756         (WebCore::HTMLConstructionSite::insertHTMLBodyElement):
1757         (WebCore::HTMLConstructionSite::insertHTMLElement):
1758         (WebCore::HTMLConstructionSite::insertSelfClosingHTMLElement):
1759         (WebCore::HTMLConstructionSite::createHTMLElement):
1760         * html/HTMLConstructionSite.h:
1761         * html/HTMLTreeBuilder.cpp:
1762         (WebCore::HTMLTreeBuilder::callTheAdoptionAgency):
1763
1764 2010-07-09  Eric Seidel  <eric@webkit.org>
1765
1766         Unreviewed.  Just renaming (discussed with Adam Barth).
1767
1768         Rename insertElement to insertHTMLElement and
1769         insertSelfClosingElement to insertSelfClosingHTMLElement
1770         to better reflect what they actually do.  This is in preparation
1771         for adding foreign content support.
1772
1773         * html/HTMLConstructionSite.cpp:
1774         (WebCore::HTMLConstructionSite::insertHTMLElement):
1775         (WebCore::HTMLConstructionSite::insertSelfClosingHTMLElement):
1776         (WebCore::HTMLConstructionSite::insertFormattingElement):
1777         (WebCore::HTMLConstructionSite::reconstructTheActiveFormattingElements):
1778         (WebCore::HTMLConstructionSite::fosterParent):
1779         * html/HTMLConstructionSite.h:
1780         * html/HTMLTreeBuilder.cpp:
1781         (WebCore::HTMLTreeBuilder::processCloseWhenNestedTag):
1782         (WebCore::HTMLTreeBuilder::processStartTagForInBody):
1783         (WebCore::HTMLTreeBuilder::processStartTagForInTable):
1784         (WebCore::HTMLTreeBuilder::processStartTag):
1785         (WebCore::HTMLTreeBuilder::callTheAdoptionAgency):
1786         (WebCore::HTMLTreeBuilder::processStartTagForInHead):
1787         (WebCore::HTMLTreeBuilder::processGenericRCDATAStartTag):
1788         (WebCore::HTMLTreeBuilder::processGenericRawTextStartTag):
1789
1790 2010-07-08  Rob Buis  <rwlbuis@gmail.com>
1791
1792         Reviewed by Eric Seidel.
1793
1794         Implement SVGSVGElement.getElementById
1795         https://bugs.webkit.org/show_bug.cgi?id=41655
1796
1797         Implement getElementById for SVGSVGElement by trying to
1798         reuse Document.getElementById. If that fails to find an
1799         SVG element in the document fragent do a subtree search.
1800
1801         Test: svg/custom/svg-getelementid.xhtml
1802
1803         * svg/SVGSVGElement.cpp:
1804         (WebCore::SVGSVGElement::getElementById):
1805         * svg/SVGSVGElement.h:
1806         * svg/SVGSVGElement.idl:
1807
1808 2010-07-08  Eric Seidel  <eric@webkit.org>
1809
1810         Reviewed by Adam Barth.
1811
1812         Unwrap a few || blocks for easier readability
1813         https://bugs.webkit.org/show_bug.cgi?id=41838
1814
1815         No functional change, thus no tests.
1816
1817         * html/HTMLTreeBuilder.cpp:
1818         (WebCore::HTMLTreeBuilder::adjustedLexerState):
1819         (WebCore::HTMLTreeBuilder::processStartTagForInBody):
1820         (WebCore::HTMLTreeBuilder::processStartTagForInTable):
1821         (WebCore::HTMLTreeBuilder::processStartTag):
1822         (WebCore::HTMLTreeBuilder::processEndTagForInBody):
1823         (WebCore::HTMLTreeBuilder::processEndTag):
1824         (WebCore::HTMLTreeBuilder::processComment):
1825         (WebCore::HTMLTreeBuilder::processStartTagForInHead):
1826
1827 2010-07-08  Sam Magnuson  <smagnuson@netflix.com>
1828
1829         Reviewed by Simon Hausmann.
1830
1831         [Qt] for debugging purposes nice I'm contributing back my FPS
1832         counter in the AnimationQtBase
1833         https://bugs.webkit.org/show_bug.cgi?id=40381
1834
1835         Simply start a timer and count frames in the AnimationQtBase and
1836         spit out the FPS count at the end of a single animation.
1837
1838         No new tests: this is a simple debugging aid.
1839
1840         * platform/graphics/qt/GraphicsLayerQt.cpp:
1841         (WebCore::AnimationQt::updateState):
1842         (WebCore::AnimationQt::updateCurrentTime):
1843         (WebCore::TransformAnimationQt::updateState):
1844         (WebCore::OpacityAnimationQt::updateState):
1845
1846 2010-07-08  Sam Magnuson  <smagnuson@netflix.com>
1847
1848         Reviewed by Simon Hausmann.
1849
1850         [Qt] instance objects created for QObjects are somtimes GC'd
1851         https://bugs.webkit.org/show_bug.cgi?id=40352
1852
1853         In markAggregate loop over all the current fields and for any that
1854         still have a dynamic property or a child, mark them as
1855         well. Otherwise the proxy instance will go away and the JS that
1856         was bound to it will be lost.
1857
1858         * bridge/qt/qt_instance.cpp:
1859         (JSC::Bindings::QtInstance::getQtInstance):
1860         (JSC::Bindings::QtInstance::removeCachedMethod):
1861         (JSC::Bindings::QtInstance::markAggregate):
1862         (JSC::Bindings::QtInstance::getPropertyNames):
1863
1864 2010-07-08  Ilya Tikhonovsky  <loislo@chromium.org>
1865
1866         Reviewed by Pavel Feldman.
1867
1868         WebInspector: migrate InspectorDOMAgent to the generated version of
1869         InspectorFrontend interface. This is the next step on the way to
1870         Remote Debugging.
1871
1872         * bindings/js/ScriptEventListener.cpp:
1873         (WebCore::eventListenerHandlerBody):
1874         (WebCore::eventListenerHandlerLocation):
1875         * bindings/js/ScriptEventListener.h:
1876         * bindings/v8/ScriptEventListener.cpp:
1877         (WebCore::eventListenerHandlerBody):
1878         (WebCore::eventListenerHandlerLocation):
1879         * bindings/v8/ScriptEventListener.h:
1880         * inspector/CodeGeneratorInspector.pm:
1881         * inspector/InspectorCSSStore.cpp:
1882         (WebCore::InspectorCSSStore::inspectorStyleSheet):
1883         * inspector/InspectorController.cpp:
1884         (WebCore::InspectorController::connectFrontend):
1885         * inspector/InspectorController.h:
1886         (WebCore::InspectorController::inspectorFrontend2):
1887         * inspector/InspectorDOMAgent.cpp:
1888         (WebCore::InspectorDOMAgent::InspectorDOMAgent):
1889         (WebCore::InspectorDOMAgent::setDocument):
1890         (WebCore::InspectorDOMAgent::handleEvent):
1891         (WebCore::InspectorDOMAgent::pushChildNodesToFrontend):
1892         (WebCore::InspectorDOMAgent::getEventListenersForNode):
1893         (WebCore::InspectorDOMAgent::buildObjectForNode):
1894         (WebCore::InspectorDOMAgent::buildArrayForElementAttributes):
1895         (WebCore::InspectorDOMAgent::buildArrayForContainerChildren):
1896         (WebCore::InspectorDOMAgent::buildObjectForEventListener):
1897         (WebCore::InspectorDOMAgent::didInsertDOMNode):
1898         (WebCore::InspectorDOMAgent::getStyles):
1899         (WebCore::InspectorDOMAgent::getAllStyles):
1900         (WebCore::InspectorDOMAgent::getStyleSheet):
1901         (WebCore::InspectorDOMAgent::getRuleRangesForStyleSheetId):
1902         (WebCore::InspectorDOMAgent::getInlineStyle):
1903         (WebCore::InspectorDOMAgent::getComputedStyle):
1904         (WebCore::InspectorDOMAgent::buildObjectForAttributeStyles):
1905         (WebCore::InspectorDOMAgent::buildArrayForCSSRules):
1906         (WebCore::InspectorDOMAgent::buildArrayForPseudoElements):
1907         (WebCore::InspectorDOMAgent::applyStyleText):
1908         (WebCore::InspectorDOMAgent::toggleStyleEnabled):
1909         (WebCore::InspectorDOMAgent::setRuleSelector):
1910         (WebCore::InspectorDOMAgent::addRule):
1911         (WebCore::InspectorDOMAgent::buildObjectForStyle):
1912         (WebCore::InspectorDOMAgent::populateObjectWithStyleProperties):
1913         (WebCore::InspectorDOMAgent::buildArrayForDisabledStyleProperties):
1914         (WebCore::InspectorDOMAgent::buildObjectForStyleSheet):
1915         (WebCore::InspectorDOMAgent::buildObjectForRule):
1916         (WebCore::InspectorDOMAgent::toArray):
1917         (WebCore::InspectorDOMAgent::reportNodesAsSearchResults):
1918         * inspector/InspectorDOMAgent.h:
1919         (WebCore::InspectorDOMAgent::create):
1920         * inspector/InspectorFrontend.cpp:
1921         * inspector/InspectorFrontend.h:
1922         * inspector/InspectorFrontend2.idl:
1923         * inspector/front-end/inspector.js:
1924         (WebInspector.dispatchMessageFromBackend):
1925
1926 2010-07-08  Simon Fraser  <simon.fraser@apple.com>
1927
1928         Reviewed by Darin Adler.
1929
1930         compositing/iframes/iframe-resize.html displays incorrectly after the resize
1931         https://bugs.webkit.org/show_bug.cgi?id=41794
1932         
1933         The clip and scroll layers of a composited iframe's RenderLayerCompositor are updated from
1934         from updateGraphicsLayerGeometry(), but this is too early to get the correct layoutWidth and
1935         layoutHeight from the FrameView which happen later in layout. So when a widget size changes,
1936         call updateAfterWidgetResize() directly on the RenderLayerBacking (if any).
1937
1938         * rendering/RenderLayerBacking.cpp:
1939         (WebCore::RenderLayerBacking::updateAfterWidgetResize): New method that updates the clip
1940         and scroll layers of the iframe's content RenderLayerCompositor.
1941         (WebCore::RenderLayerBacking::updateGraphicsLayerGeometry): Call updateAfterWidgetResize()
1942         * rendering/RenderLayerBacking.h: Add updateAfterWidgetResize().
1943         * rendering/RenderWidget.cpp:
1944         (WebCore::RenderWidget::setWidgetGeometry): Call updateAfterWidgetResize().
1945
1946 2010-07-08  Simon Fraser  <simon.fraser@apple.com>
1947
1948         Reviewed by Dan Bernstein.
1949
1950         Calling layoutTestController.layerTreeAsText() should update layout
1951         https://bugs.webkit.org/show_bug.cgi?id=41818
1952
1953         Call updateLayout() on the document before obtaining the layer tree.
1954         This will update compositing layers.
1955         
1956         * page/Frame.cpp:
1957         (WebCore::Frame::layerTreeAsText):
1958
1959 2010-07-08  Adam Barth  <abarth@webkit.org>
1960
1961         Reviewed by Eric Seidel.
1962
1963         Get my head wrapped around processCharacter
1964         https://bugs.webkit.org/show_bug.cgi?id=41812
1965
1966         The bulk of the patch is just stubbing out functions to document what
1967         the spec tells us to do.  I'll implement these functions in subsequent
1968         patches.
1969
1970         * html/HTMLConstructionSite.h:
1971         (WebCore::HTMLConstructionSite::insertLeadingWhitespace):
1972         (WebCore::HTMLConstructionSite::insertLeadingWhitespaceWithActiveFormattingElements):
1973         * html/HTMLTreeBuilder.cpp:
1974         (WebCore::HTMLTreeBuilder::processCharacter):
1975         * html/HTMLTreeBuilder.h:
1976         (WebCore::HTMLTreeBuilder::skipLeadingWhitespace):
1977
1978 2010-07-08  Tony Chang  <tony@chromium.org>
1979
1980         Reviewed by Ojan Vafai.
1981
1982         WebCore::InsertLineBreakCommand::shouldUseBreakElement ReadAV@NULL
1983         https://bugs.webkit.org/show_bug.cgi?id=30116
1984         Fixing the crash causes text insertions on hidden elements to get ignored
1985         (not a new bug).  This is https://bugs.webkit.org/show_bug.cgi?id=40342
1986
1987         Test: editing/inserting/return-key-in-hidden-textarea.html
1988
1989         * editing/InsertLineBreakCommand.cpp:
1990         (WebCore::InsertLineBreakCommand::doApply):
1991         * editing/InsertParagraphSeparatorCommand.cpp:
1992         (WebCore::InsertParagraphSeparatorCommand::doApply):
1993
1994 2010-07-08  Sam Weinig  <sam@webkit.org>
1995
1996         Reviewed by Oliver Hunt.
1997
1998         Fix for https://bugs.webkit.org/show_bug.cgi?id=41923
1999         TimeRanges should expose a JS constructor
2000
2001         * html/TimeRanges.idl: Remove [OmitConstructor].
2002         * page/DOMWindow.idl: Add constructor.
2003
2004 2010-07-08  Erik Arvidsson  <arv@chromium.org>
2005
2006         Reviewed by Ojan Vafai.
2007
2008         Add missing padding-end and margin-end CSS properties.
2009         https://bugs.webkit.org/show_bug.cgi?id=25761
2010
2011         Tests: fast/css/margin-start-end.html
2012                fast/css/padding-start-end.html
2013
2014         * css/CSSComputedStyleDeclaration.cpp:
2015         (WebCore::CSSComputedStyleDeclaration::getPropertyCSSValue):
2016         * css/CSSParser.cpp:
2017         (WebCore::CSSParser::parseValue):
2018         * css/CSSPropertyNames.in:
2019         * css/CSSStyleSelector.cpp:
2020         (WebCore::CSSStyleSelector::applyProperty):
2021         * inspector/front-end/SourceCSSTokenizer.js:
2022         (WebInspector.SourceCSSTokenizer):
2023
2024 2010-07-08  Aaron Boodman  <aa@chromium.org>
2025
2026         Reviewed by Timothy Hatcher.
2027
2028         Add the ability for user scripts and user styles to affect just the top frame.
2029
2030         https://bugs.webkit.org/show_bug.cgi?id=41529
2031
2032         Tests: userscripts/user-script-all-frames.html
2033                userscripts/user-script-top-frame-only.html
2034                userscripts/user-style-all-frames.html
2035                userscripts/user-style-top-frame-only.html
2036
2037         * WebCore.base.exp: Update PageGroup method signatures.
2038         * dom/Document.cpp:
2039         (WebCore::Document::pageGroupUserSheets): Check allFrames before injecting.
2040         * page/Frame.cpp:
2041         (WebCore::Frame::injectUserScriptsForWorld): Ditto.
2042         * page/PageGroup.cpp:
2043         (WebCore::PageGroup::addUserScriptToWorld):
2044         (WebCore::PageGroup::addUserStyleSheetToWorld):
2045         * page/PageGroup.h:
2046         * page/UserScript.h:
2047         (WebCore::UserScript::UserScript):
2048         (WebCore::UserScript::injectedFrames):
2049         * page/UserScriptTypes.h:
2050         (WebCore::):
2051         * page/UserStyleSheet.h:
2052         (WebCore::UserStyleSheet::UserStyleSheet):
2053         (WebCore::UserStyleSheet::injectedFrames):
2054
2055 2010-07-08  Adele Peterson  <adele@apple.com>
2056
2057         Reviewed by Jon Honeycutt, Adam Roben, and Darin Adler.
2058
2059         Fix for https://bugs.webkit.org/show_bug.cgi?id=41721
2060         <rdar://problem/8158561> Missing plug-in indicator should have a pressed state
2061
2062         Test: plugins/clicking-missing-plugin-fires-delegate.html
2063
2064         * html/HTMLPlugInElement.cpp:
2065         (WebCore::HTMLPlugInElement::HTMLPlugInElement): Initialize m_isCapturingMouseEvents.
