Refactoring: SpellChecker::requestCheckingFor should take Range instead of Node.
[WebKit-https.git] / Source / WebCore / ChangeLog
1 2011-11-21  Shinya Kawanaka  <shinyak@google.com>
2
3         Refactoring: SpellChecker::requestCheckingFor should take Range instead of Node.
4         https://bugs.webkit.org/show_bug.cgi?id=72847
5
6         Reviewed by Hajime Morita.
7
8         Covered by existing test.
9
10         * editing/Editor.cpp:
11         (WebCore::Editor::replaceSelectionWithFragment):
12           Passes Range to requestCheckingFor instead of Node.
13         * editing/SpellChecker.cpp:
14           Changed argument type from Node to Range.
15           The corresponding changes are also done in dependent methods.
16         (WebCore::SpellChecker::initRequest):
17         (WebCore::SpellChecker::clearRequest):
18         (WebCore::SpellChecker::canCheckAsynchronously):
19         (WebCore::SpellChecker::isBusy):
20         (WebCore::SpellChecker::isValid):
21         (WebCore::SpellChecker::isCheckable):
22         (WebCore::SpellChecker::requestCheckingFor):
23           Changed argument type from Node to Range.
24         (WebCore::SpellChecker::doRequestCheckingFor):
25         (WebCore::SpellChecker::didCheck):
26         * editing/SpellChecker.h:
27
28 2011-11-20  Kenichi Ishibashi  <bashi@chromium.org>
29
30         [Chromium] Remove old getFontFamilyForCharacters() and familyForChars() APIs.
31         https://bugs.webkit.org/show_bug.cgi?id=72844
32
33         Respects bold and italic properties that is given by fontconfig.
34
35         Reviewed by Darin Fisher.
36
37         * platform/graphics/chromium/FontCacheLinux.cpp:
38         (WebCore::FontCache::getFontDataForCharacters): Removed #if and old logic.
39
40 2011-11-20  Noel Gordon  <noel.gordon@gmail.com>
41
42         [chromium] Remove qt/QtMobileWebStyle from the gyp projects
43         https://bugs.webkit.org/show_bug.cgi?id=72843
44
45         Reviewed by Antonio Gomes.
46
47         platform/qt/QtMobileWebStyle.{h,cpp} were removed in r100123
48
49         * WebCore.gypi: remove platform/qt/QtMobileWebStyle.{h,cpp}
50
51 2011-11-20  Kentaro Hara  <haraken@chromium.org>
52
53         Unreviewed. Rebaselined run-bindings-tests results.
54
55         * bindings/scripts/test/CPP/WebDOMFloat64Array.cpp:
56         (WebDOMFloat64Array::WebDOMFloat64Array):
57         (WebDOMFloat64Array::impl):
58         (toWebCore):
59         (toWebKit):
60         * bindings/scripts/test/CPP/WebDOMFloat64Array.h:
61
62 2011-11-20  Adam Barth  <abarth@webkit.org>
63
64         REGRESSION(r100691): Safari error pages and Growl notifications fail to load stylesheets
65         https://bugs.webkit.org/show_bug.cgi?id=72836
66
67         Reviewed by Sam Weinig.
68
69         This patch removes a (minor) security mitigation.  Previously, we tried
70         sequester "directory listings" into unique origins to make it more
71         difficult for an attacker to crawl the user's local file system.
72         Unfortunately, this mitigation doesn't really buy us much security
73         because if the attacker has access to local files, we've probably lost
74         anyway.
75
76         The larger problem, however, is that this condition is overly
77         complicated and has broken in sublte ways several times in its
78         (relatively short) lifetime.  In the cases reported in this bug, we see
79         that this check affects error pages in Safari and Growl notifications,
80         even those have nothing to do with directory listings.
81
82         If we have our heart set on this directory listing mitigation, we'll
83         need a more robust way of triggering the behavior than examining URLs
84         and guess whether they contain directory listings.  For example, if we
85         implement Allow-From or Access-Control-Deny-Origin, then the embedder
86         can supply those policies along with the directory listings.  Those
87         seem like much better solutions than the in-engine hack this patch
88         removes.
89
90         * page/SecurityOrigin.cpp:
91         (WebCore::shouldTreatAsUniqueOrigin):
92
93 2011-10-17  Antonio Gomes  <agomes@rim.com>
94
95         Pass a Frame* parameter in EditorClient::respondToChangedSelection
96         https://bugs.webkit.org/show_bug.cgi?id=70248
97
98         Reviewed by Ryosuke Niwa.
99
100         Most of the port specific implementations of EditorClient::respondToChangedSelection
101         (like EditorClient{Qt,Gtk,etc}) are wrongly relying on FocusController::focusedOrMainFrame
102         method to get the Frame where the selection is happening on.
103         It is not right, since a selection can be easily triggered in an inner
104         frame that is not focused.
105
106         No new tests since it is a hard thing to test without
107         to hook into editor client in layout tests. We could
108         change the "Dumped Delegate" messages, but it would
109         require a rebasile of +1200 for a minor change.
110
111         * editing/Editor.cpp: Pass the Frame on where the selection is changing to the client.
112         (WebCore::Editor::respondToChangedSelection): Ditto.
113         (WebCore::Editor::changeSelectionAfterCommand): Ditto.
114         (WebCore::EmptyEditorClient::respondToChangedSelection):
115         * page/EditorClient.h:  Pass Frame* to indicate where the selection is happening.
116         * loader/EmptyClients.h: Changed signature as per base class change.
117
118 2011-11-17  Mark Rowe  <mrowe@apple.com>
119
120         <http://webkit.org/b/72646> Disable deprecation warnings around code where we cannot easily
121         switch away from the deprecated APIs.
122
123         Reviewed by Sam Weinig.
124
125         * platform/mac/WebCoreNSStringExtras.mm:
126         * platform/network/cf/SocketStreamHandleCFNet.cpp:
127         (WebCore::SocketStreamHandle::reportErrorToClient):
128
129 2011-11-19  Kevin Ollivier  <kevino@theolliviers.com>
130
131         [wx] C++ bindings build fix for move of array classes to WTF.
132         
133         * bindings/scripts/CodeGeneratorCPP.pm:
134         (GetCPPTypeGetter):
135         (GetNamespaceForClass):
136         (GenerateHeader):
137         (GenerateImplementation):
138
139 2011-11-19  Scott Graham  <scottmg@chromium.org>
140
141         Move gamepad to Modules/ (+ some cleanup)
142         https://bugs.webkit.org/show_bug.cgi?id=72785
143
144         Reviewed by Adam Barth.
145
146         Move main files from page/ to Modules/gamepad/. #include guard the
147         gamepad header inclusions in Navigator.cpp to avoid including for
148         ports that do not enable GAMEPAD.
149
150         * Modules/gamepad/Gamepad.cpp: Renamed from Source/WebCore/page/Gamepad.cpp.
151         (WebCore::Gamepad::Gamepad):
152         (WebCore::Gamepad::axes):
153         (WebCore::Gamepad::buttons):
154         (WebCore::Gamepad::~Gamepad):
155         * Modules/gamepad/Gamepad.h: Renamed from Source/WebCore/page/Gamepad.h.
156         * Modules/gamepad/Gamepad.idl: Renamed from Source/WebCore/page/Gamepad.idl.
157         * Modules/gamepad/GamepadList.cpp: Renamed from Source/WebCore/page/GamepadList.cpp.
158         (WebCore::GamepadList::~GamepadList):
159         (WebCore::GamepadList::set):
160         (WebCore::GamepadList::length):
161         (WebCore::GamepadList::item):
162         * Modules/gamepad/GamepadList.h: Renamed from Source/WebCore/page/GamepadList.h.
163         (WebCore::GamepadList::create):
164         (WebCore::GamepadList::GamepadList):
165         * Modules/gamepad/GamepadList.idl: Renamed from Source/WebCore/page/GamepadList.idl.
166         * WebCore.gyp/WebCore.gyp:
167         * WebCore.gypi:
168         * page/Navigator.cpp:
169
170 2011-11-19  Sheriff Bot  <webkit.review.bot@gmail.com>
171
172         Unreviewed, rolling out r100834.
173         http://trac.webkit.org/changeset/100834
174         https://bugs.webkit.org/show_bug.cgi?id=72806
175
176         this patch makes Qt run-webkit-test exit before finishing
177         (Requested by igoroliveira on #webkit).
178
179         * platform/graphics/texmap/TextureMapperNode.cpp:
180         * platform/graphics/texmap/TextureMapperNode.h:
181
182 2011-11-19  Huang Dongsung  <luxtella@company100.net>
183
184         Remove WebCore/ForwardingHeaders/runtime/JSObjectWithGlobalObject.h because
185         JSObjectWithGlobalObject.h has been removed.
186         https://bugs.webkit.org/show_bug.cgi?id=72794
187
188         r94701 removed JSObjectWithGlobalObject.h.
189
190         Reviewed by Oliver Hunt.
191
192         * ForwardingHeaders/runtime/JSObjectWithGlobalObject.h: Removed.
193
194 2011-11-19  Vsevolod Vlasov  <vsevik@chromium.org>
195
196         Web Inspector:[protocol] actual JSON messages do not conform to Inspector.json in InspectorApplicationCacheAgent.cpp
197         https://bugs.webkit.org/show_bug.cgi?id=72734
198
199         Reviewed by Timothy Hatcher.
200
201         * inspector/Inspector.json:
202
203 2011-11-19  David Barr  <davidbarr@chromium.org>
204
205         REGRESSION(r98542): Chromium: CSS text is rendered on page
206         https://bugs.webkit.org/show_bug.cgi?id=71703
207
208         Reviewed by Antti Koivisto.
209
210         Matched UA declarations uncacheable when using simpleDefaultStyleSheet.
211
212         Test: fast/css/style-tag-display-none.html
213
214         * css/CSSStyleSelector.cpp:
215         (WebCore::CSSStyleSelector::matchUARules):
216
217 2011-11-19  Jay Civelli  <jcivelli@chromium.org>
218
219         When loading a MHTML document, make sure we set the base URL before
220         we notify the frame was committed (so the document has the right base
221         URL when the notification happens). 
222         https://bugs.webkit.org/show_bug.cgi?id=72788
223
224         Reviewed by Adam Barth.
225
226         * loader/FrameLoader.cpp:
227         (WebCore::FrameLoader::receivedFirstData):
228
229 2011-11-19  Adam Barth  <abarth@webkit.org>
230
231         Integrate Source/WTF with the Chromium build system
232         https://bugs.webkit.org/show_bug.cgi?id=72790
233
234         Reviewed by Eric Seidel.
235
236         Add a dependency on the new WTF.
237
238         * WebCore.gyp/WebCore.gyp:
239
240 2011-11-18  Takashi Toyoshima  <toyoshim@chromium.org>
241
242         [Chromium] [WebSocket] export WebSocketChannel interface for plugins
243         https://bugs.webkit.org/show_bug.cgi?id=72016
244
245         Reviewed by Darin Fisher.
246
247         Add a interface to send raw binary data.
248         This interface is used by WebWebSocketChannel implementation
249         in WebKit API.
250
251         No new tests because just export a interface.
252
253         * websockets/WebSocketChannel.cpp:
254         (WebCore::WebSocketChannel::send):
255         * websockets/WebSocketChannel.h:
256
257 2011-11-18  Vineet Chaudhary  <vineet.chaudhary@motorola.com>
258
259         https://bugs.webkit.org/show_bug.cgi?id=72591
260         Remove document.width / document.height
261
262         Reviewed by Darin Adler.
263
264         Removed document.width/document.height from JS bindings,
265         but keeping the same for ObjC bindings. Also it should use
266         document.body.clientWidth and document.body.clientHeight instead.
267
268         * html/HTMLDocument.idl:
269
270 2011-11-18  Martin Robinson  <mrobinson@igalia.com>
271
272         Fix the GTK+ build.
273
274         * page/Navigator.idl: Properly disable the webkitGamepads API if gamepad
275         isn't enabled at compile time.
276
277 2011-11-18  Daniel Bates  <dbates@rim.com>
278
279         Add CMake build infrastructure for the BlackBerry port
280         https://bugs.webkit.org/show_bug.cgi?id=72768
281
282         Reviewed by Antonio Gomes.
283
284         * CMakeLists.txt: At this time the BlackBerry port doesn't support generating
285           DOM bindings from the SVG IDLs. See WebKit bug #72773.
286         * PlatformBlackBerry.cmake: Added.
287
288 2011-11-18  Sheriff Bot  <webkit.review.bot@gmail.com>
289
290         Unreviewed, rolling out r100826.
291         http://trac.webkit.org/changeset/100826
292         https://bugs.webkit.org/show_bug.cgi?id=72786
293
294         Broke Chromium Mac build (Requested by aklein on #webkit).
295
296         * page/ChromeClient.h:
297         * page/FrameView.cpp:
298         * page/FrameView.h:
299         * platform/ScrollView.cpp:
300         (WebCore::ScrollView::wheelEvent):
301         * platform/ScrollView.h:
302         * platform/ScrollableArea.h:
303         (WebCore::ScrollableArea::scrollbarStyleChanged):
304         * platform/mac/ScrollAnimatorMac.mm:
305         (WebCore::ScrollAnimatorMac::updateScrollerStyle):
306
307 2011-11-18  Alpha Lam  <hclam@chromium.org>
308
309         [chromium] composited layers are blurry with a zoom-in page scale factor
310         https://bugs.webkit.org/show_bug.cgi?id=71225
311
312         Reviewed by James Robinson.
313
314         Pass contents scale factor to the compositor such that it can:
315         1. Adjust contentBounds() of the compositoer layers with content scale.
316         2. Apply the content scale in the painter for texture update in better resolution.
317         3. Apply the content scale to the dirty rect in CanvasLayerTextureUpdater.
318
319         This change fixed blurry problem for all tiled layer types and is not limited to
320         position:fixed layers.
321
322         Tests: compositing/geometry/fixed-position-composited-page-scale-down.html
323                compositing/geometry/fixed-position-composited-page-scale.html
324                compositing/geometry/fixed-position-iframe-composited-page-scale-down.html
325                compositing/geometry/fixed-position-iframe-composited-page-scale.html
326                compositing/geometry/fixed-position-transform-composited-page-scale-down.html
327                compositing/geometry/fixed-position-transform-composited-page-scale.html
328
329         * platform/graphics/chromium/GraphicsLayerChromium.cpp:
330         (WebCore::GraphicsLayerChromium::setTransform):
331         (WebCore::GraphicsLayerChromium::updateLayerPreserves3D):
332         (WebCore::GraphicsLayerChromium::updateContentsScale):
333         (WebCore::GraphicsLayerChromium::contentsScale):
334         (WebCore::GraphicsLayerChromium::deviceOrPageScaleFactorChanged):
335         * platform/graphics/chromium/GraphicsLayerChromium.h:
336         * platform/graphics/chromium/ImageLayerChromium.cpp:
337         (WebCore::ImageLayerTextureUpdater::prepareToUpdate):
338         (WebCore::ImageLayerChromium::needsContentsScale):
339         * platform/graphics/chromium/ImageLayerChromium.h:
340         * platform/graphics/chromium/LayerChromium.cpp:
341         (WebCore::LayerChromium::LayerChromium):
342         (WebCore::LayerChromium::setContentsScale):
343         * platform/graphics/chromium/LayerChromium.h:
344         (WebCore::LayerChromium::needsContentsScale):
345         (WebCore::LayerChromium::contentsScale):
346         * platform/graphics/chromium/LayerTextureUpdater.h:
347         * platform/graphics/chromium/LayerTextureUpdaterCanvas.cpp:
348         (WebCore::LayerTextureUpdaterCanvas::paintContents):
349         (WebCore::LayerTextureUpdaterBitmap::prepareToUpdate):
350         (WebCore::LayerTextureUpdaterSkPicture::prepareToUpdate):
351         * platform/graphics/chromium/LayerTextureUpdaterCanvas.h:
352         * platform/graphics/chromium/TiledLayerChromium.cpp:
353         (WebCore::TiledLayerChromium::needsContentsScale):
354         (WebCore::TiledLayerChromium::contentBounds):
355         (WebCore::TiledLayerChromium::prepareToUpdate):
356         * platform/graphics/chromium/TiledLayerChromium.h:
357
358 2011-11-18  Igor Oliveira  <igor.oliveira@openbossa.org>
359
360         [WK2][Qt] Move Accelerated Composite animations to UIProcess
361         https://bugs.webkit.org/show_bug.cgi?id=72753
362
363         Add helper method to synchronize animations in TextureMapper.
364
365         Reviewed by Noam Rosenthal.
366
367         * platform/graphics/texmap/TextureMapperNode.cpp:
368         (WebCore::TextureMapperNode::syncAnimationsRecursively):
369         * platform/graphics/texmap/TextureMapperNode.h:
370
371 2011-11-18  Scott Graham  <scottmg@chromium.org>
372
373         IDL changes for gamepad support
374         https://bugs.webkit.org/show_bug.cgi?id=71753
375
376         Reviewed by Adam Barth.
377
378         IDL changes and associated plumbing to expose list of gamepad objects
379         on navigator object (per current spec). Full patch is
380         https://bugs.webkit.org/show_bug.cgi?id=69451. Only basic existence
381         test until more plumbing in future patches.
382
383         Test: gamepad/gamepad-api.html
384
385         * WebCore.gypi:
386         * bindings/generic/RuntimeEnabledFeatures.h:
387         (WebCore::RuntimeEnabledFeatures::setWebkitGamepadsEnabled):
388         (WebCore::RuntimeEnabledFeatures::webkitGamepadsEnabled):
389         * page/Gamepad.cpp: Added.
390         (WebCore::Gamepad::Gamepad):
391         (WebCore::Gamepad::axes):
392         (WebCore::Gamepad::buttons):
393         (WebCore::Gamepad::~Gamepad):
394         * page/Gamepad.h: Added.
395         * page/Gamepad.idl: Added.
396         * page/GamepadList.cpp: Added.
397         (WebCore::GamepadList::~GamepadList):
398         (WebCore::GamepadList::set):
399         (WebCore::GamepadList::length):
400         (WebCore::GamepadList::item):
401         * page/GamepadList.h: Added.
402         (WebCore::GamepadList::create):
403         (WebCore::GamepadList::GamepadList):
404         * page/GamepadList.idl: Added.
405         * page/Navigator.cpp:
406         (WebCore::Navigator::webkitGamepads):
407         * page/Navigator.h:
408         * page/Navigator.idl:
409
410 2011-11-18  Sheriff Bot  <webkit.review.bot@gmail.com>
411
412         Unreviewed, rolling out r100693.
413         http://trac.webkit.org/changeset/100693
414         https://bugs.webkit.org/show_bug.cgi?id=72779
415
416         This patch caused a rendering regression (see bug 72770)
417         (Requested by philip__ on #webkit).
418
419         * rendering/RenderObject.cpp:
420         (WebCore::RenderObject::setStyle):
421         * rendering/style/RenderStyle.cpp:
422         (WebCore::RenderStyle::diff):
423
424 2011-11-18  Simon Fraser  <simon.fraser@apple.com>
425
426         Reflection on composited element doesn't update if the element changes to show no content
427         https://bugs.webkit.org/show_bug.cgi?id=72774
428
429         Reviewed by Chris Marrin.
430         
431         When a style change results in a reflected element losing its backing store,
432         we need to clear the backing stores on the layer clones as well.
433
434         Test: compositing/reflections/become-simple-composited-reflection.html
435
436         * platform/graphics/ca/GraphicsLayerCA.cpp:
437         (WebCore::GraphicsLayerCA::updateLayerDrawsContent):
438
439 2011-11-18  Daniel Cheng  <dcheng@chromium.org>
440
441         [chromium] Use correct backing store for ChromiumDataObject in pasteboard writes.
442         https://bugs.webkit.org/show_bug.cgi?id=72767
443
444         Reviewed by Tony Chang.
445
446         Covered by existing tests.
447
448         * editing/chromium/EditorChromium.cpp:
449         (WebCore::Editor::newGeneralClipboard):
450         * page/chromium/EventHandlerChromium.cpp:
451         (WebCore::EventHandler::createDraggingClipboard):
452         * platform/chromium/ChromiumDataObject.cpp:
453         (WebCore::ChromiumDataObject::types):
454         (WebCore::ChromiumDataObject::getData):
455         (WebCore::ChromiumDataObject::containsFilenames):
456         (WebCore::ChromiumDataObject::ChromiumDataObject):
457         * platform/chromium/ChromiumDataObject.h:
458         (WebCore::ChromiumDataObject::createFromPasteboard):
459         (WebCore::ChromiumDataObject::create):
460         (WebCore::ChromiumDataObject::storageMode):
461         * platform/chromium/ClipboardChromium.cpp:
462         (WebCore::ClipboardChromium::hasData):
463
464 2011-11-18  Chris Evans  <cevans@google.com>
465
466         Crash with ranges across a detached, reparented node tree
467         https://bugs.webkit.org/show_bug.cgi?id=72757
468
469         Reviewed by Adam Barth.
470
471         Test: fast/dom/move-detached-child-in-range.html
472
473         * dom/RangeBoundaryPoint.h:
474         (WebCore::RangeBoundaryPoint::childBefore): protect the raw child node from getting pulled from under us.
475
476 2011-11-18  Beth Dakin  <bdakin@apple.com>
477
478         https://bugs.webkit.org/show_bug.cgi?id=72551
479         When the recommended scrollbar style changes, WKView's tracking options should 
480         adjust accordingly
481         -and corresponding-
482         <rdar://problem/10409328>
483
484         Reviewed by Darin Adler.
485
486         This new ChromeClient function is called when the recommended scrollbar style 
487         changes. This way, WebKit can respond to the change by adjusting its mouse 
488         tracking.
