d54a17c489463f6f5c2d1820aa56876a007d9ea9
[WebKit-https.git] / Source / WebCore / ChangeLog
1 2011-11-20  Kenichi Ishibashi  <bashi@chromium.org>
2
3         [Chromium] Remove old getFontFamilyForCharacters() and familyForChars() APIs.
4         https://bugs.webkit.org/show_bug.cgi?id=72844
5
6         Respects bold and italic properties that is given by fontconfig.
7
8         Reviewed by Darin Fisher.
9
10         * platform/graphics/chromium/FontCacheLinux.cpp:
11         (WebCore::FontCache::getFontDataForCharacters): Removed #if and old logic.
12
13 2011-11-20  Noel Gordon  <noel.gordon@gmail.com>
14
15         [chromium] Remove qt/QtMobileWebStyle from the gyp projects
16         https://bugs.webkit.org/show_bug.cgi?id=72843
17
18         Reviewed by Antonio Gomes.
19
20         platform/qt/QtMobileWebStyle.{h,cpp} were removed in r100123
21
22         * WebCore.gypi: remove platform/qt/QtMobileWebStyle.{h,cpp}
23
24 2011-11-20  Kentaro Hara  <haraken@chromium.org>
25
26         Unreviewed. Rebaselined run-bindings-tests results.
27
28         * bindings/scripts/test/CPP/WebDOMFloat64Array.cpp:
29         (WebDOMFloat64Array::WebDOMFloat64Array):
30         (WebDOMFloat64Array::impl):
31         (toWebCore):
32         (toWebKit):
33         * bindings/scripts/test/CPP/WebDOMFloat64Array.h:
34
35 2011-11-20  Adam Barth  <abarth@webkit.org>
36
37         REGRESSION(r100691): Safari error pages and Growl notifications fail to load stylesheets
38         https://bugs.webkit.org/show_bug.cgi?id=72836
39
40         Reviewed by Sam Weinig.
41
42         This patch removes a (minor) security mitigation.  Previously, we tried
43         sequester "directory listings" into unique origins to make it more
44         difficult for an attacker to crawl the user's local file system.
45         Unfortunately, this mitigation doesn't really buy us much security
46         because if the attacker has access to local files, we've probably lost
47         anyway.
48
49         The larger problem, however, is that this condition is overly
50         complicated and has broken in sublte ways several times in its
51         (relatively short) lifetime.  In the cases reported in this bug, we see
52         that this check affects error pages in Safari and Growl notifications,
53         even those have nothing to do with directory listings.
54
55         If we have our heart set on this directory listing mitigation, we'll
56         need a more robust way of triggering the behavior than examining URLs
57         and guess whether they contain directory listings.  For example, if we
58         implement Allow-From or Access-Control-Deny-Origin, then the embedder
59         can supply those policies along with the directory listings.  Those
60         seem like much better solutions than the in-engine hack this patch
61         removes.
62
63         * page/SecurityOrigin.cpp:
64         (WebCore::shouldTreatAsUniqueOrigin):
65
66 2011-10-17  Antonio Gomes  <agomes@rim.com>
67
68         Pass a Frame* parameter in EditorClient::respondToChangedSelection
69         https://bugs.webkit.org/show_bug.cgi?id=70248
70
71         Reviewed by Ryosuke Niwa.
72
73         Most of the port specific implementations of EditorClient::respondToChangedSelection
74         (like EditorClient{Qt,Gtk,etc}) are wrongly relying on FocusController::focusedOrMainFrame
75         method to get the Frame where the selection is happening on.
76         It is not right, since a selection can be easily triggered in an inner
77         frame that is not focused.
78
79         No new tests since it is a hard thing to test without
80         to hook into editor client in layout tests. We could
81         change the "Dumped Delegate" messages, but it would
82         require a rebasile of +1200 for a minor change.
83
84         * editing/Editor.cpp: Pass the Frame on where the selection is changing to the client.
85         (WebCore::Editor::respondToChangedSelection): Ditto.
86         (WebCore::Editor::changeSelectionAfterCommand): Ditto.
87         (WebCore::EmptyEditorClient::respondToChangedSelection):
88         * page/EditorClient.h:  Pass Frame* to indicate where the selection is happening.
89         * loader/EmptyClients.h: Changed signature as per base class change.
90
91 2011-11-17  Mark Rowe  <mrowe@apple.com>
92
93         <http://webkit.org/b/72646> Disable deprecation warnings around code where we cannot easily
94         switch away from the deprecated APIs.
95
96         Reviewed by Sam Weinig.
97
98         * platform/mac/WebCoreNSStringExtras.mm:
99         * platform/network/cf/SocketStreamHandleCFNet.cpp:
100         (WebCore::SocketStreamHandle::reportErrorToClient):
101
102 2011-11-19  Kevin Ollivier  <kevino@theolliviers.com>
103
104         [wx] C++ bindings build fix for move of array classes to WTF.
105         
106         * bindings/scripts/CodeGeneratorCPP.pm:
107         (GetCPPTypeGetter):
108         (GetNamespaceForClass):
109         (GenerateHeader):
110         (GenerateImplementation):
111
112 2011-11-19  Scott Graham  <scottmg@chromium.org>
113
114         Move gamepad to Modules/ (+ some cleanup)
115         https://bugs.webkit.org/show_bug.cgi?id=72785
116
117         Reviewed by Adam Barth.
118
119         Move main files from page/ to Modules/gamepad/. #include guard the
120         gamepad header inclusions in Navigator.cpp to avoid including for
121         ports that do not enable GAMEPAD.
122
123         * Modules/gamepad/Gamepad.cpp: Renamed from Source/WebCore/page/Gamepad.cpp.
124         (WebCore::Gamepad::Gamepad):
125         (WebCore::Gamepad::axes):
126         (WebCore::Gamepad::buttons):
127         (WebCore::Gamepad::~Gamepad):
128         * Modules/gamepad/Gamepad.h: Renamed from Source/WebCore/page/Gamepad.h.
129         * Modules/gamepad/Gamepad.idl: Renamed from Source/WebCore/page/Gamepad.idl.
130         * Modules/gamepad/GamepadList.cpp: Renamed from Source/WebCore/page/GamepadList.cpp.
131         (WebCore::GamepadList::~GamepadList):
132         (WebCore::GamepadList::set):
133         (WebCore::GamepadList::length):
134         (WebCore::GamepadList::item):
135         * Modules/gamepad/GamepadList.h: Renamed from Source/WebCore/page/GamepadList.h.
136         (WebCore::GamepadList::create):
137         (WebCore::GamepadList::GamepadList):
138         * Modules/gamepad/GamepadList.idl: Renamed from Source/WebCore/page/GamepadList.idl.
139         * WebCore.gyp/WebCore.gyp:
140         * WebCore.gypi:
141         * page/Navigator.cpp:
142
143 2011-11-19  Sheriff Bot  <webkit.review.bot@gmail.com>
144
145         Unreviewed, rolling out r100834.
146         http://trac.webkit.org/changeset/100834
147         https://bugs.webkit.org/show_bug.cgi?id=72806
148
149         this patch makes Qt run-webkit-test exit before finishing
150         (Requested by igoroliveira on #webkit).
151
152         * platform/graphics/texmap/TextureMapperNode.cpp:
153         * platform/graphics/texmap/TextureMapperNode.h:
154
155 2011-11-19  Huang Dongsung  <luxtella@company100.net>
156
157         Remove WebCore/ForwardingHeaders/runtime/JSObjectWithGlobalObject.h because
158         JSObjectWithGlobalObject.h has been removed.
159         https://bugs.webkit.org/show_bug.cgi?id=72794
160
161         r94701 removed JSObjectWithGlobalObject.h.
162
163         Reviewed by Oliver Hunt.
164
165         * ForwardingHeaders/runtime/JSObjectWithGlobalObject.h: Removed.
166
167 2011-11-19  Vsevolod Vlasov  <vsevik@chromium.org>
168
169         Web Inspector:[protocol] actual JSON messages do not conform to Inspector.json in InspectorApplicationCacheAgent.cpp
170         https://bugs.webkit.org/show_bug.cgi?id=72734
171
172         Reviewed by Timothy Hatcher.
173
174         * inspector/Inspector.json:
175
176 2011-11-19  David Barr  <davidbarr@chromium.org>
177
178         REGRESSION(r98542): Chromium: CSS text is rendered on page
179         https://bugs.webkit.org/show_bug.cgi?id=71703
180
181         Reviewed by Antti Koivisto.
182
183         Matched UA declarations uncacheable when using simpleDefaultStyleSheet.
184
185         Test: fast/css/style-tag-display-none.html
186
187         * css/CSSStyleSelector.cpp:
188         (WebCore::CSSStyleSelector::matchUARules):
189
190 2011-11-19  Jay Civelli  <jcivelli@chromium.org>
191
192         When loading a MHTML document, make sure we set the base URL before
193         we notify the frame was committed (so the document has the right base
194         URL when the notification happens). 
195         https://bugs.webkit.org/show_bug.cgi?id=72788
196
197         Reviewed by Adam Barth.
198
199         * loader/FrameLoader.cpp:
200         (WebCore::FrameLoader::receivedFirstData):
201
202 2011-11-19  Adam Barth  <abarth@webkit.org>
203
204         Integrate Source/WTF with the Chromium build system
205         https://bugs.webkit.org/show_bug.cgi?id=72790
206
207         Reviewed by Eric Seidel.
208
209         Add a dependency on the new WTF.
210
211         * WebCore.gyp/WebCore.gyp:
212
213 2011-11-18  Takashi Toyoshima  <toyoshim@chromium.org>
214
215         [Chromium] [WebSocket] export WebSocketChannel interface for plugins
216         https://bugs.webkit.org/show_bug.cgi?id=72016
217
218         Reviewed by Darin Fisher.
219
220         Add a interface to send raw binary data.
221         This interface is used by WebWebSocketChannel implementation
222         in WebKit API.
223
224         No new tests because just export a interface.
225
226         * websockets/WebSocketChannel.cpp:
227         (WebCore::WebSocketChannel::send):
228         * websockets/WebSocketChannel.h:
229
230 2011-11-18  Vineet Chaudhary  <vineet.chaudhary@motorola.com>
231
232         https://bugs.webkit.org/show_bug.cgi?id=72591
233         Remove document.width / document.height
234
235         Reviewed by Darin Adler.
236
237         Removed document.width/document.height from JS bindings,
238         but keeping the same for ObjC bindings. Also it should use
239         document.body.clientWidth and document.body.clientHeight instead.
240
241         * html/HTMLDocument.idl:
242
243 2011-11-18  Martin Robinson  <mrobinson@igalia.com>
244
245         Fix the GTK+ build.
246
247         * page/Navigator.idl: Properly disable the webkitGamepads API if gamepad
248         isn't enabled at compile time.
249
250 2011-11-18  Daniel Bates  <dbates@rim.com>
251
252         Add CMake build infrastructure for the BlackBerry port
253         https://bugs.webkit.org/show_bug.cgi?id=72768
254
255         Reviewed by Antonio Gomes.
256
257         * CMakeLists.txt: At this time the BlackBerry port doesn't support generating
258           DOM bindings from the SVG IDLs. See WebKit bug #72773.
259         * PlatformBlackBerry.cmake: Added.
260
261 2011-11-18  Sheriff Bot  <webkit.review.bot@gmail.com>
262
263         Unreviewed, rolling out r100826.
264         http://trac.webkit.org/changeset/100826
265         https://bugs.webkit.org/show_bug.cgi?id=72786
266
267         Broke Chromium Mac build (Requested by aklein on #webkit).
268
269         * page/ChromeClient.h:
270         * page/FrameView.cpp:
271         * page/FrameView.h:
272         * platform/ScrollView.cpp:
273         (WebCore::ScrollView::wheelEvent):
274         * platform/ScrollView.h:
275         * platform/ScrollableArea.h:
276         (WebCore::ScrollableArea::scrollbarStyleChanged):
277         * platform/mac/ScrollAnimatorMac.mm:
278         (WebCore::ScrollAnimatorMac::updateScrollerStyle):
279
280 2011-11-18  Alpha Lam  <hclam@chromium.org>
281
282         [chromium] composited layers are blurry with a zoom-in page scale factor
283         https://bugs.webkit.org/show_bug.cgi?id=71225
284
285         Reviewed by James Robinson.
286
287         Pass contents scale factor to the compositor such that it can:
288         1. Adjust contentBounds() of the compositoer layers with content scale.