2066         (WebCore::HTMLPlugInElement::detach): If we're still capturing when getting detached, clear the capturing node on the EventHandler.
2067         (WebCore::HTMLPlugInElement::defaultEventHandler): Call handleMissingPluginIndicatorEvent when the missing plugin indicator is showing.
2068         * html/HTMLPlugInElement.h:
2069         (WebCore::HTMLPlugInElement::isCapturingMouseEvents):
2070         (WebCore::HTMLPlugInElement::setIsCapturingMouseEvents):
2071         * page/ChromeClient.h:
2072         (WebCore::ChromeClient::shouldMissingPluginMessageBeButton): Added default implementation.
2073         * rendering/RenderEmbeddedObject.cpp:
2074         (WebCore::replacementTextRoundedRectPressedColor):
2075         (WebCore::RenderEmbeddedObject::RenderEmbeddedObject):
2076         (WebCore::RenderEmbeddedObject::setMissingPluginIndicatorIsPressed): Added.  Causes a repaint when the state changes.
2077         (WebCore::RenderEmbeddedObject::paintReplaced): Call getReplacementTextGeometry.
2078         (WebCore::RenderEmbeddedObject::getReplacementTextGeometry): Factored this out so it can be used in paintReplaced and in isInMissingPluginIndicator.
2079         (WebCore::RenderEmbeddedObject::isInMissingPluginIndicator): Hit test to see if the mouse event is in the missing plugin indicator.
2080         (WebCore::RenderEmbeddedObject::handleMissingPluginIndicatorEvent): Capture mouse events as needed and track the pressed appearance.
2081         * rendering/RenderEmbeddedObject.h:
2082
2083 2010-07-08  James Robinson  <jamesr@google.com>
2084
2085         Reviewed by Darin Fisher.
2086
2087         Allow resizing and getting the texture id from an offscreen GLES2Context
2088         https://bugs.webkit.org/show_bug.cgi?id=41828
2089
2090         When using an offscreen GLES2Context the caller needs to be able to resize the backing store
2091         managed by the embedder and get access to a texture id to pass to the compositor.  WebGL
2092         does these actions in an indirect way, it will be refactored to use this path.
2093
2094         * platform/chromium/GLES2Context.h:
2095
2096 2010-07-02  Ojan Vafai  <ojan@chromium.org>
2097
2098         Reviewed by Adam Barth.
2099
2100         Crash in RenderObject::containingBlock when clearing selection in a display:none node.
2101         https://bugs.webkit.org/show_bug.cgi?id=41523
2102
2103         updateStyleIfNeeded before clearing the selection in the RenderView. Otherwise,
2104         m_selectionStart and m_selectionEnd in RenderView point to garbage object.
2105         This fixes the crash because updateStyleIfNeeded clears the selection before
2106         clobbering nodes that contain the selection.
2107
2108         Test: editing/selection/crash-on-clear-selection.html
2109
2110         * editing/SelectionController.cpp:
2111         (WebCore::SelectionController::updateAppearance):
2112
2113 2010-07-08  Brent Fulgham  <bfulgham@webkit.org>
2114
2115         Reviewed by Xan Lopez.
2116
2117         Correct a mistake in calculating squiggle extents.  The entire
2118         quantity was being divided by two, rather than just the unitWidth.
2119         This error caused the squiggle to be about half the expected length.
2120
2121         * platform/graphics/cairo/DrawErrorUnderline.h:
2122         (drawErrorUnderline):
2123
2124 2010-07-08  John Abd-El-Malek  <jam@chromium.org>
2125
2126         Reviewed by Darin Fisher.
2127
2128         [V8] Navigation policy doesn't play nicely with pepper plugins
2129         https://bugs.webkit.org/show_bug.cgi?id=41864
2130
2131         * bindings/v8/V8Utilities.cpp:
2132         (WebCore::callingOrEnteredFrame):
2133         (WebCore::shouldAllowNavigation):
2134         (WebCore::navigateIfAllowed):
2135
2136 2010-07-08  Jon Honeycutt  <jhoneycutt@apple.com>
2137
2138         Missing plug-ins may cause an assertion failure.
2139         https://bugs.webkit.org/show_bug.cgi?id=41900
2140
2141         Reviewed by Adele Peterson.
2142
2143         Test: plugins/missing-plugin.html
2144
2145         * plugins/PluginView.cpp:
2146         (WebCore::PluginView::~PluginView):
2147         Check whether m_instance is null before trying to remove it from the
2148         map. Trying to remove a null instance results in an assertion failure.
2149         (WebCore::PluginView::PluginView):
2150         Initialize m_instance to 0, because there is an early return that may
2151         skip its being assigned its proper value.
2152
2153 2010-07-08  Darin Adler  <darin@apple.com>
2154
2155         Reviewed by Oliver Hunt.
2156
2157         Fix a test failure seen only on the Leopard bot.
2158
2159         * bindings/js/JSSharedWorkerCustom.cpp:
2160         (WebCore::JSSharedWorkerConstructor::constructJSSharedWorker):
2161         If SharedWorker::create returns an exception, don't try to create a wrapper
2162         for its return value, which can be 0.
2163
2164 2010-07-08  Albert J. Wong  <ajwong@chromium.org>
2165
2166         Not reviewed, build break fix for chromium.
2167
2168         A new MediaControlElementType enum was added which caused a warning in
2169         the Chromium build.
2170
2171         * rendering/RenderMediaControlsChromium.cpp:
2172         (WebCore::RenderMediaControlsChromium::paintMediaControlsPart):
2173             Add MediaVolumeSliderMuteButton enumartion entry.
2174
2175 2010-07-08  Dmitry Titov  <dimich@chromium.org>
2176
2177         Reviewed by David Levin.
2178
2179         Remove IDL declarations and stubs for navigator.registerProtocolHandler and navigator.registerContentHandler.
2180         https://bugs.webkit.org/show_bug.cgi?id=41878
2181
2182         This is practically a rollback of http://trac.webkit.org/changeset/50477.
2183
2184         * page/Chrome.cpp:
2185         * page/Chrome.h:
2186         * page/ChromeClient.h:
2187         * page/Navigator.cpp:
2188         * page/Navigator.h:
2189         * page/Navigator.idl:
2190
2191 2010-07-08  Xan Lopez  <xlopez@igalia.com>
2192
2193         Reviewed by Darin Adler.
2194
2195         Fix adoptRef assertion failures caused by stack-allocated ResourceHandle objects
2196         https://bugs.webkit.org/show_bug.cgi?id=41823
2197
2198         Create the ResourceHandle manually for now instead of calling
2199         ::create, since that methods does a few extra checks that make us
2200         fail a couple of HTTP tests.
2201
2202         * platform/network/soup/ResourceHandleSoup.cpp:
2203         (WebCore::):
2204
2205 2010-07-08  Alexey Proskuryakov  <ap@apple.com>
2206
2207         Reviewed by David Levin.
2208
2209         https://bugs.webkit.org/show_bug.cgi?id=41886
2210         Shorten access control failure explanation messages
2211
2212         * loader/CrossOriginAccessControl.cpp:
2213         (WebCore::passesAccessControlCheck):
2214         * loader/CrossOriginPreflightResultCache.cpp:
2215         (WebCore::CrossOriginPreflightResultCacheItem::allowsCrossOriginMethod):
2216         (WebCore::CrossOriginPreflightResultCacheItem::allowsCrossOriginHeaders):
2217         Removed "response header field" parts from error text - it's long, but doesn't add much
2218         to header field name.
2219
2220         * platform/network/ResourceErrorBase.cpp:
2221         * platform/network/ResourceErrorBase.h:
2222         * xml/XMLHttpRequest.cpp:
2223         (WebCore::XMLHttpRequest::didFail):
2224         * loader/DocumentThreadableLoader.cpp:
2225         (WebCore::DocumentThreadableLoader::DocumentThreadableLoader):
2226         (WebCore::DocumentThreadableLoader::makeSimpleCrossOriginAccessRequest):
2227         (WebCore::DocumentThreadableLoader::didReceiveResponse):
2228         (WebCore::DocumentThreadableLoader::preflightFailure):
2229         Renamed the constant for domain. WebCore shouldn't be creating errors observable by WebKit
2230         clients, because it can't create a platform error, and cannot load a localized string.
2231         We don't treat cross origin access check failures as true loading failures, so we don't tell
2232         clients about these.
2233
2234 2010-07-08  Eric Carlson  <eric.carlson@apple.com>
2235
2236         Reviewed by Dan Bernstein.
2237
2238         Mac OS X media controls should have a way to adjust volume incrementally
2239         https://bugs.webkit.org/show_bug.cgi?id=41718
2240         <rdar://problem/5679472>
2241
2242         * css/CSSPrimitiveValueMappings.h:
2243         (WebCore::CSSPrimitiveValue::CSSPrimitiveValue): Handle MediaVolumeSliderMuteButtonPart.
2244
2245         * css/CSSSelector.cpp: 
2246         (WebCore::CSSSelector::pseudoId): Handle PseudoMediaControlsVolumeSliderMuteButton.
2247         (WebCore::nameToPseudoTypeMap): Define mediaControlsVolumeSliderMuteButton.
2248         (WebCore::CSSSelector::extractPseudoType): Handle PseudoMediaControlsVolumeSliderMuteButton.
2249
2250         * css/CSSSelector.h:
2251         (WebCore::CSSSelector::): Define PseudoMediaControlsVolumeSliderMuteButton.
2252
2253         * css/CSSValueKeywords.in: Define media-volume-slider-mute-button.
2254
2255         * css/mediaControls.css: 
2256         (audio::-webkit-media-controls-volume-slider-mute-button, video::-webkit-media-controls-volume-slider-mute-button): New.
2257
2258         * css/mediaControlsQuickTime.css:
2259         (audio::-webkit-media-controls-panel, video::-webkit-media-controls-panel): Add "overflow: visible" 
2260         so volume slider pop-up will be visible.
2261
2262         (audio::-webkit-media-controls-mute-button, video::-webkit-media-controls-mute-button): Make 
2263         mute button 14x12 instead of 15x14 so it matches the size in the volume slider.
2264
2265         (audio::-webkit-media-controls-volume-slider-container, video::-webkit-media-controls-volume-slider-container):
2266         (audio::-webkit-media-controls-volume-slider, video::-webkit-media-controls-volume-slider):
2267         (audio::-webkit-media-controls-volume-slider-mute-button, video::-webkit-media-controls-volume-slider-mute-button): 
2268         New.
2269
2270         * platform/ThemeTypes.h: Define MediaVolumeSliderMuteButtonPart.
2271
2272         * rendering/MediaControlElements.cpp:
2273         (WebCore::MediaControlInputElement::MediaControlInputElement): Handle MEDIA_CONTROLS_VOLUME_SLIDER_MUTE_BUTTON.
2274         (WebCore::MediaControlMuteButtonElement::MediaControlMuteButtonElement): Add ButtonLocation 
2275         parameter both mute buttons can be created.
2276         (WebCore::MediaControlMuteButtonElement::create): Ditto.
2277
2278         * rendering/MediaControlElements.h: Define MediaVolumeSliderMuteButton and ButtonLocation.
2279
2280         * rendering/RenderMedia.cpp: 
2281         (WebCore::RenderMedia::styleDidChange): Update volume slider mute button.
2282         (WebCore::RenderMedia::createMuteButton): Pass ButtonLocation parameter to MediaControlMuteButtonElement::create.
2283         (WebCore::RenderMedia::createVolumeSliderMuteButton): New.
2284         (WebCore::RenderMedia::updateControls): Deal with volume slider mute button.
2285         (WebCore::RenderMedia::updateVolumeSliderContainer): Don't assume all ports want to position
2286         the volume slider in the same place, call new volumeSliderOffsetFromMuteButton theme function.
2287         (WebCore::RenderMedia::forwardEvent): Handle volume slider mute button.
2288         * rendering/RenderMedia.h:
2289
2290         * rendering/RenderTheme.cpp: 
2291         (WebCore::RenderTheme::paint): Handle MediaVolumeSliderMuteButtonPart.
2292         (WebCore::RenderTheme::volumeSliderOffsetFromMuteButton): New, return location previously hard
2293         coded in RenderMedia::updateVolumeSliderContainer.
2294         * rendering/RenderTheme.h:
2295
2296         * rendering/RenderThemeMac.h:
2297         * rendering/RenderThemeMac.mm:
2298         (WebCore::RenderThemeMac::adjustMediaSliderThumbSize): Deal with the volume slider thumb.
2299         (WebCore::RenderThemeMac::paintMediaVolumeSliderContainer): New, call the WKSI paint function 
2300         for this button.
2301         (WebCore::RenderThemeMac::paintMediaVolumeSliderTrack): Ditto.
2302         (WebCore::RenderThemeMac::paintMediaVolumeSliderThumb): Ditto.
2303         (WebCore::RenderThemeMac::shouldRenderMediaControlPart): Draw the volume slider parts when
2304         using the new controller UI for a movie that has audio.
2305         (WebCore::RenderThemeMac::volumeSliderOffsetFromMuteButton): New, position the volume slider 
2306         directly above the mute button.
2307
2308         * rendering/style/RenderStyleConstants.h: Define MEDIA_CONTROLS_VOLUME_SLIDER_MUTE_BUTTON.
2309
2310 2010-07-08  Antonio Gomes  <tonikitoo@webkit.org>
2311
2312         Unreviewed complementary fix for r62815
2313
2314         One last time change made me blind to a "!" in a if statment.
2315
2316         * editing/EditorCommand.cpp:
2317         (WebCore::executeToggleStyle):
2318
2319 2010-07-07  Ojan Vafai  <ojan@chromium.org>
2320
2321         Reviewed by Darin Adler.
2322
2323         Regression: Selection anchor + focus swap when arrow keys after setBaseAndExtent
2324         https://bugs.webkit.org/show_bug.cgi?id=32605
2325
2326         Only have selections be non-directional when they are set via the mouse.
2327         Otherwise, all selections are directional.
2328
2329         Test: editing/selection/extend-forward-after-set-base-and-extent.html
2330
2331         * WebCore.base.exp:
2332         * editing/SelectionController.cpp:
2333         (WebCore::SelectionController::setSelection):
2334         * editing/SelectionController.h:
2335         (WebCore::SelectionController::setSelection):
2336         * page/EventHandler.cpp:
2337         (WebCore::EventHandler::selectClosestWordFromMouseEvent):
2338         (WebCore::EventHandler::selectClosestWordOrLinkFromMouseEvent):
2339         (WebCore::EventHandler::handleMousePressEventTripleClick):
2340         (WebCore::EventHandler::handleMousePressEventSingleClick):
2341         (WebCore::EventHandler::updateSelectionForMouseDrag):
2342
2343 2010-07-08  Antonio Gomes  <tonikitoo@webkit.org>
2344
2345         Reviewed by Ojan Vafai.
2346
2347         Refactor platform dependent editing behavior code out of Settings (part II)
2348         https://bugs.webkit.org/show_bug.cgi?id=39854
2349
2350         Mac port is currently the only port relying on a style to be present on the start
2351         of a selection to consider the style as applied or not. All other ports
2352         have to have the style present in all text node of the selection for such.