489         * page/ChromeClient.h:
490         (WebCore::ChromeClient::recommendedScrollbarStyleDidChange):
491
492         Existing ScrollableArea function scrollbarStyleChanged() now takes an int 
493         indicating the new scrollbar style and a bool indicating whether it is necessary 
494         to force an update. It used to be the case that this function was ONLY used to 
495         force an update (and only called when an updated was needed), but now that it must 
496         also call into the ChromeClient, it is necessary to include a bool tracking 
497         whether we need to force an update. New implementation on FrameView is responsible 
498         for calling ChromeClient, and then that calls into the pre-existing ScrollView 
499         function for the forceUpdate part.
500         * page/FrameView.cpp:
501         (WebCore::FrameView::scrollbarStyleChanged):
502         * page/FrameView.h:
503         * platform/ScrollView.cpp:
504         (WebCore::ScrollView:: scrollbarStyleChanged):
505         * platform/ScrollView.h:
506         * platform/ScrollableArea.h:
507         (WebCore::ScrollableArea::scrollbarStyleChanged):
508         * platform/mac/ScrollAnimatorMac.mm:
509         (WebCore::ScrollAnimatorMac::updateScrollerStyle):
510
511 2011-11-18  Kelly Norton  <knorton@google.com>
512
513         Fixes several more void functions in RenderObject that return values.
514         https://bugs.webkit.org/show_bug.cgi?id=72750
515
516         Reviewed by Adam Barth.
517
518         * rendering/RenderBlock.cpp:
519         (WebCore::RenderBlock::updateBeforeAfterContent):
520         (WebCore::RenderBlock::addChildToContinuation):
521         (WebCore::RenderBlock::addChildToAnonymousColumnBlocks):
522         (WebCore::RenderBlock::addChild):
523         (WebCore::RenderBlock::addChildIgnoringContinuation):
524
525 2011-11-18  Shawn Singh  <shawnsingh@chromium.org>
526
527         [chromium] Add some useful text to existing debug dump
528         https://bugs.webkit.org/show_bug.cgi?id=72576
529
530         Reviewed by James Robinson.
531
532         No new tests needed.
533
534         * platform/graphics/chromium/cc/CCLayerImpl.cpp:
535         (WebCore::CCLayerImpl::dumpLayerProperties):
536         * platform/graphics/chromium/cc/CCRenderSurface.cpp:
537         (WebCore::CCRenderSurface::dumpSurface):
538
539 2011-11-18  Xiaomei Ji  <xji@chromium.org>
540
541         REGRESSION: rtl horizontal scrollbar / resize bug - Body shifts on resize when scrolled all the way to the left
542         https://bugs.webkit.org/show_bug.cgi?id=70395
543
544         Reviewed by Tony Chang.
545
546         This patch fixes the problem in Mac and Chromium Mac.
547    
548         The existing test fast/dom/rtl-scroll-to-leftmost-and-resize.html seems does not really work in Mac DRT
549         (the browswer window is not resized). It works in Chromium-Mac. And Chromium-Mac's code
550         is forked from Mac.
551
552         * platform/chromium/ScrollAnimatorChromiumMac.mm:
553         (WebCore::ScrollAnimatorChromiumMac::immediateScrollToPoint):
554         * platform/mac/ScrollAnimatorMac.mm:
555         (WebCore::ScrollAnimatorMac::immediateScrollToPoint):
556
557 2011-11-18  Martin Robinson  <mrobinson@igalia.com>
558
559         REGRESSION (r99924): broke 2 pasteboard tests on GTK
560         https://bugs.webkit.org/show_bug.cgi?id=72131
561
562         Reviewed by Tony Chang.
563
564         * platform/gtk/ClipboardGtk.cpp:
565         (WebCore::ClipboardGtk::clearData): Call clearAll now.
566         (WebCore::ClipboardGtk::clearAllData): Call clearAllExceptFilenames now.
567         * platform/gtk/DataObjectGtk.cpp:
568         (WebCore::DataObjectGtk::clearAllExceptFilenames): Renamed from clear.
569         (WebCore::DataObjectGtk::clearAll): Added this method which also clear filenames.
570         * platform/gtk/DataObjectGtk.h:
571         * platform/gtk/PasteboardGtk.cpp: Call clear before setting new clipboard data.
572         (WebCore::Pasteboard::writeSelection): Ditto.
573         (WebCore::Pasteboard::writePlainText): Ditto.
574         (WebCore::Pasteboard::writeURL): Ditto.
575         (WebCore::Pasteboard::writeImage): Ditto.
576         * platform/gtk/PasteboardHelper.cpp:
577         (WebCore::clearClipboardContentsCallback): Use the clearAll method now.
578
579 2011-11-18  Vineet Chaudhary  <vineet.chaudhary@motorola.com>
580
581         Access key should work on all elements.
582         https://bugs.webkit.org/show_bug.cgi?id=71854
583
584         Reviewed by Ryosuke Niwa.
585
586         All HTML elements can have the accesskey content attribute set.
587         Specification http://dev.w3.org/html5/spec/Overview.html#the-accesskey-attribute
588         Adding "accessKey" attribute to HTMLElement.idl file as [Reflect].
589
590         Test: fast/forms/access-key-for-all-elements.html
591
592         * bindings/objc/PublicDOMInterfaces.h: Moved properties form subclass to base class.
593         * html/BaseButtonInputType.cpp: 
594         (WebCore::BaseButtonInputType::accessKeyAction): Renamed variable sendToAnyElement to sendMouseEvents.
595         * html/BaseButtonInputType.h: Ditto
596         * html/BaseCheckableInputType.cpp:
597         (WebCore::BaseCheckableInputType::accessKeyAction): Ditto
598         * html/BaseCheckableInputType.h: Ditto
599         * html/HTMLAnchorElement.cpp:
600         (WebCore::HTMLAnchorElement::accessKeyAction): Ditto
601         * html/HTMLAnchorElement.h: Ditto
602         * html/HTMLAnchorElement.idl: Removed redundant IDL attribute entries.
603         * html/HTMLAreaElement.idl: Removed redundant IDL attribute entries.
604         * html/HTMLButtonElement.cpp:
605         (WebCore::HTMLButtonElement::accessKeyAction): Renamed variable sendToAnyElement to sendMouseEvents.
606         * html/HTMLButtonElement.h: Ditto
607         * html/HTMLButtonElement.idl: Removed redundant IDL attribute entries.
608         * html/HTMLElement.cpp:
609         (WebCore::HTMLElement::accessKeyAction): Renamed variable sendToAnyElement to sendMouseEvents.
610         If the element does not have a defined activation behavior, fire a click event at the element.
611         * html/HTMLElement.h: Ditto
612         * html/HTMLElement.idl: Added accessKey IDL attribute.
613         * html/HTMLInputElement.cpp:
614         (WebCore::HTMLInputElement::accessKeyAction): Renamed variable sendToAnyElement to sendMouseEvents.
615         * html/HTMLInputElement.h: Ditto
616         * html/HTMLInputElement.idl: Removed redundant IDL attribute entries.
617         * html/HTMLLabelElement.cpp:
618         (WebCore::HTMLLabelElement::accessKeyAction): Renamed variable sendToAnyElement to sendMouseEvents.
619         * html/HTMLLabelElement.h: Ditto
620         * html/HTMLLabelElement.idl: Removed redundant IDL attribute entries.
621         * html/HTMLLegendElement.cpp:
622         (WebCore::HTMLLegendElement::accessKeyAction): Renamed variable sendToAnyElement to sendMouseEvents.
623         * html/HTMLLegendElement.h: Ditto
624         * html/HTMLLegendElement.idl: Removed redundant IDL attribute entries.
625         * html/HTMLOptGroupElement.h: Renamed variable sendToAnyElement to sendMouseEvents.
626         * html/HTMLSelectElement.cpp:
627         (WebCore::HTMLSelectElement::accessKeyAction): Ditto
628         * html/HTMLSelectElement.h: Ditto
629         * html/HTMLTextAreaElement.h: Ditto
630         * html/HTMLTextAreaElement.idl: Removed redundant IDL attribute entries.
631         * html/HiddenInputType.h: Renamed variable sendToAnyElement to sendMouseEvents.
632         * html/InputType.h: Ditto
633         * html/RangeInputType.cpp:
634         (WebCore::RangeInputType::accessKeyAction): Ditto
635         * html/RangeInputType.h: Ditto
636
637 2011-11-18  Simon Fraser  <simon.fraser@apple.com>
638
639         Appearance of compound transform animations under apps linked on SnowLeopard is incorrect
640         https://bugs.webkit.org/show_bug.cgi?id=72641
641         and
642         <rdar://problem/10314267>
643
644         Reviewed by Dean Jackson.
645
646         GraphicsLayerCA contains a "linked on or after" check to account for a bug in
647         Core Animation on SnowLeopard and earlier, which is that CA would apply the list
648         of animations in reverse order.
649         
650         Our previous fix was incorrect, because it only adjusted the 'additive' property
651         of the animation list based on ordering, rather than flipping the entire list.
652         This change reverses the list of animations before giving them to CA, which fixes
653         the bug.
654                 
655         Test: animations/additive-transform-animations.html
656
657         * platform/graphics/ca/GraphicsLayerCA.cpp:
658         (WebCore::GraphicsLayerCA::appendToUncommittedAnimations):
659         (WebCore::GraphicsLayerCA::createTransformAnimationsFromKeyframes):
660         * platform/graphics/ca/GraphicsLayerCA.h:
661
662 2011-11-18  Tim Horton  <timothy_horton@apple.com>
663
664         -webkit-cross-fade rendered incorrectly in overflow divs
665         https://bugs.webkit.org/show_bug.cgi?id=72693
666         <rdar://problem/10468564>
667
668         Reviewed by Simon Fraser.
669
670         Respect the desired source rectangle when rendering the cross-fade.
671
672         Test: css3/images/cross-fade-overflow-position.html
673
674         * platform/graphics/CrossfadeGeneratedImage.cpp:
675         (WebCore::CrossfadeGeneratedImage::drawCrossfade):
676         (WebCore::CrossfadeGeneratedImage::draw):
677         (WebCore::CrossfadeGeneratedImage::drawPattern):
678         * platform/graphics/CrossfadeGeneratedImage.h:
679
680 2011-11-18  Raphael Kubo da Costa  <kubo@profusion.mobi>
681
682         Unreviewed build fix; r100686 broke the EFL build when Geolocation
683         support is enabled.
684
685         * platform/efl/GeolocationServiceEfl.cpp: Only declare
686         s_factoryFunction if CLIENT_BASED_GEOLOCATION is off, otherwise it is
687         also declared in GeolocationService.cpp.
688
689 2011-11-18  Igor Oliveira  <igor.oliveira@openbossa.org>
690
691         [TextureMapper] computePerspectiveTransformIfNeeded is called twice in TextureMapperNode::syncCompositingState
692         https://bugs.webkit.org/show_bug.cgi?id=72727
693
694         TextureMapperNode::computeAllTransforms already has a call to TextureMapperNode::computePerspectiveTransformIfNeeded,
695         so it does not need to be called again after TextureMapperNode::computeAllTransforms.
696
697         Reviewed by Noam Rosenthal.
698
699         * platform/graphics/texmap/TextureMapperNode.cpp:
700         (WebCore::TextureMapperNode::syncCompositingState):
701
702 2011-11-18  Mihnea Ovidenie  <mihnea@adobe.com>
703
704         Fix compilation warning in ComplexTextControllerCoreText.mm
705         https://bugs.webkit.org/show_bug.cgi?id=72702
706
707         Reviewed by Andreas Kling.
708
709         No functionality changed, so no new tests.
710
711         * platform/graphics/mac/ComplexTextControllerCoreText.mm:
712         (WebCore::ComplexTextController::collectComplexTextRunsForCharactersCoreText):
713
714 2011-11-18  Pavel Feldman  <pfeldman@google.com>
715
716         Not reviewed: restore front-end compilability via updating externs and JS generator.
717
718         * inspector/front-end/ElementsPanel.js:
719         (WebInspector.ElementsPanel.prototype.jumpToNextSearchResult):
720         (WebInspector.ElementsPanel.prototype.jumpToPreviousSearchResult):
721         * inspector/front-end/externs.js:
722         (WebInspector.showPanelForAnchorNavigation):
723         * inspector/generate-protocol-externs:
724
725 2011-11-17  Pavel Feldman  <pfeldman@google.com>
726
727         Web Inspector: dispatch messages from the front-end to the backend asynchronously.
728         https://bugs.webkit.org/show_bug.cgi?id=72621
729
730         We should align the way we dispatch messages from the front-end to backend across the environments:
731           - WebKit has it synchronoulsly
732           - Chromium has it asynchronously
733           - Remote debugging has it asynchronously
734         Making it asynchronous made a number of flaky Qt tests pass.
735
736         Tests uncovered console agent problem that was also fixed.
737
738         Reviewed by Yury Semikhatsky.
739
740         * inspector/InspectorConsoleAgent.cpp:
741         (WebCore::InspectorConsoleAgent::clearFrontend):
742         * inspector/InspectorFrontendClient.h:
743         * inspector/InspectorFrontendClientLocal.cpp:
744         (WebCore::InspectorBackendDispatchTask::InspectorBackendDispatchTask):
745         (WebCore::InspectorBackendDispatchTask::dispatch):
746         (WebCore::InspectorBackendDispatchTask::reset):
747         (WebCore::InspectorBackendDispatchTask::onTimer):
748         (WebCore::InspectorFrontendClientLocal::InspectorFrontendClientLocal):
749         (WebCore::InspectorFrontendClientLocal::windowObjectCleared):
750         (WebCore::InspectorFrontendClientLocal::sendMessageToBackend):
751         * inspector/InspectorFrontendClientLocal.h:
752
753 2011-11-18  Iain Merrick  <husky@google.com>
754
755         [chromium] Pass screen refresh rate into compositor.
756         https://bugs.webkit.org/show_bug.cgi?id=71040
757
758         Reviewed by Tony Gentilcore.
759
760         Covered by CCLayerTreeHostTest.
761
762         * platform/PlatformScreen.h:
763         * platform/chromium/PlatformScreenChromium.cpp:
764         (WebCore::screenRefreshRate):
765         * platform/chromium/PlatformSupport.h:
766         * platform/graphics/chromium/cc/CCLayerTreeHost.h:
767         (WebCore::CCSettings::CCSettings):
768         * platform/graphics/chromium/cc/CCThreadProxy.cpp:
769         (WebCore::CCThreadProxy::initializeImplOnImplThread):
770
771 2011-11-18  Alexandru Chiculita  <achicu@adobe.com>
772
773         [CSSShaders] Implement the computed style for mesh parameters of the custom() filter
774         https://bugs.webkit.org/show_bug.cgi?id=72478
775
776         Reviewed by Dean Jackson.
777
778         Added parsing and computed style for the mesh rows, columns, mesh box type 
779         (filter-box, border-box, content-box and padding-box) and the detached mode.
780         
781         Also fixed a case where "custom(none, 10, 10 filter-box)" was incorrectly
782         treated as "custom(none, 10)".
783
784         * css/CSSComputedStyleDeclaration.cpp:
785         (WebCore::CSSComputedStyleDeclaration::valueForFilter):
786         * css/CSSParser.cpp:
787         (WebCore::CSSParser::parseCustomFilter): Fixed a case where invalid syntax was parsed as correct syntax.
788         * css/CSSPrimitiveValueMappings.h:
789         (WebCore::CSSPrimitiveValue::CSSPrimitiveValue):
790         (WebCore::CSSPrimitiveValue::operator CustomFilterOperation::MeshBoxType):
791         * css/CSSStyleSelector.cpp:
792         (WebCore::CSSStyleSelector::createCustomFilterOperation):
793         * css/CSSValueList.h:
794         (WebCore::CSSValueListIterator::isPrimitiveValue):
795         * platform/graphics/filters/CustomFilterOperation.h:
796
797 2011-11-18  Adam Barth  <abarth@webkit.org>
798
799         Remove unneeded include (and ifdef) from V8DOMWindowCustom.cpp
800         https://bugs.webkit.org/show_bug.cgi?id=72705
801
802         Reviewed by Eric Seidel.
803
804         Death to ifdefs!
805
806         * bindings/v8/custom/V8DOMWindowCustom.cpp:
807
808 2011-11-18  Adam Barth  <abarth@webkit.org>
809
810         Move some mediastream related files into the mediastream directory
811         https://bugs.webkit.org/show_bug.cgi?id=72695
812
813         Reviewed by Eric Seidel.
814
815         Just moving the files to the proper directory.
816
817         * GNUmakefile.list.am:
818         * WebCore.gypi:
819         * mediastream/NavigatorUserMediaError.h: Renamed from Source/WebCore/page/NavigatorUserMediaError.h.
820         (WebCore::NavigatorUserMediaError::create):
821         (WebCore::NavigatorUserMediaError::~NavigatorUserMediaError):
822         (WebCore::NavigatorUserMediaError::code):
823         (WebCore::NavigatorUserMediaError::NavigatorUserMediaError):
824         * mediastream/NavigatorUserMediaError.idl: Renamed from Source/WebCore/page/NavigatorUserMediaError.idl.
825         * mediastream/NavigatorUserMediaErrorCallback.h: Renamed from Source/WebCore/page/NavigatorUserMediaErrorCallback.h.
826         (WebCore::NavigatorUserMediaErrorCallback::~NavigatorUserMediaErrorCallback):
827         * mediastream/NavigatorUserMediaErrorCallback.idl: Renamed from Source/WebCore/page/NavigatorUserMediaErrorCallback.idl.
828         * mediastream/NavigatorUserMediaSuccessCallback.h: Renamed from Source/WebCore/page/NavigatorUserMediaSuccessCallback.h.
829         (WebCore::NavigatorUserMediaSuccessCallback::~NavigatorUserMediaSuccessCallback):
830         * mediastream/NavigatorUserMediaSuccessCallback.idl: Renamed from Source/WebCore/page/NavigatorUserMediaSuccessCallback.idl.
831
832 2011-11-18  Andrey Kosyakov  <caseq@chromium.org>
833
834         Web Inspector: [Extensions API] Provide a way for extension to create a status bar icon
835         https://bugs.webkit.org/show_bug.cgi?id=45955
836
837         Reviewed by Pavel Feldman.
838
839         * inspector/front-end/ExtensionAPI.js:
840         (injectedExtensionAPI.ExtensionPanelImpl.prototype.createStatusBarButton):
841         (injectedExtensionAPI):
842         (injectedExtensionAPI.ButtonImpl.prototype.update):
843         * inspector/front-end/ExtensionPanel.js:
844         (WebInspector.ExtensionPanel):
845         (WebInspector.ExtensionPanel.prototype.get statusBarItems):
846         (WebInspector.ExtensionPanel.prototype.addStatusBarItem):
847         (WebInspector.ExtensionButton):
848         (WebInspector.ExtensionButton.prototype.update):
849         (WebInspector.ExtensionButton.prototype._onClicked):
850         * inspector/front-end/ExtensionServer.js:
851         (WebInspector.ExtensionServer):
852         (WebInspector.ExtensionServer.prototype.notifyButtonClicked):
853         (WebInspector.ExtensionServer.prototype._onCreateStatusBarButton):
854         (WebInspector.ExtensionServer.prototype._onUpdateButton):
855         (WebInspector.ExtensionServer.prototype._onCreateSidebarPane):
856         * inspector/front-end/inspector.css:
857         (button.status-bar-item.extension):
858
859 2011-11-18  Tommy Widenflycht  <tommyw@google.com>
860
861         MediaStream API: Merging the PeerConnectionHandler.h files
862         https://bugs.webkit.org/show_bug.cgi?id=72611
863
864         Reviewed by Adam Barth.
865
866         No actual code changes.
867
868         * GNUmakefile.am:
869         * GNUmakefile.list.am:
870         * WebCore.gyp/WebCore.gyp:
871         * WebCore.gypi:
872         * platform/mediastream/PeerConnectionHandler.h: Renamed from Source/WebCore/platform/mediastream/chromium/PeerConnectionHandler.h.
873         * platform/mediastream/gstreamer/PeerConnectionHandler.h: Removed.
874
875 2011-11-17  Simon Hausmann  <simon.hausmann@nokia.com>
876
877         [Qt] Layer violation: ThirdPartyCookiesQt.cpp uses QWebSettings in two places
878         https://bugs.webkit.org/show_bug.cgi?id=72597
879
880         Reviewed by Kenneth Rohde Christiansen.
881
882         Moved code that queries the third party cookie policy from QWebSettings
883         out of WebCore. Instead NetworkingContext has now the interface that allows
884         delegating the functionality into WebKit/qt.
885
886         In order to do that we need to pass the NetworkingContext around, which is used to
887         retrieve the cookie jar and the originating frame object pointer.
888
889         * platform/network/NetworkingContext.h: Add interface.
890         * platform/network/qt/QNetworkReplyHandler.cpp:
891         (WebCore::QNetworkReplyHandler::QNetworkReplyHandler): Pass NetworkingContext to ResourceRequest::toNetworkRequest.
892         (WebCore::QNetworkReplyHandler::redirect): Ditto.
893         * platform/network/qt/ResourceRequest.h: Replaced originatingObject parameter with NetworkingContext, which can
894         also provide the same plus more (cookie jar, policy callback).
895         * platform/network/qt/ResourceRequestQt.cpp:
896         (WebCore::ResourceRequest::toNetworkRequest): Ditto.
897         * platform/qt/CookieJarQt.cpp: Simplify to use NetworkingContext to get cookie jar and
898         call new thirdPartyCookiePolicyPermits API that takes the context as parameter.
899         (WebCore::networkingContext):
900         (WebCore::setCookies):
901         (WebCore::cookies):
902         (WebCore::cookieRequestHeaderFieldValue):
903         (WebCore::cookiesEnabled):
904         * platform/qt/ThirdPartyCookiesQt.cpp: Replace up-casting to QWebFrame and use of QWebSettings
905         with useage of NetworkingContext.