289         2. Apply the content scale in the painter for texture update in better resolution.
290         3. Apply the content scale to the dirty rect in CanvasLayerTextureUpdater.
291
292         This change fixed blurry problem for all tiled layer types and is not limited to
293         position:fixed layers.
294
295         Tests: compositing/geometry/fixed-position-composited-page-scale-down.html
296                compositing/geometry/fixed-position-composited-page-scale.html
297                compositing/geometry/fixed-position-iframe-composited-page-scale-down.html
298                compositing/geometry/fixed-position-iframe-composited-page-scale.html
299                compositing/geometry/fixed-position-transform-composited-page-scale-down.html
300                compositing/geometry/fixed-position-transform-composited-page-scale.html
301
302         * platform/graphics/chromium/GraphicsLayerChromium.cpp:
303         (WebCore::GraphicsLayerChromium::setTransform):
304         (WebCore::GraphicsLayerChromium::updateLayerPreserves3D):
305         (WebCore::GraphicsLayerChromium::updateContentsScale):
306         (WebCore::GraphicsLayerChromium::contentsScale):
307         (WebCore::GraphicsLayerChromium::deviceOrPageScaleFactorChanged):
308         * platform/graphics/chromium/GraphicsLayerChromium.h:
309         * platform/graphics/chromium/ImageLayerChromium.cpp:
310         (WebCore::ImageLayerTextureUpdater::prepareToUpdate):
311         (WebCore::ImageLayerChromium::needsContentsScale):
312         * platform/graphics/chromium/ImageLayerChromium.h:
313         * platform/graphics/chromium/LayerChromium.cpp:
314         (WebCore::LayerChromium::LayerChromium):
315         (WebCore::LayerChromium::setContentsScale):
316         * platform/graphics/chromium/LayerChromium.h:
317         (WebCore::LayerChromium::needsContentsScale):
318         (WebCore::LayerChromium::contentsScale):
319         * platform/graphics/chromium/LayerTextureUpdater.h:
320         * platform/graphics/chromium/LayerTextureUpdaterCanvas.cpp:
321         (WebCore::LayerTextureUpdaterCanvas::paintContents):
322         (WebCore::LayerTextureUpdaterBitmap::prepareToUpdate):
323         (WebCore::LayerTextureUpdaterSkPicture::prepareToUpdate):
324         * platform/graphics/chromium/LayerTextureUpdaterCanvas.h:
325         * platform/graphics/chromium/TiledLayerChromium.cpp:
326         (WebCore::TiledLayerChromium::needsContentsScale):
327         (WebCore::TiledLayerChromium::contentBounds):
328         (WebCore::TiledLayerChromium::prepareToUpdate):
329         * platform/graphics/chromium/TiledLayerChromium.h:
330
331 2011-11-18  Igor Oliveira  <igor.oliveira@openbossa.org>
332
333         [WK2][Qt] Move Accelerated Composite animations to UIProcess
334         https://bugs.webkit.org/show_bug.cgi?id=72753
335
336         Add helper method to synchronize animations in TextureMapper.
337
338         Reviewed by Noam Rosenthal.
339
340         * platform/graphics/texmap/TextureMapperNode.cpp:
341         (WebCore::TextureMapperNode::syncAnimationsRecursively):
342         * platform/graphics/texmap/TextureMapperNode.h:
343
344 2011-11-18  Scott Graham  <scottmg@chromium.org>
345
346         IDL changes for gamepad support
347         https://bugs.webkit.org/show_bug.cgi?id=71753
348
349         Reviewed by Adam Barth.
350
351         IDL changes and associated plumbing to expose list of gamepad objects
352         on navigator object (per current spec). Full patch is
353         https://bugs.webkit.org/show_bug.cgi?id=69451. Only basic existence
354         test until more plumbing in future patches.
355
356         Test: gamepad/gamepad-api.html
357
358         * WebCore.gypi:
359         * bindings/generic/RuntimeEnabledFeatures.h:
360         (WebCore::RuntimeEnabledFeatures::setWebkitGamepadsEnabled):
361         (WebCore::RuntimeEnabledFeatures::webkitGamepadsEnabled):
362         * page/Gamepad.cpp: Added.
363         (WebCore::Gamepad::Gamepad):
364         (WebCore::Gamepad::axes):
365         (WebCore::Gamepad::buttons):
366         (WebCore::Gamepad::~Gamepad):
367         * page/Gamepad.h: Added.
368         * page/Gamepad.idl: Added.
369         * page/GamepadList.cpp: Added.
370         (WebCore::GamepadList::~GamepadList):
371         (WebCore::GamepadList::set):
372         (WebCore::GamepadList::length):
373         (WebCore::GamepadList::item):
374         * page/GamepadList.h: Added.
375         (WebCore::GamepadList::create):
376         (WebCore::GamepadList::GamepadList):
377         * page/GamepadList.idl: Added.
378         * page/Navigator.cpp:
379         (WebCore::Navigator::webkitGamepads):
380         * page/Navigator.h:
381         * page/Navigator.idl:
382
383 2011-11-18  Sheriff Bot  <webkit.review.bot@gmail.com>
384
385         Unreviewed, rolling out r100693.
386         http://trac.webkit.org/changeset/100693
387         https://bugs.webkit.org/show_bug.cgi?id=72779
388
389         This patch caused a rendering regression (see bug 72770)
390         (Requested by philip__ on #webkit).
391
392         * rendering/RenderObject.cpp:
393         (WebCore::RenderObject::setStyle):
394         * rendering/style/RenderStyle.cpp:
395         (WebCore::RenderStyle::diff):
396
397 2011-11-18  Simon Fraser  <simon.fraser@apple.com>
398
399         Reflection on composited element doesn't update if the element changes to show no content
400         https://bugs.webkit.org/show_bug.cgi?id=72774
401
402         Reviewed by Chris Marrin.
403         
404         When a style change results in a reflected element losing its backing store,
405         we need to clear the backing stores on the layer clones as well.
406
407         Test: compositing/reflections/become-simple-composited-reflection.html
408
409         * platform/graphics/ca/GraphicsLayerCA.cpp:
410         (WebCore::GraphicsLayerCA::updateLayerDrawsContent):
411
412 2011-11-18  Daniel Cheng  <dcheng@chromium.org>
413
414         [chromium] Use correct backing store for ChromiumDataObject in pasteboard writes.
415         https://bugs.webkit.org/show_bug.cgi?id=72767
416
417         Reviewed by Tony Chang.
418
419         Covered by existing tests.
420
421         * editing/chromium/EditorChromium.cpp:
422         (WebCore::Editor::newGeneralClipboard):
423         * page/chromium/EventHandlerChromium.cpp:
424         (WebCore::EventHandler::createDraggingClipboard):
425         * platform/chromium/ChromiumDataObject.cpp:
426         (WebCore::ChromiumDataObject::types):
427         (WebCore::ChromiumDataObject::getData):
428         (WebCore::ChromiumDataObject::containsFilenames):
429         (WebCore::ChromiumDataObject::ChromiumDataObject):
430         * platform/chromium/ChromiumDataObject.h:
431         (WebCore::ChromiumDataObject::createFromPasteboard):
432         (WebCore::ChromiumDataObject::create):
433         (WebCore::ChromiumDataObject::storageMode):
434         * platform/chromium/ClipboardChromium.cpp:
435         (WebCore::ClipboardChromium::hasData):
436
437 2011-11-18  Chris Evans  <cevans@google.com>
438
439         Crash with ranges across a detached, reparented node tree
440         https://bugs.webkit.org/show_bug.cgi?id=72757
441
442         Reviewed by Adam Barth.
443
444         Test: fast/dom/move-detached-child-in-range.html
445
446         * dom/RangeBoundaryPoint.h:
447         (WebCore::RangeBoundaryPoint::childBefore): protect the raw child node from getting pulled from under us.
448
449 2011-11-18  Beth Dakin  <bdakin@apple.com>
450
451         https://bugs.webkit.org/show_bug.cgi?id=72551
452         When the recommended scrollbar style changes, WKView's tracking options should 
453         adjust accordingly
454         -and corresponding-
455         <rdar://problem/10409328>
456
457         Reviewed by Darin Adler.
458
459         This new ChromeClient function is called when the recommended scrollbar style 
460         changes. This way, WebKit can respond to the change by adjusting its mouse 
461         tracking.
462         * page/ChromeClient.h:
463         (WebCore::ChromeClient::recommendedScrollbarStyleDidChange):
464
465         Existing ScrollableArea function scrollbarStyleChanged() now takes an int 
466         indicating the new scrollbar style and a bool indicating whether it is necessary 
467         to force an update. It used to be the case that this function was ONLY used to 
468         force an update (and only called when an updated was needed), but now that it must 
469         also call into the ChromeClient, it is necessary to include a bool tracking 
470         whether we need to force an update. New implementation on FrameView is responsible 
471         for calling ChromeClient, and then that calls into the pre-existing ScrollView 
472         function for the forceUpdate part.
473         * page/FrameView.cpp:
474         (WebCore::FrameView::scrollbarStyleChanged):
475         * page/FrameView.h:
476         * platform/ScrollView.cpp:
477         (WebCore::ScrollView:: scrollbarStyleChanged):
478         * platform/ScrollView.h:
479         * platform/ScrollableArea.h:
480         (WebCore::ScrollableArea::scrollbarStyleChanged):
481         * platform/mac/ScrollAnimatorMac.mm:
482         (WebCore::ScrollAnimatorMac::updateScrollerStyle):
483
484 2011-11-18  Kelly Norton  <knorton@google.com>
485
486         Fixes several more void functions in RenderObject that return values.
487         https://bugs.webkit.org/show_bug.cgi?id=72750
488
489         Reviewed by Adam Barth.
490
491         * rendering/RenderBlock.cpp:
492         (WebCore::RenderBlock::updateBeforeAfterContent):
493         (WebCore::RenderBlock::addChildToContinuation):
494         (WebCore::RenderBlock::addChildToAnonymousColumnBlocks):
495         (WebCore::RenderBlock::addChild):
496         (WebCore::RenderBlock::addChildIgnoringContinuation):
497
498 2011-11-18  Shawn Singh  <shawnsingh@chromium.org>
499
500         [chromium] Add some useful text to existing debug dump
501         https://bugs.webkit.org/show_bug.cgi?id=72576
502
503         Reviewed by James Robinson.
504
505         No new tests needed.
506
507         * platform/graphics/chromium/cc/CCLayerImpl.cpp:
508         (WebCore::CCLayerImpl::dumpLayerProperties):
509         * platform/graphics/chromium/cc/CCRenderSurface.cpp:
510         (WebCore::CCRenderSurface::dumpSurface):
511
512 2011-11-18  Xiaomei Ji  <xji@chromium.org>
513
514         REGRESSION: rtl horizontal scrollbar / resize bug - Body shifts on resize when scrolled all the way to the left
515         https://bugs.webkit.org/show_bug.cgi?id=70395
516
517         Reviewed by Tony Chang.
518
519         This patch fixes the problem in Mac and Chromium Mac.
520    
521         The existing test fast/dom/rtl-scroll-to-leftmost-and-resize.html seems does not really work in Mac DRT
522         (the browswer window is not resized). It works in Chromium-Mac. And Chromium-Mac's code
523         is forked from Mac.