2353
2354         Patch makes situations that depend on this check like this to be controlled by
2355         the EditingBehavior class.
2356
2357         Refactoring, so no new tests needed.
2358
2359         * editing/EditingBehavior.h:
2360         (WebCore::EditingBehavior::shouldToggleStyleBasedOnStartOfSelection):
2361         * editing/EditorCommand.cpp:
2362         (WebCore::executeToggleStyle):
2363
2364 2010-07-08  Anders Carlsson  <andersca@apple.com>
2365
2366         Reviewed by Sam Weinig.
2367
2368         Add stubbed out PluginView class
2369         https://bugs.webkit.org/show_bug.cgi?id=41879
2370
2371         Export functions needed by WebKit2.
2372
2373         * WebCore.exp.in:
2374
2375 2010-07-08  Ben Murdoch  <benm@google.com>
2376
2377         Reviewed by Pavel Feldman.
2378
2379         Fix build break in V8ConsoleCustom.cpp
2380         https://bugs.webkit.org/show_bug.cgi?id=40825
2381
2382         No new tests as just fixing a build break.
2383
2384         * bindings/v8/custom/V8ConsoleCustom.cpp: Guard profilesAccessorGetter
2385             appropriately.
2386
2387 2010-07-08  Sam Weinig  <sam@webkit.org>
2388
2389         Reviewed by Anders Carlsson.
2390
2391         Pass Page to BackForwardControllerClient::createBackForwardList since it
2392         may be called before implementations of BackForwardControllerClient have
2393         access to a Page.
2394
2395         * history/BackForwardController.cpp:
2396         (WebCore::BackForwardController::BackForwardController):
2397         * history/BackForwardControllerClient.h:
2398
2399 2010-07-08  Sheriff Bot  <webkit.review.bot@gmail.com>
2400
2401         Unreviewed, rolling out r62778.
2402         http://trac.webkit.org/changeset/62778
2403         https://bugs.webkit.org/show_bug.cgi?id=41866
2404
2405         Broke svg/W3C-SVG-1.1/text-intro-05-t.svg in debug (Requested
2406         by pfeldman on #webkit).
2407
2408         * platform/graphics/chromium/FontLinux.cpp:
2409         (WebCore::TextRunWalker::TextRunWalker):
2410         (WebCore::TextRunWalker::~TextRunWalker):
2411
2412 2010-07-08  Vitaly Repeshko  <vitalyr@chromium.org>
2413
2414         Reviewed by Pavel Feldman.
2415
2416         Fix adoptRef usage violation in IDBObjectStoreRequest
2417         https://bugs.webkit.org/show_bug.cgi?id=41869
2418
2419         * storage/IDBObjectStoreRequest.cpp:
2420         (WebCore::IDBObjectStoreRequest::IDBObjectStoreRequest):
2421
2422 2010-07-08  Sam Weinig  <sam@webkit.org>
2423
2424         Another chromium build fix.
2425
2426         * history/BackForwardListChromium.cpp:
2427         (WebCore::BackForwardListImpl::goBack):
2428         (WebCore::BackForwardListImpl::goForward):
2429         (WebCore::BackForwardListImpl::backListWithLimit):
2430         (WebCore::BackForwardListImpl::forwardListWithLimit):
2431         (WebCore::BackForwardListImpl::containsItem):
2432         (WebCore::BackForwardListImpl::removeItem):
2433
2434 2010-07-08  Sam Weinig  <sam@webkit.org>
2435
2436         Fix Chromium build.
2437
2438         * WebCore.gyp/WebCore.gyp:
2439
2440 2010-07-08  Sam Weinig  <sam@webkit.org>
2441
2442         Reviewed by Anders Carlsson.
2443
2444         Patch for https://bugs.webkit.org/show_bug.cgi?id=41826
2445         Convert BackForwardList to an abstract base class and add BackForwardListImpl
2446         as the concrete implementation of it.
2447
2448         * CMakeLists.txt:
2449         * GNUmakefile.am:
2450         * WebCore.exp.in:
2451         * WebCore.gypi:
2452         * WebCore.pro:
2453         * WebCore.vcproj/WebCore.vcproj:
2454         * WebCore.xcodeproj/project.pbxproj:
2455         * history/BackForwardController.cpp:
2456         (WebCore::BackForwardController::BackForwardController):
2457         * history/BackForwardController.h:
2458         * history/BackForwardList.cpp: Removed.
2459         * history/BackForwardList.h:
2460         (WebCore::BackForwardList::~BackForwardList):
2461         (WebCore::BackForwardList::isBackForwardListImpl):
2462         (WebCore::BackForwardList::BackForwardList):
2463         * history/BackForwardListChromium.cpp:
2464         * history/BackForwardListImpl.cpp: Copied from WebCore/history/BackForwardList.cpp.
2465         * history/BackForwardListImpl.h: Copied from WebCore/history/BackForwardList.h.
2466         (WebCore::BackForwardListImpl::isBackForwardListImpl):
2467
2468 2010-07-08  Vitaly Repeshko  <vitalyr@chromium.org>
2469
2470         Reviewed by Pavel Feldman.
2471
2472         Fix adoptRef usage violations (mostly in chromium)
2473         https://bugs.webkit.org/show_bug.cgi?id=41863
2474
2475         * bindings/v8/V8DOMWrapper.cpp:
2476         (WebCore::V8DOMWrapper::wrapNativeNodeFilter):
2477         * bindings/v8/V8NodeFilterCondition.h:
2478         (WebCore::V8NodeFilterCondition::create):
2479         * bindings/v8/custom/V8HTMLAllCollectionCustom.cpp:
2480         (WebCore::getNamedItems):
2481         * bindings/v8/custom/V8HTMLCollectionCustom.cpp:
2482         (WebCore::getNamedItems):
2483         * bindings/v8/custom/V8HTMLFormElementCustom.cpp:
2484         (WebCore::V8HTMLFormElement::namedPropertyGetter):
2485         * bindings/v8/custom/V8HTMLSelectElementCustom.cpp:
2486         (WebCore::V8HTMLSelectElement::namedPropertyGetter):
2487         * bindings/v8/custom/V8NamedNodesCollection.h:
2488         (WebCore::V8NamedNodesCollection::create):
2489         (WebCore::V8NamedNodesCollection::V8NamedNodesCollection):
2490         * storage/IDBDatabaseRequest.cpp:
2491         (WebCore::IDBDatabaseRequest::IDBDatabaseRequest):
2492         * storage/IndexedDatabaseRequest.cpp:
2493         (WebCore::IndexedDatabaseRequest::IndexedDatabaseRequest):
2494
2495 2010-07-08  Xan Lopez  <xlopez@igalia.com>
2496
2497         Reviewed by Gustavo Noronha.
2498
2499         Silence a few noisy build rules.
2500
2501         * GNUmakefile.am:
2502
2503 2010-07-08  Andreas Kling  <andreas.kling@nokia.com>
2504
2505         Reviewed by Simon Hausmann.
2506
2507         [Qt] Canvas putImageData() resets painter state
2508         https://bugs.webkit.org/show_bug.cgi?id=41827
2509
2510         Use drawImage() to copy pixels in putImageData() instead of QPixmap::operator=
2511
2512         Test: fast/canvas/canvas-state-intact-after-putImageData.html
2513
2514         * platform/graphics/qt/ImageBufferQt.cpp:
2515         (WebCore::putImageData):
2516
2517 2010-07-08  Andrey Kosyakov  <caseq@chromium.org>
2518
2519         Reviewed by Pavel Feldman.
2520
2521         Web Inspector: Do not invoke shortcuts popup upon bare '?' if it's typed
2522         into an input field.
2523         https://bugs.webkit.org/show_bug.cgi?id=41760
2524
2525         * inspector/front-end/inspector.js:
2526         (WebInspector.documentKeyDown):
2527
2528 2010-07-08  Patrick Gansterer  <paroga@paroga.com>
2529
2530         Reviewed by Kent Tamura.
2531
2532         [WINCE] Buildfix for EventHandler
2533         https://bugs.webkit.org/show_bug.cgi?id=41829
2534
2535         SM_MENUDROPALIGNMENT isn't supported on WinCE.
2536
2537         * page/EventHandler.cpp:
2538         (WebCore::EventHandler::sendContextMenuEventForKey):
2539
2540 2010-07-08  Xiaomei Ji  <xji@chromium.org>
2541
2542         Reviewed by David Levin.
2543
2544         Fix characters with unicode-bidi-mirror property are not correctly
2545         mirrored in Linux.
2546         https://bugs.webkit.org/show_bug.cgi?id=41305
2547
2548         Since harfbuzz does not do mirroring, chromium should iterate each
2549         character in the string and mirror it if needed before passing the
2550         string to harfbuzz for shaping.
2551
2552         Test: fast/text/international/bidi-mirror-he-ar.html
2553
2554         * platform/graphics/chromium/FontLinux.cpp:
2555         (WebCore::TextRunWalker::TextRunWalker):
2556         (WebCore::TextRunWalker::~TextRunWalker):
2557
2558 2010-07-08  Nikolas Zimmermann  <nzimmermann@rim.com>
2559
2560         Reviewed by Rob Buis & Dirk Schulze.
2561
2562         SVG text transformed incorrectly when a transform and gradient applied
2563         https://bugs.webkit.org/show_bug.cgi?id=41563
2564
2565         When rendering filled/stroked text with a gradient, RenderSVGResourceGradient creates a mask image,
2566         renders the text into the image, and then clips the current context against that image buffer,
2567         and filling a rect afterwards. This happened in the wrong coordinate space so far. Don't actually
2568         try to compute the right transformation matrix (which failed), but extract it directly from the
2569         GraphicsContext and apply this transformation to the mask image context.
2570
2571         Fixes pixellation in svg/batik/text/textEffect3.svg and makes the new test svg/custom/text-rotated-gradient.svg pass.
2572         Only affects CoreGraphics platforms.
2573
2574         Test: svg/custom/text-rotated-gradient.svg
2575
2576         * rendering/RenderSVGResourceGradient.cpp:
2577         (WebCore::absoluteTransformFromContext):
2578         (WebCore::createMaskAndSwapContextForTextGradient):
2579         (WebCore::clipToTextMask):
2580
2581 2010-07-07  Alexander Pavlov  <apavlov@chromium.org>
2582
2583         Reviewed by Darin Fisher.
2584
2585         [Chromium] Crash when re-entering message loop
2586         https://bugs.webkit.org/show_bug.cgi?id=41697
2587
2588         A Chromium-specific test case will be submitted into Chromium shortly.
2589
2590         * page/PageGroupLoadDeferrer.cpp:
2591         (WebCore::PageGroupLoadDeferrer::PageGroupLoadDeferrer):
2592         * page/PageGroupLoadDeferrer.h:
2593         (WebCore::PageGroupLoadDeferrer::nextDeferrer):
2594
2595 2010-07-07  Yury Semikhatsky  <yurys@chromium.org>
2596
2597         Reviewed by Pavel Feldman.
2598
2599         Web Inspector: skip breakpoints when script is already paused
2600         https://bugs.webkit.org/show_bug.cgi?id=41768
2601
2602         Test: inspector/debugger-no-nested-pause.html
2603
2604         * bindings/v8/ScriptDebugServer.cpp:
2605         (WebCore::ScriptDebugServer::handleV8DebugEvent):
2606
2607 2010-07-08  Eric Seidel  <eric@webkit.org>
2608
2609         Reviewed by Adam Barth.
2610
2611         WebCore/benchmarks/parser/html-parser.html spends a lot of time in deprecatedParseURL
2612         https://bugs.webkit.org/show_bug.cgi?id=41807
2613
2614         Wow.  This was an awful bug.  We were always taking the slow case
2615         every time we parsed a URL.  This is about a 10% speedup on our
2616         parsing benchmark, and might cause as much as a 1% speedup for Apple's
2617         PLT (even though I can't run that).
2618
2619         We still spend a lot of time in deprecatedParseURL.  We might consider
2620         inlining it if its being kept around much longer.
2621
2622         No behavioral change, just fixing a broken optimization.
2623
2624         * css/CSSHelper.cpp:
2625         (WebCore::deprecatedParseURL):
2626          - We only need to strip characters <= '\r', not >.
2627
2628 2010-07-08  Yury Semikhatsky  <yurys@chromium.org>
2629
2630         Reviewed by Pavel Feldman.
2631
2632         Web Inspector: factor out common part of debugger tests
2633         https://bugs.webkit.org/show_bug.cgi?id=41836
2634
2635         * bindings/js/ScriptDebugServer.cpp:
2636         (WebCore::ScriptDebugServer::recompileAllJSFunctions): postpone script recompilation if JS stack is not empty.
2637
2638 2010-07-08  Patrick Gansterer  <paroga@paroga.com>
2639
2640         Reviewed by Kent Tamura.
2641
2642         Buildfix for !ENABLE(SVG_ANIMATION) after r51567.
2643         https://bugs.webkit.org/show_bug.cgi?id=41803
2644
2645         * svg/SVGDocumentExtensions.cpp:
2646         (WebCore::SVGDocumentExtensions::sampleAnimationAtTime):
2647
2648 2010-07-07  Pavel Podivilov  <podivilov@chromium.org>
2649
2650         Reviewed by Nate Chapin.
2651
2652         [V8] Fix document wrapper memory leak in bindings.
2653         https://bugs.webkit.org/show_bug.cgi?id=41771
2654
2655         * bindings/v8/custom/V8HTMLDocumentCustom.cpp:
2656         (WebCore::V8HTMLDocument::WrapInShadowObject): Do not create static persistent handle
2657         to shadowConstructor because it keeps the first context alive forever.
2658
2659 2010-07-07  Andreas Kling  <andreas.kling@nokia.com>
2660
2661         Reviewed by Tor Arne Vestbø.
2662
2663         [Qt] Enable smooth pixmap transforms by default
2664         https://bugs.webkit.org/show_bug.cgi?id=41774
2665
2666         * platform/graphics/qt/GraphicsContextQt.cpp:
2667         (WebCore::GraphicsContextPlatformPrivate::GraphicsContextPlatformPrivate):
2668         (WebCore::GraphicsContext::setImageInterpolationQuality):
2669
2670 2010-07-07  Jesus Sanchez-Palencia  <jesus@webkit.org>
2671
2672         Reviewed by Antti Koivisto.
2673
2674         [Qt] Missing include to build QtWebKit with WebKit2 in MediaPlayerPrivatePhonon.cpp
2675         https://bugs.webkit.org/show_bug.cgi?id=41767
2676
2677         * platform/graphics/qt/MediaPlayerPrivatePhonon.cpp: Add Logging.h include
2678
2679 2010-07-07  Pavel Podivilov  <podivilov@chromium.org>
2680
2681         Reviewed by Yury Semikhatsky.
2682
2683         Web Inspector: Move v8-related script offset conversion to DebuggerScript.js
2684         https://bugs.webkit.org/show_bug.cgi?id=41755
2685
2686         * bindings/v8/ScriptDebugServer.cpp:
2687         (WebCore::ScriptDebugServer::dispatchDidParseSource):
2688
2689 2010-07-07  Nicolas Weber  <thakis@chromium.org>
2690
2691         Reviewed by Dimitri Glazkov.
2692         https://bugs.webkit.org/show_bug.cgi?id=41580
2693
2694         Fix rendering of radial gradients in skia if both points of the
2695         gradient are the same and r0 > 0.