906         (WebCore::thirdPartyCookiePolicyPermits):
907         * platform/qt/ThirdPartyCookiesQt.h:
908
909 2011-11-17  Kenichi Ishibashi  <bashi@chromium.org>
910
911         crash: WebCore::FontPlatformData::roundsGlyphAdvances on Lion
912         https://bugs.webkit.org/show_bug.cgi?id=71997
913
914         Reviewed by Dan Bernstein.
915
916         The cause is a null dereference of a fontData that is stored in
917         ComplexTextRun. The fontData is initialized by using the
918         fontCache, but it could be null when the font is in fallback
919         list. The reason a font from the fallback list might not be in the
920         font Cache is that it may be a web font. Before looking up the
921         fontCache, try to see whether the font is in the fallback list.
922
923         No new tests. We don't have webfonts that can produce the problem.
924
925         * platform/graphics/mac/ComplexTextControllerCoreText.mm:
926         (WebCore::ComplexTextController::collectComplexTextRunsForCharactersCoreText): See fallback list first, then lookup cache.
927
928 2011-11-17  Kaustubh Atrawalkar  <kaustubh@motorola.com>
929
930         Remove initProgressEvent method
931         https://bugs.webkit.org/show_bug.cgi?id=71340
932
933         Reviewed by Adam Barth.
934
935         This method has been removed from the spec draft.
936         http://www.w3.org/TR/progress-events/#interface-progressevent
937
938         No new tests. Removed method.
939
940         * dom/ProgressEvent.cpp:
941         * dom/ProgressEvent.h:
942         * dom/ProgressEvent.idl:
943
944 2011-11-17  Eunmi Lee  <eunmi15.lee@samsung.com>
945
946         [EFL] Move keyIdentifierForEvasKeyName() and windowsKeyCodeForEvasKeyName() to the
947         EflKeyboardUtilities.cpp to use in the WebKit2
948         https://bugs.webkit.org/show_bug.cgi?id=62451
949
950         Reviewed by Martin Robinson.
951
952         The keyIdentifierForEvasKeyName() and windowsKeyCodeForEvasKeyName() were static functions
953         in the PlatformKeyboardEventEfl.cpp. But they are also needed in the WebKit2 EFL port, so I
954         moved them to the separated file - EflKeyboardUtilities.cpp.
955
956         * PlatformEfl.cmake:
957         * platform/efl/EflKeyboardUtilities.cpp: Copied from Source/WebCore/platform/efl/PlatformKeyboardEventEfl.cpp.
958         (WebCore::createKeyMap):
959         (WebCore::createWindowsKeyMap):
960         (WebCore::keyIdentifierForEvasKeyName):
961         (WebCore::windowsKeyCodeForEvasKeyName):
962         * platform/efl/EflKeyboardUtilities.h: Added.
963         * platform/efl/PlatformKeyboardEventEfl.cpp:
964
965 2011-11-17  Martin Robinson  <mrobinson@igalia.com>
966
967         [GTK] The process freezes when you right click on windowless Flash
968         https://bugs.webkit.org/show_bug.cgi?id=69123
969
970         Reviewed by Xan Lopez.
971
972         No new tests. I tried to create a layout test that exercised this
973         issue, but it appears that EventSender clicks do not trigger
974         it. This is covered by the manual tests containing Flash.
975
976         * plugins/PluginPackage.cpp:
977         (WebCore::PluginPackage::determineQuirks): Always activate the
978         windowless Flash quirk if on x86_64 and X11.
979         * plugins/gtk/PluginViewGtk.cpp:
980         (WebCore::PluginView::handleMouseEvent): Avoid sending right-click
981         events if we have the quirk.
982
983 2011-11-17  Peter Rybin  <peter.rybin@gmail.com>
984
985         Web Inspector: clear fixme in generator script
986         https://bugs.webkit.org/show_bug.cgi?id=71372
987
988         Remove unnecessary field name map and update license year number.
989
990         Reviewed by Pavel Feldman.
991
992         * inspector/CodeGeneratorInspector.py:
993
994 2011-11-17  Raphael Kubo da Costa  <kubo@profusion.mobi>
995
996         [EFL] Clean up the use of DATA_DIR in the buildsystem
997         https://bugs.webkit.org/show_bug.cgi?id=72681
998
999         Reviewed by Daniel Bates.
1000
1001         Add the -DDATA_DIR definition here instead of defining it globally in
1002         OptionsEfl.cmake, as WebCore is the only place which needs it.
1003
1004         No new tests, this is a buildsystem change.
1005
1006         * PlatformEfl.cmake:
1007
1008 2011-11-17  Adam Klein  <adamk@chromium.org>
1009
1010         Move JS recursion counter from V8Proxy to V8BindingPerIsolateData
1011         https://bugs.webkit.org/show_bug.cgi?id=72645
1012
1013         Reviewed by Adam Barth.
1014
1015         With the JS recursion level stored as a member of V8Proxy, it's tied
1016         to a frame. But this is incorrect, as there's no reason that a JS call
1017         stack need be restricted to a single frame (see my new test case for
1018         an example of code going across frames).
1019
1020         In order to get the correct accounting of JS recursion level, per-Isolate
1021         is the right granularity (per dslomov), which is what this patch accomplishes.
1022
1023         Test: storage/indexeddb/transaction-abort-with-js-recursion-cross-frame.html
1024
1025         * bindings/v8/V8Binding.cpp:
1026         (WebCore::V8BindingPerIsolateData::V8BindingPerIsolateData):
1027         * bindings/v8/V8Binding.h:
1028         (WebCore::V8BindingPerIsolateData::recursionLevel):
1029         (WebCore::V8BindingPerIsolateData::incrementRecursionLevel):
1030         (WebCore::V8BindingPerIsolateData::decrementRecursionLevel):
1031         (WebCore::V8RecursionScope::V8RecursionScope):
1032         (WebCore::V8RecursionScope::~V8RecursionScope):
1033         * bindings/v8/V8Proxy.cpp:
1034         (WebCore::incrementRecursionLevel):
1035         (WebCore::decrementRecursionLevel):
1036         (WebCore::recursionLevel):
1037         (WebCore::V8Proxy::V8Proxy):
1038         (WebCore::V8Proxy::runScript):
1039         (WebCore::V8Proxy::callFunction):
1040         (WebCore::V8Proxy::didLeaveScriptContext):
1041         * bindings/v8/V8Proxy.h:
1042
1043 2011-11-17  Robin Cao  <robin.cao@torchmobile.com.cn>
1044
1045         [chromium] Font::drawComplexText can not draw a segment of text run
1046         https://bugs.webkit.org/show_bug.cgi?id=72095
1047
1048         Reviewed by Adam Barth.
1049
1050         drawComplexText() should respect the 'from' and 'to' arguments.
1051         Drawing the whole text run may result in text overlapping.
1052
1053         Test: platform/chromium-linux/fast/text/international/draw-complex-text-from-to.html
1054
1055         * platform/graphics/chromium/ComplexTextControllerLinux.cpp:
1056         (WebCore::ComplexTextController::glyphsForRange):
1057         * platform/graphics/chromium/ComplexTextControllerLinux.h:
1058         * platform/graphics/chromium/FontLinux.cpp:
1059         (WebCore::Font::drawComplexText):
1060
1061 2011-11-17  Adam Barth  <abarth@webkit.org>
1062
1063         Unique SecurityOrigins shouldn't remember their old schemes and hosts
1064         https://bugs.webkit.org/show_bug.cgi?id=71745
1065
1066         Reviewed by Darin Adler.
1067
1068         This is the final step in this series of patches.
1069
1070         This patch removes the forceUnique flag from SecurityOrigin::create.
1071         Now, we create unique origins without passing in the document's URL,
1072         preventing information from the document's URL from leaking into the
1073         unique origin.
1074
1075         * WebCore.exp.in:
1076         * dom/Document.cpp:
1077         (WebCore::Document::setIsViewSource):
1078         (WebCore::Document::initSecurityContext):
1079         * loader/cache/MemoryCache.cpp:
1080         (WebCore::MemoryCache::getOriginsWithCache):
1081             - Update this callsite to use createFromString, which does exactly
1082               what this code is doing manually.
1083         * page/SecurityOrigin.cpp:
1084         (WebCore::SecurityOrigin::create):
1085         * page/SecurityOrigin.h:
1086
1087 2011-11-17  Vincent Scheib  <scheib@chromium.org>
1088
1089         Pointer Lock: Page Settings unecessary
1090         https://bugs.webkit.org/show_bug.cgi?id=72662
1091
1092         Reviewed by Darin Fisher.
1093
1094         No new tests.
1095
1096         * page/Settings.cpp:
1097         (WebCore::Settings::Settings):
1098         * page/Settings.h:
1099
1100 2011-11-17  James Robinson  <jamesr@chromium.org>
1101
1102         [chromium] Disable incremental uploading in threaded compositing path
1103         https://bugs.webkit.org/show_bug.cgi?id=72669
1104
1105         Reviewed by Kenneth Russell.
1106
1107         Since we don't currently support atomic incremental uploads, incremental uploads in the threaded path result in
1108         very strange-looking texture popping as tiles come in on pages that require more than 16 tile uploads per frame.
1109         This disables that logic (by setting the upload limit per frame to 99999) until we handle the incremental
1110         updates in an atomic fashion.
1111
1112         * platform/graphics/chromium/cc/CCThreadProxy.cpp:
1113         (WebCore::CCThreadProxy::scheduledActionUpdateMoreResources):
1114
1115 2011-11-06  Nat Duca  <nduca@chromium.org>
1116
1117         [chromium] Fix handling of setNeedsCommit and setNeedsAnimate in threaded mode
1118         https://bugs.webkit.org/show_bug.cgi?id=71638
1119
1120         Reviewed by James Robinson.
1121
1122         * platform/graphics/chromium/cc/CCThreadProxy.cpp:
1123         (WebCore::CCThreadProxy::CCThreadProxy):
1124         (WebCore::CCThreadProxy::setNeedsAnimate):
1125         (WebCore::CCThreadProxy::beginFrameAndCommit):
1126
1127 2011-11-17  Adam Barth  <abarth@webkit.org>
1128
1129         Remove bogus ASSERT.
1130
1131         * page/SecurityOrigin.cpp:
1132         (WebCore::SecurityOrigin::SecurityOrigin):
1133
1134 2011-11-17  Peter Kasting  <pkasting@google.com>
1135
1136         Unreviewed, rolling out r100698.
1137         http://trac.webkit.org/changeset/100698
1138         https://bugs.webkit.org/show_bug.cgi?id=72239
1139
1140         This change wasn't the problem either.
1141
1142         * dom/Document.cpp:
1143         (WebCore::Document::implicitClose):
1144
1145 2011-11-17  David Reveman  <reveman@chromium.org>
1146
1147         [Chromium] Calls to paintContentsIfDirty() and updateCompositorResources() should be balanced.
1148         https://bugs.webkit.org/show_bug.cgi?id=72630
1149
1150         Reviewed by James Robinson.
1151
1152         Layer property changes during paintContent() can leave the layer
1153         in an invalid state as paintContentsIfDirty() has been called
1154         without a matching updateCompositorResources() call. Removing
1155         conditionals around these calls ensure they are balanced.
1156
1157         This patch is tested by the following unit test:
1158         - CCLayerTreeHostTestOpacityChange.runMultiThread
1159
1160         * platform/graphics/chromium/ContentLayerChromium.cpp:
1161         (WebCore::ContentLayerChromium::paintContentsIfDirty):
1162         * platform/graphics/chromium/VideoLayerChromium.cpp:
1163         (WebCore::VideoLayerChromium::updateCompositorResources):
1164         * platform/graphics/chromium/cc/CCLayerTreeHost.cpp:
1165         (WebCore::CCLayerTreeHost::paintMaskAndReplicaForRenderSurface):
1166         (WebCore::CCLayerTreeHost::paintLayerContents):
1167         (WebCore::CCLayerTreeHost::updateCompositorResources):
1168         * platform/graphics/chromium/cc/CCLayerTreeHost.h:
1169
1170 2011-11-17  Peter Kasting  <pkasting@google.com>
1171
1172         Unreviewed, rolling out r100584.
1173         http://trac.webkit.org/changeset/100584
1174         https://bugs.webkit.org/show_bug.cgi?id=72239
1175
1176         See if this change caused Linux dbg crashes.
1177
1178         * dom/Document.cpp:
1179         (WebCore::Document::implicitClose):
1180
1181 2011-11-17  Adam Barth  <abarth@webkit.org>
1182
1183         Remove cargo-cult copy/pasting of ScriptExecutionContext namespace
1184         https://bugs.webkit.org/show_bug.cgi?id=72676
1185
1186         Reviewed by Eric Seidel.
1187
1188         It looks like this cargo-cult started with initDNSPrefetch being hacked
1189         into setSecurityOrigin.  I've removed this hack along with the
1190         copy/paste code.
1191
1192         * dom/DOMImplementation.cpp:
1193         * dom/Document.cpp:
1194         (WebCore::Document::setIsViewSource):
1195         (WebCore::Document::open):
1196         (WebCore::Document::initSecurityContext):
1197         (WebCore::Document::setSecurityOrigin):
1198         * dom/Document.h:
1199         * xml/XSLTProcessor.cpp:
1200
1201 2011-11-17  Peter Kasting  <pkasting@google.com>
1202
1203         Unreviewed, rolling out r100676.
1204         http://trac.webkit.org/changeset/100676
1205         https://bugs.webkit.org/show_bug.cgi?id=72393
1206
1207         Looks like r100572 was not the source of the crashes.
1208
1209         * bindings/js/ScriptDebugServer.cpp:
1210         (WebCore::ScriptDebugServer::canSetScriptSource):
1211         * bindings/js/ScriptDebugServer.h:
1212         * bindings/v8/ScriptDebugServer.cpp:
1213         (WebCore::ScriptDebugServer::stepOutOfFunction):
1214         (WebCore::ScriptDebugServer::canSetScriptSource):
1215         * bindings/v8/ScriptDebugServer.h:
1216         * inspector/CodeGeneratorInspector.py:
1217         * inspector/Inspector.json:
1218         * inspector/InspectorDebuggerAgent.cpp:
1219         (WebCore::InspectorDebuggerAgent::getCapabilities):
1220         * inspector/InspectorDebuggerAgent.h:
1221         * inspector/front-end/DebuggerModel.js:
1222         (WebInspector.DebuggerModel):
1223         (WebInspector.DebuggerModel.prototype.enableDebugger):
1224         (WebInspector.DebuggerModel.prototype.disableDebugger):
1225         (WebInspector.DebuggerModel.prototype.canSetScriptSource):
1226         * inspector/front-end/DebuggerPresentationModel.js:
1227         (WebInspector.DebuggerPresentationModel.prototype.canEditScriptSource):
1228         * inspector/front-end/Settings.js:
1229
1230 2011-11-17  Konstantin Scheglov  <scheglov@google.com>
1231
1232         Absolute child is not repainted when parent opacity changes
1233         https://bugs.webkit.org/show_bug.cgi?id=68777
1234
1235         Reviewed by Simon Fraser.
1236
1237         First time when we change opacity for parent we don't have layer, so
1238         diff=StyleDifferenceRepaint is used instead of diff=StyleDifferenceRepaintLayer.
1239         Layer is created later, in styleDidChange().
1240         So, when we recalculate later diff, we now check for diff=StyleDifferenceRepaintLayer and
1241         performs repaintIncludingDescendants().
1242
1243         Test: fast/layers/layer-absolute-parent-opacity.html
1244
1245         * rendering/RenderObject.cpp:
1246         (WebCore::RenderObject::setStyle): Call repaintIncludingDescendants() instead of repaint().
1247         * rendering/style/RenderStyle.cpp:
1248         (WebCore::RenderStyle::diff): Add ContextSensitivePropertyOpacity when change opacity.
1249
1250 2011-11-17  Nate Chapin  <japhet@chromium.org>
1251
1252         r100311 dropped a RefPtr that is very
1253         helpful. Add it back.
1254         https://bugs.webkit.org/show_bug.cgi?id=72647
1255
1256         Reviewed by Adam Barth.
1257
1258         http/tests/misc/xslt-bad-import.html should
1259         stop crashing in full chromium builds.
1260
1261         * loader/cache/CachedResourceLoader.cpp:
1262         (WebCore::CachedResourceLoader::loadDone):
1263
1264 2011-11-17  Adam Barth  <abarth@webkit.org>
1265
1266         Refactor SecurityOrigin::create to be easier to understand
1267         https://bugs.webkit.org/show_bug.cgi?id=72342
1268
1269         Reviewed by Eric Seidel.
1270
1271         Over time, the SecurityOrigin constructor has grown a bit out of
1272         control.  This patch attempts to separate the different concerns into
1273         free functions.  The general approach is to put more logic in the
1274         "create" function and introduce a simple constructor for unique
1275         origins.
1276
1277         This patch shouldn't change any behavior.
1278
1279         * page/SecurityOrigin.cpp:
1280         (WebCore::schemeRequiresAuthority):
1281         (WebCore::shouldUseInnerURL):
1282         (WebCore::extractInnerURL):
1283         (WebCore::isDirectory):
1284         (WebCore::shouldTreatAsUniqueOrigin):
1285         (WebCore::SecurityOrigin::SecurityOrigin):
1286         (WebCore::SecurityOrigin::create):
1287         (WebCore::SecurityOrigin::createUnique):
1288         (WebCore::SecurityOrigin::databaseIdentifier):
1289         * page/SecurityOrigin.h:
1290
1291 2011-11-17  Shawn Singh  <shawnsingh@chromium.org>
1292
1293         [chromium] Fix minor style nit in CCLayerImpl
1294         https://bugs.webkit.org/show_bug.cgi?id=71070
1295
1296         Reviewed by James Robinson.
1297
1298         Fixes if-statements on several settors to follow WebKit
1299         conventions. Existing CCLayerImplTest unit tests already cover
1300         this change.
1301
1302         * platform/graphics/chromium/cc/CCLayerImpl.cpp:
1303         (WebCore::CCLayerImpl::setBounds):
1304         (WebCore::CCLayerImpl::setMaskLayer):
1305         (WebCore::CCLayerImpl::setReplicaLayer):
1306         (WebCore::CCLayerImpl::setDrawsContent):
1307         (WebCore::CCLayerImpl::setAnchorPoint):
1308         (WebCore::CCLayerImpl::setAnchorPointZ):
1309         (WebCore::CCLayerImpl::setBackgroundColor):
1310         (WebCore::CCLayerImpl::setMasksToBounds):
1311         (WebCore::CCLayerImpl::setOpaque):
1312         (WebCore::CCLayerImpl::setOpacity):
1313         (WebCore::CCLayerImpl::setPosition):
1314         (WebCore::CCLayerImpl::setPreserves3D):
1315         (WebCore::CCLayerImpl::setZoomAnimatorTransform):
1316         (WebCore::CCLayerImpl::setSublayerTransform):
1317         (WebCore::CCLayerImpl::setTransform):
1318         (WebCore::CCLayerImpl::setDebugBorderColor):
1319         (WebCore::CCLayerImpl::setDebugBorderWidth):
1320         (WebCore::CCLayerImpl::setContentBounds):
1321         (WebCore::CCLayerImpl::setScrollPosition):
1322         (WebCore::CCLayerImpl::setScrollDelta):
1323         (WebCore::CCLayerImpl::setScaleDelta):
1324         (WebCore::CCLayerImpl::setDoubleSided):
1325
1326 2011-11-17  Michael Saboff  <msaboff@apple.com>
1327
1328         Leaks seen in MemoryPressureHandlerMac.mm on Leaks bot
1329         https://bugs.webkit.org/show_bug.cgi?id=72416
1330
1331         Added code to release _cache_event_source and _timer_event_source
1332         after they are canceled.  Also added defensive code to
1333         clean up the _timer_event_source in uninstall().
1334
1335         Reviewed by Geoffrey Garen.
1336
1337         No new tests, fixing leaks caught by leaks bot.
1338
1339         * platform/mac/MemoryPressureHandlerMac.mm:
1340         (WebCore::MemoryPressureHandler::uninstall):
1341         (WebCore::MemoryPressureHandler::holdOff):
1342
1343 2011-11-17  Ken Buchanan <kenrb@chromium.org>
1344
1345         Crash from positioned generated content under run-in
1346         https://bugs.webkit.org/show_bug.cgi?id=70456
1347
1348         Reviewed by David Hyatt.
1349
1350         Modified handling of run-in children to clear generated children
1351         before removing the parent from the render tree. This caused problems
1352         with absolute positioned children being not properly removed from the
1353         positioned object list of the RenderView.
1354
1355         * rendering/RenderBlock.cpp:
1356         (WebCore::RenderBlock::handleRunInChild):
1357
1358 2011-11-17  Peter Kasting  <pkasting@google.com>
1359
1360         Unreviewed, rolling out r100572.
1361         https://bugs.webkit.org/show_bug.cgi?id=72393
1362         https://bugs.webkit.org/show_bug.cgi?id=72651
1363
1364         May have caused seg faults on Chromium Linux dbg bot.
1365
1366         * bindings/js/ScriptDebugServer.cpp:
1367         * bindings/js/ScriptDebugServer.h:
1368         * bindings/v8/ScriptDebugServer.cpp:
1369         (WebCore::ScriptDebugServer::stepOutOfFunction):
1370         * bindings/v8/ScriptDebugServer.h:
1371         * inspector/CodeGeneratorInspector.py:
1372         * inspector/Inspector.json:
1373         * inspector/InspectorDebuggerAgent.cpp:
1374         * inspector/InspectorDebuggerAgent.h:
1375         * inspector/front-end/DebuggerModel.js:
1376         (WebInspector.DebuggerModel):
1377         (WebInspector.DebuggerModel.prototype.enableDebugger):
1378         (WebInspector.DebuggerModel.prototype.disableDebugger):
1379         * inspector/front-end/DebuggerPresentationModel.js:
1380         (WebInspector.DebuggerPresentationModel.prototype.canEditScriptSource):
1381         * inspector/front-end/Settings.js:
1382
1383 2011-11-17  Sheriff Bot  <webkit.review.bot@gmail.com>
1384
1385         Unreviewed, rolling out r100652.
1386         http://trac.webkit.org/changeset/100652
1387         https://bugs.webkit.org/show_bug.cgi?id=72648
1388
1389         "Caused outline-offset-min-assert.html to assert on debug
1390         builds" (Requested by mwenge2 on #webkit).