524
525         * platform/chromium/ScrollAnimatorChromiumMac.mm:
526         (WebCore::ScrollAnimatorChromiumMac::immediateScrollToPoint):
527         * platform/mac/ScrollAnimatorMac.mm:
528         (WebCore::ScrollAnimatorMac::immediateScrollToPoint):
529
530 2011-11-18  Martin Robinson  <mrobinson@igalia.com>
531
532         REGRESSION (r99924): broke 2 pasteboard tests on GTK
533         https://bugs.webkit.org/show_bug.cgi?id=72131
534
535         Reviewed by Tony Chang.
536
537         * platform/gtk/ClipboardGtk.cpp:
538         (WebCore::ClipboardGtk::clearData): Call clearAll now.
539         (WebCore::ClipboardGtk::clearAllData): Call clearAllExceptFilenames now.
540         * platform/gtk/DataObjectGtk.cpp:
541         (WebCore::DataObjectGtk::clearAllExceptFilenames): Renamed from clear.
542         (WebCore::DataObjectGtk::clearAll): Added this method which also clear filenames.
543         * platform/gtk/DataObjectGtk.h:
544         * platform/gtk/PasteboardGtk.cpp: Call clear before setting new clipboard data.
545         (WebCore::Pasteboard::writeSelection): Ditto.
546         (WebCore::Pasteboard::writePlainText): Ditto.
547         (WebCore::Pasteboard::writeURL): Ditto.
548         (WebCore::Pasteboard::writeImage): Ditto.
549         * platform/gtk/PasteboardHelper.cpp:
550         (WebCore::clearClipboardContentsCallback): Use the clearAll method now.
551
552 2011-11-18  Vineet Chaudhary  <vineet.chaudhary@motorola.com>
553
554         Access key should work on all elements.
555         https://bugs.webkit.org/show_bug.cgi?id=71854
556
557         Reviewed by Ryosuke Niwa.
558
559         All HTML elements can have the accesskey content attribute set.
560         Specification http://dev.w3.org/html5/spec/Overview.html#the-accesskey-attribute
561         Adding "accessKey" attribute to HTMLElement.idl file as [Reflect].
562
563         Test: fast/forms/access-key-for-all-elements.html
564
565         * bindings/objc/PublicDOMInterfaces.h: Moved properties form subclass to base class.
566         * html/BaseButtonInputType.cpp: 
567         (WebCore::BaseButtonInputType::accessKeyAction): Renamed variable sendToAnyElement to sendMouseEvents.
568         * html/BaseButtonInputType.h: Ditto
569         * html/BaseCheckableInputType.cpp:
570         (WebCore::BaseCheckableInputType::accessKeyAction): Ditto
571         * html/BaseCheckableInputType.h: Ditto
572         * html/HTMLAnchorElement.cpp:
573         (WebCore::HTMLAnchorElement::accessKeyAction): Ditto
574         * html/HTMLAnchorElement.h: Ditto
575         * html/HTMLAnchorElement.idl: Removed redundant IDL attribute entries.
576         * html/HTMLAreaElement.idl: Removed redundant IDL attribute entries.
577         * html/HTMLButtonElement.cpp:
578         (WebCore::HTMLButtonElement::accessKeyAction): Renamed variable sendToAnyElement to sendMouseEvents.
579         * html/HTMLButtonElement.h: Ditto
580         * html/HTMLButtonElement.idl: Removed redundant IDL attribute entries.
581         * html/HTMLElement.cpp:
582         (WebCore::HTMLElement::accessKeyAction): Renamed variable sendToAnyElement to sendMouseEvents.
583         If the element does not have a defined activation behavior, fire a click event at the element.
584         * html/HTMLElement.h: Ditto
585         * html/HTMLElement.idl: Added accessKey IDL attribute.
586         * html/HTMLInputElement.cpp:
587         (WebCore::HTMLInputElement::accessKeyAction): Renamed variable sendToAnyElement to sendMouseEvents.
588         * html/HTMLInputElement.h: Ditto
589         * html/HTMLInputElement.idl: Removed redundant IDL attribute entries.
590         * html/HTMLLabelElement.cpp:
591         (WebCore::HTMLLabelElement::accessKeyAction): Renamed variable sendToAnyElement to sendMouseEvents.
592         * html/HTMLLabelElement.h: Ditto
593         * html/HTMLLabelElement.idl: Removed redundant IDL attribute entries.
594         * html/HTMLLegendElement.cpp:
595         (WebCore::HTMLLegendElement::accessKeyAction): Renamed variable sendToAnyElement to sendMouseEvents.
596         * html/HTMLLegendElement.h: Ditto
597         * html/HTMLLegendElement.idl: Removed redundant IDL attribute entries.
598         * html/HTMLOptGroupElement.h: Renamed variable sendToAnyElement to sendMouseEvents.
599         * html/HTMLSelectElement.cpp:
600         (WebCore::HTMLSelectElement::accessKeyAction): Ditto
601         * html/HTMLSelectElement.h: Ditto
602         * html/HTMLTextAreaElement.h: Ditto
603         * html/HTMLTextAreaElement.idl: Removed redundant IDL attribute entries.
604         * html/HiddenInputType.h: Renamed variable sendToAnyElement to sendMouseEvents.
605         * html/InputType.h: Ditto
606         * html/RangeInputType.cpp:
607         (WebCore::RangeInputType::accessKeyAction): Ditto
608         * html/RangeInputType.h: Ditto
609
610 2011-11-18  Simon Fraser  <simon.fraser@apple.com>
611
612         Appearance of compound transform animations under apps linked on SnowLeopard is incorrect
613         https://bugs.webkit.org/show_bug.cgi?id=72641
614         and
615         <rdar://problem/10314267>
616
617         Reviewed by Dean Jackson.
618
619         GraphicsLayerCA contains a "linked on or after" check to account for a bug in
620         Core Animation on SnowLeopard and earlier, which is that CA would apply the list
621         of animations in reverse order.
622         
623         Our previous fix was incorrect, because it only adjusted the 'additive' property
624         of the animation list based on ordering, rather than flipping the entire list.
625         This change reverses the list of animations before giving them to CA, which fixes
626         the bug.
627                 
628         Test: animations/additive-transform-animations.html
629
630         * platform/graphics/ca/GraphicsLayerCA.cpp:
631         (WebCore::GraphicsLayerCA::appendToUncommittedAnimations):
632         (WebCore::GraphicsLayerCA::createTransformAnimationsFromKeyframes):
633         * platform/graphics/ca/GraphicsLayerCA.h:
634
635 2011-11-18  Tim Horton  <timothy_horton@apple.com>
636
637         -webkit-cross-fade rendered incorrectly in overflow divs
638         https://bugs.webkit.org/show_bug.cgi?id=72693
639         <rdar://problem/10468564>
640
641         Reviewed by Simon Fraser.
642
643         Respect the desired source rectangle when rendering the cross-fade.
644
645         Test: css3/images/cross-fade-overflow-position.html
646
647         * platform/graphics/CrossfadeGeneratedImage.cpp:
648         (WebCore::CrossfadeGeneratedImage::drawCrossfade):
649         (WebCore::CrossfadeGeneratedImage::draw):
650         (WebCore::CrossfadeGeneratedImage::drawPattern):
651         * platform/graphics/CrossfadeGeneratedImage.h:
652
653 2011-11-18  Raphael Kubo da Costa  <kubo@profusion.mobi>
654
655         Unreviewed build fix; r100686 broke the EFL build when Geolocation
656         support is enabled.
657
658         * platform/efl/GeolocationServiceEfl.cpp: Only declare
659         s_factoryFunction if CLIENT_BASED_GEOLOCATION is off, otherwise it is
660         also declared in GeolocationService.cpp.
661
662 2011-11-18  Igor Oliveira  <igor.oliveira@openbossa.org>
663
664         [TextureMapper] computePerspectiveTransformIfNeeded is called twice in TextureMapperNode::syncCompositingState
665         https://bugs.webkit.org/show_bug.cgi?id=72727
666
667         TextureMapperNode::computeAllTransforms already has a call to TextureMapperNode::computePerspectiveTransformIfNeeded,
668         so it does not need to be called again after TextureMapperNode::computeAllTransforms.
669
670         Reviewed by Noam Rosenthal.
671
672         * platform/graphics/texmap/TextureMapperNode.cpp:
673         (WebCore::TextureMapperNode::syncCompositingState):
674
675 2011-11-18  Mihnea Ovidenie  <mihnea@adobe.com>
676
677         Fix compilation warning in ComplexTextControllerCoreText.mm
678         https://bugs.webkit.org/show_bug.cgi?id=72702
679
680         Reviewed by Andreas Kling.
681
682         No functionality changed, so no new tests.
683
684         * platform/graphics/mac/ComplexTextControllerCoreText.mm:
685         (WebCore::ComplexTextController::collectComplexTextRunsForCharactersCoreText):
686
687 2011-11-18  Pavel Feldman  <pfeldman@google.com>
688
689         Not reviewed: restore front-end compilability via updating externs and JS generator.
690
691         * inspector/front-end/ElementsPanel.js:
692         (WebInspector.ElementsPanel.prototype.jumpToNextSearchResult):
693         (WebInspector.ElementsPanel.prototype.jumpToPreviousSearchResult):
694         * inspector/front-end/externs.js:
695         (WebInspector.showPanelForAnchorNavigation):
696         * inspector/generate-protocol-externs:
697
698 2011-11-17  Pavel Feldman  <pfeldman@google.com>
699
700         Web Inspector: dispatch messages from the front-end to the backend asynchronously.
701         https://bugs.webkit.org/show_bug.cgi?id=72621
702
703         We should align the way we dispatch messages from the front-end to backend across the environments:
704           - WebKit has it synchronoulsly
705           - Chromium has it asynchronously
706           - Remote debugging has it asynchronously
707         Making it asynchronous made a number of flaky Qt tests pass.
708
709         Tests uncovered console agent problem that was also fixed.
710
711         Reviewed by Yury Semikhatsky.
712
713         * inspector/InspectorConsoleAgent.cpp:
714         (WebCore::InspectorConsoleAgent::clearFrontend):
715         * inspector/InspectorFrontendClient.h:
716         * inspector/InspectorFrontendClientLocal.cpp:
717         (WebCore::InspectorBackendDispatchTask::InspectorBackendDispatchTask):
718         (WebCore::InspectorBackendDispatchTask::dispatch):
719         (WebCore::InspectorBackendDispatchTask::reset):
720         (WebCore::InspectorBackendDispatchTask::onTimer):
721         (WebCore::InspectorFrontendClientLocal::InspectorFrontendClientLocal):
722         (WebCore::InspectorFrontendClientLocal::windowObjectCleared):
723         (WebCore::InspectorFrontendClientLocal::sendMessageToBackend):
724         * inspector/InspectorFrontendClientLocal.h:
725
726 2011-11-18  Iain Merrick  <husky@google.com>
727
728         [chromium] Pass screen refresh rate into compositor.
729         https://bugs.webkit.org/show_bug.cgi?id=71040
730
731         Reviewed by Tony Gentilcore.
732
733         Covered by CCLayerTreeHostTest.
734
735         * platform/PlatformScreen.h:
736         * platform/chromium/PlatformScreenChromium.cpp:
737         (WebCore::screenRefreshRate):
738         * platform/chromium/PlatformSupport.h:
739         * platform/graphics/chromium/cc/CCLayerTreeHost.h:
740         (WebCore::CCSettings::CCSettings):
741         * platform/graphics/chromium/cc/CCThreadProxy.cpp:
742         (WebCore::CCThreadProxy::initializeImplOnImplThread):
743
744 2011-11-18  Alexandru Chiculita  <achicu@adobe.com>
745
746         [CSSShaders] Implement the computed style for mesh parameters of the custom() filter
747         https://bugs.webkit.org/show_bug.cgi?id=72478
748
749         Reviewed by Dean Jackson.
750
751         Added parsing and computed style for the mesh rows, columns, mesh box type 
752         (filter-box, border-box, content-box and padding-box) and the detached mode.
753         
754         Also fixed a case where "custom(none, 10, 10 filter-box)" was incorrectly
755         treated as "custom(none, 10)".