2696
2697         Test: fast/gradients/radial-centered.html
2698
2699         * platform/graphics/skia/GradientSkia.cpp:
2700         (WebCore::Gradient::platformGradient):
2701
2702 2010-06-18  MORITA Hajime  <morrita@google.com>
2703
2704         Reviewed by Ojan Vafai.
2705
2706         https://bugs.webkit.org/show_bug.cgi?id=26526
2707         Add support for input events (oninput) to contentEditable elements
2708         
2709         Made a default event handler on the Node to dispatch an input event when
2710         webkitEditableContentChanged arrived.
2711         
2712         Test: fast/events/event-input-contentEditable.html
2713
2714         * dom/Node.cpp:
2715         (WebCore::Node::defaultEventHandler):
2716
2717 2010-07-07  Darin Adler  <darin@apple.com>
2718
2719         Reviewed by Anders Carlsson.
2720
2721         Fix adoptRef assertion failures caused by stack-allocated ResourceHandle objects
2722         https://bugs.webkit.org/show_bug.cgi?id=41823
2723
2724         * platform/network/android/ResourceHandleAndroid.cpp:
2725         (WebCore::ResourceHandle::loadResourceSynchronously): Use adoptRef and new instead
2726         of allocating an object on the stack.
2727         * platform/network/curl/ResourceHandleCurl.cpp:
2728         (WebCore::ResourceHandle::loadResourceSynchronously): Ditto.
2729         * platform/network/qt/ResourceHandleQt.cpp:
2730         (WebCore::ResourceHandle::loadResourceSynchronously): Ditto.
2731
2732         * platform/network/soup/ResourceHandleSoup.cpp:
2733         (WebCore::ResourceHandle::loadResourceSynchronously): Use create instead of
2734         allocating an object on the stack.
2735
2736 2010-07-07  Joseph Pecoraro  <joepeck@webkit.org>
2737
2738         Rubber-stamped by Pavel Feldman.
2739
2740         Web Inspector: All RefPtr arguments and return values should be changed to PassRefPtr according to common practice.
2741         https://bugs.webkit.org/show_bug.cgi?id=41759
2742
2743         * inspector/CodeGeneratorInspector.pm:
2744
2745 2010-07-07  Chris Fleizach  <cfleizach@apple.com>
2746
2747         Reviewed by Darin Adler.
2748
2749         AX: TextArea should return AXSelectedTextRange of 0,0 if the cursor is not in the text area
2750         https://bugs.webkit.org/show_bug.cgi?id=41810
2751
2752         Test: platform/mac/accessibility/selected-text-range-for-empty-textarea.html
2753
2754         * accessibility/mac/AccessibilityObjectWrapper.mm:
2755         (-[AccessibilityObjectWrapper accessibilityAttributeValue:]):
2756
2757 2010-07-07  Chris Fleizach  <cfleizach@apple.com>
2758
2759         Reviewed by Beth Dakin.
2760
2761         AX: Data table heuristics: assume data table if at least one row or column of TH cells exist
2762         https://bugs.webkit.org/show_bug.cgi?id=41806
2763
2764         Test: platform/mac/accessibility/table-with-row-col-of-headers.html
2765
2766         * accessibility/AccessibilityTable.cpp:
2767         (WebCore::AccessibilityTable::isTableExposableThroughAccessibility):
2768
2769 2010-07-07  Mark Rowe  <mrowe@apple.com>
2770
2771         Fix the build.
2772
2773         Xcode decided to change the reference type from "Relative to Build Product" to "Relative to Group"
2774         when I moved the file between folders. That's incredibly confusing so I'll file a bug against Xcode.
2775
2776         * WebCore.xcodeproj/project.pbxproj:
2777
2778 2010-07-07  Mark Rowe  <mrowe@apple.com>
2779
2780         Fix the MathML build.
2781
2782         * html/HTMLTreeBuilder.cpp:
2783         (WebCore::HTMLTreeBuilder::resetInsertionModeAppropriately): Fix a typo in a variable name.
2784
2785 2010-07-07  James Hawkins  <jhawkins@chromium.org>
2786
2787         Reviewed by Dimitri Glazkov.
2788
2789         Move setSuggestedValue() and suggestedValue() back to public as it's
2790         now used by WebInputElement.
2791         https://bugs.webkit.org/show_bug.cgi?id=41785
2792
2793         No new tests, as the method is only being used by Chromium's WebKit
2794         API.
2795
2796         * html/HTMLInputElement.h:
2797
2798 2010-07-07  Mark Rowe  <mrowe@apple.com>
2799
2800         Rubber-stamped by Sam Weinig.
2801
2802         Clean up the project file slightly.
2803
2804         Remove obsolete .exp files, add the new .exp.in file, and move ExportFileGenerator.cpp in to the Exports collection.
2805
2806         * WebCore.xcodeproj/project.pbxproj:
2807
2808 2010-07-07  James Robinson  <jamesr@chromium.org> and Vincent Scheib <schieb@chromium.org>
2809
2810         Reviewed by Dimitri Glazkov.
2811
2812         WebCore::GLES2Context should allow creating onscreen and offscreen contexts
2813         https://bugs.webkit.org/show_bug.cgi?id=41492
2814
2815         You should be able to create a GLES2Context for rendering both on screen
2816         and off screen (to a texture).  Chromium's layer renderer (used for compositing)
2817         does the former.  Currently WebGL does the latter by side stepping the
2818         WebCore::GLES2Context API completely.  Longer term it should move over to using this.
2819
2820         * platform/chromium/GLES2Context.h:
2821         * platform/graphics/chromium/LayerRendererChromium.cpp:
2822         (WebCore::LayerRendererChromium::initGL):
2823
2824 2010-07-07  Simon Fraser  <simon.fraser@apple.com>
2825
2826         Reviewed by Dan Bernstein.
2827
2828         Document::ownerElement() should return an HTMLFrameOwnerElement
2829         https://bugs.webkit.org/show_bug.cgi?id=41789
2830
2831         Change return type of RenderLayerCompositor::enclosingIFrameElement() to
2832         HTMLFrameOwnerElement*. No behavioral changes.
2833
2834         * rendering/RenderLayerCompositor.cpp:
2835         (WebCore::RenderLayerCompositor::shouldPropagateCompositingToEnclosingIFrame):
2836         (WebCore::RenderLayerCompositor::enclosingIFrameElement):
2837         * rendering/RenderLayerCompositor.h:
2838
2839 2010-07-07  Adam Barth  <abarth@webkit.org>
2840
2841         Reviewed by Sam Weinig.
2842
2843         Add reverseFind to Vector and deploy in HTML5 parser
2844         https://bugs.webkit.org/show_bug.cgi?id=41778
2845
2846         This patch moves reverseFind from begin an
2847         HTMLFormattingElementList-specific concept to begin a general Vector
2848         concept.  Also, instead of using Entry as the type for operator==, we
2849         now use elements directly.  The old code compiled because the Entry
2850         constructor wasn't explicit, which means we were churning refs on every
2851         comparison!
2852
2853         * html/HTMLFormattingElementList.cpp:
2854         (WebCore::HTMLFormattingElementList::find):
2855         (WebCore::HTMLFormattingElementList::bookmarkFor):
2856         (WebCore::HTMLFormattingElementList::insertAt):
2857         (WebCore::HTMLFormattingElementList::remove):
2858         * html/HTMLFormattingElementList.h:
2859         (WebCore::HTMLFormattingElementList::Entry::operator==):
2860         (WebCore::HTMLFormattingElementList::Entry::operator!=):
2861
2862 2010-07-07  Darin Adler  <darin@apple.com>
2863
2864         Reviewed by Adam Barth.
2865
2866         Turn on adoptRef assertion for RefCounted
2867         https://bugs.webkit.org/show_bug.cgi?id=41547
2868
2869         The WebCore part of this fixes all the assertions I saw in testing.
2870
2871         * html/FileReader.cpp:
2872         (WebCore::FileReader::readAsBinaryString): Added null checks.
2873         Callers from JavaScript can pass the wrong type, which becomes null.
2874         (WebCore::FileReader::readAsText): Ditto.
2875         (WebCore::FileReader::readAsDataURL): Ditto.
2876
2877         * html/FileStreamClient.h: Removed unneeded include.
2878
2879         * html/FileStreamProxy.cpp:
2880         (WebCore::FileStreamProxy::FileStreamProxy): Made inline and moved
2881         some of the code, including the ref, into the create function.
2882         (WebCore::FileStreamProxy::create): Moved some of the code from
2883         the constructor here. It's safe to ref once the object has been
2884         created and adopted.
2885
2886         * html/FileStreamProxy.h: Changed create function to no longer be
2887         inlined. Also removed an unneeded include.
2888
2889         * page/EventSource.cpp:
2890         (WebCore::EventSource::EventSource): Made inline, changed arguments
2891         and moved code that involves the need to ref this object into the
2892         create function. Also moved failure handling out there since it's
2893         cleaner to have a function that fails than a constructor. For
2894         example, the function can return 0.
2895         (WebCore::EventSource::create): Moved some of the code from the
2896         constructor here.
2897
2898         * page/EventSource.h: Removed unneeded includes. Made the
2899         creation function non-inline. Changed the arguments to the constructor.
2900
2901         * storage/StorageAreaSync.cpp:
2902         (WebCore::StorageAreaSync::StorageAreaSync): Made inline. Moved
2903         code that requires ref'ing this object out to the create function.
2904         (WebCore::StorageAreaSync::create): Moved some of the code from the
2905         constructor here.
2906
2907         * storage/StorageAreaSync.h: Removed unneeded includes. Changed
2908         the type of one of the constructor arguments from String to
2909         const String&.
2910
2911         * workers/SharedWorker.cpp:
2912         (WebCore::SharedWorker::SharedWorker): Made inline. Moved most of
2913         the setup code out of here into the create function.
2914         (WebCore::SharedWorker::create): Moved the code here.
2915
2916         * workers/SharedWorker.h: Removed unneeded includes. Made the
2917         create function non-inline. Marked the toSharedWorker override private
2918         to catch people doing an unnecessary virtual function call if they
2919         already have a SharedWorker*.
2920
2921         * workers/Worker.cpp:
2922         (WebCore::Worker::Worker): Made inline. Moved most of the setup code
2923         out of here into the create function.
2924         (WebCore::Worker::create): Moved the code here.
2925
2926         * workers/Worker.h: Made the create function non-inline. Changed
2927         the arguments to the constructor.
2928
2929 2010-07-07  Chris Fleizach  <cfleizach@apple.com>
2930
2931         Reviewed by Beth Dakin.
2932
2933         AX: when a node's role changes, the AX tree might need to be updated
2934         https://bugs.webkit.org/show_bug.cgi?id=41784
2935
2936         Test: platform/mac/accessibility/update-children-when-aria-role-changes.html
2937
2938         * accessibility/AccessibilityRenderObject.cpp:
2939         (WebCore::AccessibilityRenderObject::AccessibilityRenderObject):
2940         (WebCore::AccessibilityRenderObject::updateAccessibilityRole):
2941
2942 2010-07-07  Chris Fleizach  <cfleizach@apple.com>
2943
2944         Reviewed by Beth Dakin.
2945
2946         AX: when an element uses role="text" but no aria-label, it should default to textUnderElement()
2947         https://bugs.webkit.org/show_bug.cgi?id=41780
2948
2949         Test: platform/mac/accessibility/static-text-role-uses-text-under-element.html
2950
2951         * accessibility/AccessibilityRenderObject.cpp:
2952         (WebCore::AccessibilityRenderObject::stringValue):
2953            Fallback to textUnderElement() if text() does not return any explicitly set text.
2954         (WebCore::AccessibilityRenderObject::accessibilityIsIgnored): 
2955            Remove erroneous check to ignore text elements that didn't have a explicit text value.
2956
2957 2010-07-07  Simon Fraser  <simon.fraser@apple.com>
2958
2959         Reviewed by Darin Adler.
2960
2961         Document::ownerElement() should return an HTMLFrameOwnerElement
2962         https://bugs.webkit.org/show_bug.cgi?id=41789
2963
2964         Change the return type of Document::ownerElement() to an HTMLFrameOwnerElement, for better
2965         type checking. No behavioral changes.
2966
2967         * accessibility/mac/AccessibilityObjectWrapper.mm:
2968         * dom/Document.cpp:
2969         (WebCore::Document::ownerElement):
2970         * dom/Document.h:
2971         * editing/SelectionController.cpp:
2972         * rendering/RenderBoxModelObject.cpp:
2973         * rendering/RenderLayer.cpp:
2974         * rendering/RenderView.cpp:
2975
2976 2010-07-07  Simon Fraser  <simon.fraser@apple.com>
2977
2978         Reviewed by Dan Bernstein.
2979
2980         REGRESSION: GMail becomes blank after closing other tabs
2981         https://bugs.webkit.org/show_bug.cgi?id=40421
2982
2983         RenderLayerCompositor's attach/detachRootPlatformLayer methods use
2984         setNeedsStyleRecalc() in order to trigger acclerated compositing layers
2985         to be hooked together across iframe boundaries. However, it was possible
2986         for these to get called while inside of Document::recalcStyle(), which
2987         is bad because it can cause the recalc to fail to get processed.
2988         
2989         Fix this by using the existing queuePostAttachCallback() functionality
2990         to delay the call to setNeedsStyleRecalc() if post-attach callbacks 
2991         are suspended (indicating that we're inside recalcStyle()).
2992         
2993         No new tests because I wasn't able to make a test that shows the problem.
2994
2995         * dom/ContainerNode.h: Make queuePostAttachCallback() public.
2996         * dom/ContainerNode.cpp:
2997         (WebCore::ContainerNode::postAttachCallbacksAreSuspended): Added; returns
2998         whether s_attachDepth is non-zero.
2999
3000         * rendering/RenderLayerCompositor.cpp:
3001         (WebCore::RenderLayerCompositor::attachRootPlatformLayer): Call scheduleNeedsStyleRecalc()
3002         instead of setNeedsStyleRecalc().
3003         (WebCore::RenderLayerCompositor::detachRootPlatformLayer): Ditto.
3004         (WebCore::needsStyleRecalcCallback): Here we call setNeedsStyleRecalc().
3005         (WebCore::RenderLayerCompositor::scheduleNeedsStyleRecalc):
3006         (WebCore::RenderLayerCompositor::notifyIFramesOfCompositingChange):
3007         * rendering/RenderLayerCompositor.h: Add scheduleNeedsStyleRecalc().
3008
3009 2010-07-06  Simon Fraser  <simon.fraser@apple.com>
3010
3011         Reviewed by Dan Bernstein.
3012
3013         Incorrect compositing order with negative z-index
3014         https://bugs.webkit.org/show_bug.cgi?id=38959
3015         
3016         When painting the contents of compositing layers whose phase is "GraphicsLayerPaintBackground"
3017         (indicating that they are used for the background of elements with negative z-index children),
3018         we need to paint the non-composited negative-z-order descendants at the end of the background phase, so
3019         they appear behind composited negative-z-order descendants.