1391
1392         * rendering/RenderObject.cpp:
1393         (WebCore::RenderObject::paintOutline):
1394
1395 2011-11-17  Mark Rowe  <mrowe@apple.com>
1396
1397         <http://webkit.org/b/72637> Stop performing runtime version checks on OS versions where the check can never possibly fail.
1398
1399         Reviewed by Simon Fraser.
1400
1401         * platform/graphics/cg/ImageBufferDataCG.cpp:
1402         (WebCore::haveVImageRoundingErrorFix): When not targeting Snow Leopard we always have the fix.
1403
1404 2011-11-17  Chris Fleizach  <cfleizach@apple.com>
1405
1406         AX: The scrollArea is not correctly returning the scrollbars
1407         https://bugs.webkit.org/show_bug.cgi?id=70247
1408
1409         Reviewed by Beth Dakin.
1410
1411         A few bugs that were preventing this from working.
1412            1) In accessibilityAttributeValue: when an element did not have a renderer() we were returning earlier than we should have
1413            2) We were not updating and clearing the scrollbars correctly when children were cleared or when asked for.
1414
1415         Test: platform/mac/accessibility/scrollbars.html
1416
1417         * accessibility/AccessibilityScrollView.cpp:
1418         (WebCore::AccessibilityScrollView::scrollBar):
1419         (WebCore::AccessibilityScrollView::clearChildren):
1420         * accessibility/AccessibilityScrollView.h:
1421         * accessibility/mac/WebAccessibilityObjectWrapper.mm:
1422         (-[WebAccessibilityObjectWrapper accessibilityAttributeValue:]):
1423
1424 2011-11-14  Adam Barth  <abarth@webkit.org>
1425
1426         Unique origins shouldn't remember their scheme, host, or port
1427         https://bugs.webkit.org/show_bug.cgi?id=72308
1428
1429         Reviewed by Eric Seidel.
1430
1431         This patch contains the bulk (all?) of the behavior differences in this
1432         patch series.  Unique origins shouldn't remember their schemes.  Doing
1433         so causes some privileges (e.g., local access) to leak into unique
1434         origins.
1435
1436         * page/SecurityOrigin.cpp:
1437         (WebCore::SecurityOrigin::SecurityOrigin):
1438             - Explicitly clear out the protocol, host, and port for unique
1439               origins.  A future patch will refactor all this code to be more
1440               elegant.
1441         * platform/SchemeRegistry.cpp:
1442         (WebCore::schemesWithUniqueOrigins):
1443             - Merge "about" and "javascript" in with the general case now that
1444               we don't have a separate notion of an empty origin.
1445
1446 2011-11-17  Chris Fleizach  <cfleizach@apple.com>
1447
1448         AX: cleanup style and naming and code in accessibility search mechanism
1449         https://bugs.webkit.org/show_bug.cgi?id=72570
1450
1451         Reviewed by Beth Dakin.
1452
1453         Cleanup the naming and code style within the element searching mechanism.
1454
1455         * accessibility/AccessibilityObject.cpp:
1456         (WebCore::AccessibilityObject::isAccessibilityObjectSearchMatch):
1457         (WebCore::AccessibilityObject::isAccessibilityTextSearchMatch):
1458         (WebCore::AccessibilityObject::firstAccessibleObjectFromNode):
1459         (WebCore::AccessibilityObject::findMatchingObjects):
1460         * accessibility/AccessibilityObject.h:
1461         * accessibility/mac/WebAccessibilityObjectWrapper.mm:
1462         (-[WebAccessibilityObjectWrapper accessibilityAttributeValue:forParameter:]):
1463
1464 2011-11-17  Julien Chaffraix  <jchaffraix@webkit.org>
1465
1466         CSS table with 100% width can overflow their containing block
1467         https://bugs.webkit.org/show_bug.cgi?id=72180
1468
1469         Reviewed by David Hyatt.
1470
1471         Tests: fast/table/table-in-table-percent-width-collapsing-border-quirks-mode.html
1472                fast/table/table-in-table-percent-width-collapsing-border.html
1473                fast/table/table-in-table-percent-width-quirks-mode.html
1474                fast/table/table-in-table-percent-width.html
1475
1476         After r97555, we would include the borders in a CSS table's logical width even if the 'width'
1477         property was a percent. This does not match what Firefox and IE are doing. Thus don't apply
1478         this behavior to percent 'width'.
1479
1480         * rendering/RenderTable.cpp:
1481         (WebCore::RenderTable::computeLogicalWidth):
1482
1483 2011-11-17  Adrienne Walker  <enne@google.com>
1484
1485         [chromium] Implicitly skip render surfaces that won't be drawn
1486         https://bugs.webkit.org/show_bug.cgi?id=71038
1487
1488         Rather than having redundant checks in three places for how to walk
1489         through a render surface list, instead don't add render surfaces that
1490         don't need to get rendered to the render surface list.
1491
1492         Reviewed by James Robinson.
1493
1494         Covered by existing layout tests and unit tests.
1495
1496         * platform/graphics/chromium/LayerRendererChromium.cpp:
1497         (WebCore::LayerRendererChromium::drawLayersOntoRenderSurfaces):
1498         * platform/graphics/chromium/cc/CCLayerTreeHost.cpp:
1499         (WebCore::CCLayerTreeHost::paintLayerContents):
1500         (WebCore::CCLayerTreeHost::updateCompositorResources):
1501         * platform/graphics/chromium/cc/CCLayerTreeHostCommon.cpp:
1502         (WebCore::calculateDrawTransformsAndVisibilityInternal):
1503
1504 2011-11-17  Simon Hausmann  <simon.hausmann@nokia.com>
1505
1506         [Qt] Layer violation: qt_runtime.cpp accesses QWebElement and QTDRTNode
1507         https://bugs.webkit.org/show_bug.cgi?id=72595
1508
1509         Reviewed by Noam Rosenthal.
1510
1511         Removed QWebElement and QtDRTNode usage that reached from WebCore into
1512         WebKit/qt and replaced it with the ability to register custom JSValue
1513         conversion functions. The old code has been moved to WebKit/qt.
1514
1515         * bridge/qt/qt_instance.cpp:
1516         (JSC::Bindings::QtInstance::QtInstance): Remove unnecessary meta type registration
1517         (now done in QtWebElementRuntime::initialize in WebKit/qt).
1518         * bridge/qt/qt_runtime.cpp:
1519         (JSC::Bindings::registerCustomType):
1520         (JSC::Bindings::convertValueToQVariant):
1521         (JSC::Bindings::convertQVariantToValue):
1522         * bridge/qt/qt_runtime.h:
1523
1524 2011-11-17  Fady Samuel  <fsamuel@chromium.org>
1525
1526         Pass Aspect Ratio to RenderStyle
1527         https://bugs.webkit.org/show_bug.cgi?id=72350
1528
1529         Reviewed by Ojan Vafai.
1530
1531         * css/CSSComputedStyleDeclaration.cpp:
1532         (WebCore::CSSComputedStyleDeclaration::getPropertyCSSValue):
1533         * css/CSSStyleApplyProperty.cpp:
1534         (WebCore::ApplyPropertyAspectRatio::applyInheritValue):
1535         (WebCore::ApplyPropertyAspectRatio::applyInitialValue):
1536         (WebCore::ApplyPropertyAspectRatio::applyValue):
1537         (WebCore::ApplyPropertyAspectRatio::createHandler):
1538         (WebCore::CSSStyleApplyProperty::CSSStyleApplyProperty):
1539         * css/CSSStyleSelector.cpp:
1540         (WebCore::CSSStyleSelector::applyProperty):
1541         * css/CSSValue.h:
1542         (WebCore::CSSValue::isAspectRatioValue):
1543         * rendering/style/RenderStyle.h:
1544         (WebCore::InheritedFlags::hasAspectRatio):
1545         (WebCore::InheritedFlags::aspectRatio):
1546         (WebCore::InheritedFlags::aspectRatioDenominator):
1547         (WebCore::InheritedFlags::aspectRatioNumerator):
1548         (WebCore::InheritedFlags::setHasAspectRatio):
1549         (WebCore::InheritedFlags::setAspectRatioDenominator):
1550         (WebCore::InheritedFlags::setAspectRatioNumerator):
1551         (WebCore::InheritedFlags::initialHasAspectRatio):
1552         (WebCore::InheritedFlags::initialAspectRatioDenominator):
1553         (WebCore::InheritedFlags::initialAspectRatioNumerator):
1554         * rendering/style/StyleRareNonInheritedData.cpp:
1555         (WebCore::StyleRareNonInheritedData::StyleRareNonInheritedData):
1556         (WebCore::StyleRareNonInheritedData::operator==):
1557         * rendering/style/StyleRareNonInheritedData.h:
1558
1559 2011-11-09  Robert Hogan  <robert@webkit.org>
1560
1561         CSS 2.1 failure: outline-color-* tests fail
1562         https://bugs.webkit.org/show_bug.cgi?id=71931
1563
1564         Reviewed by Julien Chaffraix.
1565
1566         WebKit wasn't displaying the top block in these tests because it did not paint the outline
1567         of divs with zero size.
1568
1569         * rendering/RenderObject.cpp:
1570         (WebCore::RenderObject::paintOutline): paint the outline even when the block has zero size
1571
1572 2011-11-17  Sergio Villar Senin  <svillar@igalia.com>
1573
1574         [GTK] plugins/get-url-notify-with-url-that-fails-to-load.html on bots after r100466
1575         https://bugs.webkit.org/show_bug.cgi?id=72613
1576
1577         Reviewed by Martin Robinson.
1578
1579         Do not assume that a SoupRequest always exists, it is not
1580         generated for example when the provided URL is
1581         invalid. ResourceHandle::platformSetDefersLoading was crashing
1582         because of that.
1583
1584         * platform/network/soup/ResourceHandleSoup.cpp:
1585         (WebCore::ResourceHandle::platformSetDefersLoading):
1586
1587 2011-11-17  Vsevolod Vlasov  <vsevik@chromium.org>
1588
1589         Web Inspector: Remove onlineDetectionEnabled from Preferences
1590         https://bugs.webkit.org/show_bug.cgi?id=72617
1591
1592         Reviewed by Pavel Feldman.
1593
1594         * inspector/front-end/ApplicationCacheItemsView.js:
1595         (WebInspector.ApplicationCacheItemsView):
1596         (WebInspector.ApplicationCacheItemsView.prototype.get statusBarItems):
1597         * inspector/front-end/Settings.js:
1598
1599 2011-11-17  Ken Buchanan  <kenrb@chromium.org>
1600
1601         Crash from nested tables with generated content
1602         https://bugs.webkit.org/show_bug.cgi?id=68811
1603
1604         Reviewed by David Hyatt.
1605
1606         When adding a child to a table that has generated content, this change
1607         ensures that we leave alone any generated content renderers that belong
1608         to descendants in the tree. They don't need to be touched, and doing so
1609         can create confusion about who the content belongs to.
1610
1611         This patch also simplifies some existing code for finding pseudoelement
1612         renderers. 
1613
1614         * rendering/RenderBlock.cpp:
1615         (WebCore::RenderBlock::addChildIgnoringAnonymousColumnBlocks):
1616         * rendering/RenderObject.cpp:
1617         (WebCore::RenderObject::addChild):
1618         (WebCore::RenderObject::isBeforeAfterContentGeneratedByAncestor): Added
1619         * rendering/RenderObject.h:
1620         (WebCore::RenderObject::findAfterContentRenderer): Deleted
1621         (WebCore::RenderObject::findBeforeContentRenderer): Deleted
1622         * rendering/RenderObjectChildList.cpp:
1623         (WebCore::RenderObjectChildList::beforePseudoElementRenderer):
1624         (WebCore::RenderObjectChildList::afterPseudoElementRenderer):
1625         * rendering/RenderTable.cpp:
1626         (WebCore::RenderTable::addChild):
1627         * rendering/RenderTableRow.cpp:
1628         (WebCore::RenderTableRow::addChild):
1629         * rendering/RenderTableSection.cpp:
1630         (WebCore::RenderTableSection::addChild):
1631
1632 2011-11-17  Patrick Gansterer  <paroga@webkit.org>
1633
1634         Unreviewed WinCE build fix for r94119.
1635
1636         MSVC throws multiply defined symbols linker error when using local class in inline function.
1637
1638         * bindings/js/JSDictionary.h:
1639         (WebCore::JSDictionary::IdentitySetter::identitySetter):
1640         (WebCore::JSDictionary::tryGetProperty):
1641
1642 2011-11-17  Eric Carlson  <eric.carlson@apple.com>
1643
1644         TextTrackList not sorted correctly
1645         https://bugs.webkit.org/show_bug.cgi?id=72545
1646
1647         Reviewed by Darin Adler.
1648
1649         Test: media/track/track-texttracks.html
1650
1651         * WebCore.xcodeproj/project.pbxproj: Add TextTrack.h to WebCore private headers because 
1652             it is included by HTMLMediaElement.h.
1653         * html/LoadableTextTrack.cpp:
1654         (WebCore::LoadableTextTrack::LoadableTextTrack): Pass track type to base class constructor.
1655         (WebCore::LoadableTextTrack::trackElementIndex): New, return the <track> element's tree order
1656             for sorting.
1657         * html/LoadableTextTrack.h:
1658
1659         * html/TextTrack.cpp:
1660         (WebCore::TextTrack::TextTrack): Set track type.
1661         * html/TextTrack.h:
1662         (WebCore::TextTrack::create): Ditto.
1663         (WebCore::TextTrack::trackType): Ditto.
1664
1665         * html/track/TextTrackList.cpp:
1666         (TextTrackList::length): Update to deal with two TextTrack vectors.
1667         (TextTrackList::item): Ditto.
1668         (TextTrackList::append): Ditto.
1669         (TextTrackList::remove): Ditto
1670         * html/track/TextTrackList.h: Store the two types of TextTracks in separate Vectors to make
1671             it simpler to keep them in the correct order.
1672
1673 2011-11-17  Simon Hausmann  <simon.hausmann@nokia.com>
1674
1675         [Qt] Layer violation: WebCore::dnsPrefetch uses QWebSettings::globalSettings()
1676         https://bugs.webkit.org/show_bug.cgi?id=72596
1677
1678         Reviewed by Kenneth Rohde Christiansen.
1679
1680         We don't need to use QWebSettings here, because we now propagate the DNS prefetch
1681         setting to WebCore::Settings and the rest of WebCore checks the setting before calling
1682         WebCore::prefetchDNS.
1683
1684         * platform/network/qt/DnsPrefetchHelper.cpp:
1685         (WebCore::prefetchDNS): Removed the use of QWebSettings.
1686         * platform/network/qt/DnsPrefetchHelper.h: Remove offending qwebsettings.h inclusion.
1687
1688 2011-11-17  Simon Hausmann  <simon.hausmann@nokia.com>
1689
1690         [Qt] Layer violation: Image::loadPlatformResource uses QWebSettings::webGraphic
1691         https://bugs.webkit.org/show_bug.cgi?id=72594
1692
1693         Reviewed by Kenneth Rohde Christiansen.
1694
1695         Move the cache for the resource pixmaps into ImageQt.cpp.
1696
1697         * platform/graphics/Image.h: Add Qt specific setter for resource pixmaps.
1698         * platform/graphics/qt/ImageQt.cpp: Moved resource pixmap hash from qwebsettings.
1699         (earlyClearGraphics):
1700         (graphics):
1701         (loadResourcePixmap):
1702         (WebCore::Image::setPlatformResource):
1703
1704 2011-11-17  Zeno Albisser  <zeno@webkit.org>
1705
1706         [Qt][WK2] Touch/Mouse events are delivered with wrong coordinates.
1707         https://bugs.webkit.org/show_bug.cgi?id=72604
1708
1709         When using the QtViewportInterationEngine for zooming/panning,
1710         no additional scroll offset should be applied to input events
1711         by the ScrollView.
1712
1713         This patch is based on work by Andreas Kling.
1714
1715         Reviewed by Kenneth Rohde Christiansen.
1716
1717         * platform/ScrollView.cpp:
1718         (WebCore::ScrollView::windowToContents):
1719         (WebCore::ScrollView::contentsToWindow):
1720
1721 2011-11-17  Ben Murdoch  <benm@google.com>
1722
1723         Unreviewed build fix.
1724
1725         Fix the Windows builds by adding WorkerEventQueue.cpp|h to the
1726         vcproj. Build break was introduced in
1727         https://bugs.webkit.org/show_bug.cgi?id=72528
1728
1729         * WebCore.vcproj/WebCore.vcproj: Add missing files.
1730
1731 2011-11-17  Mihnea Ovidenie  <mihnea@adobe.com>
1732
1733         CSS Exclusions: parse the wrap-margin and wrap-padding properties
1734         https://bugs.webkit.org/show_bug.cgi?id=71900
1735
1736         Reviewed by Dean Jackson.
1737
1738         Tests: fast/exclusions/wrap-margin-parsing.html
1739                fast/exclusions/wrap-padding-parsing.html
1740
1741         * css/CSSComputedStyleDeclaration.cpp:
1742         (WebCore::CSSComputedStyleDeclaration::getPropertyCSSValue):
1743         * css/CSSParser.cpp:
1744         (WebCore::isSimpleLengthPropertyID):
1745         (WebCore::CSSParser::parseValue):
1746         * css/CSSProperty.cpp:
1747         (WebCore::CSSProperty::isInheritedProperty):
1748         * css/CSSPropertyNames.in:
1749         * css/CSSStyleApplyProperty.cpp:
1750         (WebCore::CSSStyleApplyProperty::CSSStyleApplyProperty):
1751         * css/CSSStyleSelector.cpp:
1752         (WebCore::CSSStyleSelector::applyProperty):
1753         * rendering/style/RenderStyle.cpp:
1754         (WebCore::RenderStyle::diff):
1755         * rendering/style/RenderStyle.h:
1756         (WebCore::InheritedFlags::wrapPadding):
1757         (WebCore::InheritedFlags::setWrapPadding):
1758         (WebCore::InheritedFlags::initialWrapPadding):
1759         (WebCore::InheritedFlags::wrapMargin):
1760         (WebCore::InheritedFlags::setWrapMargin):
1761         (WebCore::InheritedFlags::initialWrapMargin):
1762         * rendering/style/StyleRareNonInheritedData.cpp:
1763         (WebCore::StyleRareNonInheritedData::StyleRareNonInheritedData):
1764         (WebCore::StyleRareNonInheritedData::operator==):
1765         * rendering/style/StyleRareNonInheritedData.h:
1766
1767 2011-11-17  Kenichi Ishibashi  <bashi@chromium.org>
1768
1769         [chromium] don't call fontconfig twice in complex text path
1770         https://bugs.webkit.org/show_bug.cgi?id=38701
1771
1772         Adds a new API for getting font family. For now, FontCacheLinux calls the new API, but don't use additional properties for compatibility. The old API will be removed when Chromium is ready to use new API.
1773
1774         Reviewed by Tony Chang.
1775
1776         No new tests. No behavior changes for now.
1777
1778         * platform/chromium/PlatformSupport.h: Added FontFamily struct and changed the declaration of getFontFamilyForCharacters().
1779         * platform/graphics/chromium/FontCacheLinux.cpp:
1780         (WebCore::FontCache::getFontDataForCharacters): Uses new PlatformSupport::getFontFamilyForCharacters().
1781         * platform/graphics/chromium/FontPlatformDataLinux.h:
1782         (WebCore::FontPlatformData::setFakeBold): Added.
1783         (WebCore::FontPlatformData::setFakeItalic): Added.
1784
1785 2011-11-17  Mario Sanchez Prada  <msanchez@igalia.com>
1786
1787         [GTK] Consider parent AtkObject in webkit_accessible_get_parent(), if already set
1788         https://bugs.webkit.org/show_bug.cgi?id=72525
1789
1790         Reviewed by Xan Lopez.
1791
1792         * accessibility/gtk/AccessibilityObjectWrapperAtk.cpp:
1793         (webkit_accessible_get_parent): Call to the implementation of
1794         atk_object_get_parent in AtkObject class to check whether a parent
1795         AtkObject has been previously set, before trying to find one.
1796
1797 2011-11-17  Pavel Feldman  <pfeldman@google.com>
1798
1799         Not reviewed: fix IE user agents strings in the inspector.
1800
1801         * inspector/front-end/SettingsScreen.js:
1802         (WebInspector.SettingsScreen.prototype._createUserAgentSelectRowElement.get const):
1803
1804 2011-11-16  Kenneth Rohde Christiansen  <kenneth@webkit.org>
1805
1806         Make use-fixed-layout work reliable
1807         https://bugs.webkit.org/show_bug.cgi?id=72511
1808
1809         Reviewed by Simon Hausmann.
1810
1811         Always send a viewport update per page load as we depend on that,
1812         to reset all viewport handling before doing layout.
1813
1814         * page/Page.cpp:
1815         (WebCore::Page::updateViewportArguments):
1816
1817 2011-11-16  Alexander Pavlov  <apavlov@chromium.org>
1818
1819         Web Inspector: inspector follows javascript: hrefs as relative
1820         https://bugs.webkit.org/show_bug.cgi?id=72373
1821
1822         javascript: hrefs should never be linkified for security.