756
757         * css/CSSComputedStyleDeclaration.cpp:
758         (WebCore::CSSComputedStyleDeclaration::valueForFilter):
759         * css/CSSParser.cpp:
760         (WebCore::CSSParser::parseCustomFilter): Fixed a case where invalid syntax was parsed as correct syntax.
761         * css/CSSPrimitiveValueMappings.h:
762         (WebCore::CSSPrimitiveValue::CSSPrimitiveValue):
763         (WebCore::CSSPrimitiveValue::operator CustomFilterOperation::MeshBoxType):
764         * css/CSSStyleSelector.cpp:
765         (WebCore::CSSStyleSelector::createCustomFilterOperation):
766         * css/CSSValueList.h:
767         (WebCore::CSSValueListIterator::isPrimitiveValue):
768         * platform/graphics/filters/CustomFilterOperation.h:
769
770 2011-11-18  Adam Barth  <abarth@webkit.org>
771
772         Remove unneeded include (and ifdef) from V8DOMWindowCustom.cpp
773         https://bugs.webkit.org/show_bug.cgi?id=72705
774
775         Reviewed by Eric Seidel.
776
777         Death to ifdefs!
778
779         * bindings/v8/custom/V8DOMWindowCustom.cpp:
780
781 2011-11-18  Adam Barth  <abarth@webkit.org>
782
783         Move some mediastream related files into the mediastream directory
784         https://bugs.webkit.org/show_bug.cgi?id=72695
785
786         Reviewed by Eric Seidel.
787
788         Just moving the files to the proper directory.
789
790         * GNUmakefile.list.am:
791         * WebCore.gypi:
792         * mediastream/NavigatorUserMediaError.h: Renamed from Source/WebCore/page/NavigatorUserMediaError.h.
793         (WebCore::NavigatorUserMediaError::create):
794         (WebCore::NavigatorUserMediaError::~NavigatorUserMediaError):
795         (WebCore::NavigatorUserMediaError::code):
796         (WebCore::NavigatorUserMediaError::NavigatorUserMediaError):
797         * mediastream/NavigatorUserMediaError.idl: Renamed from Source/WebCore/page/NavigatorUserMediaError.idl.
798         * mediastream/NavigatorUserMediaErrorCallback.h: Renamed from Source/WebCore/page/NavigatorUserMediaErrorCallback.h.
799         (WebCore::NavigatorUserMediaErrorCallback::~NavigatorUserMediaErrorCallback):
800         * mediastream/NavigatorUserMediaErrorCallback.idl: Renamed from Source/WebCore/page/NavigatorUserMediaErrorCallback.idl.
801         * mediastream/NavigatorUserMediaSuccessCallback.h: Renamed from Source/WebCore/page/NavigatorUserMediaSuccessCallback.h.
802         (WebCore::NavigatorUserMediaSuccessCallback::~NavigatorUserMediaSuccessCallback):
803         * mediastream/NavigatorUserMediaSuccessCallback.idl: Renamed from Source/WebCore/page/NavigatorUserMediaSuccessCallback.idl.
804
805 2011-11-18  Andrey Kosyakov  <caseq@chromium.org>
806
807         Web Inspector: [Extensions API] Provide a way for extension to create a status bar icon
808         https://bugs.webkit.org/show_bug.cgi?id=45955
809
810         Reviewed by Pavel Feldman.
811
812         * inspector/front-end/ExtensionAPI.js:
813         (injectedExtensionAPI.ExtensionPanelImpl.prototype.createStatusBarButton):
814         (injectedExtensionAPI):
815         (injectedExtensionAPI.ButtonImpl.prototype.update):
816         * inspector/front-end/ExtensionPanel.js:
817         (WebInspector.ExtensionPanel):
818         (WebInspector.ExtensionPanel.prototype.get statusBarItems):
819         (WebInspector.ExtensionPanel.prototype.addStatusBarItem):
820         (WebInspector.ExtensionButton):
821         (WebInspector.ExtensionButton.prototype.update):
822         (WebInspector.ExtensionButton.prototype._onClicked):
823         * inspector/front-end/ExtensionServer.js:
824         (WebInspector.ExtensionServer):
825         (WebInspector.ExtensionServer.prototype.notifyButtonClicked):
826         (WebInspector.ExtensionServer.prototype._onCreateStatusBarButton):
827         (WebInspector.ExtensionServer.prototype._onUpdateButton):
828         (WebInspector.ExtensionServer.prototype._onCreateSidebarPane):
829         * inspector/front-end/inspector.css:
830         (button.status-bar-item.extension):
831
832 2011-11-18  Tommy Widenflycht  <tommyw@google.com>
833
834         MediaStream API: Merging the PeerConnectionHandler.h files
835         https://bugs.webkit.org/show_bug.cgi?id=72611
836
837         Reviewed by Adam Barth.
838
839         No actual code changes.
840
841         * GNUmakefile.am:
842         * GNUmakefile.list.am:
843         * WebCore.gyp/WebCore.gyp:
844         * WebCore.gypi:
845         * platform/mediastream/PeerConnectionHandler.h: Renamed from Source/WebCore/platform/mediastream/chromium/PeerConnectionHandler.h.
846         * platform/mediastream/gstreamer/PeerConnectionHandler.h: Removed.
847
848 2011-11-17  Simon Hausmann  <simon.hausmann@nokia.com>
849
850         [Qt] Layer violation: ThirdPartyCookiesQt.cpp uses QWebSettings in two places
851         https://bugs.webkit.org/show_bug.cgi?id=72597
852
853         Reviewed by Kenneth Rohde Christiansen.
854
855         Moved code that queries the third party cookie policy from QWebSettings
856         out of WebCore. Instead NetworkingContext has now the interface that allows
857         delegating the functionality into WebKit/qt.
858
859         In order to do that we need to pass the NetworkingContext around, which is used to
860         retrieve the cookie jar and the originating frame object pointer.
861
862         * platform/network/NetworkingContext.h: Add interface.
863         * platform/network/qt/QNetworkReplyHandler.cpp:
864         (WebCore::QNetworkReplyHandler::QNetworkReplyHandler): Pass NetworkingContext to ResourceRequest::toNetworkRequest.
865         (WebCore::QNetworkReplyHandler::redirect): Ditto.
866         * platform/network/qt/ResourceRequest.h: Replaced originatingObject parameter with NetworkingContext, which can
867         also provide the same plus more (cookie jar, policy callback).
868         * platform/network/qt/ResourceRequestQt.cpp:
869         (WebCore::ResourceRequest::toNetworkRequest): Ditto.
870         * platform/qt/CookieJarQt.cpp: Simplify to use NetworkingContext to get cookie jar and
871         call new thirdPartyCookiePolicyPermits API that takes the context as parameter.
872         (WebCore::networkingContext):
873         (WebCore::setCookies):
874         (WebCore::cookies):
875         (WebCore::cookieRequestHeaderFieldValue):
876         (WebCore::cookiesEnabled):
877         * platform/qt/ThirdPartyCookiesQt.cpp: Replace up-casting to QWebFrame and use of QWebSettings
878         with useage of NetworkingContext.
879         (WebCore::thirdPartyCookiePolicyPermits):
880         * platform/qt/ThirdPartyCookiesQt.h:
881
882 2011-11-17  Kenichi Ishibashi  <bashi@chromium.org>
883
884         crash: WebCore::FontPlatformData::roundsGlyphAdvances on Lion
885         https://bugs.webkit.org/show_bug.cgi?id=71997
886
887         Reviewed by Dan Bernstein.
888
889         The cause is a null dereference of a fontData that is stored in
890         ComplexTextRun. The fontData is initialized by using the
891         fontCache, but it could be null when the font is in fallback
892         list. The reason a font from the fallback list might not be in the
893         font Cache is that it may be a web font. Before looking up the
894         fontCache, try to see whether the font is in the fallback list.
895
896         No new tests. We don't have webfonts that can produce the problem.
897
898         * platform/graphics/mac/ComplexTextControllerCoreText.mm:
899         (WebCore::ComplexTextController::collectComplexTextRunsForCharactersCoreText): See fallback list first, then lookup cache.
900
901 2011-11-17  Kaustubh Atrawalkar  <kaustubh@motorola.com>
902
903         Remove initProgressEvent method
904         https://bugs.webkit.org/show_bug.cgi?id=71340
905
906         Reviewed by Adam Barth.
907
908         This method has been removed from the spec draft.
909         http://www.w3.org/TR/progress-events/#interface-progressevent
910
911         No new tests. Removed method.
912
913         * dom/ProgressEvent.cpp:
914         * dom/ProgressEvent.h:
915         * dom/ProgressEvent.idl:
916
917 2011-11-17  Eunmi Lee  <eunmi15.lee@samsung.com>
918
919         [EFL] Move keyIdentifierForEvasKeyName() and windowsKeyCodeForEvasKeyName() to the
920         EflKeyboardUtilities.cpp to use in the WebKit2
921         https://bugs.webkit.org/show_bug.cgi?id=62451
922
923         Reviewed by Martin Robinson.
924
925         The keyIdentifierForEvasKeyName() and windowsKeyCodeForEvasKeyName() were static functions
926         in the PlatformKeyboardEventEfl.cpp. But they are also needed in the WebKit2 EFL port, so I
927         moved them to the separated file - EflKeyboardUtilities.cpp.
928
929         * PlatformEfl.cmake:
930         * platform/efl/EflKeyboardUtilities.cpp: Copied from Source/WebCore/platform/efl/PlatformKeyboardEventEfl.cpp.
931         (WebCore::createKeyMap):
932         (WebCore::createWindowsKeyMap):
933         (WebCore::keyIdentifierForEvasKeyName):
934         (WebCore::windowsKeyCodeForEvasKeyName):
935         * platform/efl/EflKeyboardUtilities.h: Added.
936         * platform/efl/PlatformKeyboardEventEfl.cpp:
937
938 2011-11-17  Martin Robinson  <mrobinson@igalia.com>
939
940         [GTK] The process freezes when you right click on windowless Flash
941         https://bugs.webkit.org/show_bug.cgi?id=69123
942
943         Reviewed by Xan Lopez.
944
945         No new tests. I tried to create a layout test that exercised this
946         issue, but it appears that EventSender clicks do not trigger
947         it. This is covered by the manual tests containing Flash.
948
949         * plugins/PluginPackage.cpp:
950         (WebCore::PluginPackage::determineQuirks): Always activate the
951         windowless Flash quirk if on x86_64 and X11.
952         * plugins/gtk/PluginViewGtk.cpp:
953         (WebCore::PluginView::handleMouseEvent): Avoid sending right-click
954         events if we have the quirk.
955
956 2011-11-17  Peter Rybin  <peter.rybin@gmail.com>
957
958         Web Inspector: clear fixme in generator script
959         https://bugs.webkit.org/show_bug.cgi?id=71372
960
961         Remove unnecessary field name map and update license year number.
962
963         Reviewed by Pavel Feldman.
964
965         * inspector/CodeGeneratorInspector.py:
966
967 2011-11-17  Raphael Kubo da Costa  <kubo@profusion.mobi>
968
969         [EFL] Clean up the use of DATA_DIR in the buildsystem
970         https://bugs.webkit.org/show_bug.cgi?id=72681
971
972         Reviewed by Daniel Bates.
973
974         Add the -DDATA_DIR definition here instead of defining it globally in
975         OptionsEfl.cmake, as WebCore is the only place which needs it.
976
977         No new tests, this is a buildsystem change.
978
979         * PlatformEfl.cmake:
980
981 2011-11-17  Adam Klein  <adamk@chromium.org>
982
983         Move JS recursion counter from V8Proxy to V8BindingPerIsolateData
984         https://bugs.webkit.org/show_bug.cgi?id=72645
985
986         Reviewed by Adam Barth.
987
988         With the JS recursion level stored as a member of V8Proxy, it's tied
989         to a frame. But this is incorrect, as there's no reason that a JS call
990         stack need be restricted to a single frame (see my new test case for
991         an example of code going across frames).
992
993         In order to get the correct accounting of JS recursion level, per-Isolate
994         is the right granularity (per dslomov), which is what this patch accomplishes.