3020
3021         Test: compositing/z-order/negative-z-index.html
3022
3023         * rendering/RenderLayerBacking.cpp:
3024         (WebCore::RenderLayerBacking::paintIntoLayer):
3025
3026 2010-07-06  Simon Fraser  <simon.fraser@apple.com>
3027
3028         Reviewed by Dan Bernstein.
3029
3030         Incorrect compositing order with negative z-index
3031         https://bugs.webkit.org/show_bug.cgi?id=38959
3032
3033         Prerequisite: adopt new paintList() method that was added when painting compositing layer contents.
3034         No behavioral changes.
3035
3036         * rendering/RenderLayerBacking.cpp:
3037         (WebCore::RenderLayerBacking::paintIntoLayer):
3038
3039 2010-07-07  Adam Barth  <abarth@webkit.org>
3040
3041         Unreviewed.
3042
3043         Fix ASSERT in Eric's previous patch.  I think he tested his patch in
3044         Release and not Debug.
3045
3046         * html/HTMLFormattingElementList.h:
3047         (WebCore::HTMLFormattingElementList::Entry::Entry):
3048         (WebCore::HTMLFormattingElementList::findIndex):
3049
3050 2010-07-07  Eric Seidel  <eric@webkit.org>
3051
3052         Reviewed by Adam Barth.
3053
3054         Grease the TreeBuilder's lightning
3055         https://bugs.webkit.org/show_bug.cgi?id=41756
3056
3057         Brings the new TreeBuilder from 7s to 3s on the parser benchmark.
3058         This makes performance comparable to the old parser.
3059
3060         We have not begun to fight!  There is so much fat left on these bones.
3061
3062         * html/HTMLFormattingElementList.cpp:
3063         (WebCore::HTMLFormattingElementList::find):
3064         (WebCore::HTMLFormattingElementList::bookmarkFor):
3065         (WebCore::HTMLFormattingElementList::insertAt):
3066         (WebCore::HTMLFormattingElementList::remove):
3067         * html/HTMLFormattingElementList.h:
3068         (WebCore::HTMLFormattingElementList::findIndex):
3069
3070 2010-07-07  Eric Seidel  <eric@webkit.org>
3071
3072         Reviewed by Adam Barth.
3073
3074         HTMLTreeBuilder is way too slow
3075         https://bugs.webkit.org/show_bug.cgi?id=41754
3076
3077         This takes us from 14s to 7s on our parsing benchmark.
3078         That's still much slower than the old tree builder, but there
3079         is a huge amount of fat left to trim.
3080
3081         Vector<T> wasn't able to inline all the Entry functions when
3082         they were buried in the cpp.  Turns out the active formatting elements
3083         list is very hot.
3084
3085         I'm not sure Vector<T> is going to be the right data structure for us
3086         in the end, but it has done alright for bring-up.
3087
3088         * html/HTMLFormattingElementList.cpp:
3089         * html/HTMLFormattingElementList.h:
3090         (WebCore::HTMLFormattingElementList::Entry::Entry):
3091         (WebCore::HTMLFormattingElementList::Entry::~Entry):
3092         (WebCore::HTMLFormattingElementList::Entry::isMarker):
3093         (WebCore::HTMLFormattingElementList::Entry::element):
3094         (WebCore::HTMLFormattingElementList::Entry::replaceElement):
3095         (WebCore::HTMLFormattingElementList::Entry::operator==):
3096         (WebCore::HTMLFormattingElementList::Entry::operator!=):
3097
3098 2010-07-06  Darin Adler  <darin@apple.com>
3099
3100         Reviewed by Adam Barth.
3101
3102         More OwnPtr work, including making clear set the pointer to 0 before deletion
3103         https://bugs.webkit.org/show_bug.cgi?id=41727
3104
3105         * WebCore.exp.in: Updated.
3106
3107         * css/CSSSelector.h:
3108         (WebCore::CSSSelector::RareData::RareData): Use adoptPtr.
3109         (WebCore::CSSSelector::createRareData): Ditto.
3110         * dom/SpaceSplitString.h:
3111         (WebCore::SpaceSplitString::SpaceSplitString): Ditto.
3112         (WebCore::SpaceSplitString::set): Ditto.
3113         * history/CachedFrame.cpp:
3114         (WebCore::CachedFrame::CachedFrame): Ditto.
3115         (WebCore::CachedFrame::setCachedFramePlatformData): Ditto.
3116
3117         * history/CachedFrame.h: Use PassOwnPtr.
3118
3119         * loader/appcache/ApplicationCacheGroup.cpp:
3120         (WebCore::CallCacheListenerTask::create): Use adoptPtr.
3121         * loader/appcache/ApplicationCacheStorage.cpp:
3122         (WebCore::ApplicationCacheStorage::storeCopyOfCache): Ditto.
3123         * platform/PurgeableBuffer.h:
3124         (WebCore::PurgeableBuffer::create): Ditto.
3125         * platform/graphics/GlyphMetricsMap.h:
3126         (WebCore::::locatePageSlowCase): Ditto.
3127         * platform/graphics/GraphicsLayer.h:
3128         (WebCore::AnimationValue::AnimationValue): Ditto.
3129         (WebCore::TransformAnimationValue::TransformAnimationValue): Ditto.
3130         * platform/graphics/MediaPlayer.h:
3131         (WebCore::MediaPlayer::create): Ditto.
3132         * platform/graphics/SimpleFontData.h:
3133         (WebCore::SimpleFontData::boundsForGlyph): Ditto.
3134         * platform/mac/PurgeableBufferMac.cpp:
3135         (WebCore::PurgeableBuffer::create): Ditto.
3136         * rendering/InlineFlowBox.h:
3137         (WebCore::InlineFlowBox::setHorizontalOverflowPositions): Ditto.
3138         (WebCore::InlineFlowBox::setVerticalOverflowPositions): Ditto.
3139         * rendering/RootInlineBox.h:
3140         (WebCore::RootInlineBox::floats): Ditto.
3141         * rendering/style/RenderStyle.h:
3142         (WebCore::InheritedFlags::inheritAnimations): Ditto.
3143         (WebCore::InheritedFlags::inheritTransitions): Ditto.
3144
3145         * rendering/style/SVGRenderStyleDefs.h: Use PassOwnPtr.
3146
3147 2010-07-07  Sam Weinig  <sam@webkit.org>
3148
3149         Reviewed by Anders Carlsson.
3150
3151         Patch for https://bugs.webkit.org/show_bug.cgi?id=41772
3152         Add basic piping for BackForwardControllerClient.
3153
3154         * WebCore.exp.in:
3155         * WebCore.xcodeproj/project.pbxproj:
3156         * history/BackForwardController.cpp:
3157         (WebCore::BackForwardController::BackForwardController):
3158         * history/BackForwardControllerClient.h:
3159         * page/Page.cpp:
3160         (WebCore::Page::Page):
3161         * page/Page.h:
3162         * svg/graphics/SVGImage.cpp:
3163         (WebCore::SVGImage::dataChanged):
3164
3165 2010-07-07  Lucas De Marchi  <lucas.demarchi@profusion.mobi>
3166
3167         Unreviewed build fix after r62577.
3168
3169         [EFL] build fix after r62577
3170         https://bugs.webkit.org/show_bug.cgi?id=41764
3171
3172         No new functionality so no new tests.
3173
3174 2010-07-07  Gustavo Noronha Silva  <gustavo.noronha@collabora.co.uk>
3175
3176         Reviewed by Xan Lopez.
3177
3178         [GTK] Scrollbars sometimes go dead and stop scrolling the view
3179         https://bugs.webkit.org/show_bug.cgi?id=41711
3180
3181         Rework the setGtkAdjustment function slightly, make it have an
3182         early return, and be more readable.
3183
3184         * platform/gtk/ScrollViewGtk.cpp:
3185         (WebCore::ScrollView::setGtkAdjustments):
3186
3187 2010-07-07  Kristian Monsen  <kristianm@google.com>
3188
3189         Reviewed by Steve Block.
3190
3191         Updating Android to use the UserGestureIndicator class instead of Android specific code.
3192         The rest of the changes are in parts of the Android port which are yet to be upstreamed.
3193
3194         No change in behavior, so no new test.
3195
3196         * platform/network/android/ResourceRequest.h:
3197         (WebCore::ResourceRequest::ResourceRequest):
3198         (WebCore::ResourceRequest::doUpdateResourceRequest):
3199
3200 2010-07-07  Andy Estes  <aestes@apple.com>
3201
3202         Reviewed by Adam Barth.
3203
3204         Allow a beforeload listener to prevent loading of images in <object> tags.
3205         https://bugs.webkit.org/show_bug.cgi?id=41027
3206         <rdar://problem/8120596>
3207
3208         Tests: fast/dom/beforeload/image-object-before-load-innerHTML.html
3209                fast/dom/beforeload/image-object-before-load.html
3210
3211         * html/HTMLObjectElement.cpp:
3212         (WebCore::HTMLObjectElement::attach): Do not call
3213         RenderImage::setCachedImage() at the end of attach(). Instead, allow
3214         this to happen conditionally after beforeload is dispatched.
3215         * loader/ImageLoader.cpp:
3216         (WebCore::ImageLoader::dispatchPendingBeforeLoadEvent): Render fallback
3217         content if an object's load was cancelled.
3218
3219 2010-07-06  Nikolas Zimmermann  <nzimmermann@rim.com>
3220
3221         Reviewed by Dirk Schulze.
3222
3223         <use> on <font-face> causes crashes, if SVGUseElement gets detached
3224         https://bugs.webkit.org/show_bug.cgi?id=41621
3225
3226         Do not call removeFromMappedElementSheet() from the SVGFontFaceElement destructor,
3227         as that can potentially cause the element to be reattached while destructing.
3228
3229         In order to fix the crash in the testcase, the order of calling the base-class detach
3230         method in SVGUseElement and the instance/shadow tree destruction has to be reversed,
3231         matching the order in removedFromDocument().
3232
3233         Test: svg/custom/use-font-face-crash.svg
3234
3235         * svg/SVGFontFaceElement.cpp:
3236         (WebCore::SVGFontFaceElement::~SVGFontFaceElement): Remove removeFromMappedElementSheet() call.
3237         * svg/SVGUseElement.cpp:
3238         (WebCore::SVGUseElement::detach): Reverse order of calling base-class detach method and instance/shadow tree destruction.
3239
3240 2010-07-07  Steve Block  <steveblock@google.com>
3241
3242         Reviewed by Adam Barth.
3243
3244         document.createEvent() should support DeviceOrientationEvent
3245         https://bugs.webkit.org/show_bug.cgi?id=41618
3246
3247         Tests: fast/dom/DeviceOrientation/create-event.html
3248
3249         * dom/Document.cpp:
3250         (WebCore::Document::createEvent):
3251
3252 2010-07-07  Yury Semikhatsky  <yurys@chromium.org>
3253
3254         Reviewed by Pavel Feldman.
3255
3256         Web Inspector: show functions from internal browser scripts on the call stack if they
3257         are there.
3258         https://bugs.webkit.org/show_bug.cgi?id=41762
3259
3260         * English.lproj/localizedStrings.js:
3261         * inspector/front-end/CallStackSidebarPane.js:
3262         (WebInspector.CallStackSidebarPane.prototype.update):
3263
3264 2010-07-07  Ilya Tikhonovsky  <loislo@chromium.org>
3265
3266         Reviewed by Pavel Feldman.
3267
3268         Web Inspector: All RefPtr arguments and return values should be
3269         changed to PassRefPtr whenever it is possible according to common practice.
3270         https://bugs.webkit.org/show_bug.cgi?id=41759
3271
3272         * inspector/CodeGeneratorInspector.pm:
3273         * inspector/InspectorTimelineAgent.cpp:
3274         (WebCore::InspectorTimelineAgent::pushGCEventRecords):
3275         (WebCore::InspectorTimelineAgent::didInstallTimer):
3276         (WebCore::InspectorTimelineAgent::didRemoveTimer):
3277         (WebCore::InspectorTimelineAgent::didScheduleResourceRequest):
3278         (WebCore::InspectorTimelineAgent::willSendResourceRequest):
3279         (WebCore::InspectorTimelineAgent::didFinishLoadingResource):
3280         (WebCore::InspectorTimelineAgent::didMarkTimeline):
3281         (WebCore::InspectorTimelineAgent::didMarkDOMContentEvent):
3282         (WebCore::InspectorTimelineAgent::didMarkLoadEvent):
3283         (WebCore::InspectorTimelineAgent::addRecordToTimeline):
3284         (WebCore::InspectorTimelineAgent::setHeapSizeStatistic):
3285         (WebCore::InspectorTimelineAgent::pushCurrentRecord):
3286         * inspector/InspectorTimelineAgent.h:
3287         (WebCore::InspectorTimelineAgent::TimelineRecordEntry::TimelineRecordEntry):
3288         * inspector/TimelineRecordFactory.cpp:
3289         (WebCore::TimelineRecordFactory::createGenericRecord):
3290         (WebCore::TimelineRecordFactory::createGCEventData):
3291         (WebCore::TimelineRecordFactory::createFunctionCallData):
3292         (WebCore::TimelineRecordFactory::createEventDispatchData):
3293         (WebCore::TimelineRecordFactory::createGenericTimerData):
3294         (WebCore::TimelineRecordFactory::createTimerInstallData):
3295         (WebCore::TimelineRecordFactory::createXHRReadyStateChangeData):
3296         (WebCore::TimelineRecordFactory::createXHRLoadData):
3297         (WebCore::TimelineRecordFactory::createEvaluateScriptData):
3298         (WebCore::TimelineRecordFactory::createMarkTimelineData):
3299         (WebCore::TimelineRecordFactory::createScheduleResourceRequestData):
3300         (WebCore::TimelineRecordFactory::createResourceSendRequestData):
3301         (WebCore::TimelineRecordFactory::createResourceReceiveResponseData):
3302         (WebCore::TimelineRecordFactory::createResourceFinishData):
3303         (WebCore::TimelineRecordFactory::createReceiveResourceData):
3304         (WebCore::TimelineRecordFactory::createPaintData):
3305         (WebCore::TimelineRecordFactory::createParseHTMLData):
3306         * inspector/TimelineRecordFactory.h:
3307
3308 2010-07-07  Antti Koivisto  <koivisto@iki.fi>
3309
3310         Reviewed by Kenneth Rohde Christiansen.
3311         
3312         [Qt] Initial WebKit2 implementation
3313         https://bugs.webkit.org/show_bug.cgi?id=40233
3314
3315         Export some event conversion functions needed for WebKit2.
3316
3317         * platform/PlatformKeyboardEvent.h:
3318         * platform/qt/PlatformKeyboardEventQt.cpp:
3319         (WebCore::keyIdentifierForQtKeyCode):
3320         (WebCore::windowsKeyCodeForKeyEvent):
3321
3322 2010-07-06  Steve Block  <steveblock@google.com>
3323
3324         Reviewed by Adam Barth.
3325
3326         DeviceOrientationEventConstructor should be exposed at window.DeviceOrientationEvent
3327         https://bugs.webkit.org/show_bug.cgi?id=41685
3328
3329         Test: fast/dom/Window/window-properties-device-orientation.html
3330               fast/dom/DeviceOrientation/window-property.txt
3331
3332         * page/DOMWindow.idl:
3333
3334 2010-07-07  Sheriff Bot  <webkit.review.bot@gmail.com>
3335
3336         Unreviewed, rolling out r62645.
3337         http://trac.webkit.org/changeset/62645
3338         https://bugs.webkit.org/show_bug.cgi?id=41757
3339
3340         "Broken inspector tests on several platforms" (Requested by
3341         eseidel on #webkit).