1823
1824         Reviewed by Yury Semikhatsky.
1825
1826         * inspector/front-end/ElementsTreeOutline.js:
1827         (WebInspector.ElementsTreeElement.prototype._buildAttributeDOM):
1828         * inspector/front-end/ResourceUtils.js:
1829         (WebInspector.completeURL):
1830
1831 2011-11-17  Nikolas Zimmermann  <nzimmermann@rim.com>
1832
1833         Not reviewed. Fix 32bit builds.
1834
1835         * platform/ClockGeneric.cpp:
1836         (ClockGeneric::now): Use narrowPrecisionToFloat.
1837         * rendering/FilterEffectRenderer.cpp:
1838         (WebCore::FilterEffectRenderer::build): Use it correctly.
1839
1840 2011-11-17  Dominic Mazzoni  <dmazzoni@google.com>
1841
1842         Accessibility: Chromium requires an AX notification when an iframe loads.
1843         https://bugs.webkit.org/show_bug.cgi?id=72239
1844
1845         When a document finishes loading, we were sending an AXLoadComplete if it
1846         was the top document. Now, if it's a document in an iframe, send an
1847         AXLayoutComplete on the iframe.
1848
1849         Reviewed by Chris Fleizach.
1850
1851         Test: accessibility/loading-iframe-sends-notification.html
1852
1853         * dom/Document.cpp:
1854         (WebCore::Document::implicitClose):
1855
1856 2011-11-16  Mark Rowe  <mrowe@apple.com>
1857
1858         <http://webkit.org/b/72574> Remove unnecessary use of CarbonCore APIs from Audio code
1859
1860         Reviewed by Andy Estes.
1861
1862         * platform/audio/mac/AudioDestinationMac.cpp:
1863         (WebCore::AudioDestinationMac::AudioDestinationMac): Switch from using the Carbon Component Manager
1864         to using AudioUnit's own component interface.
1865         (WebCore::AudioDestinationMac::~AudioDestinationMac): Ditto.
1866         * platform/audio/mac/AudioFileReaderMac.cpp:
1867         (WebCore::AudioFileReader::AudioFileReader): Remove an unncessary trip through the Carbon File Manager
1868         when converting a char* path to a CFURLRef representing the same.
1869
1870 2011-11-17  Adam Barth  <abarth@webkit.org>
1871
1872         CSP report-only mode doesn't work from an HTTP header
1873         https://bugs.webkit.org/show_bug.cgi?id=71958
1874
1875         Reviewed by Eric Seidel.
1876
1877         "It's tested or it's broken." -- Adam Leventhal
1878
1879         Test: http/tests/security/contentSecurityPolicy/report-only-from-header.php
1880
1881         * loader/FrameLoader.cpp:
1882         (WebCore::FrameLoader::didBeginDocument):
1883
1884 2011-11-15  Pavel Feldman  <pfeldman@google.com>
1885
1886         Web Inspector: introduce Debugger domain capabilities concept.
1887         https://bugs.webkit.org/show_bug.cgi?id=72393
1888
1889         Many of the Preferences that we have in Settings.js and override in DevTools.js
1890         are really not preferences, but capabilities. Protocol clients should have a way
1891         of figuring out whether some capability is present before using it.
1892
1893         Reviewed by Yury Semikhatsky.
1894
1895         * bindings/js/ScriptDebugServer.cpp:
1896         (WebCore::ScriptDebugServer::canSetScriptSource):
1897         * bindings/js/ScriptDebugServer.h:
1898         * bindings/v8/ScriptDebugServer.cpp:
1899         (WebCore::ScriptDebugServer::canSetScriptSource):
1900         * bindings/v8/ScriptDebugServer.h:
1901         * inspector/CodeGeneratorInspector.py:
1902         * inspector/Inspector.json:
1903         * inspector/InspectorDebuggerAgent.cpp:
1904         (WebCore::InspectorDebuggerAgent::getCapabilities):
1905         * inspector/InspectorDebuggerAgent.h:
1906         * inspector/front-end/DebuggerModel.js:
1907         (WebInspector.DebuggerModel):
1908         (WebInspector.DebuggerModel.prototype.enableDebugger):
1909         (WebInspector.DebuggerModel.prototype.canSetScriptSource):
1910         * inspector/front-end/DebuggerPresentationModel.js:
1911         (WebInspector.DebuggerPresentationModel.prototype.canEditScriptSource):
1912         * inspector/front-end/Settings.js:
1913
1914 2011-11-17  Mihnea Ovidenie  <mihnea@adobe.com>
1915
1916         CSS exclusions: parse the wrap-flow and wrap-through properties
1917         https://bugs.webkit.org/show_bug.cgi?id=71904
1918
1919         Reviewed by Dean Jackson.
1920
1921         Tests: fast/exclusions/wrap-flow-parsing.html
1922                fast/exclusions/wrap-through-parsing.html
1923
1924         * css/CSSComputedStyleDeclaration.cpp:
1925         (WebCore::CSSComputedStyleDeclaration::getPropertyCSSValue):
1926         * css/CSSParser.cpp:
1927         (WebCore::CSSParser::parseValue):
1928         * css/CSSPrimitiveValueMappings.h:
1929         (WebCore::CSSPrimitiveValue::CSSPrimitiveValue):
1930         (WebCore::CSSPrimitiveValue::operator WrapFlow):
1931         (WebCore::CSSPrimitiveValue::operator WrapThrough):
1932         * css/CSSProperty.cpp:
1933         (WebCore::CSSProperty::isInheritedProperty):
1934         * css/CSSPropertyNames.in:
1935         * css/CSSStyleSelector.cpp:
1936         (WebCore::CSSStyleSelector::applyProperty):
1937         * css/CSSValueKeywords.in:
1938         * rendering/style/RenderStyle.cpp:
1939         (WebCore::RenderStyle::diff):
1940         * rendering/style/RenderStyle.h:
1941         (WebCore::InheritedFlags::wrapFlow):
1942         (WebCore::InheritedFlags::wrapThrough):
1943         (WebCore::InheritedFlags::setWrapFlow):
1944         (WebCore::InheritedFlags::setWrapThrough):
1945         (WebCore::InheritedFlags::initialWrapFlow):
1946         (WebCore::InheritedFlags::initialWrapThrough):
1947         * rendering/style/RenderStyleConstants.h:
1948         * rendering/style/StyleRareNonInheritedData.cpp:
1949         (WebCore::StyleRareNonInheritedData::StyleRareNonInheritedData):
1950         (WebCore::StyleRareNonInheritedData::operator==):
1951         * rendering/style/StyleRareNonInheritedData.h:
1952
1953 2011-11-16  Dean Jackson  <dino@apple.com>
1954
1955         Implement filter function shorthands
1956         https://bugs.webkit.org/show_bug.cgi?id=68475
1957
1958         Reviewed by Simon Fraser.
1959
1960         Implement the shorthand functions for filter effects.
1961         This includes grayscale, sepia, invert, hue-rotate, saturate,
1962         opacity, gamma, drop-shadow and blur. At the moment sharpen
1963         and url are not supported.
1964
1965         CSSParser needed to be updated because it was mistakenly
1966         clamping saturation values to [0,1]. Any positive number
1967         is allowed so you can produce super-saturated images.
1968
1969         The biggest change was the API to FilterEffectRenderer. It now
1970         builds a list of effects and applies the filter itself.
1971
1972         Note that the drop-shadow and blur operations don't yet
1973         provide accurate results because they produce an output image
1974         that is larger than the input. See
1975         https://bugs.webkit.org/show_bug.cgi?id=71929
1976         https://bugs.webkit.org/show_bug.cgi?id=71930
1977
1978         While I was there, I fixed a small style issue in
1979         CustomFilterOperation.
1980
1981         Tests: css3/filters/effect-blur.html
1982                css3/filters/effect-combined.html
1983                css3/filters/effect-drop-shadow.html
1984                css3/filters/effect-gamma.html
1985                css3/filters/effect-grayscale.html
1986                css3/filters/effect-hue-rotate.html
1987                css3/filters/effect-invert.html
1988                css3/filters/effect-opacity.html
1989                css3/filters/effect-saturate.html
1990                css3/filters/effect-sepia.html
1991
1992         * WebCore.xcodeproj/project.pbxproj: Add StyleShader.h to
1993         the project (missing from earlier commit).
1994         * css/CSSParser.cpp:
1995         (WebCore::CSSParser::isValidFilterArgument): Don't clamp
1996         saturate to [0,1]
1997         * platform/graphics/filters/CustomFilterOperation.h:
1998         * rendering/FilterEffectRenderer.cpp:
1999         (WebCore::endMatrixRow):
2000         (WebCore::lastMatrixRow):
2001         (WebCore::FilterEffectRenderer::FilterEffectRenderer):
2002         (WebCore::FilterEffectRenderer::inputContext):
2003         (WebCore::FilterEffectRenderer::build):
2004         (WebCore::FilterEffectRenderer::prepare):
2005         (WebCore::FilterEffectRenderer::apply):
2006         * rendering/FilterEffectRenderer.h:
2007         (WebCore::FilterEffectRenderer::setSourceImageRect):
2008         (WebCore::FilterEffectRenderer::output):
2009         (WebCore::FilterEffectRenderer::setMaxEffectRects):
2010         (WebCore::FilterEffectRenderer::lastEffect):
2011         * rendering/RenderLayer.cpp:
2012         (WebCore::RenderLayer::paintLayer):
2013         (WebCore::RenderLayer::updateOrRemoveFilterEffect):
2014         (WebCore::RenderLayer::updateFilterBackingStore):
2015
2016 2011-11-16  Kentaro Hara  <haraken@chromium.org>
2017
2018         Remove all custom constructors of Events from JSC
2019         https://bugs.webkit.org/show_bug.cgi?id=72577
2020
2021         Reviewed by Adam Barth.
2022
2023         - Makes CodeGeneratorJS.pm generate Event constructors
2024         if [ConstructorTemplate=Event] IDL is specified.
2025         - Removes EventConstructors.h and JSEventConstructors.cpp.
2026         - Replaces all JSC custom constructors of Events
2027         with the generated code by [ConstructorTemplate=Event] IDL.
2028
2029         Tests: fast/events/constructors/before-load-event-constructor.html
2030                fast/events/constructors/close-event-constructor.html
2031                fast/events/constructors/custom-event-constructor.html
2032                fast/events/constructors/error-event-constructor.html
2033                fast/events/constructors/event-constructors.html
2034                fast/events/constructors/hash-change-event-constructor.html
2035                fast/events/constructors/message-event-constructor.html
2036                fast/events/constructors/overflow-event-constructor.html
2037                fast/events/constructors/page-transition-event-constructor.html
2038                fast/events/constructors/pop-state-event-constructor.html
2039                fast/events/constructors/progress-event-constructor.html
2040                fast/events/constructors/track-event-constructor.html
2041                fast/events/constructors/webkit-animation-event-constructor.html
2042                fast/events/constructors/webkit-transition-event-constructor.html
2043
2044         * bindings/generic/EventConstructors.h: Removed. This is what we wanted to do in this patch.
2045         * bindings/js/JSEventConstructors.cpp: Ditto.
2046         * GNUmakefile.list.am: Removed EventConstructors.h and JSEventConstructors.cpp.
2047         * Target.pri: Ditto.
2048         * UseJSC.cmake: Ditto.
2049         * WebCore.gypi: Ditto.
2050         * WebCore.vcproj/WebCore.vcproj: Ditto.
2051         * WebCore.xcodeproj/project.pbxproj: Ditto.
2052         * bindings/js/JSBindingsAllInOne.cpp: Ditto.
2053
2054         * bindings/scripts/CodeGeneratorJS.pm:
2055         (GenerateHeader): Added JSDictionary.h.
2056         (GenerateConstructorDeclaration): Added a header for fillXXXXEventInit(...).
2057         (GenerateConstructorDefinition): Generates an Event constructor. The generated code is the same as the code that had been written in JSEventConstructors.cpp.
2058         (IsConstructable): Judges if a given interface is constructable.
2059         (IsConstructorTemplate): Judges if a given interface has a given template, e.g. judges if an interface has [ConstructorTemplate=Event].
2060
2061         * bindings/scripts/test/TestEventConstructor.idl: Changed 'CustomConstructor=Event' to 'ConstructorTemplate=Event'. We should have changed this in r100108.
2062         * bindings/scripts/test/JS/JSTestEventConstructor.cpp: Updated a run-bindings-tests result.
2063         (WebCore::JSTestEventConstructorConstructor::constructJSTestEventConstructor):
2064         (WebCore::fillTestEventConstructorInit):
2065         * bindings/scripts/test/JS/JSTestEventConstructor.h: Ditto.
2066         * bindings/scripts/test/V8/V8TestEventConstructor.cpp: Ditto.
2067         (WebCore::V8TestEventConstructor::constructorCallback):
2068         (WebCore::fillTestEventConstructorInit):
2069         * bindings/scripts/test/V8/V8TestEventConstructor.h: Ditto.
2070
2071         * dom/BeforeLoadEvent.idl: In essence, replaced [JSCustomConstructor] IDL with [JSConstructorTemplate=Event] IDL.
2072         * dom/CustomEvent.idl: Ditto.
2073         * dom/ErrorEvent.idl: Ditto.
2074         * dom/Event.idl: Ditto.
2075         * dom/HashChangeEvent.idl: Ditto.
2076         * dom/MessageEvent.idl: Ditto.
2077         * dom/OverflowEvent.idl: Ditto.
2078         * dom/PageTransitionEvent.idl: Ditto.
2079         * dom/PopStateEvent.idl: Ditto.
2080         * dom/ProgressEvent.idl: Ditto.
2081         * dom/WebKitAnimationEvent.idl: Ditto.
2082         * dom/WebKitTransitionEvent.idl: Ditto.
2083         * html/track/TrackEvent.idl: Ditto.
2084         * websockets/CloseEvent.idl: Ditto.
2085
2086 2011-11-16  Dean Jackson  <dino@apple.com>
2087
2088         Build fix for Apple WebKit due to r100560.
2089         When I removed the violating header file it lost the
2090         reference to Color.
2091
2092         No review.
2093
2094         * platform/graphics/filters/FilterOperation.h:
2095
2096 2011-11-16  Dean Jackson  <dino@apple.com>
2097
2098         DropShadowFilterOperation violates platform isolation
2099         https://bugs.webkit.org/show_bug.cgi?id=72544
2100
2101         Reviewed by Simon Fraser.
2102
2103         Move ShadowData properties into the DropShadowFilterOperation
2104         to avoid depending on something outside platform.
2105
2106         * css/CSSComputedStyleDeclaration.cpp:
2107         (WebCore::CSSComputedStyleDeclaration::valueForFilter):
2108         * css/CSSStyleSelector.cpp:
2109         (WebCore::CSSStyleSelector::createFilterOperations):
2110         * platform/graphics/filters/FilterOperation.h:
2111         (WebCore::DropShadowFilterOperation::create):
2112         (WebCore::DropShadowFilterOperation::x):
2113         (WebCore::DropShadowFilterOperation::y):
2114         (WebCore::DropShadowFilterOperation::stdDeviation):
2115         (WebCore::DropShadowFilterOperation::color):
2116         (WebCore::DropShadowFilterOperation::operator==):
2117         (WebCore::DropShadowFilterOperation::DropShadowFilterOperation):
2118
2119 2011-11-16  Adam Bergkvist  <adam.bergkvist@ericsson.com>
2120
2121         Use a simple page client for user consent in getUserMedia()
2122         https://bugs.webkit.org/show_bug.cgi?id=70897
2123
2124         Reviewed by Adam Barth.
2125
2126         This is one in a series of patches that update the MediaStream feature
2127         to use WebCore platform interfaces.
2128
2129         Covered by existing tests.
2130
2131         * GNUmakefile.list.am:
2132         * WebCore.gypi:
2133         * mediastream/MediaStreamClient.h: Removed.
2134         * mediastream/MediaStreamController.cpp: Removed.
2135         * mediastream/MediaStreamController.h: Removed.
2136         * mediastream/MediaStreamFrameController.cpp: Removed.
2137         * mediastream/MediaStreamFrameController.h: Removed.
2138         * mediastream/UserMediaClient.h: Added.
2139         (WebCore::UserMediaClient::~UserMediaClient):
2140         * mediastream/UserMediaRequest.cpp: Added.
2141         (WebCore::UserMediaRequest::create):
2142         (WebCore::UserMediaRequest::UserMediaRequest):
2143         (WebCore::UserMediaRequest::~UserMediaRequest):
2144         (WebCore::UserMediaRequest::start):
2145         (WebCore::UserMediaRequest::mediaStreamSourcesQueryCompleted):
2146         (WebCore::UserMediaRequest::succeed):
2147         (WebCore::UserMediaRequest::fail):
2148         (WebCore::UserMediaRequest::contextDestroyed):
2149         (WebCore::UserMediaRequest::parseOptions):
2150         * mediastream/UserMediaRequest.h: Added.
2151         (WebCore::UserMediaRequest::audio):
2152         (WebCore::UserMediaRequest::video):
2153         (WebCore::UserMediaRequest::cameraPreferenceUser):
2154         (WebCore::UserMediaRequest::cameraPreferenceEnvironment):
2155         (WebCore::UserMediaRequest::successCallback):
2156         (WebCore::UserMediaRequest::errorCallback):
2157         * page/CallbackTask.h: Removed.
2158         * page/Frame.cpp:
2159         (WebCore::Frame::Frame):
2160         (WebCore::Frame::~Frame):
2161         (WebCore::Frame::pageDestroyed):
2162         (WebCore::Frame::transferChildFrameToNewDocument):
2163         * page/Frame.h:
2164         * page/Navigator.cpp:
2165         (WebCore::Navigator::webkitGetUserMedia):
2166         * page/NavigatorUserMediaErrorCallback.h:
2167         * page/Page.cpp:
2168         (WebCore::Page::Page):
2169         (WebCore::Page::~Page):
2170         (WebCore::Page::PageClients::PageClients):
2171         * page/Page.h:
2172         (WebCore::Page::userMediaClient):
2173
2174 2011-11-16  David Grogan  <dgrogan@chromium.org>
2175
2176         instantiate WorkerEventQueue in WorkerContext
2177         https://bugs.webkit.org/show_bug.cgi?id=72528
2178
2179         Reviewed by David Levin.
2180
2181         No new tests - nothing uses it yet.
2182
2183         * workers/WorkerContext.cpp:
2184         (WebCore::WorkerContext::WorkerContext):
2185         Instantiate WorkerEventQueue in InitializerList
2186
2187         * workers/WorkerEventQueue.h:
2188         Don't inherit from RefCounted, WorkerContext has an OwnPtr to it
2189
2190 2011-11-16  Kentaro Hara  <haraken@chromium.org>
2191
2192         Unreviewed. Rebaselined run-bindings-tests results.
2193
2194         * bindings/scripts/test/JS/JSTestEventConstructor.cpp:
2195         (WebCore::JSTestEventConstructor::JSTestEventConstructor):
2196         (WebCore::JSTestEventConstructorOwner::finalize):
2197         * bindings/scripts/test/JS/JSTestEventConstructor.h:
2198         (WebCore::JSTestEventConstructor::impl):
2199         (WebCore::JSTestEventConstructor::releaseImpl):
2200         * bindings/scripts/test/JS/JSTestInterface.cpp:
2201         (WebCore::JSTestInterface::JSTestInterface):
2202         (WebCore::JSTestInterfaceOwner::finalize):
2203         * bindings/scripts/test/JS/JSTestInterface.h:
2204         (WebCore::JSTestInterface::impl):
2205         (WebCore::JSTestInterface::releaseImpl):
2206         * bindings/scripts/test/JS/JSTestMediaQueryListListener.cpp:
2207         (WebCore::JSTestMediaQueryListListener::JSTestMediaQueryListListener):
2208         (WebCore::JSTestMediaQueryListListenerOwner::finalize):
2209         * bindings/scripts/test/JS/JSTestMediaQueryListListener.h:
2210         (WebCore::JSTestMediaQueryListListener::impl):
2211         (WebCore::JSTestMediaQueryListListener::releaseImpl):
2212         * bindings/scripts/test/JS/JSTestNamedConstructor.cpp:
2213         (WebCore::JSTestNamedConstructor::JSTestNamedConstructor):
2214         (WebCore::JSTestNamedConstructorOwner::finalize):
2215         * bindings/scripts/test/JS/JSTestNamedConstructor.h:
2216         (WebCore::JSTestNamedConstructor::impl):
2217         (WebCore::JSTestNamedConstructor::releaseImpl):
2218         * bindings/scripts/test/JS/JSTestObj.cpp:
2219         (WebCore::JSTestObj::JSTestObj):
2220         (WebCore::JSTestObjOwner::finalize):
2221         * bindings/scripts/test/JS/JSTestObj.h:
2222         (WebCore::JSTestObj::impl):
2223         (WebCore::JSTestObj::releaseImpl):
2224         * bindings/scripts/test/JS/JSTestSerializedScriptValueInterface.cpp:
2225         (WebCore::JSTestSerializedScriptValueInterface::JSTestSerializedScriptValueInterface):
2226         (WebCore::JSTestSerializedScriptValueInterfaceOwner::finalize):
2227         * bindings/scripts/test/JS/JSTestSerializedScriptValueInterface.h:
2228         (WebCore::JSTestSerializedScriptValueInterface::impl):
2229         (WebCore::JSTestSerializedScriptValueInterface::releaseImpl):
2230
2231 2011-11-16  Jay Civelli  <jcivelli@chromium.org>
2232
2233         Make sure MHTML documents use the domain of the MHTML file.
2234         https://bugs.webkit.org/show_bug.cgi?id=72445
2235
2236         Reviewed by Adam Barth.
2237
2238         * dom/Document.h:
2239         (WebCore::Document::setBaseURL):
2240         * loader/FrameLoader.cpp:
2241         (WebCore::FrameLoader::receivedFirstData):
2242
2243 2011-11-16  Julien Chaffraix  <jchaffraix@webkit.org>
2244
2245         Update supported display list after -webkit-grid and -webkit-inline-grid addition
2246         https://bugs.webkit.org/show_bug.cgi?id=72559
2247
2248         Reviewed by Tony Chang.
2249
2250         * css/CSSParser.cpp:
2251         (WebCore::CSSParser::parseValue): Added the 2 new supported
2252         values (forgotten in the previous change).