995
996         Test: storage/indexeddb/transaction-abort-with-js-recursion-cross-frame.html
997
998         * bindings/v8/V8Binding.cpp:
999         (WebCore::V8BindingPerIsolateData::V8BindingPerIsolateData):
1000         * bindings/v8/V8Binding.h:
1001         (WebCore::V8BindingPerIsolateData::recursionLevel):
1002         (WebCore::V8BindingPerIsolateData::incrementRecursionLevel):
1003         (WebCore::V8BindingPerIsolateData::decrementRecursionLevel):
1004         (WebCore::V8RecursionScope::V8RecursionScope):
1005         (WebCore::V8RecursionScope::~V8RecursionScope):
1006         * bindings/v8/V8Proxy.cpp:
1007         (WebCore::incrementRecursionLevel):
1008         (WebCore::decrementRecursionLevel):
1009         (WebCore::recursionLevel):
1010         (WebCore::V8Proxy::V8Proxy):
1011         (WebCore::V8Proxy::runScript):
1012         (WebCore::V8Proxy::callFunction):
1013         (WebCore::V8Proxy::didLeaveScriptContext):
1014         * bindings/v8/V8Proxy.h:
1015
1016 2011-11-17  Robin Cao  <robin.cao@torchmobile.com.cn>
1017
1018         [chromium] Font::drawComplexText can not draw a segment of text run
1019         https://bugs.webkit.org/show_bug.cgi?id=72095
1020
1021         Reviewed by Adam Barth.
1022
1023         drawComplexText() should respect the 'from' and 'to' arguments.
1024         Drawing the whole text run may result in text overlapping.
1025
1026         Test: platform/chromium-linux/fast/text/international/draw-complex-text-from-to.html
1027
1028         * platform/graphics/chromium/ComplexTextControllerLinux.cpp:
1029         (WebCore::ComplexTextController::glyphsForRange):
1030         * platform/graphics/chromium/ComplexTextControllerLinux.h:
1031         * platform/graphics/chromium/FontLinux.cpp:
1032         (WebCore::Font::drawComplexText):
1033
1034 2011-11-17  Adam Barth  <abarth@webkit.org>
1035
1036         Unique SecurityOrigins shouldn't remember their old schemes and hosts
1037         https://bugs.webkit.org/show_bug.cgi?id=71745
1038
1039         Reviewed by Darin Adler.
1040
1041         This is the final step in this series of patches.
1042
1043         This patch removes the forceUnique flag from SecurityOrigin::create.
1044         Now, we create unique origins without passing in the document's URL,
1045         preventing information from the document's URL from leaking into the
1046         unique origin.
1047
1048         * WebCore.exp.in:
1049         * dom/Document.cpp:
1050         (WebCore::Document::setIsViewSource):
1051         (WebCore::Document::initSecurityContext):
1052         * loader/cache/MemoryCache.cpp:
1053         (WebCore::MemoryCache::getOriginsWithCache):
1054             - Update this callsite to use createFromString, which does exactly
1055               what this code is doing manually.
1056         * page/SecurityOrigin.cpp:
1057         (WebCore::SecurityOrigin::create):
1058         * page/SecurityOrigin.h:
1059
1060 2011-11-17  Vincent Scheib  <scheib@chromium.org>
1061
1062         Pointer Lock: Page Settings unecessary
1063         https://bugs.webkit.org/show_bug.cgi?id=72662
1064
1065         Reviewed by Darin Fisher.
1066
1067         No new tests.
1068
1069         * page/Settings.cpp:
1070         (WebCore::Settings::Settings):
1071         * page/Settings.h:
1072
1073 2011-11-17  James Robinson  <jamesr@chromium.org>
1074
1075         [chromium] Disable incremental uploading in threaded compositing path
1076         https://bugs.webkit.org/show_bug.cgi?id=72669
1077
1078         Reviewed by Kenneth Russell.
1079
1080         Since we don't currently support atomic incremental uploads, incremental uploads in the threaded path result in
1081         very strange-looking texture popping as tiles come in on pages that require more than 16 tile uploads per frame.
1082         This disables that logic (by setting the upload limit per frame to 99999) until we handle the incremental
1083         updates in an atomic fashion.
1084
1085         * platform/graphics/chromium/cc/CCThreadProxy.cpp:
1086         (WebCore::CCThreadProxy::scheduledActionUpdateMoreResources):
1087
1088 2011-11-06  Nat Duca  <nduca@chromium.org>
1089
1090         [chromium] Fix handling of setNeedsCommit and setNeedsAnimate in threaded mode
1091         https://bugs.webkit.org/show_bug.cgi?id=71638
1092
1093         Reviewed by James Robinson.
1094
1095         * platform/graphics/chromium/cc/CCThreadProxy.cpp:
1096         (WebCore::CCThreadProxy::CCThreadProxy):
1097         (WebCore::CCThreadProxy::setNeedsAnimate):
1098         (WebCore::CCThreadProxy::beginFrameAndCommit):
1099
1100 2011-11-17  Adam Barth  <abarth@webkit.org>
1101
1102         Remove bogus ASSERT.
1103
1104         * page/SecurityOrigin.cpp:
1105         (WebCore::SecurityOrigin::SecurityOrigin):
1106
1107 2011-11-17  Peter Kasting  <pkasting@google.com>
1108
1109         Unreviewed, rolling out r100698.
1110         http://trac.webkit.org/changeset/100698
1111         https://bugs.webkit.org/show_bug.cgi?id=72239
1112
1113         This change wasn't the problem either.
1114
1115         * dom/Document.cpp:
1116         (WebCore::Document::implicitClose):
1117
1118 2011-11-17  David Reveman  <reveman@chromium.org>
1119
1120         [Chromium] Calls to paintContentsIfDirty() and updateCompositorResources() should be balanced.
1121         https://bugs.webkit.org/show_bug.cgi?id=72630
1122
1123         Reviewed by James Robinson.
1124
1125         Layer property changes during paintContent() can leave the layer
1126         in an invalid state as paintContentsIfDirty() has been called
1127         without a matching updateCompositorResources() call. Removing
1128         conditionals around these calls ensure they are balanced.
1129
1130         This patch is tested by the following unit test:
1131         - CCLayerTreeHostTestOpacityChange.runMultiThread
1132
1133         * platform/graphics/chromium/ContentLayerChromium.cpp:
1134         (WebCore::ContentLayerChromium::paintContentsIfDirty):
1135         * platform/graphics/chromium/VideoLayerChromium.cpp:
1136         (WebCore::VideoLayerChromium::updateCompositorResources):
1137         * platform/graphics/chromium/cc/CCLayerTreeHost.cpp:
1138         (WebCore::CCLayerTreeHost::paintMaskAndReplicaForRenderSurface):
1139         (WebCore::CCLayerTreeHost::paintLayerContents):
1140         (WebCore::CCLayerTreeHost::updateCompositorResources):
1141         * platform/graphics/chromium/cc/CCLayerTreeHost.h:
1142
1143 2011-11-17  Peter Kasting  <pkasting@google.com>
1144
1145         Unreviewed, rolling out r100584.
1146         http://trac.webkit.org/changeset/100584
1147         https://bugs.webkit.org/show_bug.cgi?id=72239
1148
1149         See if this change caused Linux dbg crashes.
1150
1151         * dom/Document.cpp:
1152         (WebCore::Document::implicitClose):
1153
1154 2011-11-17  Adam Barth  <abarth@webkit.org>
1155
1156         Remove cargo-cult copy/pasting of ScriptExecutionContext namespace
1157         https://bugs.webkit.org/show_bug.cgi?id=72676
1158
1159         Reviewed by Eric Seidel.
1160
1161         It looks like this cargo-cult started with initDNSPrefetch being hacked
1162         into setSecurityOrigin.  I've removed this hack along with the
1163         copy/paste code.
1164
1165         * dom/DOMImplementation.cpp:
1166         * dom/Document.cpp:
1167         (WebCore::Document::setIsViewSource):
1168         (WebCore::Document::open):
1169         (WebCore::Document::initSecurityContext):
1170         (WebCore::Document::setSecurityOrigin):
1171         * dom/Document.h:
1172         * xml/XSLTProcessor.cpp:
1173
1174 2011-11-17  Peter Kasting  <pkasting@google.com>
1175
1176         Unreviewed, rolling out r100676.
1177         http://trac.webkit.org/changeset/100676
1178         https://bugs.webkit.org/show_bug.cgi?id=72393
1179
1180         Looks like r100572 was not the source of the crashes.
1181
1182         * bindings/js/ScriptDebugServer.cpp:
1183         (WebCore::ScriptDebugServer::canSetScriptSource):
1184         * bindings/js/ScriptDebugServer.h:
1185         * bindings/v8/ScriptDebugServer.cpp:
1186         (WebCore::ScriptDebugServer::stepOutOfFunction):
1187         (WebCore::ScriptDebugServer::canSetScriptSource):
1188         * bindings/v8/ScriptDebugServer.h:
1189         * inspector/CodeGeneratorInspector.py:
1190         * inspector/Inspector.json:
1191         * inspector/InspectorDebuggerAgent.cpp:
1192         (WebCore::InspectorDebuggerAgent::getCapabilities):
1193         * inspector/InspectorDebuggerAgent.h:
1194         * inspector/front-end/DebuggerModel.js:
1195         (WebInspector.DebuggerModel):
1196         (WebInspector.DebuggerModel.prototype.enableDebugger):
1197         (WebInspector.DebuggerModel.prototype.disableDebugger):
1198         (WebInspector.DebuggerModel.prototype.canSetScriptSource):
1199         * inspector/front-end/DebuggerPresentationModel.js:
1200         (WebInspector.DebuggerPresentationModel.prototype.canEditScriptSource):
1201         * inspector/front-end/Settings.js:
1202
1203 2011-11-17  Konstantin Scheglov  <scheglov@google.com>
1204
1205         Absolute child is not repainted when parent opacity changes
1206         https://bugs.webkit.org/show_bug.cgi?id=68777
1207
1208         Reviewed by Simon Fraser.
1209
1210         First time when we change opacity for parent we don't have layer, so
1211         diff=StyleDifferenceRepaint is used instead of diff=StyleDifferenceRepaintLayer.
1212         Layer is created later, in styleDidChange().
1213         So, when we recalculate later diff, we now check for diff=StyleDifferenceRepaintLayer and
1214         performs repaintIncludingDescendants().
1215
1216         Test: fast/layers/layer-absolute-parent-opacity.html
1217
1218         * rendering/RenderObject.cpp:
1219         (WebCore::RenderObject::setStyle): Call repaintIncludingDescendants() instead of repaint().
1220         * rendering/style/RenderStyle.cpp:
1221         (WebCore::RenderStyle::diff): Add ContextSensitivePropertyOpacity when change opacity.
1222
1223 2011-11-17  Nate Chapin  <japhet@chromium.org>
1224
1225         r100311 dropped a RefPtr that is very
1226         helpful. Add it back.
1227         https://bugs.webkit.org/show_bug.cgi?id=72647
1228
1229         Reviewed by Adam Barth.
1230
1231         http/tests/misc/xslt-bad-import.html should
1232         stop crashing in full chromium builds.
1233
1234         * loader/cache/CachedResourceLoader.cpp:
1235         (WebCore::CachedResourceLoader::loadDone):
1236
1237 2011-11-17  Adam Barth  <abarth@webkit.org>
1238
1239         Refactor SecurityOrigin::create to be easier to understand
1240         https://bugs.webkit.org/show_bug.cgi?id=72342
1241
1242         Reviewed by Eric Seidel.
1243
1244         Over time, the SecurityOrigin constructor has grown a bit out of
1245         control.  This patch attempts to separate the different concerns into
1246         free functions.  The general approach is to put more logic in the
1247         "create" function and introduce a simple constructor for unique
1248         origins.