3342
3343         * inspector/front-end/ScriptView.js:
3344         (WebInspector.ScriptView.prototype._addBreakpoint):
3345         * inspector/front-end/ScriptsPanel.js:
3346         (WebInspector.ScriptsPanel):
3347         (WebInspector.ScriptsPanel.prototype._resourceLoadingFinished):
3348         (WebInspector.ScriptsPanel.prototype._breakpointAdded):
3349         (WebInspector.ScriptsPanel.prototype._scriptOrResourceForURLAndLine):
3350         (WebInspector.ScriptsPanel.prototype._showScriptOrResource):
3351         (WebInspector.ScriptsPanel.prototype._toggleBreakpointsClicked):
3352         * inspector/front-end/SourceView.js:
3353         (WebInspector.SourceView.prototype._addBreakpoint):
3354
3355 2010-07-07  Steve Block  <steveblock@google.com>
3356
3357         Reviewed by Adam Barth.
3358
3359         JavaInstanceV8 needs to push a local reference frame to avoid table overflow.
3360         https://bugs.webkit.org/show_bug.cgi?id=41516
3361
3362         This change is required to help prevent the Java virtual machine from
3363         running out of local references. The Java virtual machine supports only
3364         a limited number of local references. Normally, local references are
3365         cleared when the native method returns.
3366
3367         This change adds calls to PushLocalFrame() and PopLocalFrame() around
3368         each invocation of a method on JavaInstance. These calls instruct the
3369         virtual machine to clear local references when the frame is popped.
3370         This means that in the case where many calls to JavaInstance methods
3371         are made within the same native call, local references are cleared as
3372         soon as they are no longer needed, thus conserving references.
3373
3374         This change mirrors exactly JavaInstanceJSC.
3375
3376         No new tests.
3377
3378         * bridge/jni/v8/JavaInstanceV8.cpp:
3379         (JavaInstance::virtualBegin):
3380         (JavaInstance::virtualEnd):
3381         * bridge/jni/v8/JavaInstanceV8.h:
3382
3383 2010-07-06  Yury Semikhatsky  <yurys@chromium.org>
3384
3385         Reviewed by Pavel Feldman.
3386
3387         Inspector should remember the size of sidebars set by the user
3388         https://bugs.webkit.org/show_bug.cgi?id=19207
3389
3390         * inspector/front-end/AbstractTimelinePanel.js:
3391         (WebInspector.AbstractTimelinePanel):
3392         * inspector/front-end/AuditsPanel.js:
3393         (WebInspector.AuditsPanel):
3394         * inspector/front-end/ConsolePanel.js:
3395         (WebInspector.ConsolePanel):
3396         * inspector/front-end/ElementsPanel.js:
3397         (WebInspector.ElementsPanel.prototype.rightSidebarResizerDragEnd):
3398         * inspector/front-end/Panel.js:
3399         (WebInspector.Panel):
3400         (WebInspector.Panel.prototype.get toolbarItemClass):
3401         (WebInspector.Panel.prototype.show):
3402         (WebInspector.Panel.prototype.createSidebar):
3403         (WebInspector.Panel.prototype.get _sidebarWidthSettingName):
3404         (WebInspector.Panel.prototype._endSidebarDragging):
3405         (WebInspector.Panel.prototype.updateSidebarWidth):
3406         (WebInspector.Panel.prototype.restoreSidebarWidth):
3407         (WebInspector.Panel.prototype.saveSidebarWidth):
3408         * inspector/front-end/ProfilesPanel.js:
3409         (WebInspector.ProfilesPanel):
3410         * inspector/front-end/ResourcesPanel.js:
3411         (WebInspector.ResourcesPanel):
3412         (WebInspector.ResourcesPanel.prototype.resourceTrackingWasEnabled):
3413         * inspector/front-end/ScriptsPanel.js:
3414         (WebInspector.ScriptsPanel):
3415         (WebInspector.ScriptsPanel.prototype._endSidebarResizeDrag):
3416         (WebInspector.ScriptsPanel.prototype._sidebarResizeDrag):
3417         (WebInspector.ScriptsPanel.prototype.setSidebarWidth):
3418         * inspector/front-end/Settings.js:
3419         (WebInspector.populateApplicationSettings):
3420         * inspector/front-end/StoragePanel.js:
3421         (WebInspector.StoragePanel):
3422         * inspector/front-end/TimelinePanel.js:
3423         (WebInspector.TimelinePanel):
3424
3425 2010-07-05  Steve Block  <steveblock@google.com>
3426
3427         Reviewed by Adam Barth.
3428
3429         DeviceOrientation should be renamed to DeviceOrientationController
3430         https://bugs.webkit.org/show_bug.cgi?id=41608
3431
3432         No new tests, renaming only.
3433
3434         * Android.mk:
3435         * CMakeLists.txt:
3436         * GNUmakefile.am:
3437         * WebCore.gypi:
3438         * WebCore.pro:
3439         * WebCore.vcproj/WebCore.vcproj:
3440         * WebCore.xcodeproj/project.pbxproj:
3441         * dom/DeviceOrientation.cpp: Removed.
3442         * dom/DeviceOrientation.h: Removed.
3443         * dom/DeviceOrientationController.cpp: Copied from WebCore/dom/DeviceOrientation.cpp.
3444         (WebCore::DeviceOrientationController::DeviceOrientationController):
3445         (WebCore::DeviceOrientationController::onDeviceOrientationChange):
3446         * dom/DeviceOrientationController.h: Copied from WebCore/dom/DeviceOrientation.h.
3447         * page/Page.cpp:
3448         (WebCore::Page::Page):
3449         * page/Page.h:
3450         (WebCore::Page::deviceOrientationController):
3451
3452 2010-07-07  Pavel Podivilov  <podivilov@chromium.org>
3453
3454         Reviewed by Yury Semikhatsky.
3455
3456         Web Inspector: do not activate all breakpoints on page reload
3457         https://bugs.webkit.org/show_bug.cgi?id=41461
3458
3459         Test: inspector/debugger-breakpoints-not-activated-on-reload.html
3460
3461         * inspector/front-end/ScriptView.js:
3462         (WebInspector.ScriptView.prototype._addBreakpoint):
3463         * inspector/front-end/ScriptsPanel.js:
3464         (WebInspector.ScriptsPanel):
3465         (WebInspector.ScriptsPanel.prototype._breakpointAdded):
3466         (WebInspector.ScriptsPanel.prototype.toggleBreakpointsClicked):
3467         * inspector/front-end/SourceView.js:
3468         (WebInspector.SourceView.prototype._addBreakpoint):
3469
3470 2010-07-07  Nikolas Zimmermann  <nzimmermann@rim.com>
3471
3472         Not reviewed. Rollout r62633 - test crashes after some last-minute modifications.
3473
3474         * svg/SVGFontFaceElement.cpp:
3475         (WebCore::SVGFontFaceElement::~SVGFontFaceElement):
3476
3477 2010-07-07  Eric Seidel  <eric@webkit.org>
3478
3479         Reviewed by Adam Barth.
3480
3481         Fix </html> handling in before head mode
3482         https://bugs.webkit.org/show_bug.cgi?id=41752
3483
3484         Covered by html5lib/runner.html.
3485
3486         * html/HTMLTreeBuilder.cpp:
3487         (WebCore::HTMLTreeBuilder::processEndTag):
3488
3489 2010-07-07  Eric Seidel  <eric@webkit.org>
3490
3491         Reviewed by Adam Barth.
3492
3493         </div> pops too many <div> elements in body
3494         https://bugs.webkit.org/show_bug.cgi?id=41751
3495
3496         Another missing return!
3497
3498         While tracking this down I added some debugging code to both the
3499         open elements stack and the active formatting elements list.
3500
3501         I also unwrapped a very long || chain to make it more readable.
3502
3503         * html/HTMLElementStack.cpp:
3504         (WebCore::HTMLElementStack::show):
3505         * html/HTMLElementStack.h:
3506         * html/HTMLFormattingElementList.cpp:
3507         (WebCore::HTMLFormattingElementList::show):
3508         * html/HTMLFormattingElementList.h:
3509         * html/HTMLTreeBuilder.cpp:
3510         (WebCore::HTMLTreeBuilder::processEndTagForInBody):
3511
3512 2010-07-06  Eric Seidel  <eric@webkit.org>
3513
3514         Reviewed by Adam Barth.
3515
3516         HTMLTreeBuilder needs to handle <table><input>
3517         https://bugs.webkit.org/show_bug.cgi?id=41744
3518
3519         * html/HTMLTreeBuilder.cpp:
3520         (WebCore::HTMLTreeBuilder::processStartTagForInTable):
3521
3522 2010-07-06  Eric Seidel  <eric@webkit.org>
3523
3524         Reviewed by Adam Barth.
3525
3526         Teach HTMLTreeBuilder how to handle <button><button>
3527         https://bugs.webkit.org/show_bug.cgi?id=41743
3528
3529         * html/HTMLTreeBuilder.cpp:
3530         (WebCore::HTMLTreeBuilder::processStartTagForInBody):
3531
3532 2010-07-06  Eric Seidel  <eric@webkit.org>
3533
3534         Reviewed by Adam Barth.
3535
3536         Teach TreeBuilder how to handle <form><form>
3537         https://bugs.webkit.org/show_bug.cgi?id=41741
3538
3539         * html/HTMLTreeBuilder.cpp:
3540         (WebCore::HTMLTreeBuilder::processStartTagForInBody):
3541
3542 2010-07-06  Eric Seidel  <eric@webkit.org>
3543
3544         Reviewed by Adam Barth.
3545
3546         <table> should auto-close <p> when not in quirks mode
3547         https://bugs.webkit.org/show_bug.cgi?id=41740
3548
3549         This is already covered by multiple tests in html5lib/runner.html
3550
3551         * html/HTMLTreeBuilder.cpp:
3552         (WebCore::HTMLTreeBuilder::processStartTagForInBody):
3553
3554 2010-07-06  Eric Seidel  <eric@webkit.org>
3555
3556         Reviewed by Adam Barth.
3557
3558         Make "in body" character handling reconstruct active formatting elements
3559         https://bugs.webkit.org/show_bug.cgi?id=41739
3560
3561         This fixes a bunch of tests, and due to some other bug in our
3562         formatting code, regresses two.  I'll fix that bug in a separate
3563         patch.
3564
3565         * html/HTMLTreeBuilder.cpp:
3566         (WebCore::HTMLTreeBuilder::processCharacter):
3567
3568 2010-07-06  Nikolas Zimmermann  <nzimmermann@rim.com>
3569
3570         Reviewed by Darin Adler.
3571
3572         <use> on <font-face> causes crashes, if SVGUseElement gets detached
3573         https://bugs.webkit.org/show_bug.cgi?id=41621
3574
3575         Do not call removeFromMappedElementSheet() from the destructor, as the call to document()->updateStyleSelector() that can potentially
3576         cause the element to be reattached while destructing. It's not needed at all, because removedFromDocument() is called before destruction,
3577         which already calls removeFromMappedElementSheet() - at this point it's still safe to update the style selector.
3578
3579         The crash is reproducable when using <use> on <font-face>.
3580
3581         Test: svg/custom/use-font-face-crash.svg
3582
3583         * svg/SVGFontFaceElement.cpp:
3584         (WebCore::SVGFontFaceElement::~SVGFontFaceElement):
3585
3586 2010-07-07  Hayato Ito  <hayato@chromium.org>
3587
3588         Reviewed by Darin Adler.
3589
3590         Rolling out 'page-break-inside:avoid' part of the r54929.
3591         Rebased the related layout tests, which are now expected to fail, as well.
3592
3593         https://bugs.webkit.org/show_bug.cgi?id=41532
3594
3595         * rendering/RenderBlock.cpp:
3596         (WebCore::RenderBlock::paintChildren):
3597
3598 2010-07-07  Mark Rowe  <mrowe@apple.com>
3599
3600         Fix failures in a handful of Java-related tests.
3601
3602         * WebCore.exp.in: Remove an extraneous ENABLE_ from an #if so that the condition
3603         evalutes to true when we expect it to.
3604
3605 2010-07-06  Pavel Feldman  <pfeldman@chromium.org>
3606
3607         Reviewed by Joseph Pecoraro.
3608
3609         Web Inspector: provide line numbers for inline styles.
3610
3611         https://bugs.webkit.org/show_bug.cgi?id=41691
3612
3613         Test: inspector/styles-source-lines-inline.html
3614
3615         * css/CSSParser.cpp:
3616         (WebCore::CSSParser::parseSheet):
3617         * css/CSSParser.h:
3618         * css/CSSStyleSheet.cpp:
3619         (WebCore::CSSStyleSheet::parseString):
3620         (WebCore::CSSStyleSheet::parseStringAtLine):
3621         * css/CSSStyleSheet.h:
3622         * dom/StyleElement.cpp:
3623         (WebCore::StyleElement::process):
3624         (WebCore::StyleElement::createSheet):
3625         * dom/StyleElement.h:
3626         * html/HTMLStyleElement.cpp:
3627         (WebCore::HTMLStyleElement::HTMLStyleElement):
3628         (WebCore::HTMLStyleElement::finishParsingChildren):
3629         * html/HTMLStyleElement.h:
3630         * inspector/InspectorCSSStore.cpp:
3631         (WebCore::InspectorCSSStore::getRuleRangesForStyleSheet):
3632         * inspector/InspectorDOMAgent.cpp:
3633         (WebCore::InspectorDOMAgent::buildObjectForRule):
3634         * inspector/front-end/DOMAgent.js:
3635         (WebInspector.CSSStyleDeclaration.parseRule):
3636         * inspector/front-end/StylesSidebarPane.js:
3637         (WebInspector.StylePropertiesSection.else.linkifyUncopyable):
3638         (WebInspector.StylePropertiesSection):
3639
3640 2010-07-06  Dan Bernstein  <mitz@apple.com>
3641
3642         Reviewed by Jon Honeycutt.
3643
3644         <rdar://problem/8163651> Wrong value type being supplied for kCTTypesetterOptionForcedEmbeddingLevel key
3645
3646         No change in behavior, so no new test.
3647
3648         * platform/graphics/mac/ComplexTextControllerCoreText.cpp:
3649         (WebCore::ComplexTextController::collectComplexTextRunsForCharactersCoreText): Use CFNumber instances
3650         as the values of the kCTTypesetterOptionForcedEmbeddingLevel key, instead of CFBooleans.
3651
3652 2010-07-06  Mark Rowe  <mrowe@apple.com>
3653
3654         Fix production builds on Mac OS X.
3655
3656         * DerivedSources.make: Don't try and generate the export file during the installhdrs build phase.
3657         It's not useful until much later in the build process, and it's not possible to generate it until
3658         after the tool has been compiled anyway.
3659         * WebCore.xcodeproj/project.pbxproj: Tweak how the ICU and forwarding headers are copied so that it
3660         works in production builds once more.
3661
3662 2010-07-06  Steve Falkenburg  <sfalken@apple.com>
3663
3664         Reviewed by Simon Fraser.
3665
3666         Expose URL matching from WebUserContentURLPattern
3667         https://bugs.webkit.org/show_bug.cgi?id=41726
3668         <rdar://problem/7910144>
3669
3670         * WebCore.exp.in: Export UserContentURLPattern::matches for use in WebKit.
3671
3672 2010-07-06  Eric Seidel  <eric@webkit.org>
3673
3674         Reviewed by Adam Barth.
3675
3676         Fix <nobr><nobr> case in HTMLTreeBuilder
3677         https://bugs.webkit.org/show_bug.cgi?id=41735
3678
3679         We were both not handling <nobr> correctly, as well as
3680         never hitting the <nobr> case because our formatting
3681         elements check was overzealous.