2253
2254 2011-11-11  Adrienne Walker  <enne@google.com>
2255
2256         [chromium] Expose mock scrollbars to window.internals
2257         https://bugs.webkit.org/show_bug.cgi?id=72195
2258
2259         Reviewed by James Robinson.
2260
2261         * testing/Internals.cpp:
2262         (WebCore::Internals::setMockScrollbarsEnabled):
2263         * testing/Internals.h:
2264         * testing/Internals.idl:
2265
2266 2011-11-16  Michael Nordman  <michaeln@google.com>
2267
2268         ApplicationCache manifest should work with any MIME type.
2269         https://bugs.webkit.org/show_bug.cgi?id=72082
2270
2271         Reviewed by Alexey Proskuryakov.
2272
2273         * loader/appcache/ApplicationCacheGroup.cpp:
2274         (WebCore::ApplicationCacheGroup::didReceiveManifestResponse): Remove the test for a particular type.
2275
2276 2011-11-16  Daniel Sievers  <sievers@chromium.org>
2277
2278         [Chromium] Avoid color mask operations for root layers
2279         https://bugs.webkit.org/show_bug.cgi?id=72452
2280
2281         Instead of relying on the combination of clearing the surface and initializing
2282         the alpha channel to 1.0 followed by disabling alpha in the color mask when
2283         rendering the root layer tiles, add shaders to support writing out opaque layers
2284         (alpha channel values written as 1.0).
2285
2286         Reviewed by James Robinson.
2287
2288         No functional change made that requires new tests.
2289
2290         * platform/graphics/chromium/LayerRendererChromium.cpp:
2291         (WebCore::LayerRendererChromium::drawLayersInternal):
2292         (WebCore::LayerRendererChromium::initializeSharedObjects):
2293         (WebCore::LayerRendererChromium::tilerProgramOpaque):
2294         (WebCore::LayerRendererChromium::tilerProgramOpaqueAA):
2295         (WebCore::LayerRendererChromium::tilerProgramSwizzleOpaque):
2296         (WebCore::LayerRendererChromium::tilerProgramSwizzleOpaqueAA):
2297         (WebCore::LayerRendererChromium::cleanupSharedObjects):
2298         * platform/graphics/chromium/LayerRendererChromium.h:
2299         * platform/graphics/chromium/ShaderChromium.cpp:
2300         (WebCore::FragmentTexOpaqueBinding::FragmentTexOpaqueBinding):
2301         (WebCore::FragmentTexOpaqueBinding::init):
2302         (WebCore::FragmentShaderRGBATexOpaque::getShaderString):
2303         (WebCore::FragmentShaderRGBATexSwizzleOpaque::getShaderString):
2304         (WebCore::FragmentTexClampOpaqueAABinding::FragmentTexClampOpaqueAABinding):
2305         (WebCore::FragmentTexClampOpaqueAABinding::init):
2306         (WebCore::FragmentShaderRGBATexClampOpaqueAA::getShaderString):
2307         (WebCore::FragmentShaderRGBATexClampSwizzleOpaqueAA::getShaderString):
2308         * platform/graphics/chromium/ShaderChromium.h:
2309         (WebCore::FragmentTexOpaqueBinding::alphaLocation):
2310         (WebCore::FragmentTexOpaqueBinding::samplerLocation):
2311         (WebCore::FragmentTexClampOpaqueAABinding::alphaLocation):
2312         (WebCore::FragmentTexClampOpaqueAABinding::samplerLocation):
2313         (WebCore::FragmentTexClampOpaqueAABinding::fragmentTexTransformLocation):
2314         (WebCore::FragmentTexClampOpaqueAABinding::edgeLocation):
2315         * platform/graphics/chromium/cc/CCTiledLayerImpl.cpp:
2316         (WebCore::CCTiledLayerImpl::draw):
2317         * platform/graphics/chromium/cc/CCTiledLayerImpl.h:
2318
2319 2011-11-16  Tim Horton  <timothy_horton@apple.com>
2320
2321         Implement CSS3 Images cross-fade() image function
2322         https://bugs.webkit.org/show_bug.cgi?id=52162
2323         <rdar://problem/10209254>
2324
2325         Reviewed by Simon Fraser.
2326
2327         Render -webkit-cross-fade. Only cross-fades entirely composed of images will render for now,
2328         cross-fades involving generated images are not yet implemented.
2329
2330         Reorganize GeneratedImage to be the base class for GeneratorGeneratedImage and CrossfadeGeneratedImage.
2331
2332         Add a pending state to CSSImageGeneratorValue, which is used to enable the pending-images loading
2333         mechanism for -webkit-cross-fade's sub-images. Rework the logic in CSSStyleSelector to support pending
2334         generated images.
2335         
2336         Support parsing fractional values for the cross-fade amount (for example, 0.5 = 50%). Clamp cross-fade
2337         amount to 0-1 range.
2338
2339         Tests: css3/images/cross-fade-invalidation.html
2340                css3/images/cross-fade-simple.html
2341                css3/images/cross-fade-sizing.html
2342                css3/images/cross-fade-tiled.html
2343
2344         * CMakeLists.txt:
2345         * GNUmakefile.list.am:
2346         * Target.pri:
2347         * WebCore.gypi:
2348         * WebCore.xcodeproj/project.pbxproj:
2349         * css/CSSCanvasValue.h:
2350         (WebCore::CSSCanvasValue::isPending):
2351         (WebCore::CSSCanvasValue::loadSubimages):
2352         * css/CSSCrossfadeValue.cpp:
2353         (WebCore::CSSCrossfadeValue::isPending):
2354         (WebCore::CSSCrossfadeValue::loadSubimages):
2355         (WebCore::subimageIsPending):
2356         (WebCore::loadSubimage):
2357         (WebCore::cachedImageForCSSValue):
2358         (WebCore::CSSCrossfadeValue::image):
2359         (WebCore::CSSCrossfadeValue::crossfadeChanged):
2360         * css/CSSCrossfadeValue.h:
2361         (WebCore::CSSCrossfadeValue::create):
2362         (WebCore::CSSCrossfadeValue::~CSSCrossfadeValue):
2363         (WebCore::CSSCrossfadeValue::fixedSize):
2364         (WebCore::CSSCrossfadeValue::CSSCrossfadeValue):
2365         (WebCore::CSSCrossfadeValue::CrossfadeObserverProxy::CrossfadeObserverProxy):
2366         * css/CSSGradientValue.cpp:
2367         (WebCore::CSSGradientValue::image):
2368         * css/CSSGradientValue.h:
2369         (WebCore::CSSGradientValue::isPending):
2370         (WebCore::CSSGradientValue::loadSubimages):
2371         * css/CSSImageGeneratorValue.cpp:
2372         (WebCore::CSSImageGeneratorValue::generatedOrPendingImage):
2373         (WebCore::CSSImageGeneratorValue::generatedImage):
2374         (WebCore::CSSImageGeneratorValue::isPending):
2375         (WebCore::CSSImageGeneratorValue::loadSubimages):
2376         * css/CSSImageGeneratorValue.h:
2377         * css/CSSParser.cpp:
2378         (WebCore::CSSParser::parseCrossfade):
2379         * css/CSSStyleSelector.cpp:
2380         (WebCore::CSSStyleSelector::styleImage):
2381         (WebCore::CSSStyleSelector::generatedOrPendingFromValue):
2382         (WebCore::CSSStyleSelector::loadPendingImage):
2383         (WebCore::CSSStyleSelector::loadPendingImages):
2384         * css/CSSStyleSelector.h:
2385         * platform/graphics/BitmapImage.h:
2386         * platform/graphics/CrossfadeGeneratedImage.cpp: Added.
2387         (WebCore::CrossfadeGeneratedImage::CrossfadeGeneratedImage):
2388         (WebCore::CrossfadeGeneratedImage::~CrossfadeGeneratedImage):
2389         (WebCore::CrossfadeGeneratedImage::drawCrossfade):
2390         (WebCore::CrossfadeGeneratedImage::draw):
2391         (WebCore::CrossfadeGeneratedImage::drawPattern):
2392         (WebCore::CrossfadeGeneratedImage::imageChanged):
2393         * platform/graphics/CrossfadeGeneratedImage.h: Added.
2394         (WebCore::CrossfadeGeneratedImage::create):
2395         (WebCore::CrossfadeSubimageObserverProxy::CrossfadeSubimageObserverProxy):
2396         (WebCore::CrossfadeSubimageObserverProxy::setReady):
2397         * platform/graphics/GeneratedImage.h:
2398         (WebCore::GeneratedImage::GeneratedImage):
2399         * platform/graphics/GeneratorGeneratedImage.cpp: Renamed from Source/WebCore/platform/graphics/GeneratedImage.cpp.
2400         (WebCore::GeneratorGeneratedImage::draw):
2401         (WebCore::GeneratorGeneratedImage::drawPattern):
2402         (WebCore::GeneratedImage::computeIntrinsicDimensions):
2403         * platform/graphics/GeneratorGeneratedImage.h: Copied from Source/WebCore/platform/graphics/GeneratedImage.h.
2404         (WebCore::GeneratorGeneratedImage::create):
2405         (WebCore::GeneratorGeneratedImage::~GeneratorGeneratedImage):
2406         (WebCore::GeneratorGeneratedImage::GeneratorGeneratedImage):
2407         * platform/graphics/Image.h:
2408         * platform/graphics/ImageBuffer.h:
2409         * rendering/style/StylePendingImage.h:
2410         (WebCore::StylePendingImage::create):
2411         (WebCore::StylePendingImage::data):
2412         (WebCore::StylePendingImage::cssImageValue):
2413         (WebCore::StylePendingImage::cssImageGeneratorValue):
2414         (WebCore::StylePendingImage::StylePendingImage):
2415
2416
2417 2011-11-16  Dan Bernstein  <mitz@apple.com>
2418
2419         WebCore part of <rdar://problem/10262242> Add API for paginated display
2420         https://bugs.webkit.org/show_bug.cgi?id=72537
2421
2422         Reviewed by Anders Carlsson.
2423
2424         * WebCore.exp.in: Exported Page::pageCount().
2425         * page/Page.cpp:
2426         (WebCore::Page::pageCount): Added this getter.
2427         * page/Page.h:
2428
2429 2011-11-16  Shawn Singh  <shawnsingh@chromium.org>
2430
2431         [chromium] Track property changes for render surfaces.
2432         https://bugs.webkit.org/show_bug.cgi?id=72521
2433
2434         Reviewed by James Robinson.
2435
2436         Created CCRenderSurfaceTest for testing.
2437
2438         * platform/graphics/chromium/LayerRendererChromium.cpp:
2439         (WebCore::LayerRendererChromium::drawLayersOntoRenderSurfaces):
2440         * platform/graphics/chromium/cc/CCLayerImpl.cpp:
2441         (WebCore::CCLayerImpl::resetPropertyChangedFlagForSubtree):
2442         * platform/graphics/chromium/cc/CCLayerImpl.h:
2443         * platform/graphics/chromium/cc/CCRenderSurface.cpp:
2444         (WebCore::CCRenderSurface::CCRenderSurface):
2445         (WebCore::CCRenderSurface::setClipRect):
2446         (WebCore::CCRenderSurface::setContentRect):
2447         (WebCore::CCRenderSurface::surfacePropertyChanged):
2448         * platform/graphics/chromium/cc/CCRenderSurface.h:
2449         (WebCore::CCRenderSurface::resetPropertyChangedFlag):
2450
2451 2011-11-16  Ben Wells  <benwells@chromium.org>
2452
2453         Seaming on border corners with mixed colour alpha borders
2454         https://bugs.webkit.org/show_bug.cgi?id=70471
2455
2456         Reviewed by Simon Fraser.
2457
2458         Seaming is fixed by antialiasing mitred corners for edges that have alpha and are joining
2459         a side that is of a different color.
2460
2461         Test: fast/borders/border-mixed-alpha.html
2462
2463         * rendering/RenderBoxModelObject.cpp:
2464         (WebCore::colorNeedsAntiAliasAtCorner):
2465         (WebCore::RenderBoxModelObject::paintOneBorderSide):
2466
2467 2011-11-16  Sam Weinig  <sam@webkit.org>
2468
2469         JS wrappers of DOM objects should have no-op constructors
2470         https://bugs.webkit.org/show_bug.cgi?id=72556
2471
2472         Reviewed by Geoffrey Garen.
2473
2474         Stop using a RefPtr to hold DOM objects contained by JavaScript
2475         wrappers and instead use a raw pointer. We were already releasing
2476         the underlying object before the destructor ran (via the finalizer)
2477         so the default behavior of destroying the RefPtr is always unnecessary
2478         busy work. 
2479
2480         * bindings/js/JSCSSValueCustom.cpp:
2481         (WebCore::JSCSSValueOwner::finalize):
2482         * bindings/js/JSNodeCustom.cpp:
2483         (WebCore::JSNodeOwner::finalize):
2484         (WebCore::JSNode::visitChildren):
2485         Call releaseImpl() instead of clearImpl().
2486
2487         * bindings/scripts/CodeGeneratorJS.pm:
2488         (GenerateHeader):
2489         Stop storing m_impl in a RefPtr and instead use a raw pointer. Switch
2490         clearImpl() to releaseImpl(), which explicitly derefs the pointer and
2491         clear it.
2492
2493         (GenerateImplementation):
2494         Use leakPtr() to explicitly adopt the PassRefPtr into the raw pointer.
2495         Change default finalize to call releaseImpl() instead of clearImpl().
2496
2497 2011-11-16  Michael Saboff  <msaboff@apple.com>
2498
2499         Enable 8 Bit Strings in JavaScriptCore
2500         https://bugs.webkit.org/show_bug.cgi?id=71337
2501
2502         This patch turns on 8 bit strings in StringImpl and enables
2503         their use in JavaScriptCore. Some of the changes are to
2504         turn on code that had been staged (Lexer.cpp, Identifier.cpp,
2505         SmallStrings.cpp and some of StringImpl.{h,cpp}).
2506         Other changes are minor fixes to make 8 bit strings work
2507         (UString.h, StringImpl::getData16SlowCase()).
2508         Changed StringBuffer to be a templated class based on character
2509         type.  This change riplled into WebCore code as well.
2510
2511         Reviewed by Geoffrey Garen.
2512
2513         No new tests - Changes in response to refactoring StringBuffer to
2514         be a template on character type.
2515
2516         * css/CSSParser.cpp:
2517         (WebCore::quoteCSSString):
2518         * css/CSSPrimitiveValue.cpp:
2519         (WebCore::formatNumber):
2520         * dom/Document.cpp:
2521         (WebCore::canonicalizedTitle):
2522         * platform/Length.cpp:
2523         (WebCore::newCoordsArray):
2524         * platform/sql/SQLiteStatement.cpp:
2525         (WebCore::SQLiteStatement::prepare):
2526         * platform/text/TextCodecUTF16.cpp:
2527         (WebCore::TextCodecUTF16::decode):
2528         * platform/text/TextCodecUTF8.cpp:
2529         (WebCore::TextCodecUTF8::decode):
2530         * rendering/RenderText.cpp:
2531         (WebCore::makeCapitalized):
2532         * xml/XSLTProcessorLibxslt.cpp:
2533         (WebCore::writeToStringBuilder):
2534
2535 2011-11-16  Alexandre Elias  <aelias@google.com>
2536
2537         [chromium] Improvements for page scale delta during commit
2538         https://bugs.webkit.org/show_bug.cgi?id=72471
2539
2540         Reviewed by James Robinson.
2541
2542         Page scale now follows the same commit flow as scroll position:
2543         the delta is folded into m_pageScale at BFAC time, and a "sent" value
2544         is preserved for temporary use until the commit finishes.
2545
2546         I also merged setPageScaleFactor and setPageScaleFactorLimits into one
2547         function on the impl side.  The reason is that setPageFactor must
2548         be applied after the limits are updated, but on the other hand setting
2549         the limits first may cause an unnecessary clamp of the scale delta.
2550         Merging the methods avoids this bind.
2551
2552         No new tests. (planning to add later: https://bugs.webkit.org/show_bug.cgi?id=71529)
2553
2554         * platform/graphics/chromium/cc/CCLayerTreeHost.cpp:
2555         (WebCore::CCLayerTreeHost::finishCommitOnImplThread):
2556         (WebCore::CCLayerTreeHost::applyScrollAndScale):
2557         * platform/graphics/chromium/cc/CCLayerTreeHostCommon.h:
2558         * platform/graphics/chromium/cc/CCLayerTreeHostImpl.cpp:
2559         (WebCore::CCLayerTreeHostImpl::CCLayerTreeHostImpl):
2560         (WebCore::CCLayerTreeHostImpl::setPageScaleFactorAndLimits):
2561         (WebCore::CCLayerTreeHostImpl::adjustScrollsForPageScaleChange):
2562         (WebCore::CCLayerTreeHostImpl::setScaleDelta):
2563         (WebCore::CCLayerTreeHostImpl::applyScaleDeltaToScrollLayer):
2564         (WebCore::CCLayerTreeHostImpl::scrollRootLayer):
2565         (WebCore::CCLayerTreeHostImpl::processScrollDeltas):
2566         * platform/graphics/chromium/cc/CCLayerTreeHostImpl.h:
2567
2568 2011-11-16  Joshua Bell  <jsbell@chromium.org>
2569
2570         IndexedDB: Reduce nested key depth threshold, re-enable flaky test
2571         https://bugs.webkit.org/show_bug.cgi?id=72529
2572
2573         Reviewed by Adam Barth.
2574
2575         Drop maximum array key depth from 20k to 2k.
2576
2577         * bindings/v8/IDBBindingUtilities.cpp:
2578
2579 2011-11-16  Nate Chapin  <japhet@chromium.org>
2580
2581         Fix incorrect multipart handling in r100311.
2582         SubresourceLoader::didReceiveData() is getting called
2583         twice, which has unintended side effects.
2584         https://bugs.webkit.org/show_bug.cgi?id=72436
2585
2586         Reviewed by Adam Barth.
2587
2588         http/tests/multipart/invalid-image-data.html should stop
2589         asserting after this.
2590
2591         * loader/SubresourceLoader.cpp:
2592         (WebCore::SubresourceLoader::didReceiveResponse):
2593         (WebCore::SubresourceLoader::didReceiveData):
2594         (WebCore::SubresourceLoader::sendDataToResource):
2595         * loader/SubresourceLoader.h:
2596
2597 2011-11-16  Justin Schuh  <jschuh@chromium.org>
2598
2599         Clear SVG filter client when its node is detached
2600         https://bugs.webkit.org/show_bug.cgi?id=71741
2601
2602         Reviewed by Eric Seidel.
2603
2604         Test: svg/filters/reparent-animated-filter-target.html
2605
2606         * rendering/svg/SVGResourcesCache.cpp:
2607         (WebCore::SVGResourcesCache::clientDestroyed):
2608
2609 2011-11-16  John Bates  <jbates@google.com>
2610
2611         Page/layer flashes after GPU-accelerated CSS transition
2612         https://bugs.webkit.org/show_bug.cgi?id=72343
2613
2614         LayerRendererChromium was resizing the window to 1x1 at initialization.
2615         In some cases, there is no drawLayers before switching back to
2616         software rendering. This left the window resized to 1x1 and the
2617         following software paints would therefore not be visible. This change
2618         moves the reshape call into drawLayers so that it will only be called
2619         if rendering will occur.
2620
2621         Reviewed by James Robinson.
2622
2623         New test: CCLayerTreeHostImplTest.reshapeNotCalledUntilDraw.
2624
2625         * platform/graphics/chromium/LayerRendererChromium.cpp:
2626         (WebCore::LayerRendererChromium::viewportChanged):
2627         (WebCore::LayerRendererChromium::doViewportChanged):
2628         (WebCore::LayerRendererChromium::drawLayersInternal):
2629         * platform/graphics/chromium/LayerRendererChromium.h:
2630
2631 2011-11-16  Alexandre Elias  <aelias@google.com>
2632
2633         [chromium] Add null pointer check in setDeviceScaleFactor
2634         https://bugs.webkit.org/show_bug.cgi?id=72464
2635
2636         Reviewed by James Robinson.
2637
2638         No new tests. (Tiny fix.)
2639
2640         * page/Page.cpp:
2641         (WebCore::Page::setDeviceScaleFactor):
2642
2643 2011-11-16  Sheriff Bot  <webkit.review.bot@gmail.com>
2644
2645         Unreviewed, rolling out r100438.
2646         http://trac.webkit.org/changeset/100438
2647         https://bugs.webkit.org/show_bug.cgi?id=72536
2648
2649         Broke unit tests (Requested by jamesr_ on #webkit).
2650
2651         * platform/PlatformScreen.h:
2652         * platform/chromium/PlatformScreenChromium.cpp:
2653         * platform/chromium/PlatformSupport.h:
2654         * platform/graphics/chromium/cc/CCLayerTreeHost.h:
2655         (WebCore::CCSettings::CCSettings):
2656         * platform/graphics/chromium/cc/CCThreadProxy.cpp:
2657         (WebCore::CCThreadProxy::initializeImplOnImplThread):
2658
2659 2011-11-16  Andy Estes  <aestes@apple.com>
2660
2661         Assertion failure in LayerFlushScheduler::resume() when running some layout tests in WebKitTestRunner
2662         https://bugs.webkit.org/show_bug.cgi?id=72535
2663
2664         Reviewed by Anders Carlsson.
2665
2666         LayerFlushScheduler attempted to use a counter strategy for calls to
2667         suspend() and resume(), which allowed us to assert that these calls
2668         were balanced. Unfortunately it is hard to guarantee this in WebKit2,
2669         where we sometimes try to call suspend() before we've entered
2670         compositing mode (hence before we have a LayerTreeHost and a
2671         LayerFlushScheduler). When we later call resume(), this call ends up
2672         being unbalanced and asserts.