1249
1250         This patch shouldn't change any behavior.
1251
1252         * page/SecurityOrigin.cpp:
1253         (WebCore::schemeRequiresAuthority):
1254         (WebCore::shouldUseInnerURL):
1255         (WebCore::extractInnerURL):
1256         (WebCore::isDirectory):
1257         (WebCore::shouldTreatAsUniqueOrigin):
1258         (WebCore::SecurityOrigin::SecurityOrigin):
1259         (WebCore::SecurityOrigin::create):
1260         (WebCore::SecurityOrigin::createUnique):
1261         (WebCore::SecurityOrigin::databaseIdentifier):
1262         * page/SecurityOrigin.h:
1263
1264 2011-11-17  Shawn Singh  <shawnsingh@chromium.org>
1265
1266         [chromium] Fix minor style nit in CCLayerImpl
1267         https://bugs.webkit.org/show_bug.cgi?id=71070
1268
1269         Reviewed by James Robinson.
1270
1271         Fixes if-statements on several settors to follow WebKit
1272         conventions. Existing CCLayerImplTest unit tests already cover
1273         this change.
1274
1275         * platform/graphics/chromium/cc/CCLayerImpl.cpp:
1276         (WebCore::CCLayerImpl::setBounds):
1277         (WebCore::CCLayerImpl::setMaskLayer):
1278         (WebCore::CCLayerImpl::setReplicaLayer):
1279         (WebCore::CCLayerImpl::setDrawsContent):
1280         (WebCore::CCLayerImpl::setAnchorPoint):
1281         (WebCore::CCLayerImpl::setAnchorPointZ):
1282         (WebCore::CCLayerImpl::setBackgroundColor):
1283         (WebCore::CCLayerImpl::setMasksToBounds):
1284         (WebCore::CCLayerImpl::setOpaque):
1285         (WebCore::CCLayerImpl::setOpacity):
1286         (WebCore::CCLayerImpl::setPosition):
1287         (WebCore::CCLayerImpl::setPreserves3D):
1288         (WebCore::CCLayerImpl::setZoomAnimatorTransform):
1289         (WebCore::CCLayerImpl::setSublayerTransform):
1290         (WebCore::CCLayerImpl::setTransform):
1291         (WebCore::CCLayerImpl::setDebugBorderColor):
1292         (WebCore::CCLayerImpl::setDebugBorderWidth):
1293         (WebCore::CCLayerImpl::setContentBounds):
1294         (WebCore::CCLayerImpl::setScrollPosition):
1295         (WebCore::CCLayerImpl::setScrollDelta):
1296         (WebCore::CCLayerImpl::setScaleDelta):
1297         (WebCore::CCLayerImpl::setDoubleSided):
1298
1299 2011-11-17  Michael Saboff  <msaboff@apple.com>
1300
1301         Leaks seen in MemoryPressureHandlerMac.mm on Leaks bot
1302         https://bugs.webkit.org/show_bug.cgi?id=72416
1303
1304         Added code to release _cache_event_source and _timer_event_source
1305         after they are canceled.  Also added defensive code to
1306         clean up the _timer_event_source in uninstall().
1307
1308         Reviewed by Geoffrey Garen.
1309
1310         No new tests, fixing leaks caught by leaks bot.
1311
1312         * platform/mac/MemoryPressureHandlerMac.mm:
1313         (WebCore::MemoryPressureHandler::uninstall):
1314         (WebCore::MemoryPressureHandler::holdOff):
1315
1316 2011-11-17  Ken Buchanan <kenrb@chromium.org>
1317
1318         Crash from positioned generated content under run-in
1319         https://bugs.webkit.org/show_bug.cgi?id=70456
1320
1321         Reviewed by David Hyatt.
1322
1323         Modified handling of run-in children to clear generated children
1324         before removing the parent from the render tree. This caused problems
1325         with absolute positioned children being not properly removed from the
1326         positioned object list of the RenderView.
1327
1328         * rendering/RenderBlock.cpp:
1329         (WebCore::RenderBlock::handleRunInChild):
1330
1331 2011-11-17  Peter Kasting  <pkasting@google.com>
1332
1333         Unreviewed, rolling out r100572.
1334         https://bugs.webkit.org/show_bug.cgi?id=72393
1335         https://bugs.webkit.org/show_bug.cgi?id=72651
1336
1337         May have caused seg faults on Chromium Linux dbg bot.
1338
1339         * bindings/js/ScriptDebugServer.cpp:
1340         * bindings/js/ScriptDebugServer.h:
1341         * bindings/v8/ScriptDebugServer.cpp:
1342         (WebCore::ScriptDebugServer::stepOutOfFunction):
1343         * bindings/v8/ScriptDebugServer.h:
1344         * inspector/CodeGeneratorInspector.py:
1345         * inspector/Inspector.json:
1346         * inspector/InspectorDebuggerAgent.cpp:
1347         * inspector/InspectorDebuggerAgent.h:
1348         * inspector/front-end/DebuggerModel.js:
1349         (WebInspector.DebuggerModel):
1350         (WebInspector.DebuggerModel.prototype.enableDebugger):
1351         (WebInspector.DebuggerModel.prototype.disableDebugger):
1352         * inspector/front-end/DebuggerPresentationModel.js:
1353         (WebInspector.DebuggerPresentationModel.prototype.canEditScriptSource):
1354         * inspector/front-end/Settings.js:
1355
1356 2011-11-17  Sheriff Bot  <webkit.review.bot@gmail.com>
1357
1358         Unreviewed, rolling out r100652.
1359         http://trac.webkit.org/changeset/100652
1360         https://bugs.webkit.org/show_bug.cgi?id=72648
1361
1362         "Caused outline-offset-min-assert.html to assert on debug
1363         builds" (Requested by mwenge2 on #webkit).
1364
1365         * rendering/RenderObject.cpp:
1366         (WebCore::RenderObject::paintOutline):
1367
1368 2011-11-17  Mark Rowe  <mrowe@apple.com>
1369
1370         <http://webkit.org/b/72637> Stop performing runtime version checks on OS versions where the check can never possibly fail.
1371
1372         Reviewed by Simon Fraser.
1373
1374         * platform/graphics/cg/ImageBufferDataCG.cpp:
1375         (WebCore::haveVImageRoundingErrorFix): When not targeting Snow Leopard we always have the fix.
1376
1377 2011-11-17  Chris Fleizach  <cfleizach@apple.com>
1378
1379         AX: The scrollArea is not correctly returning the scrollbars
1380         https://bugs.webkit.org/show_bug.cgi?id=70247
1381
1382         Reviewed by Beth Dakin.
1383
1384         A few bugs that were preventing this from working.
1385            1) In accessibilityAttributeValue: when an element did not have a renderer() we were returning earlier than we should have
1386            2) We were not updating and clearing the scrollbars correctly when children were cleared or when asked for.
1387
1388         Test: platform/mac/accessibility/scrollbars.html
1389
1390         * accessibility/AccessibilityScrollView.cpp:
1391         (WebCore::AccessibilityScrollView::scrollBar):
1392         (WebCore::AccessibilityScrollView::clearChildren):
1393         * accessibility/AccessibilityScrollView.h:
1394         * accessibility/mac/WebAccessibilityObjectWrapper.mm:
1395         (-[WebAccessibilityObjectWrapper accessibilityAttributeValue:]):
1396
1397 2011-11-14  Adam Barth  <abarth@webkit.org>
1398
1399         Unique origins shouldn't remember their scheme, host, or port
1400         https://bugs.webkit.org/show_bug.cgi?id=72308
1401
1402         Reviewed by Eric Seidel.
1403
1404         This patch contains the bulk (all?) of the behavior differences in this
1405         patch series.  Unique origins shouldn't remember their schemes.  Doing
1406         so causes some privileges (e.g., local access) to leak into unique
1407         origins.
1408
1409         * page/SecurityOrigin.cpp:
1410         (WebCore::SecurityOrigin::SecurityOrigin):
1411             - Explicitly clear out the protocol, host, and port for unique
1412               origins.  A future patch will refactor all this code to be more
1413               elegant.
1414         * platform/SchemeRegistry.cpp:
1415         (WebCore::schemesWithUniqueOrigins):
1416             - Merge "about" and "javascript" in with the general case now that
1417               we don't have a separate notion of an empty origin.
1418
1419 2011-11-17  Chris Fleizach  <cfleizach@apple.com>
1420
1421         AX: cleanup style and naming and code in accessibility search mechanism
1422         https://bugs.webkit.org/show_bug.cgi?id=72570
1423
1424         Reviewed by Beth Dakin.
1425
1426         Cleanup the naming and code style within the element searching mechanism.
1427
1428         * accessibility/AccessibilityObject.cpp:
1429         (WebCore::AccessibilityObject::isAccessibilityObjectSearchMatch):
1430         (WebCore::AccessibilityObject::isAccessibilityTextSearchMatch):
1431         (WebCore::AccessibilityObject::firstAccessibleObjectFromNode):
1432         (WebCore::AccessibilityObject::findMatchingObjects):
1433         * accessibility/AccessibilityObject.h:
1434         * accessibility/mac/WebAccessibilityObjectWrapper.mm:
1435         (-[WebAccessibilityObjectWrapper accessibilityAttributeValue:forParameter:]):
1436
1437 2011-11-17  Julien Chaffraix  <jchaffraix@webkit.org>
1438
1439         CSS table with 100% width can overflow their containing block
1440         https://bugs.webkit.org/show_bug.cgi?id=72180
1441
1442         Reviewed by David Hyatt.
1443
1444         Tests: fast/table/table-in-table-percent-width-collapsing-border-quirks-mode.html
1445                fast/table/table-in-table-percent-width-collapsing-border.html
1446                fast/table/table-in-table-percent-width-quirks-mode.html
1447                fast/table/table-in-table-percent-width.html
1448
1449         After r97555, we would include the borders in a CSS table's logical width even if the 'width'
1450         property was a percent. This does not match what Firefox and IE are doing. Thus don't apply
1451         this behavior to percent 'width'.
1452
1453         * rendering/RenderTable.cpp:
1454         (WebCore::RenderTable::computeLogicalWidth):
1455
1456 2011-11-17  Adrienne Walker  <enne@google.com>
1457
1458         [chromium] Implicitly skip render surfaces that won't be drawn
1459         https://bugs.webkit.org/show_bug.cgi?id=71038
1460
1461         Rather than having redundant checks in three places for how to walk
1462         through a render surface list, instead don't add render surfaces that
1463         don't need to get rendered to the render surface list.
1464
1465         Reviewed by James Robinson.
1466
1467         Covered by existing layout tests and unit tests.
1468
1469         * platform/graphics/chromium/LayerRendererChromium.cpp:
1470         (WebCore::LayerRendererChromium::drawLayersOntoRenderSurfaces):
1471         * platform/graphics/chromium/cc/CCLayerTreeHost.cpp:
1472         (WebCore::CCLayerTreeHost::paintLayerContents):
1473         (WebCore::CCLayerTreeHost::updateCompositorResources):
1474         * platform/graphics/chromium/cc/CCLayerTreeHostCommon.cpp:
1475         (WebCore::calculateDrawTransformsAndVisibilityInternal):
1476
1477 2011-11-17  Simon Hausmann  <simon.hausmann@nokia.com>
1478
1479         [Qt] Layer violation: qt_runtime.cpp accesses QWebElement and QTDRTNode
1480         https://bugs.webkit.org/show_bug.cgi?id=72595
1481
1482         Reviewed by Noam Rosenthal.
1483
1484         Removed QWebElement and QtDRTNode usage that reached from WebCore into
1485         WebKit/qt and replaced it with the ability to register custom JSValue
1486         conversion functions. The old code has been moved to WebKit/qt.
1487
1488         * bridge/qt/qt_instance.cpp:
1489         (JSC::Bindings::QtInstance::QtInstance): Remove unnecessary meta type registration
1490         (now done in QtWebElementRuntime::initialize in WebKit/qt).