3682
3683         * html/HTMLTreeBuilder.cpp:
3684         (WebCore::HTMLTreeBuilder::processStartTagForInBody):
3685
3686 2010-07-06  Adam Barth  <abarth@webkit.org>
3687
3688         Reviewed by Eric Seidel.
3689
3690         Implement start table tag in table mode
3691         https://bugs.webkit.org/show_bug.cgi?id=41736
3692
3693         * html/HTMLTreeBuilder.cpp:
3694         (WebCore::HTMLTreeBuilder::processStartTagForInTable):
3695         (WebCore::HTMLTreeBuilder::processTableEndTagForInTable):
3696         (WebCore::HTMLTreeBuilder::processEndTagForInTable):
3697         * html/HTMLTreeBuilder.h:
3698
3699 2010-07-06  Adam Barth  <abarth@webkit.org>
3700
3701         Reviewed by Eric Seidel.
3702
3703         Implement "act as if the banana has been peeled" (for lack of a better term)
3704         https://bugs.webkit.org/show_bug.cgi?id=41734
3705
3706         The spec has a tricky passage that says "whenever a node would be
3707         inserted into the current node, it must instead be foster parented."
3708         This patch attempts to implement that requirement.
3709
3710         * html/HTMLConstructionSite.cpp:
3711         (WebCore::HTMLConstructionSite::attach):
3712         (WebCore::HTMLConstructionSite::HTMLConstructionSite):
3713         (WebCore::HTMLConstructionSite::insertHTMLHtmlElement):
3714         (WebCore::HTMLConstructionSite::insertHTMLHeadElement):
3715         (WebCore::HTMLConstructionSite::insertHTMLBodyElement):
3716         (WebCore::HTMLConstructionSite::insertScriptElement):
3717         (WebCore::HTMLConstructionSite::fosterParent):
3718         * html/HTMLConstructionSite.h:
3719         (WebCore::HTMLConstructionSite::RedirectToFosterParentGuard::RedirectToFosterParentGuard):
3720         (WebCore::HTMLConstructionSite::RedirectToFosterParentGuard::~RedirectToFosterParentGuard):
3721         * html/HTMLTreeBuilder.cpp:
3722         (WebCore::HTMLTreeBuilder::processStartTagForInTable):
3723         (WebCore::HTMLTreeBuilder::processEndTagForInTable):
3724         * html/HTMLTreeBuilder.h:
3725
3726 2010-07-06  Eric Seidel  <eric@webkit.org>
3727
3728         Reviewed by Adam Barth.
3729
3730         Fix </optgroup> in "in select" mode
3731         https://bugs.webkit.org/show_bug.cgi?id=41733
3732
3733         I had to add a oneBelowTop() accessor.
3734         I added some ASSERTs after looking at the fragment case
3735         documentation.  I'm now convinced that top() and oneBeforeTop()
3736         should never be NULL, so we just ASSERT they aren't.
3737
3738         This is a very obscure corner of the HTML spec, but at least
3739         we have a test for it.  This makes one more html5lib test pass.
3740
3741         * html/HTMLConstructionSite.h:
3742         (WebCore::HTMLConstructionSite::oneBelowTop):
3743         * html/HTMLElementStack.cpp:
3744         (WebCore::HTMLElementStack::topRecord):
3745         (WebCore::HTMLElementStack::top):
3746         (WebCore::HTMLElementStack::oneBelowTop):
3747         * html/HTMLElementStack.h:
3748         * html/HTMLTreeBuilder.cpp:
3749         (WebCore::HTMLTreeBuilder::processEndTag):
3750
3751 2010-07-06  Eric Seidel  <eric@webkit.org>
3752
3753         Reviewed by Adam Barth.
3754
3755         Fix "in row" "Anything else" handling of end tags
3756         https://bugs.webkit.org/show_bug.cgi?id=41731
3757
3758         I also fixed </br> handling to follow the spec more closely
3759         since it implementation was old and didn't use all our
3760         new processFake* hotness.
3761
3762         This is already covered by 3 tests, however we don't
3763         pass them yet, due to lack of table foster parenting
3764         code (which Adam is working on as we speak).
3765
3766         * html/HTMLTreeBuilder.cpp:
3767         (WebCore::HTMLTreeBuilder::processEndTagForInBody):
3768         (WebCore::HTMLTreeBuilder::processEndTag):
3769
3770 2010-07-06  Mark Rowe  <mrowe@apple.com>
3771
3772         Fix clean builds on Mac OS X.
3773
3774         Since "Copy Forwarding and ICU Headers" was split in to a separate target
3775         from the WebCore framework it no longer had automatica access to the
3776         PRIVATE_HEADERS_FOLDER_PATH variable. This caused it to copy the header
3777         files in to the wrong location. Fix that by giving PRIVATE_HEADERS_FOLDER_PATH
3778         its correct value and ensuring that the path exists.
3779
3780         * WebCore.xcodeproj/project.pbxproj:
3781
3782 2010-07-06  Sam Weinig  <sam@webkit.org>
3783
3784         Fix Gtk build.
3785
3786         * GNUmakefile.am:
3787
3788 2010-07-06  Mark Rowe  <mrowe@apple.com>
3789
3790         Tiger build fix.
3791
3792         Don't use WebCore's LDFLAGS when building WebCoreExportFileGenerator.
3793
3794         * WebCore.xcodeproj/project.pbxproj:
3795
3796 2010-07-06  Eric Seidel  <eric@webkit.org>
3797
3798         Reviewed by Adam Barth.
3799
3800         "In cell" does not correctly handle <td><tr> or <td><td>
3801         https://bugs.webkit.org/show_bug.cgi?id=41729
3802
3803         This change is mostly cleanup to try and prevent forgetting
3804         tag name checks in the future by using inlines to reduce
3805         copy/paste code.
3806
3807         3 little bugs in InCellMode:
3808          - Missing trTag from the long or statement (reason for the cleanup)
3809          - Used || instead of &&
3810          - Forgot to reprocess the tag after closeTheCell()
3811
3812         * html/HTMLTreeBuilder.cpp:
3813         (WebCore::HTMLTreeBuilder::processStartTagForInBody):
3814         (WebCore::HTMLTreeBuilder::processStartTagForInTable):
3815         (WebCore::HTMLTreeBuilder::processStartTag):
3816         (WebCore::HTMLTreeBuilder::processEndTagForInTable):
3817         (WebCore::HTMLTreeBuilder::processEndTag):
3818
3819 2010-07-06  Adam Barth  <abarth@webkit.org>
3820
3821         Reviewed by Eric Seidel.
3822
3823         Move fosterParent to HTMLConstructionSite
3824         https://bugs.webkit.org/show_bug.cgi?id=41728
3825
3826         This paves the way for handling the default case of InTableMode
3827         properly.
3828
3829         * html/HTMLConstructionSite.cpp:
3830         (WebCore::HTMLConstructionSite::fosterParent):
3831         * html/HTMLConstructionSite.h:
3832         * html/HTMLTreeBuilder.cpp:
3833         (WebCore::HTMLTreeBuilder::callTheAdoptionAgency):
3834
3835 2010-07-06  Adam Barth  <abarth@webkit.org>
3836
3837         Rubber-stamped by Eric Seidel.
3838
3839         Move HTMLConstructionSite to its own file.  No behavior change.
3840
3841         * Android.mk:
3842         * CMakeLists.txt:
3843         * GNUmakefile.am:
3844         * WebCore.gypi:
3845         * WebCore.pro:
3846         * WebCore.vcproj/WebCore.vcproj:
3847         * WebCore.xcodeproj/project.pbxproj:
3848         * html/HTMLConstructionSite.cpp: Copied from WebCore/html/HTMLTreeBuilder.cpp.
3849         (WebCore::HTMLNames::hasImpliedEndTag):
3850         (WebCore::HTMLConstructionSite::attach):
3851         (WebCore::HTMLConstructionSite::HTMLConstructionSite):
3852         (WebCore::HTMLConstructionSite::~HTMLConstructionSite):
3853         (WebCore::HTMLConstructionSite::insertHTMLHtmlStartTagBeforeHTML):
3854         (WebCore::HTMLConstructionSite::mergeAttributesFromTokenIntoElement):
3855         (WebCore::HTMLConstructionSite::insertHTMLHtmlStartTagInBody):
3856         (WebCore::HTMLConstructionSite::insertHTMLBodyStartTagInBody):
3857         * html/HTMLConstructionSite.h: Copied from WebCore/html/HTMLTreeBuilder.h.
3858         * html/HTMLTreeBuilder.cpp:
3859         * html/HTMLTreeBuilder.h:
3860
3861 2010-07-06  Mark Rowe  <mrowe@apple.com>
3862
3863         Build fix.
3864
3865         The path to the generated file needs to be relative to the built product directory,
3866         not to the project file.
3867
3868         * WebCore.xcodeproj/project.pbxproj:
3869
3870 2010-07-06  Mark Rowe  <mrowe@apple.com>
3871
3872         Reviewed by Darin Adler.
3873
3874         <http://webkit.org/b/41725> REGRESSION(r62283): No longer possible to build for Tiger from SnowLeopard due to export file shenanigans
3875
3876         Change how the exports file is generated. Rather than having the logic for which symbols
3877         to include be encoded in a Makefile that concatenated separate files, the logic is now
3878         encoded as #if's in the base exports file. This ensures that the exact same preprocessor
3879         settings are used when determining which symbols to export as when compiling the code that
3880         defines the symbols.
3881
3882         * DerivedSources.make: Remove logic that determines which export files to concatenate.
3883         Replace with an invocation of WebCoreExportFileGenerator.
3884         * WebCore.3DRendering.exp: Removed.
3885         * WebCore.AcceleratedCompositing.exp: Removed.
3886         * WebCore.ClientBasedGeolocation.exp: Removed.
3887         * WebCore.ContextMenus.exp: Removed.
3888         * WebCore.DashboardSupport.exp: Removed.
3889         * WebCore.DragSupport.exp: Removed.
3890         * WebCore.Geolocation.exp: Removed.
3891         * WebCore.Inspector.exp: Removed.
3892         * WebCore.JNI.exp: Removed.
3893         * WebCore.NPAPI.exp: Removed.
3894         * WebCore.OrientationEvents.exp: Removed.
3895         * WebCore.PluginHostProcess.exp: Removed.
3896         * WebCore.ProtectionSpaceAuthCallback.exp: Removed.
3897         * WebCore.SVG.Animation.exp: Removed.
3898         * WebCore.SVG.Filters.exp: Removed.
3899         * WebCore.SVG.ForeignObject.exp: Removed.
3900         * WebCore.SVG.exp: Removed.
3901         * WebCore.Tiger.exp: Removed.
3902         * WebCore.Video.exp: Removed.
3903         * WebCore.VideoProxy.exp: Removed.
3904         * WebCore.exp.in: Renamed from WebCore/WebCore.base.exp. Merge the individual .exp
3905         files in, wrapping the appropriate parts with the necessary #if's.
3906         * WebCore.xcodeproj/project.pbxproj:
3907         * make-exports-file-generator: Added.
3908
3909 2010-07-06  Sam Weinig  <sam@webkit.org>
3910
3911         Reviewed by Anders Carlsson.
3912
3913         Patch for https://bugs.webkit.org/show_bug.cgi?id=41723
3914         Add BackForwardController class
3915
3916         - Add BackForwardController which currently only forwards to the existing
3917           BackForwardList implementation. This is a first step toward allowing
3918           for a proxying BackForwardList implementation.
3919
3920         * CMakeLists.txt:
3921         * GNUmakefile.am:
3922         * WebCore.base.exp:
3923         * WebCore.gypi:
3924         * WebCore.pro:
3925         * WebCore.vcproj/WebCore.vcproj:
3926         * WebCore.xcodeproj/project.pbxproj:
3927         * history/BackForwardController.cpp: Added.
3928         (WebCore::BackForwardController::BackForwardController):
3929         (WebCore::BackForwardController::~BackForwardController):
3930         * history/BackForwardController.h: Added.
3931         (WebCore::BackForwardController::client):
3932         (WebCore::BackForwardController::list):
3933         * history/BackForwardControllerClient.h: Added.
3934         (WebCore::BackForwardControllerClient::~BackForwardControllerClient):
3935         * page/Page.cpp:
3936         (WebCore::Page::Page):
3937         (WebCore::Page::~Page):
3938         (WebCore::Page::backForwardList):
3939         (WebCore::Page::goBack):
3940         (WebCore::Page::goForward):
3941         (WebCore::Page::canGoBackOrForward):
3942         (WebCore::Page::goBackOrForward):
3943         (WebCore::Page::getHistoryLength):
3944         * page/Page.h:
3945
3946 2010-07-06  Gustavo Noronha Silva  <gustavo.noronha@collabora.co.uk>
3947
3948         Unreviewed.
3949
3950         Fixes regression on API test by disabling the scrollbars before
3951         configuring the adjustments.
3952
3953         * platform/gtk/ScrollViewGtk.cpp:
3954         (WebCore::ScrollView::setGtkAdjustments):
3955
3956 2010-07-06  Eric Seidel  <eric@webkit.org>
3957
3958         Reviewed by Adam Barth.
3959
3960         Add support for <li>, <dd> and <dt> nested tag closing
3961         https://bugs.webkit.org/show_bug.cgi?id=41720
3962
3963         li, dd, and dt have nearly identical logic in HTML5, so I
3964         attempted to make them share as much code as possible.
3965
3966         This fixes a bunch of subtests for html5lib/runner.html
3967
3968         * html/HTMLTreeBuilder.cpp:
3969         (WebCore::HTMLTreeBuilder::processCloseWhenNestedTag):
3970         (WebCore::HTMLTreeBuilder::processStartTagForInBody):
3971         (WebCore::HTMLTreeBuilder::processAnyOtherEndTagForInBody):
3972         (WebCore::HTMLTreeBuilder::furthestBlockForFormattingElement):
3973         * html/HTMLTreeBuilder.h:
3974
3975 2010-07-06  Adam Barth  <abarth@webkit.org>
3976
3977         Reviewed by Eric Seidel.
3978
3979         Factor HTMLConstructionSite out of HTMLTreeBuilder
3980         https://bugs.webkit.org/show_bug.cgi?id=41716
3981
3982         The HTMLContructionSite is the model object on which the
3983         HTMLTreeBuilder (a controller) acts.
3984
3985         No behavior change.  I'll move this class into its own file in a
3986         followup patch.