2673
2674         For now, remove the assertions and allow unbalanced calls to suspend()
2675         and resume().
2676
2677         * platform/graphics/ca/LayerFlushScheduler.cpp:
2678         (WebCore::LayerFlushScheduler::suspend):
2679         (WebCore::LayerFlushScheduler::resume):
2680         * platform/graphics/ca/LayerFlushScheduler.h:
2681         * platform/graphics/ca/mac/LayerFlushSchedulerMac.cpp:
2682         (WebCore::LayerFlushScheduler::LayerFlushScheduler):
2683         (WebCore::LayerFlushScheduler::runLoopObserverCallback):
2684         (WebCore::LayerFlushScheduler::schedule):
2685
2686 2011-11-16  Sheriff Bot  <webkit.review.bot@gmail.com>
2687
2688         Unreviewed, rolling out r100473.
2689         http://trac.webkit.org/changeset/100473
2690         https://bugs.webkit.org/show_bug.cgi?id=72534
2691
2692         "Broke the Mac Build" (Requested by mwenge2 on #webkit).
2693
2694         * CMakeLists.txt:
2695         * GNUmakefile.list.am:
2696         * Target.pri:
2697         * WebCore.gypi:
2698         * rendering/RenderBlock.cpp:
2699         (WebCore::RenderBlock::addChildIgnoringAnonymousColumnBlocks):
2700         * rendering/RenderObject.cpp:
2701         (WebCore::RenderObject::createObject):
2702         (WebCore::RenderObject::addChild):
2703         * rendering/RenderObject.h:
2704         * rendering/RenderTable.cpp:
2705         (WebCore::RenderTable::addChild):
2706         * rendering/RenderTable.h:
2707         * rendering/RenderTableCaption.cpp: Removed.
2708         * rendering/RenderTableCaption.h: Removed.
2709
2710 2011-11-16  Sheriff Bot  <webkit.review.bot@gmail.com>
2711
2712         Unreviewed, rolling out r100479.
2713         http://trac.webkit.org/changeset/100479
2714         https://bugs.webkit.org/show_bug.cgi?id=72533
2715
2716         "Broke the Mac Build" (Requested by mwenge2 on #webkit).
2717
2718         * WebCore.vcproj/WebCore.vcproj:
2719         * WebCore.xcodeproj/project.pbxproj:
2720         * rendering/RenderingAllInOne.cpp:
2721
2722 2011-11-16  Chris Fleizach  <cfleizach@apple.com>
2723
2724         WebKitTestRunner needs to support accessibility-related DRT APIs
2725         https://bugs.webkit.org/show_bug.cgi?id=42131
2726
2727         Minor changes needed to support AX testing in WKTestRunner.
2728
2729         Reviewed by Beth Dakin.
2730
2731         * WebCore.exp.in:
2732              Expose focusedUIElementForPage so that the WK2 injected bundle can retrieve it.
2733         * accessibility/mac/WebAccessibilityObjectWrapper.mm:
2734         (accessibilitySearchKeyForString):
2735             Remove an unncessary assert that was causing issues with the WK2 test run.
2736
2737 2011-11-16  Sergio Villar Senin  <svillar@igalia.com>
2738
2739         [Soup] Somet tests fail with FAIL Unexpected response data received: Wrong method: GET
2740         https://bugs.webkit.org/show_bug.cgi?id=69219
2741
2742         Reviewed by Martin Robinson.
2743
2744         Do not stop appending data to the request body if any of the blob
2745         items to upload is not accesible.
2746
2747         * platform/network/soup/ResourceHandleSoup.cpp:
2748         (WebCore::addEncodedBlobToSoupMessageBody):
2749
2750 2011-11-16  Beth Dakin  <bdakin@apple.com>
2751
2752         https://bugs.webkit.org/show_bug.cgi?id=72400
2753         Scrollbar uiStateTransitionProgress requires tracking the mouse all the time
2754         -and corresponding-
2755         <rdar://problem/10409328>
2756
2757         Reviewed by Darin Adler.
2758
2759         This patch makes it so we track the mouse all the time when we have legacy 
2760         scrollbars (rather than only tracking the mouse when the window is key). When 
2761         we're in that mode, we want to do as little work as possible when handling the 
2762         mouseMoved event so that this extra tracking has little to no performance impact. 
2763         Also, we don't want to change basic behaviors by having normal web content hover 
2764         effects start happening when a window is in the background. So this patch also 
2765         introduces a way to handle a mouseMoved event that will only affect scrollbars.
2766
2767         EventHandler::mouseMoved() and EventHandler::handleMouseEvent() both now take a 
2768         boolean parameter that indicates if we are only updating scrollbars. If that is 
2769         the case, then we make our HitTestRequest ReadOnly, and we return early once 
2770         updateLastScrollbarUnderMouse() is called.
2771         * WebCore.exp.in:
2772         * page/EventHandler.cpp:
2773         (WebCore::EventHandler::mouseMoved):
2774         (WebCore::EventHandler::handleMouseMoveEvent):
2775
2776         In addition to calling Scrollbar::mouseExited() when appropriate, this function 
2777         now calls a new function, Scrollbar::mouseEntered() when appropriate. 
2778         (WebCore::EventHandler::updateLastScrollbarUnderMouse):
2779         * page/EventHandler.h:
2780
2781         Scrollbar::mouseMoved() used to be responsible for calling 
2782         ScrollAnimator::mouseEnteredScrollbar(), but now Scrollbar::mouseEntered() takes 
2783         care of that instead, much like how Scrollbar::mouseExited() takes care of calling 
2784         the animator's exit function.
2785         * platform/Scrollbar.cpp:
2786         (WebCore::Scrollbar::mouseMoved):
2787         (WebCore::Scrollbar::mouseEntered):
2788         * platform/Scrollbar.h:
2789
2790 2011-11-16  Andreas Kling  <kling@webkit.org>
2791
2792         CSSValue: isInheritedValue() doesn't need a dedicated bit.
2793         <http://webkit.org/b/72514>
2794
2795         Reviewed by Antti Koivisto.
2796
2797         Remove CSSValue::m_isInherited and have isInheritedValue() check the
2798         class type instead. There's no compelling reason for CSSInheritedValue
2799         to have a dedicated bit, since nobody subclasses it anyway.
2800
2801         * css/CSSValue.h:
2802         (WebCore::CSSValue::isInheritedValue):
2803         (WebCore::CSSValue::CSSValue):
2804
2805 2011-11-16  Robert Hogan  <robert@webkit.org>
2806
2807         Fix build on Windows and Mac after r100473
2808
2809         Unreviewed, fix build.
2810
2811         * WebCore.vcproj/WebCore.vcproj:
2812         * WebCore.xcodeproj/project.pbxproj:
2813         * rendering/RenderingAllInOne.cpp:
2814
2815 2011-11-16  Philip Rogers  <pdr@google.com>
2816
2817         Remove extra GraphicsContextStateSaver restore() call.
2818         https://bugs.webkit.org/show_bug.cgi?id=72497
2819
2820         Reviewed by Andreas Kling.
2821
2822         * html/canvas/CanvasRenderingContext2D.cpp:
2823         (WebCore::CanvasRenderingContext2D::drawTextInternal):
2824
2825 2011-11-08  Robert Hogan  <robert@webkit.org>
2826
2827         CSS 2.1 failure: border-collapse-offset-002.htm fails
2828         https://bugs.webkit.org/show_bug.cgi?id=71705
2829
2830         Table captions are implemented as children of the table but have a special
2831         requirement to expand to the full width of the table rather than just the 'available'
2832         width, i.e. the full width minus padding and borders.
2833
2834         To accomodate this create a RenderTableCaption object that reimplements containingBlockLogicalWidthForContent()
2835         to return the full width of the containing block (i.e. the table) rather than the available width.
2836
2837         Reviewed by Antti Koivisto.
2838
2839         * CMakeLists.txt: Add RenderTableCaption.[cpp|h]
2840         * GNUmakefile.list.am: Add RenderTableCaption.[cpp|h]
2841         * Target.pri: Add RenderTableCaption.[cpp|h]
2842         * WebCore.gypi: Add RenderTableCaption.[cpp|h]
2843         * rendering/RenderBlock.cpp:
2844         (WebCore::RenderBlock::addChildIgnoringAnonymousColumnBlocks): Use RenderTableCaption
2845         * rendering/RenderObject.cpp:
2846         (WebCore::RenderObject::createObject): Add RenderTableCaption.[cpp|h]
2847         (WebCore::RenderObject::addChild): ditto
2848         * rendering/RenderObject.h:
2849         (WebCore::RenderObject::isTableCaption):
2850         * rendering/RenderTable.cpp:
2851         (WebCore::RenderTable::addChild):
2852         * rendering/RenderTable.h:
2853         * rendering/RenderTableCaption.cpp: Added.
2854         (WebCore::RenderTableCaption::RenderTableCaption): Implement RenderTableCaption
2855         (WebCore::RenderTableCaption::~RenderTableCaption):
2856         (WebCore::RenderTableCaption::containingBlockLogicalWidthForContent): Return the containing block's full width rather than it's available width.
2857         * rendering/RenderTableCaption.h: Added.
2858         (WebCore::RenderTableCaption::isTableCaption):
2859         (WebCore::toRenderTableCaption):
2860
2861 2011-11-16  Antaryami Pandia  <antaryami.pandia@motorola.com>
2862
2863         [Gtk] display:none has no effect on <option> element.
2864         https://bugs.webkit.org/show_bug.cgi?id=72370
2865
2866         Reviewed by Martin Robinson.
2867
2868         * platform/gtk/GtkPopupMenu.cpp:
2869         (WebCore::GtkPopupMenu::appendItem):
2870         * platform/gtk/PopupMenuGtk.cpp:
2871         (WebCore::PopupMenuGtk::createGtkActionForMenuItem):
2872
2873 2011-11-16  Dan Winship  <danw@gnome.org>
2874
2875         [GTK] Fix platformDefersLoading to handle non-http requests, and
2876         to not use broken-ish libsoup APIs.
2877         https://bugs.webkit.org/show_bug.cgi?id=72227
2878
2879         Reviewed by Martin Robinson.
2880
2881         * platform/network/ResourceHandleInternal.h:
2882         * platform/network/soup/ResourceHandleSoup.cpp:
2883         (WebCore::sendRequestCallback):
2884         (WebCore::startHTTPRequest):
2885         (WebCore::hasBeenSent):
2886         (WebCore::ResourceHandle::platformSetDefersLoading):
2887         (WebCore::readCallback): rather than deferring by using
2888         soup_session_pause_message(), let the read complete, but just don't
2889         process the result until we're no longer deferred.
2890         (WebCore::startNonHTTPRequest): Don't start the request if
2891         it's deferred.
2892
2893 2011-11-16  Vsevolod Vlasov  <vsevik@chromium.org>
2894
2895         Web Inspector: Application cache status should be updated after swapCache().
2896         https://bugs.webkit.org/show_bug.cgi?id=72123
2897
2898         Reviewed by Pavel Feldman.
2899
2900         Application cache view resources and status are now updated after swapCache() call.
2901         Refresh button removed from application cache view.
2902         Application cache inspector tests are moved to their own folder.
2903
2904         Tests: http/tests/inspector/appcache/appcache-iframe-manifests.html
2905                http/tests/inspector/appcache/appcache-manifest-with-non-existing-file.html
2906                http/tests/inspector/appcache/appcache-swap.html
2907
2908         * inspector/front-end/ApplicationCacheItemsView.js:
2909         (WebInspector.ApplicationCacheItemsView):
2910         (WebInspector.ApplicationCacheItemsView.prototype.get statusBarItems):
2911         (WebInspector.ApplicationCacheItemsView.prototype.wasShown):
2912         (WebInspector.ApplicationCacheItemsView.prototype._maybeUpdate):
2913         (WebInspector.ApplicationCacheItemsView.prototype._markDirty):
2914         (WebInspector.ApplicationCacheItemsView.prototype.updateStatus):
2915         (WebInspector.ApplicationCacheItemsView.prototype._updateCallback):
2916         (WebInspector.ApplicationCacheItemsView.prototype._deleteCallback):
2917         * loader/appcache/ApplicationCacheGroup.cpp:
2918         (WebCore::ApplicationCacheGroup::setNewestCache):
2919         (WebCore::ApplicationCacheGroup::makeObsolete):
2920         (WebCore::ApplicationCacheGroup::setUpdateStatus):
2921         * loader/appcache/ApplicationCacheHost.cpp:
2922         (WebCore::ApplicationCacheHost::notifyDOMApplicationCache):
2923         (WebCore::ApplicationCacheHost::swapCache):
2924
2925 2011-11-16  Eric Carlson  <eric.carlson@apple.com>
2926
2927         addTrack() must throw an exception if 'kind' is unknown
2928         https://bugs.webkit.org/show_bug.cgi?id=71915
2929
2930         Reviewed by Philippe Normand.
2931
2932         Tests: media/track/track-addtrack-kind.html
2933                media/track/track-kind.html
2934
2935         * html/HTMLMediaElement.cpp:
2936         (WebCore::HTMLMediaElement::textTrackKindChanged): New. Will be implemented for 62885.
2937         (WebCore::HTMLMediaElement::addTrack): Throw if  'kind' is not a known value.
2938         (WebCore::HTMLMediaElement::addTextTrack): Call textTracks(), it will allocate the track
2939             list object if necessary.
2940         (WebCore::HTMLMediaElement::textTracks): Never return NULL, a TextTrackList with no tracks
2941             is allowed.
2942         * html/HTMLMediaElement.h:
2943         (WebCore::HTMLMediaElement::addTrack): Add variants to deal with optional parameters plus
2944             a mandatory ExceptionCode parameter.
2945         * html/HTMLMediaElement.idl: addTrack can generate an exception.
2946
2947         * html/HTMLTrackElement.cpp:
2948         (WebCore::HTMLTrackElement::attributeChanged): kind, label, and srclang attribute changes should
2949             percolate down the the TextTrack.
2950         (WebCore::HTMLTrackElement::kind): Return the TextTrack kind because it is not necessarily the
2951             same as the attribute value.
2952         (WebCore::HTMLTrackElement::ensureTrack): Only pass legal 'kind' keywords to create a TextTrack.
2953         (WebCore::HTMLTrackElement::textTrackKindChanged): Notify parent element, if any.
2954         * html/HTMLTrackElement.h:
2955
2956         * html/TextTrack.cpp:
2957         (WebCore::TextTrack::subtitlesKeyword): New, return legal kind attribute value.
2958         (WebCore::TextTrack::captionsKeyword): Ditto.
2959         (WebCore::TextTrack::descriptionsKeyword): Ditto.
2960         (WebCore::TextTrack::chaptersKeyword): Ditto.
2961         (WebCore::TextTrack::metadataKeyword): Ditto.
2962         (WebCore::TextTrack::TextTrack): Call setKind to make sure m_kind is always set to legal value.
2963         (WebCore::TextTrack::isValidKindKeyword): New, validate 'kind' value.
2964         (WebCore::TextTrack::setKind): Only allow legal values.
2965         * html/TextTrack.h:
2966         (WebCore::TextTrack::kind):
2967         (WebCore::TextTrack::label):
2968         (WebCore::TextTrack::setLabel):
2969         (WebCore::TextTrack::language):
2970         (WebCore::TextTrack::setLanguage):
2971         (WebCore::TextTrack::readyState):
2972         (WebCore::TextTrack::mode):
2973
2974 2011-11-16  Andreas Kling  <kling@webkit.org>
2975
2976         CSSValue: Clean up initial value construction.
2977         <http://webkit.org/b/72502>
2978
2979         Reviewed by Antti Koivisto.
2980
2981         Instead of determining whether a given CSSInitialValue is 'implicit' or not by
2982         querying the CSSValue::ClassType (InitialClass vs. ImplicitInitialClass),
2983         add a protected CSSValue member and set it from the CSSInitialValue constructor.
2984
2985         Also get rid of the CSSValue::m_isInitial bit since we can now replace the
2986         checks by classType() == InitialClass.
2987
2988         No new tests, this is a cleanup.
2989
2990         * css/CSSInitialValue.h:
2991         (WebCore::CSSInitialValue::CSSInitialValue):
2992
2993             Poke 'implicit' constructor argument into CSSValue::m_isImplicit.
2994
2995         * css/CSSValue.cpp:
2996         (WebCore::CSSValue::cssText):
2997         (WebCore::CSSValue::destroy):
2998
2999             Remove ImplicitInitialClass cases.
3000
3001         * css/CSSValue.h:
3002         (WebCore::CSSValue::isImplicitInitialValue):
3003         (WebCore::CSSValue::isInitialValue):
3004         (WebCore::CSSValue::CSSValue):
3005
3006 2011-11-16  Antaryami Pandia  <antaryami.pandia@motorola.com>
3007
3008         Remove unnecessary if check from RenderListBox::paintItemForeground.
3009         https://bugs.webkit.org/show_bug.cgi?id=72488
3010
3011         Reviewed by Andreas Kling.
3012
3013         * rendering/RenderListBox.cpp:
3014         (WebCore::RenderListBox::paintItemForeground):
3015
3016 2011-11-15  Simon Hausmann  <simon.hausmann@nokia.com>
3017
3018         [Qt] Centralize hide_symbols and ensure all libs are built with symbol visibility & bsymbolic_functions
3019
3020         Reviewed by Tor Arne Vestbø.
3021
3022         * Target.pri: Eliminate duplicated symbol stuff that lives now in default_post.prf.
3023
3024 2011-11-16  Iain Merrick  <husky@google.com>
3025
3026         [chromium] Pass screen refresh rate into compositor.
3027         https://bugs.webkit.org/show_bug.cgi?id=71040
3028
3029         Reviewed by Tony Gentilcore.
3030
3031         * platform/PlatformScreen.h:
3032         * platform/chromium/PlatformScreenChromium.cpp:
3033         (WebCore::screenRefreshRate):
3034         * platform/chromium/PlatformSupport.h:
3035         * platform/graphics/chromium/cc/CCLayerTreeHost.h:
3036         (WebCore::CCSettings::CCSettings):
3037         * platform/graphics/chromium/cc/CCThreadProxy.cpp:
3038         (WebCore::CCThreadProxy::initializeImplOnImplThread):
3039
3040 2011-11-15  Andrey Kosyakov  <caseq@chromium.org>
3041
3042         Web Inspector: [Extensions API] drop ExtensionSidebarPane.onUpdated, use callbacks instead
3043         https://bugs.webkit.org/show_bug.cgi?id=72388
3044
3045         Reviewed by Pavel Feldman.
3046
3047         Tests: inspector/extensions/extensions-panel.html
3048                inspector/extensions/extensions-sidebar.html
3049
3050         - fire ExtensionSidebarPane.on{Hidden,Shown} for non-iframe content (experssions/objects);
3051         - drop ExtensionsSidebarPane.onUpdated, provide callback for setObject()/setExpression() instead;
3052         - fix an exception when a page is replaced with object/expression.
3053
3054          inspector/front-end/ExtensionAPI.js:
3055         (injectedExtensionAPI.ExtensionSidebarPaneImpl):
3056         (injectedExtensionAPI.ExtensionSidebarPaneImpl.prototype.setExpression):
3057         (injectedExtensionAPI.ExtensionSidebarPaneImpl.prototype.setObject):
3058         * inspector/front-end/ExtensionPanel.js:
3059         (WebInspector.ExtensionNotifierView):
3060         (WebInspector.ExtensionNotifierView.prototype.wasShown):
3061         (WebInspector.ExtensionNotifierView.prototype.willHide):
3062         (WebInspector.ExtensionSidebarPane.prototype.setObject):
3063         (WebInspector.ExtensionSidebarPane.prototype.setExpression):
3064         (WebInspector.ExtensionSidebarPane.prototype.setPage):
3065         (WebInspector.ExtensionSidebarPane.prototype._onEvaluate):
3066         (WebInspector.ExtensionSidebarPane.prototype._makeObjectPropertiesView):
3067         (WebInspector.ExtensionSidebarPane.prototype._setObject):
3068         * inspector/front-end/ExtensionServer.js:
3069         (WebInspector.ExtensionServer.prototype._onSetSidebarContent.callback):
3070         (WebInspector.ExtensionServer.prototype._onSetSidebarContent):
3071         (WebInspector.ExtensionServer.prototype._dispatchCallback):
3072         * inspector/front-end/View.js:
3073         (WebInspector.View.prototype.detach):
3074
3075 2011-11-16  Simon Hausmann  <simon.hausmann@nokia.com>
3076
3077         Unreviewed, rolling out r100266.
3078         http://trac.webkit.org/changeset/100266
3079
3080         Broke WTR.
3081
3082         * Target.pri:
3083
3084 2011-11-16  Per-Erik Brodin  <per-erik.brodin@ericsson.com>
3085
3086         [GTK] fast/events/event-creation.html fails creating MediaStreamEvent
3087         https://bugs.webkit.org/show_bug.cgi?id=70720
3088
3089         Reviewed by Philippe Normand.
3090
3091         Added missing overriding of Event::interfaceName()
3092
3093         * mediastream/MediaStreamEvent.cpp:
3094         (WebCore::MediaStreamEvent::stream): Changed return value to raw pointer.
3095         (WebCore::MediaStreamEvent::interfaceName): Added back from r98044.
3096         * mediastream/MediaStreamEvent.h:
3097         * mediastream/MediaStreamEvent.idl: Changed module name to "events".