1491         * bridge/qt/qt_runtime.cpp:
1492         (JSC::Bindings::registerCustomType):
1493         (JSC::Bindings::convertValueToQVariant):
1494         (JSC::Bindings::convertQVariantToValue):
1495         * bridge/qt/qt_runtime.h:
1496
1497 2011-11-17  Fady Samuel  <fsamuel@chromium.org>
1498
1499         Pass Aspect Ratio to RenderStyle
1500         https://bugs.webkit.org/show_bug.cgi?id=72350
1501
1502         Reviewed by Ojan Vafai.
1503
1504         * css/CSSComputedStyleDeclaration.cpp:
1505         (WebCore::CSSComputedStyleDeclaration::getPropertyCSSValue):
1506         * css/CSSStyleApplyProperty.cpp:
1507         (WebCore::ApplyPropertyAspectRatio::applyInheritValue):
1508         (WebCore::ApplyPropertyAspectRatio::applyInitialValue):
1509         (WebCore::ApplyPropertyAspectRatio::applyValue):
1510         (WebCore::ApplyPropertyAspectRatio::createHandler):
1511         (WebCore::CSSStyleApplyProperty::CSSStyleApplyProperty):
1512         * css/CSSStyleSelector.cpp:
1513         (WebCore::CSSStyleSelector::applyProperty):
1514         * css/CSSValue.h:
1515         (WebCore::CSSValue::isAspectRatioValue):
1516         * rendering/style/RenderStyle.h:
1517         (WebCore::InheritedFlags::hasAspectRatio):
1518         (WebCore::InheritedFlags::aspectRatio):
1519         (WebCore::InheritedFlags::aspectRatioDenominator):
1520         (WebCore::InheritedFlags::aspectRatioNumerator):
1521         (WebCore::InheritedFlags::setHasAspectRatio):
1522         (WebCore::InheritedFlags::setAspectRatioDenominator):
1523         (WebCore::InheritedFlags::setAspectRatioNumerator):
1524         (WebCore::InheritedFlags::initialHasAspectRatio):
1525         (WebCore::InheritedFlags::initialAspectRatioDenominator):
1526         (WebCore::InheritedFlags::initialAspectRatioNumerator):
1527         * rendering/style/StyleRareNonInheritedData.cpp:
1528         (WebCore::StyleRareNonInheritedData::StyleRareNonInheritedData):
1529         (WebCore::StyleRareNonInheritedData::operator==):
1530         * rendering/style/StyleRareNonInheritedData.h:
1531
1532 2011-11-09  Robert Hogan  <robert@webkit.org>
1533
1534         CSS 2.1 failure: outline-color-* tests fail
1535         https://bugs.webkit.org/show_bug.cgi?id=71931
1536
1537         Reviewed by Julien Chaffraix.
1538
1539         WebKit wasn't displaying the top block in these tests because it did not paint the outline
1540         of divs with zero size.
1541
1542         * rendering/RenderObject.cpp:
1543         (WebCore::RenderObject::paintOutline): paint the outline even when the block has zero size
1544
1545 2011-11-17  Sergio Villar Senin  <svillar@igalia.com>
1546
1547         [GTK] plugins/get-url-notify-with-url-that-fails-to-load.html on bots after r100466
1548         https://bugs.webkit.org/show_bug.cgi?id=72613
1549
1550         Reviewed by Martin Robinson.
1551
1552         Do not assume that a SoupRequest always exists, it is not
1553         generated for example when the provided URL is
1554         invalid. ResourceHandle::platformSetDefersLoading was crashing
1555         because of that.
1556
1557         * platform/network/soup/ResourceHandleSoup.cpp:
1558         (WebCore::ResourceHandle::platformSetDefersLoading):
1559
1560 2011-11-17  Vsevolod Vlasov  <vsevik@chromium.org>
1561
1562         Web Inspector: Remove onlineDetectionEnabled from Preferences
1563         https://bugs.webkit.org/show_bug.cgi?id=72617
1564
1565         Reviewed by Pavel Feldman.
1566
1567         * inspector/front-end/ApplicationCacheItemsView.js:
1568         (WebInspector.ApplicationCacheItemsView):
1569         (WebInspector.ApplicationCacheItemsView.prototype.get statusBarItems):
1570         * inspector/front-end/Settings.js:
1571
1572 2011-11-17  Ken Buchanan  <kenrb@chromium.org>
1573
1574         Crash from nested tables with generated content
1575         https://bugs.webkit.org/show_bug.cgi?id=68811
1576
1577         Reviewed by David Hyatt.
1578
1579         When adding a child to a table that has generated content, this change
1580         ensures that we leave alone any generated content renderers that belong
1581         to descendants in the tree. They don't need to be touched, and doing so
1582         can create confusion about who the content belongs to.
1583
1584         This patch also simplifies some existing code for finding pseudoelement
1585         renderers. 
1586
1587         * rendering/RenderBlock.cpp:
1588         (WebCore::RenderBlock::addChildIgnoringAnonymousColumnBlocks):
1589         * rendering/RenderObject.cpp:
1590         (WebCore::RenderObject::addChild):
1591         (WebCore::RenderObject::isBeforeAfterContentGeneratedByAncestor): Added
1592         * rendering/RenderObject.h:
1593         (WebCore::RenderObject::findAfterContentRenderer): Deleted
1594         (WebCore::RenderObject::findBeforeContentRenderer): Deleted
1595         * rendering/RenderObjectChildList.cpp:
1596         (WebCore::RenderObjectChildList::beforePseudoElementRenderer):
1597         (WebCore::RenderObjectChildList::afterPseudoElementRenderer):
1598         * rendering/RenderTable.cpp:
1599         (WebCore::RenderTable::addChild):
1600         * rendering/RenderTableRow.cpp:
1601         (WebCore::RenderTableRow::addChild):
1602         * rendering/RenderTableSection.cpp:
1603         (WebCore::RenderTableSection::addChild):
1604
1605 2011-11-17  Patrick Gansterer  <paroga@webkit.org>
1606
1607         Unreviewed WinCE build fix for r94119.
1608
1609         MSVC throws multiply defined symbols linker error when using local class in inline function.
1610
1611         * bindings/js/JSDictionary.h:
1612         (WebCore::JSDictionary::IdentitySetter::identitySetter):
1613         (WebCore::JSDictionary::tryGetProperty):
1614
1615 2011-11-17  Eric Carlson  <eric.carlson@apple.com>
1616
1617         TextTrackList not sorted correctly
1618         https://bugs.webkit.org/show_bug.cgi?id=72545
1619
1620         Reviewed by Darin Adler.
1621
1622         Test: media/track/track-texttracks.html
1623
1624         * WebCore.xcodeproj/project.pbxproj: Add TextTrack.h to WebCore private headers because 
1625             it is included by HTMLMediaElement.h.
1626         * html/LoadableTextTrack.cpp:
1627         (WebCore::LoadableTextTrack::LoadableTextTrack): Pass track type to base class constructor.
1628         (WebCore::LoadableTextTrack::trackElementIndex): New, return the <track> element's tree order
1629             for sorting.
1630         * html/LoadableTextTrack.h:
1631
1632         * html/TextTrack.cpp:
1633         (WebCore::TextTrack::TextTrack): Set track type.
1634         * html/TextTrack.h:
1635         (WebCore::TextTrack::create): Ditto.
1636         (WebCore::TextTrack::trackType): Ditto.
1637
1638         * html/track/TextTrackList.cpp:
1639         (TextTrackList::length): Update to deal with two TextTrack vectors.
1640         (TextTrackList::item): Ditto.
1641         (TextTrackList::append): Ditto.
1642         (TextTrackList::remove): Ditto
1643         * html/track/TextTrackList.h: Store the two types of TextTracks in separate Vectors to make
1644             it simpler to keep them in the correct order.
1645
1646 2011-11-17  Simon Hausmann  <simon.hausmann@nokia.com>
1647
1648         [Qt] Layer violation: WebCore::dnsPrefetch uses QWebSettings::globalSettings()
1649         https://bugs.webkit.org/show_bug.cgi?id=72596
1650
1651         Reviewed by Kenneth Rohde Christiansen.
1652
1653         We don't need to use QWebSettings here, because we now propagate the DNS prefetch
1654         setting to WebCore::Settings and the rest of WebCore checks the setting before calling
1655         WebCore::prefetchDNS.
1656
1657         * platform/network/qt/DnsPrefetchHelper.cpp:
1658         (WebCore::prefetchDNS): Removed the use of QWebSettings.
1659         * platform/network/qt/DnsPrefetchHelper.h: Remove offending qwebsettings.h inclusion.
1660
1661 2011-11-17  Simon Hausmann  <simon.hausmann@nokia.com>
1662
1663         [Qt] Layer violation: Image::loadPlatformResource uses QWebSettings::webGraphic
1664         https://bugs.webkit.org/show_bug.cgi?id=72594
1665
1666         Reviewed by Kenneth Rohde Christiansen.
1667
1668         Move the cache for the resource pixmaps into ImageQt.cpp.
1669
1670         * platform/graphics/Image.h: Add Qt specific setter for resource pixmaps.
1671         * platform/graphics/qt/ImageQt.cpp: Moved resource pixmap hash from qwebsettings.
1672         (earlyClearGraphics):
1673         (graphics):
1674         (loadResourcePixmap):
1675         (WebCore::Image::setPlatformResource):
1676
1677 2011-11-17  Zeno Albisser  <zeno@webkit.org>
1678
1679         [Qt][WK2] Touch/Mouse events are delivered with wrong coordinates.
1680         https://bugs.webkit.org/show_bug.cgi?id=72604
1681
1682         When using the QtViewportInterationEngine for zooming/panning,
1683         no additional scroll offset should be applied to input events
1684         by the ScrollView.
1685
1686         This patch is based on work by Andreas Kling.
1687
1688         Reviewed by Kenneth Rohde Christiansen.
1689
1690         * platform/ScrollView.cpp:
1691         (WebCore::ScrollView::windowToContents):
1692         (WebCore::ScrollView::contentsToWindow):
1693
1694 2011-11-17  Ben Murdoch  <benm@google.com>
1695
1696         Unreviewed build fix.
1697
1698         Fix the Windows builds by adding WorkerEventQueue.cpp|h to the
1699         vcproj. Build break was introduced in
1700         https://bugs.webkit.org/show_bug.cgi?id=72528
1701
1702         * WebCore.vcproj/WebCore.vcproj: Add missing files.
1703
1704 2011-11-17  Mihnea Ovidenie  <mihnea@adobe.com>
1705
1706         CSS Exclusions: parse the wrap-margin and wrap-padding properties
1707         https://bugs.webkit.org/show_bug.cgi?id=71900
1708
1709         Reviewed by Dean Jackson.
1710
1711         Tests: fast/exclusions/wrap-margin-parsing.html
1712                fast/exclusions/wrap-padding-parsing.html
1713
1714         * css/CSSComputedStyleDeclaration.cpp:
1715         (WebCore::CSSComputedStyleDeclaration::getPropertyCSSValue):
1716         * css/CSSParser.cpp:
1717         (WebCore::isSimpleLengthPropertyID):
1718         (WebCore::CSSParser::parseValue):
1719         * css/CSSProperty.cpp:
1720         (WebCore::CSSProperty::isInheritedProperty):
1721         * css/CSSPropertyNames.in:
1722         * css/CSSStyleApplyProperty.cpp:
1723         (WebCore::CSSStyleApplyProperty::CSSStyleApplyProperty):
1724         * css/CSSStyleSelector.cpp:
1725         (WebCore::CSSStyleSelector::applyProperty):
1726         * rendering/style/RenderStyle.cpp:
1727         (WebCore::RenderStyle::diff):
1728         * rendering/style/RenderStyle.h:
1729         (WebCore::InheritedFlags::wrapPadding):
1730         (WebCore::InheritedFlags::setWrapPadding):
1731         (WebCore::InheritedFlags::initialWrapPadding):
1732         (WebCore::InheritedFlags::wrapMargin):
1733         (WebCore::InheritedFlags::setWrapMargin):
1734         (WebCore::InheritedFlags::initialWrapMargin):
1735         * rendering/style/StyleRareNonInheritedData.cpp:
1736         (WebCore::StyleRareNonInheritedData::StyleRareNonInheritedData):
1737         (WebCore::StyleRareNonInheritedData::operator==):
1738         * rendering/style/StyleRareNonInheritedData.h:
1739
1740 2011-11-17  Kenichi Ishibashi  <bashi@chromium.org>
1741
1742         [chromium] don't call fontconfig twice in complex text path
1743         https://bugs.webkit.org/show_bug.cgi?id=38701
1744
1745         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.