3987
3988         * html/HTMLFormattingElementList.h:
3989         (WebCore::HTMLFormattingElementList::at):
3990         * html/HTMLTreeBuilder.cpp:
3991         (WebCore::HTMLTreeBuilder::HTMLTreeBuilder):
3992         (WebCore::HTMLConstructionSite::HTMLConstructionSite):
3993         (WebCore::HTMLTreeBuilder::passTokenToLegacyParser):
3994         (WebCore::HTMLTreeBuilder::processDoctypeToken):
3995         (WebCore::HTMLConstructionSite::insertHTMLHtmlStartTagBeforeHTML):
3996         (WebCore::HTMLConstructionSite::mergeAttributesFromTokenIntoElement):
3997         (WebCore::HTMLConstructionSite::insertHTMLHtmlStartTagInBody):
3998         (WebCore::HTMLConstructionSite::insertHTMLBodyStartTagInBody):
3999         (WebCore::HTMLTreeBuilder::processFakePEndTagIfPInScope):
4000         (WebCore::HTMLTreeBuilder::processIsindexStartTagForInBody):
4001         (WebCore::HTMLTreeBuilder::processStartTagForInBody):
4002         (WebCore::HTMLTreeBuilder::processColgroupEndTagForInColumnGroup):
4003         (WebCore::HTMLTreeBuilder::closeTheCell):
4004         (WebCore::HTMLTreeBuilder::processStartTagForInTable):
4005         (WebCore::HTMLTreeBuilder::processStartTag):
4006         (WebCore::HTMLTreeBuilder::processBodyEndTagForInBody):
4007         (WebCore::HTMLTreeBuilder::processAnyOtherEndTagForInBody):
4008         (WebCore::HTMLTreeBuilder::furthestBlockForFormattingElement):
4009         (WebCore::HTMLTreeBuilder::findFosterParentFor):
4010         (WebCore::HTMLTreeBuilder::callTheAdoptionAgency):
4011         (WebCore::HTMLTreeBuilder::resetInsertionModeAppropriately):
4012         (WebCore::HTMLTreeBuilder::processEndTagForInBody):
4013         (WebCore::HTMLTreeBuilder::processCaptionEndTagForInCaption):
4014         (WebCore::HTMLTreeBuilder::processTrEndTagForInRow):
4015         (WebCore::HTMLTreeBuilder::processEndTagForInTable):
4016         (WebCore::HTMLTreeBuilder::processEndTag):
4017         (WebCore::HTMLTreeBuilder::processComment):
4018         (WebCore::HTMLTreeBuilder::processCharacter):
4019         (WebCore::HTMLTreeBuilder::processEndOfFile):
4020         (WebCore::HTMLTreeBuilder::processDefaultForBeforeHTMLMode):
4021         (WebCore::HTMLTreeBuilder::processStartTagForInHead):
4022         (WebCore::HTMLConstructionSite::insertDoctype):
4023         (WebCore::HTMLConstructionSite::insertComment):
4024         (WebCore::HTMLConstructionSite::insertCommentOnDocument):
4025         (WebCore::HTMLConstructionSite::insertCommentOnHTMLHtmlElement):
4026         (WebCore::HTMLConstructionSite::createElementAndAttachToCurrent):
4027         (WebCore::HTMLConstructionSite::insertHTMLHtmlElement):
4028         (WebCore::HTMLConstructionSite::insertHTMLHeadElement):
4029         (WebCore::HTMLConstructionSite::insertHTMLBodyElement):
4030         (WebCore::HTMLConstructionSite::insertElement):
4031         (WebCore::HTMLConstructionSite::insertSelfClosingElement):
4032         (WebCore::HTMLConstructionSite::insertFormattingElement):
4033         (WebCore::HTMLTreeBuilder::processGenericRCDATAStartTag):
4034         (WebCore::HTMLTreeBuilder::processGenericRawTextStartTag):
4035         (WebCore::HTMLConstructionSite::insertScriptElement):
4036         (WebCore::HTMLTreeBuilder::processScriptStartTag):
4037         (WebCore::HTMLConstructionSite::insertTextNode):
4038         (WebCore::HTMLConstructionSite::createElement):
4039         (WebCore::HTMLTreeBuilder::indexOfFirstUnopenFormattingElement):
4040         (WebCore::HTMLTreeBuilder::reconstructTheActiveFormattingElements):
4041         (WebCore::HTMLTreeBuilder::generateImpliedEndTagsWithExclusion):
4042         (WebCore::HTMLTreeBuilder::generateImpliedEndTags):
4043         * html/HTMLTreeBuilder.h:
4044         (WebCore::HTMLConstructionSite::currentElement):
4045         (WebCore::HTMLConstructionSite::openElements):
4046         (WebCore::HTMLConstructionSite::activeFormattingElements):
4047         (WebCore::HTMLConstructionSite::head):
4048         (WebCore::HTMLConstructionSite::form):
4049         (WebCore::HTMLConstructionSite::releaseForm):
4050         (WebCore::HTMLConstructionSite::setForm):
4051         (WebCore::HTMLConstructionSite::fragmentScriptingPermission):
4052         (WebCore::HTMLConstructionSite::attach):
4053
4054 2010-07-06  Gustavo Noronha Silva  <gustavo.noronha@collabora.co.uk>
4055
4056         Reviewed by Xan Lopez.
4057
4058         [GTK] Scrollbars sometimes go dead and stop scrolling the view
4059         https://bugs.webkit.org/show_bug.cgi?id=41711
4060
4061         Fixes GtkScrolledWindow scrollbars not actually scrolling the page
4062         in certain conditions. No tests because it is hard to reproduce,
4063         and depends on interaction with a widget that is outside of the
4064         WebView, which is tricky.
4065
4066         * platform/gtk/ScrollViewGtk.cpp:
4067         (WebCore::ScrollView::setGtkAdjustments):
4068
4069 2010-07-06  Darin Adler  <darin@apple.com>
4070
4071         Reviewed by Geoffrey Garen.
4072
4073         Simplify access to element attribute map, removing unneeded functions
4074         https://bugs.webkit.org/show_bug.cgi?id=41703
4075
4076         * css/CSSStyleSelector.cpp:
4077         (WebCore::CSSStyleSelector::canShareStyleWithElement): Use the
4078         attributeMap function instead of the mappedAttributes function.
4079         (WebCore::CSSStyleSelector::styleForElement): Ditto.
4080
4081         * dom/Element.cpp:
4082         (WebCore::Element::~Element): Renamed namedAttrMap to m_attributeMap.
4083         (WebCore::Element::removeAttribute): Ditto.
4084         (WebCore::Element::getAttribute): Ditto.
4085         (WebCore::Element::setAttribute): Ditto.
4086         (WebCore::Element::setAttributeMap): Ditto.
4087         (WebCore::Element::hasAttributes): Ditto.
4088         (WebCore::Element::createAttributeMap): Ditto.
4089         (WebCore::Element::insertedIntoDocument): Ditto.
4090         (WebCore::Element::removedFromDocument): Ditto.
4091         (WebCore::Element::getURLAttribute): Ditto.
4092
4093         * dom/Element.h: Removed unneeded virtual attributes function.
4094         Made read-only argument to attributes function default to false.
4095         (Later, we'll eliminate this.) Renamed namedAttrMap to m_attributeMap
4096         and made it private instead of protected.
4097
4098         * dom/StyledElement.cpp:
4099         (WebCore::StyledElement::attributeChanged): Use attributeMap function
4100         instead of namedAttrMap data member or mappedAttributes function.
4101         (WebCore::StyledElement::classAttributeChanged): Ditto.
4102         (WebCore::StyledElement::parseMappedAttribute): Ditto.
4103
4104         * dom/StyledElement.h: Removed mappedAttributes function and changed
4105         callers to use attributeMap function instead.
4106
4107         * html/HTMLInputElement.cpp:
4108         (WebCore::HTMLInputElement::setInputType): Use attributeMap instead of
4109         mappedAttributes.
4110         * svg/SVGStyledElement.cpp:
4111         (WebCore::SVGStyledElement::getPresentationAttribute): Ditto.
4112
4113 2010-07-06  Anders Carlsson  <andersca@apple.com>
4114
4115         Reviewed by Sam Weinig.
4116
4117         Add a build rule that removes the old generated Obj-C bindings for PluginArray and MimeTypeArray.
4118
4119         * DerivedSources.make:
4120
4121 2010-07-06  Adam Barth  <abarth@webkit.org>
4122
4123         Unreviewed.
4124
4125         Make the remaining insertion modes explicit.
4126
4127         * html/HTMLTreeBuilder.cpp:
4128         (WebCore::HTMLTreeBuilder::processStartTag):
4129         (WebCore::HTMLTreeBuilder::processEndTag):
4130         (WebCore::HTMLTreeBuilder::processCharacter):
4131         (WebCore::HTMLTreeBuilder::processEndOfFile):
4132
4133 2010-07-06  Alexey Proskuryakov  <ap@apple.com>
4134
4135         Not reviewed, fixing a simple copy/paste mistake.
4136
4137         https://bugs.webkit.org/show_bug.cgi?id=41156
4138         Cross origin XMLHttpRequest should log the reason why connection failed
4139
4140         Covered by xmlhttprequest/access-control-basic-non-simple-deny-cached.html
4141
4142         * loader/CrossOriginPreflightResultCache.cpp:
4143         (WebCore::CrossOriginPreflightResultCacheItem::allowsCrossOriginHeaders): Fix incorrectly
4144         copy/pasted header field name.
4145
4146 2010-07-06  Xan Lopez  <xlopez@igalia.com>
4147
4148         GTK+ fixes for the DOM plug-in renaming Apocalypse.
4149
4150         * GNUmakefile.am:
4151
4152 2010-07-06  Anders Carlsson  <andersca@apple.com>
4153
4154         Yet another Windows build fix.
4155
4156         * DerivedSources.cpp:
4157
4158 2010-07-06  Anders Carlsson  <andersca@apple.com>
4159
4160         Fix Windows build.
4161
4162         * WebCore.vcproj/WebCore.vcproj:
4163
4164 2010-07-06  Anders Carlsson  <andersca@apple.com>
4165
4166         More build fixes.
4167
4168         * WebCore.gypi:
4169         * WebCore.pro:
4170
4171 2010-07-06  Anders Carlsson  <andersca@apple.com>
4172
4173         Try to fix Qt build.
4174
4175         * WebCore.pro:
4176
4177 2010-07-06  Anders Carlsson  <andersca@apple.com>
4178
4179         Attempt to fix GTK+ build.
4180
4181         * GNUmakefile.am:
4182
4183 2010-07-06  Adam Barth  <abarth@webkit.org>
4184
4185         Address reviewer comments from https://bugs.webkit.org/show_bug.cgi?id=41671
4186
4187         Deploy isTableBodyContextTag where possible.  No behavior change.
4188
4189         * html/HTMLTreeBuilder.cpp:
4190         (WebCore::HTMLTreeBuilder::processStartTagForInBody):
4191         (WebCore::HTMLTreeBuilder::processStartTag):
4192         (WebCore::HTMLTreeBuilder::processEndTagForInTable):
4193         (WebCore::HTMLTreeBuilder::processEndTag):
4194
4195 2010-07-06  Anders Carlsson  <andersca@apple.com>
4196
4197         Build fixes.
4198
4199         * WebCore.gypi:
4200         * bindings/js/JSBindingsAllInOne.cpp:
4201
4202 2010-07-06  Anders Carlsson  <andersca@apple.com>
4203
4204         Reviewed by Sam Weinig.
4205
4206         DOM plug-in related renames
4207         https://bugs.webkit.org/show_bug.cgi?id=41709
4208
4209         This is in preparation for using "Plugin" for something other than a DOM object.
4210
4211         MimeType => DOMMimeType
4212         MimeTypeArray => DOMMimeTypeArray
4213         Plugin => DOMPlugin
4214         PluginArray => DOMPluginArray
4215
4216         * CMakeLists.txt:
4217         * DerivedSources.make:
4218         * GNUmakefile.am:
4219         * WebCore.pri:
4220         * WebCore.vcproj/WebCore.vcproj:
4221         * WebCore.xcodeproj/project.pbxproj:
4222         * bindings/js/JSDOMMimeTypeArrayCustom.cpp: Added.
4223         (WebCore::JSDOMMimeTypeArray::canGetItemsForName):
4224         (WebCore::JSDOMMimeTypeArray::nameGetter):
4225         * bindings/js/JSDOMPluginArrayCustom.cpp: Added.
4226         (WebCore::JSDOMPluginArray::canGetItemsForName):
4227         (WebCore::JSDOMPluginArray::nameGetter):
4228         * bindings/js/JSDOMPluginCustom.cpp: Added.
4229         (WebCore::JSDOMPlugin::canGetItemsForName):
4230         (WebCore::JSDOMPlugin::nameGetter):
4231         * bindings/js/JSMimeTypeArrayCustom.cpp: Removed.
4232         * bindings/js/JSPluginArrayCustom.cpp: Removed.
4233         * bindings/js/JSPluginCustom.cpp: Removed.
4234         * bindings/scripts/CodeGeneratorJS.pm:
4235         * page/DOMWindow.idl:
4236         * page/Navigator.cpp:
4237         (WebCore::Navigator::plugins):
4238         (WebCore::Navigator::mimeTypes):
4239         * page/Navigator.h:
4240         * page/Navigator.idl:
4241         * plugins/DOMMimeType.cpp: Added.
4242         (WebCore::DOMMimeType::DOMMimeType):
4243         (WebCore::DOMMimeType::~DOMMimeType):
4244         (WebCore::DOMMimeType::type):
4245         (WebCore::DOMMimeType::suffixes):
4246         (WebCore::DOMMimeType::description):
4247         (WebCore::DOMMimeType::enabledPlugin):
4248         * plugins/DOMMimeType.h: Added.
4249         (WebCore::DOMMimeType::create):
4250         (WebCore::DOMMimeType::mimeClassInfo):
4251         * plugins/DOMMimeType.idl: Added.
4252         * plugins/DOMMimeTypeArray.cpp: Added.
4253         (WebCore::DOMMimeTypeArray::DOMMimeTypeArray):
4254         (WebCore::DOMMimeTypeArray::~DOMMimeTypeArray):
4255         (WebCore::DOMMimeTypeArray::length):
4256         (WebCore::DOMMimeTypeArray::item):
4257         (WebCore::DOMMimeTypeArray::canGetItemsForName):
4258         (WebCore::DOMMimeTypeArray::namedItem):
4259         (WebCore::DOMMimeTypeArray::getPluginData):
4260         * plugins/DOMMimeTypeArray.h: Added.
4261         (WebCore::DOMMimeTypeArray::create):
4262         (WebCore::DOMMimeTypeArray::disconnectFrame):
4263         * plugins/DOMMimeTypeArray.idl: Added.
4264         * plugins/DOMPlugin.cpp: Added.
4265         (WebCore::DOMPlugin::DOMPlugin):
4266         (WebCore::DOMPlugin::~DOMPlugin):
4267         (WebCore::DOMPlugin::name):
4268         (WebCore::DOMPlugin::filename):
4269         (WebCore::DOMPlugin::description):
4270         (WebCore::DOMPlugin::length):
4271         (WebCore::DOMPlugin::item):
4272         (WebCore::DOMPlugin::canGetItemsForName):
4273         (WebCore::DOMPlugin::namedItem):
4274         * plugins/DOMPlugin.h: Added.
4275         (WebCore::DOMPlugin::create):
4276         (WebCore::DOMPlugin::pluginInfo):
4277         * plugins/DOMPlugin.idl: Added.
4278         * plugins/DOMPluginArray.cpp: Added.
4279         (WebCore::DOMPluginArray::DOMPluginArray):
4280         (WebCore::DOMPluginArray::~DOMPluginArray):
4281         (WebCore::DOMPluginArray::length):
4282         (WebCore::DOMPluginArray::item):
4283         (WebCore::DOMPluginArray::canGetItemsForName):
4284         (WebCore::DOMPluginArray::namedItem):
4285         (WebCore::DOMPluginArray::refresh):
4286         (WebCore::DOMPluginArray::pluginData):
4287         * plugins/DOMPluginArray.h: Added.
4288         (WebCore::DOMPluginArray::create):
4289         (WebCore::DOMPluginArray::disconnectFrame):
4290         * plugins/DOMPluginArray.idl: Added.
4291         * plugins/MimeType.cpp: Removed.
4292         * plugins/MimeType.h: Removed.
4293         * plugins/MimeType.idl: Removed.
4294         * plugins/MimeTypeArray.cpp: Removed.
4295         * plugins/MimeTypeArray.h: Removed.
4296         * plugins/MimeTypeArray.idl: Removed.
4297         * plugins/Plugin.cpp: Removed.
4298         * plugins/Plugin.h: Removed.
4299         * plugins/Plugin.idl: Removed.
4300         * plugins/PluginArray.cpp: Removed.
4301         * plugins/PluginArray.h: Removed.
4302         * plugins/PluginArray.idl: Removed.
4303
4304 2010-07-06  Alexey Proskuryakov  <ap@apple.com>
4305
4306         Reviewed by Darin Adler.