3098
3099 2011-11-16  Mario Sanchez Prada  <msanchez@igalia.com>
3100
3101         [GTK] Use GQuark's in the ATK wrapper to get and set arbitrary data
3102         https://bugs.webkit.org/show_bug.cgi?id=72394
3103
3104         Reviewed by Martin Robinson.
3105
3106         No new functionality, no new tests needed.
3107
3108         * accessibility/gtk/AccessibilityObjectWrapperAtk.cpp:
3109         (webkit_accessible_class_init): Initialize GQuarks.
3110         (getGailTextUtilForAtk): Use gailTextUtilQuark.
3111         (getPangoLayoutForAtk): Remove unused call to
3112         g_object_set_data_full, since that data is no used anywhere.
3113         (webkitAccessibleHyperlinkImplGetHyperlink): Use hyperlinkObjectQuark.
3114
3115 2011-11-15  Alexandru Chiculita  <achicu@adobe.com>
3116
3117         [CSSShaders] Implement the style cached resources and computed style for the shader urls
3118         https://bugs.webkit.org/show_bug.cgi?id=72378
3119
3120         Reviewed by Dean Jackson.
3121
3122         Test: css3/filters/custom-filter-property-computed-style.html
3123
3124         * GNUmakefile.list.am:
3125         * Target.pri:
3126         * WebCore.gypi:
3127         * WebCore.vcproj/WebCore.vcproj:
3128         * WebCore.xcodeproj/project.pbxproj:
3129         * css/CSSComputedStyleDeclaration.cpp:
3130         (WebCore::CSSComputedStyleDeclaration::valueForFilter):
3131         * css/CSSStyleSelector.cpp:
3132         (WebCore::CSSStyleSelector::CSSStyleSelector):
3133         (WebCore::CSSStyleSelector::styleForKeyframe):
3134         (WebCore::CSSStyleSelector::pseudoStyleForElement):
3135         (WebCore::CSSStyleSelector::styleForPage):
3136         (WebCore::CSSStyleSelector::applyMatchedDeclarations):
3137         (WebCore::CSSStyleSelector::styleShader):
3138         (WebCore::CSSStyleSelector::cachedOrPendingStyleShaderFromValue):
3139         (WebCore::CSSStyleSelector::loadPendingShaders):
3140         (WebCore::CSSStyleSelector::createCustomFilterOperation):
3141         (WebCore::CSSStyleSelector::createFilterOperations):
3142         * css/CSSStyleSelector.h:
3143         * css/WebKitCSSShaderValue.cpp:
3144         (WebCore::WebKitCSSShaderValue::WebKitCSSShaderValue):
3145         (WebCore::WebKitCSSShaderValue::~WebKitCSSShaderValue):
3146         (WebCore::WebKitCSSShaderValue::cachedShader):
3147         (WebCore::WebKitCSSShaderValue::cachedOrPendingShader):
3148         * css/WebKitCSSShaderValue.h:
3149         * loader/cache/CachedResource.cpp:
3150         (WebCore::defaultPriorityForResourceType):
3151         * loader/cache/CachedResource.h:
3152         * loader/cache/CachedResourceLoader.cpp:
3153         (WebCore::createResource):
3154         (WebCore::CachedResourceLoader::requestShader):
3155         (WebCore::CachedResourceLoader::checkInsecureContent):
3156         (WebCore::CachedResourceLoader::canRequest):
3157         * loader/cache/CachedResourceLoader.h:
3158         * loader/cache/CachedShader.cpp:
3159         (WebCore::CachedShader::CachedShader):
3160         (WebCore::CachedShader::~CachedShader):
3161         * loader/cache/CachedShader.h:
3162         * platform/graphics/filters/CustomFilterOperation.h:
3163         * rendering/style/RenderStyle.h:
3164         (WebCore::InheritedFlags::filter):
3165         * rendering/style/StyleCachedShader.cpp:
3166         (WebCore::StyleCachedShader::StyleCachedShader):
3167         (WebCore::StyleCachedShader::cssValue):
3168         * rendering/style/StyleCachedShader.h:
3169         (WebCore::StyleCachedShader::create):
3170         * rendering/style/StylePendingShader.h:
3171         (WebCore::StylePendingShader::create):
3172         (WebCore::StylePendingShader::cssValue):
3173         (WebCore::StylePendingShader::cssShaderValue):
3174         (WebCore::StylePendingShader::StylePendingShader):
3175         * rendering/style/StyleShader.h:
3176         (WebCore::StyleShader::~StyleShader):
3177         (WebCore::StyleShader::isCachedShader):
3178         (WebCore::StyleShader::isPendingShader):
3179         (WebCore::StyleShader::StyleShader):
3180
3181 2011-11-15  Sergio Villar Senin  <svillar@igalia.com>
3182
3183         [WK2] [GTK] fast/css/webkit-mask-crash-fieldset-legend.html asserts WebKitWebProcess
3184         https://bugs.webkit.org/show_bug.cgi?id=69510
3185
3186         Reviewed by Simon Fraser.
3187
3188         End the current transparency layer before early returning from
3189         paintMask() when there is a maskBoxImage which is still being
3190         loaded. This will balance the previous call to
3191         beginTransparencyLayer().
3192
3193         * rendering/InlineFlowBox.cpp:
3194         (WebCore::InlineFlowBox::paintMask):
3195
3196 2011-11-15  Darin Adler  <darin@apple.com>
3197
3198         Incorrect type checks in RenderTheme media code
3199         https://bugs.webkit.org/show_bug.cgi?id=72184
3200
3201         Reviewed by Eric Carlson.
3202
3203         No tests added. Ideally this patch should be revised to add tests!
3204
3205         * accessibility/AccessibilityMediaControls.cpp:
3206         (WebCore::AccessibilityMediaControl::create): Use mediaControlElementType.
3207         (WebCore::AccessibilityMediaControl::controlType): Ditto.
3208         (WebCore::AccessibilityMediaTimeline::valueDescription): Use early return
3209         rather than an assertion to check type of input element.
3210
3211         * html/shadow/MediaControlElements.cpp:
3212         (WebCore::mediaControlElementType): Added. A type-safe way to get the
3213         media control element type after checking isMediaControlElement but with
3214         no other assumptions.
3215         * html/shadow/MediaControlElements.h: Added mediaControlElementType.
3216
3217         * platform/efl/RenderThemeEfl.cpp:
3218         (WebCore::RenderThemeEfl::paintMediaPlayButton): Use mediaControlElementType.
3219         (WebCore::RenderThemeEfl::paintMediaSeekBackButton): Use mediaControlElementType.
3220         (WebCore::RenderThemeEfl::paintMediaSeekForwardButton): Use mediaControlElementType.
3221         * platform/gtk/RenderThemeGtk.cpp:
3222         (WebCore::RenderThemeGtk::paintMediaPlayButton): Check isMediaControlElement and
3223         use mediaControlElementType.
3224         * rendering/RenderThemeMac.mm:
3225         (WebCore::RenderThemeMac::paintMediaMuteButton): Ditto. Also remove uneeded
3226         redundant null check.
3227         (WebCore::RenderThemeMac::paintMediaPlayButton): Ditto.
3228         (WebCore::RenderThemeMac::paintMediaToggleClosedCaptionsButton): Ditto.
3229
3230 2011-11-15  Jeff Timanus  <twiz@chromium.org>
3231
3232         [chromium] During tear down, prevent the WebGLLayerChromium instance from attempting to stop a timer for a NULL context.
3233         https://bugs.webkit.org/show_bug.cgi?id=72423
3234
3235         Reviewed by Kenneth Russell.
3236
3237         * platform/graphics/chromium/WebGLLayerChromium.cpp:
3238         (WebCore::WebGLLayerChromium::setDrawingBuffer):
3239
3240 2011-11-15  Mark Hahnenberg  <mhahnenberg@apple.com>
3241
3242         Rebaseline generated WebCore bindings
3243
3244         Unreviewed build fix
3245
3246         No new tests.
3247
3248         * bindings/scripts/test/JS/JSTestEventConstructor.cpp:
3249         (WebCore::isObservable):
3250         (WebCore::JSTestEventConstructorOwner::isReachableFromOpaqueRoots):
3251         (WebCore::JSTestEventConstructorOwner::finalize):
3252         * bindings/scripts/test/JS/JSTestEventConstructor.h:
3253         (WebCore::JSTestEventConstructor::clearImpl):
3254         (WebCore::wrapperOwner):
3255         (WebCore::wrapperContext):
3256         * bindings/scripts/test/JS/JSTestInterface.cpp:
3257         (WebCore::JSTestInterfaceOwner::finalize):
3258         * bindings/scripts/test/JS/JSTestInterface.h:
3259         (WebCore::JSTestInterface::clearImpl):
3260         * bindings/scripts/test/JS/JSTestMediaQueryListListener.cpp:
3261         (WebCore::isObservable):
3262         (WebCore::JSTestMediaQueryListListenerOwner::isReachableFromOpaqueRoots):
3263         (WebCore::JSTestMediaQueryListListenerOwner::finalize):
3264         * bindings/scripts/test/JS/JSTestMediaQueryListListener.h:
3265         (WebCore::JSTestMediaQueryListListener::clearImpl):
3266         (WebCore::wrapperOwner):
3267         (WebCore::wrapperContext):
3268         * bindings/scripts/test/JS/JSTestNamedConstructor.cpp:
3269         (WebCore::JSTestNamedConstructorOwner::finalize):
3270         * bindings/scripts/test/JS/JSTestNamedConstructor.h:
3271         (WebCore::JSTestNamedConstructor::clearImpl):
3272         * bindings/scripts/test/JS/JSTestObj.cpp:
3273         (WebCore::isObservable):
3274         (WebCore::JSTestObjOwner::isReachableFromOpaqueRoots):
3275         (WebCore::JSTestObjOwner::finalize):
3276         * bindings/scripts/test/JS/JSTestObj.h:
3277         (WebCore::JSTestObj::clearImpl):
3278         (WebCore::wrapperOwner):
3279         (WebCore::wrapperContext):
3280         * bindings/scripts/test/JS/JSTestSerializedScriptValueInterface.cpp:
3281         (WebCore::isObservable):
3282         (WebCore::JSTestSerializedScriptValueInterfaceOwner::isReachableFromOpaqueRoots):
3283         (WebCore::JSTestSerializedScriptValueInterfaceOwner::finalize):
3284         * bindings/scripts/test/JS/JSTestSerializedScriptValueInterface.h:
3285         (WebCore::JSTestSerializedScriptValueInterface::clearImpl):
3286         (WebCore::wrapperOwner):
3287         (WebCore::wrapperContext):
3288
3289 2011-11-15  Jeff Timanus  <twiz@chromium.org>
3290
3291         Patch removing duplicated code in the setup of the DrawingBuffer used
3292         to host the back-buffer for WebGL contents.
3293         https://bugs.webkit.org/show_bug.cgi?id=72327
3294
3295         Reviewed by Kenneth Russell.
3296
3297         * html/canvas/WebGLRenderingContext.cpp:
3298         (WebCore::WebGLRenderingContext::copyTexImage2D):
3299         (WebCore::WebGLRenderingContext::copyTexSubImage2D):
3300         (WebCore::WebGLRenderingContext::readPixels):
3301
3302 2011-11-15  Mark Hahnenberg  <mhahnenberg@apple.com>
3303
3304         JS DOM wrappers depend on destructor to deref impl RefPtrs
3305         https://bugs.webkit.org/show_bug.cgi?id=72341
3306
3307         Reviewed by Sam Weinig.
3308
3309         No new tests.
3310
3311         Added clearing of impl RefPtrs to JS DOM wrapper nodes and removed the default 
3312         wrapperOwner function in favor of generating all WeakHandleOwners and wrapperOwner functions.
3313
3314         * bindings/js/JSCSSValueCustom.cpp:
3315         (WebCore::JSCSSValueOwner::finalize):
3316         * bindings/js/JSDOMBinding.h:
3317         * bindings/js/JSNodeCustom.cpp:
3318         (WebCore::JSNodeOwner::finalize):
3319         * bindings/scripts/CodeGeneratorJS.pm:
3320         (GenerateHeader):
3321         (GenerateImplementation):
3322
3323 2011-11-15  Joseph Pecoraro  <pecoraro@apple.com>
3324
3325         Web Inspector: Share Highlight Code for Drawing Outlined Quad
3326         https://bugs.webkit.org/show_bug.cgi?id=72451
3327
3328         Reviewed by Timothy Hatcher.
3329
3330         * inspector/DOMNodeHighlighter.cpp:
3331
3332 2011-11-15  Sheriff Bot  <webkit.review.bot@gmail.com>
3333
3334         Unreviewed, rolling out r100308.
3335         http://trac.webkit.org/changeset/100308
3336         https://bugs.webkit.org/show_bug.cgi?id=72450
3337
3338         Introduces WebGL conformance test regressions. (Requested by
3339         twiz on #webkit).
3340
3341         * html/canvas/WebGLRenderingContext.cpp:
3342         (WebCore::WebGLRenderingContext::copyTexImage2D):
3343         (WebCore::WebGLRenderingContext::copyTexSubImage2D):
3344         (WebCore::WebGLRenderingContext::readPixels):
3345
3346 2011-11-15  James Robinson  <jamesr@chromium.org>
3347
3348         Rollout http://trac.webkit.org/changeset/99813, caused some crashes in
3349         TiledLayerChromium::updateCompositorResources()
3350
3351         * WebCore.gypi:
3352         * platform/graphics/chromium/BitmapCanvasLayerTextureUpdater.cpp: Removed.
3353         * platform/graphics/chromium/BitmapCanvasLayerTextureUpdater.h: Removed.
3354         * platform/graphics/chromium/CanvasLayerTextureUpdater.cpp: Removed.
3355         * platform/graphics/chromium/CanvasLayerTextureUpdater.h: Removed.
3356         * platform/graphics/chromium/ContentLayerChromium.cpp:
3357         (WebCore::ContentLayerChromium::createTextureUpdater):
3358         * platform/graphics/chromium/FrameBufferSkPictureCanvasLayerTextureUpdater.h: Removed.
3359         * platform/graphics/chromium/ImageLayerChromium.cpp:
3360         (WebCore::ImageLayerTextureUpdater::prepareToUpdate):
3361         * platform/graphics/chromium/LayerRendererChromium.cpp:
3362         * platform/graphics/chromium/LayerTextureUpdater.h:
3363         * platform/graphics/chromium/LayerTextureUpdaterCanvas.cpp: Renamed from Source/WebCore/platform/graphics/chromium/FrameBufferSkPictureCanvasLayerTextureUpdater.cpp.
3364         (WebCore::FrameBuffer::FrameBuffer::FrameBuffer):
3365         (WebCore::FrameBuffer::FrameBuffer::~FrameBuffer):
3366         (WebCore::FrameBuffer::FrameBuffer::initialize):
3367         (WebCore::LayerTextureUpdaterCanvas::LayerTextureUpdaterCanvas):
3368         (WebCore::LayerTextureUpdaterCanvas::paintContents):
3369         (WebCore::LayerTextureUpdaterBitmap::create):
3370         (WebCore::LayerTextureUpdaterBitmap::LayerTextureUpdaterBitmap):
3371         (WebCore::LayerTextureUpdaterBitmap::sampledTexelFormat):
3372         (WebCore::LayerTextureUpdaterBitmap::prepareToUpdate):
3373         (WebCore::LayerTextureUpdaterBitmap::updateTextureRect):
3374         (WebCore::LayerTextureUpdaterSkPicture::create):
3375         (WebCore::LayerTextureUpdaterSkPicture::LayerTextureUpdaterSkPicture):
3376         (WebCore::LayerTextureUpdaterSkPicture::~LayerTextureUpdaterSkPicture):
3377         (WebCore::LayerTextureUpdaterSkPicture::sampledTexelFormat):
3378         (WebCore::LayerTextureUpdaterSkPicture::prepareToUpdate):
3379         (WebCore::LayerTextureUpdaterSkPicture::updateTextureRect):
3380         * platform/graphics/chromium/LayerTextureUpdaterCanvas.h: Added.
3381         (WebCore::LayerTextureUpdaterCanvas::contentRect):
3382         (WebCore::LayerTextureUpdaterBitmap::orientation):
3383         (WebCore::LayerTextureUpdaterSkPicture::orientation):
3384         * platform/graphics/chromium/SkPictureCanvasLayerTextureUpdater.cpp: Removed.
3385         * platform/graphics/chromium/SkPictureCanvasLayerTextureUpdater.h: Removed.
3386         * platform/graphics/chromium/TiledLayerChromium.cpp:
3387         (WebCore::UpdatableTile::UpdatableTile):
3388         (WebCore::UpdatableTile::texture):
3389         (WebCore::TiledLayerChromium::updateCompositorResources):
3390         (WebCore::TiledLayerChromium::pushPropertiesTo):
3391         (WebCore::TiledLayerChromium::createTile):
3392         (WebCore::TiledLayerChromium::protectTileTextures):
3393         (WebCore::TiledLayerChromium::prepareToUpdate):
3394         * platform/graphics/chromium/cc/CCTextureUpdater.cpp:
3395         (WebCore::CCTextureUpdater::append):
3396         (WebCore::CCTextureUpdater::update):
3397         * platform/graphics/chromium/cc/CCTextureUpdater.h:
3398
3399 2011-11-15  Julien Chaffraix  <jchaffraix@webkit.org>
3400
3401         Add the needed plumbing to parse display: -webkit-inline-grid
3402         https://bugs.webkit.org/show_bug.cgi?id=72438
3403
3404         Reviewed by Tony Chang.
3405
3406         Test: fast/css-grid-layout/display-grid-set-get.html
3407
3408         Added the needed constants and plugged everything together.
3409         Again we treat display: -webkit-inline-grid like display: none
3410         for the moment.
3411
3412         * css/CSSParser.cpp:
3413         (WebCore::CSSParser::parseValue):
3414         * css/CSSPrimitiveValueMappings.h:
3415         (WebCore::CSSPrimitiveValue::CSSPrimitiveValue):
3416         * css/CSSValueKeywords.in:
3417         * rendering/RenderObject.cpp:
3418         (WebCore::RenderObject::createObject):
3419         * rendering/style/RenderStyleConstants.h:
3420
3421 2011-11-15  W. James MacLean  <wjmaclean@chromium.org>
3422
3423         [chromium] Move setVisibleRect() calls into calculateDrawTransformAndVisibility()
3424         https://bugs.webkit.org/show_bug.cgi?id=72162
3425
3426         Reviewed by Kenneth Russell.
3427
3428         Refactoring of existing functionality, so uses existing tests.
3429
3430         * platform/graphics/chromium/LayerRendererChromium.cpp:
3431         (WebCore::LayerRendererChromium::drawLayer):
3432         * platform/graphics/chromium/cc/CCLayerTreeHost.cpp:
3433         (WebCore::paintContentsIfDirty):
3434         (WebCore::CCLayerTreeHost::paintMaskAndReplicaForRenderSurface):
3435         (WebCore::CCLayerTreeHost::paintLayerContents):
3436         * platform/graphics/chromium/cc/CCLayerTreeHostCommon.cpp:
3437         (WebCore::walkLayersAndCalculateVisibleLayerRects):
3438         (WebCore::CCLayerTreeHostCommon::calculateDrawTransformsAndVisibility):
3439
3440 2011-11-15  Sheriff Bot  <webkit.review.bot@gmail.com>
3441
3442         Unreviewed, rolling out r100340.
3443         http://trac.webkit.org/changeset/100340
3444         https://bugs.webkit.org/show_bug.cgi?id=72448
3445
3446         Caused assertion failure in Win dbg canary. (Requested by
3447         pkasting on #webkit).
3448
3449         * platform/graphics/chromium/LayerRendererChromium.cpp:
3450         (WebCore::LayerRendererChromium::LayerRendererChromium):
3451         (WebCore::LayerRendererChromium::viewportChanged):
3452         (WebCore::LayerRendererChromium::drawLayersInternal):
3453         * platform/graphics/chromium/LayerRendererChromium.h:
3454
3455 2011-11-15  Julien Chaffraix  <jchaffraix@webkit.org>
3456
3457         Switch table indexing to unsigned
3458         https://bugs.webkit.org/show_bug.cgi?id=72083
3459
3460         Reviewed by Darin Adler.
3461
3462         No expected change in behavior.
3463
3464         All of the code is now using unsigned for indexing!
3465
3466         * rendering/FixedTableLayout.cpp:
3467         (WebCore::FixedTableLayout::layout):
3468         * rendering/RenderTable.cpp:
3469         (WebCore::RenderTable::colElement):
3470         (WebCore::RenderTable::cellAbove):
3471         * rendering/RenderTableSection.cpp:
3472         (WebCore::RenderTableSection::splitColumn):
3473         Added some ASSERTs to make sure we don't underflow. Looking at how
3474         the different variables are populated, they should not be reached.
3475
3476         * rendering/RenderTableCell.cpp:
3477         (WebCore::RenderTableCell::colSpan):
3478         (WebCore::RenderTableCell::rowSpan):
3479         Those 2 functions promotes HTMLTableCellElement's int to unsigned
3480         which should be fine as we make sure their are positive. Also HTML5
3481         makes those 2 fields "unsigned long" which goes in the same direction.
3482
3483         * rendering/AutoTableLayout.cpp:
3484         (WebCore::AutoTableLayout::layout):
3485         * rendering/RenderTableSection.cpp:
3486         (WebCore::RenderTableSection::nodeAtPoint):
3487         Rewrote a couple of reverse iterating to be able to use unsigned
3488         without overflowing.
3489
3490         * rendering/AutoTableLayout.cpp:
3491         (WebCore::AutoTableLayout::recalcColumn):
3492         (WebCore::AutoTableLayout::fullRecalc):
3493         (WebCore::AutoTableLayout::calcEffectiveLogicalWidth):
3494         (WebCore::AutoTableLayout::insertSpanCell):
3495         * rendering/AutoTableLayout.h:
3496         * rendering/FixedTableLayout.cpp:
3497         (WebCore::FixedTableLayout::calcWidthArray):
3498         * rendering/Ren