1746
1747         Reviewed by Tony Chang.
1748
1749         No new tests. No behavior changes for now.
1750
1751         * platform/chromium/PlatformSupport.h: Added FontFamily struct and changed the declaration of getFontFamilyForCharacters().
1752         * platform/graphics/chromium/FontCacheLinux.cpp:
1753         (WebCore::FontCache::getFontDataForCharacters): Uses new PlatformSupport::getFontFamilyForCharacters().
1754         * platform/graphics/chromium/FontPlatformDataLinux.h:
1755         (WebCore::FontPlatformData::setFakeBold): Added.
1756         (WebCore::FontPlatformData::setFakeItalic): Added.
1757
1758 2011-11-17  Mario Sanchez Prada  <msanchez@igalia.com>
1759
1760         [GTK] Consider parent AtkObject in webkit_accessible_get_parent(), if already set
1761         https://bugs.webkit.org/show_bug.cgi?id=72525
1762
1763         Reviewed by Xan Lopez.
1764
1765         * accessibility/gtk/AccessibilityObjectWrapperAtk.cpp:
1766         (webkit_accessible_get_parent): Call to the implementation of
1767         atk_object_get_parent in AtkObject class to check whether a parent
1768         AtkObject has been previously set, before trying to find one.
1769
1770 2011-11-17  Pavel Feldman  <pfeldman@google.com>
1771
1772         Not reviewed: fix IE user agents strings in the inspector.
1773
1774         * inspector/front-end/SettingsScreen.js:
1775         (WebInspector.SettingsScreen.prototype._createUserAgentSelectRowElement.get const):
1776
1777 2011-11-16  Kenneth Rohde Christiansen  <kenneth@webkit.org>
1778
1779         Make use-fixed-layout work reliable
1780         https://bugs.webkit.org/show_bug.cgi?id=72511
1781
1782         Reviewed by Simon Hausmann.
1783
1784         Always send a viewport update per page load as we depend on that,
1785         to reset all viewport handling before doing layout.
1786
1787         * page/Page.cpp:
1788         (WebCore::Page::updateViewportArguments):
1789
1790 2011-11-16  Alexander Pavlov  <apavlov@chromium.org>
1791
1792         Web Inspector: inspector follows javascript: hrefs as relative
1793         https://bugs.webkit.org/show_bug.cgi?id=72373
1794
1795         javascript: hrefs should never be linkified for security.
1796
1797         Reviewed by Yury Semikhatsky.
1798
1799         * inspector/front-end/ElementsTreeOutline.js:
1800         (WebInspector.ElementsTreeElement.prototype._buildAttributeDOM):
1801         * inspector/front-end/ResourceUtils.js:
1802         (WebInspector.completeURL):
1803
1804 2011-11-17  Nikolas Zimmermann  <nzimmermann@rim.com>
1805
1806         Not reviewed. Fix 32bit builds.
1807
1808         * platform/ClockGeneric.cpp:
1809         (ClockGeneric::now): Use narrowPrecisionToFloat.
1810         * rendering/FilterEffectRenderer.cpp:
1811         (WebCore::FilterEffectRenderer::build): Use it correctly.
1812
1813 2011-11-17  Dominic Mazzoni  <dmazzoni@google.com>
1814
1815         Accessibility: Chromium requires an AX notification when an iframe loads.
1816         https://bugs.webkit.org/show_bug.cgi?id=72239
1817
1818         When a document finishes loading, we were sending an AXLoadComplete if it
1819         was the top document. Now, if it's a document in an iframe, send an
1820         AXLayoutComplete on the iframe.
1821
1822         Reviewed by Chris Fleizach.
1823
1824         Test: accessibility/loading-iframe-sends-notification.html
1825
1826         * dom/Document.cpp:
1827         (WebCore::Document::implicitClose):
1828
1829 2011-11-16  Mark Rowe  <mrowe@apple.com>
1830
1831         <http://webkit.org/b/72574> Remove unnecessary use of CarbonCore APIs from Audio code
1832
1833         Reviewed by Andy Estes.
1834
1835         * platform/audio/mac/AudioDestinationMac.cpp:
1836         (WebCore::AudioDestinationMac::AudioDestinationMac): Switch from using the Carbon Component Manager
1837         to using AudioUnit's own component interface.
1838         (WebCore::AudioDestinationMac::~AudioDestinationMac): Ditto.
1839         * platform/audio/mac/AudioFileReaderMac.cpp:
1840         (WebCore::AudioFileReader::AudioFileReader): Remove an unncessary trip through the Carbon File Manager
1841         when converting a char* path to a CFURLRef representing the same.
1842
1843 2011-11-17  Adam Barth  <abarth@webkit.org>
1844
1845         CSP report-only mode doesn't work from an HTTP header
1846         https://bugs.webkit.org/show_bug.cgi?id=71958
1847
1848         Reviewed by Eric Seidel.
1849
1850         "It's tested or it's broken." -- Adam Leventhal
1851
1852         Test: http/tests/security/contentSecurityPolicy/report-only-from-header.php
1853
1854         * loader/FrameLoader.cpp:
1855         (WebCore::FrameLoader::didBeginDocument):
1856
1857 2011-11-15  Pavel Feldman  <pfeldman@google.com>
1858
1859         Web Inspector: introduce Debugger domain capabilities concept.
1860         https://bugs.webkit.org/show_bug.cgi?id=72393
1861
1862         Many of the Preferences that we have in Settings.js and override in DevTools.js
1863         are really not preferences, but capabilities. Protocol clients should have a way
1864         of figuring out whether some capability is present before using it.
1865
1866         Reviewed by Yury Semikhatsky.
1867
1868         * bindings/js/ScriptDebugServer.cpp:
1869         (WebCore::ScriptDebugServer::canSetScriptSource):
1870         * bindings/js/ScriptDebugServer.h:
1871         * bindings/v8/ScriptDebugServer.cpp:
1872         (WebCore::ScriptDebugServer::canSetScriptSource):
1873         * bindings/v8/ScriptDebugServer.h:
1874         * inspector/CodeGeneratorInspector.py:
1875         * inspector/Inspector.json:
1876         * inspector/InspectorDebuggerAgent.cpp:
1877         (WebCore::InspectorDebuggerAgent::getCapabilities):
1878         * inspector/InspectorDebuggerAgent.h:
1879         * inspector/front-end/DebuggerModel.js:
1880         (WebInspector.DebuggerModel):
1881         (WebInspector.DebuggerModel.prototype.enableDebugger):
1882         (WebInspector.DebuggerModel.prototype.canSetScriptSource):
1883         * inspector/front-end/DebuggerPresentationModel.js:
1884         (WebInspector.DebuggerPresentationModel.prototype.canEditScriptSource):
1885         * inspector/front-end/Settings.js:
1886
1887 2011-11-17  Mihnea Ovidenie  <mihnea@adobe.com>
1888
1889         CSS exclusions: parse the wrap-flow and wrap-through properties
1890         https://bugs.webkit.org/show_bug.cgi?id=71904
1891
1892         Reviewed by Dean Jackson.
1893
1894         Tests: fast/exclusions/wrap-flow-parsing.html
1895                fast/exclusions/wrap-through-parsing.html
1896
1897         * css/CSSComputedStyleDeclaration.cpp:
1898         (WebCore::CSSComputedStyleDeclaration::getPropertyCSSValue):
1899         * css/CSSParser.cpp:
1900         (WebCore::CSSParser::parseValue):
1901         * css/CSSPrimitiveValueMappings.h:
1902         (WebCore::CSSPrimitiveValue::CSSPrimitiveValue):
1903         (WebCore::CSSPrimitiveValue::operator WrapFlow):
1904         (WebCore::CSSPrimitiveValue::operator WrapThrough):
1905         * css/CSSProperty.cpp:
1906         (WebCore::CSSProperty::isInheritedProperty):
1907         * css/CSSPropertyNames.in:
1908         * css/CSSStyleSelector.cpp:
1909         (WebCore::CSSStyleSelector::applyProperty):
1910         * css/CSSValueKeywords.in:
1911         * rendering/style/RenderStyle.cpp:
1912         (WebCore::RenderStyle::diff):
1913         * rendering/style/RenderStyle.h:
1914         (WebCore::InheritedFlags::wrapFlow):
1915         (WebCore::InheritedFlags::wrapThrough):
1916         (WebCore::InheritedFlags::setWrapFlow):
1917         (WebCore::InheritedFlags::setWrapThrough):
1918         (WebCore::InheritedFlags::initialWrapFlow):
1919         (WebCore::InheritedFlags::initialWrapThrough):
1920         * rendering/style/RenderStyleConstants.h:
1921         * rendering/style/StyleRareNonInheritedData.cpp:
1922         (WebCore::StyleRareNonInheritedData::StyleRareNonInheritedData):
1923         (WebCore::StyleRareNonInheritedData::operator==):
1924         * rendering/style/StyleRareNonInheritedData.h:
1925
1926 2011-11-16  Dean Jackson  <dino@apple.com>
1927
1928         Implement filter function shorthands
1929         https://bugs.webkit.org/show_bug.cgi?id=68475
1930
1931         Reviewed by Simon Fraser.
1932
1933         Implement the shorthand functions for filter effects.
1934         This includes grayscale, sepia, invert, hue-rotate, saturate,
1935         opacity, gamma, drop-shadow and blur. At the moment sharpen
1936         and url are not supported.
1937
1938         CSSParser needed to be updated because it was mistakenly
1939         clamping saturation values to [0,1]. Any positive number
1940         is allowed so you can produce super-saturated images.
1941
1942         The biggest change was the API to FilterEffectRenderer. It now
1943         builds a list of effects and applies the filter itself.
1944
1945         Note that the drop-shadow and blur operations don't yet
1946         provide accurate results because they produce an output image
1947         that is larger than the input. See
1948         https://bugs.webkit.org/show_bug.cgi?id=71929
1949         https://bugs.webkit.org/show_bug.cgi?id=71930
1950
1951         While I was there, I fixed a small style issue in
1952         CustomFilterOperation.
1953
1954         Tests: css3/filters/effect-blur.html
1955                css3/filters/effect-combined.html
1956                css3/filters/effect-drop-shadow.html
1957                css3/filters/effect-gamma.html
1958                css3/filters/effect-grayscale.html
1959                css3/filters/effect-hue-rotate.html
1960                css3/filters/effect-invert.html
1961                css3/filters/effect-opacity.html
1962                css3/filters/effect-saturate.html
1963                css3/filters/effect-sepia.html
1964
1965         * WebCore.xcodeproj/project.pbxproj: Add StyleShader.h to
1966         the project (missing from earlier commit).
1967         * css/CSSParser.cpp:
1968         (WebCore::CSSParser::isValidFilterArgument): Don't clamp
1969         saturate to [0,1]
1970         * platform/graphics/filters/CustomFilterOperation.h:
1971         * rendering/FilterEffectRenderer.cpp:
1972         (WebCore::endMatrixRow):
1973         (WebCore::lastMatrixRow):
1974         (WebCore::FilterEffectRenderer::FilterEffectRenderer):
1975         (WebCore::FilterEffectRenderer::inputContext):
1976         (WebCore::FilterEffectRenderer::build):
1977         (WebCore::FilterEffectRenderer::prepare):
1978         (WebCore::FilterEffectRenderer::apply):
1979         * rendering/FilterEffectRenderer.h:
1980         (WebCore::FilterEffectRenderer::setSourceImageRect):
1981         (WebCore::FilterEffectRenderer::output):
1982         (WebCore::FilterEffectRenderer::setMaxEffectRects):
1983         (WebCore::FilterEffectRenderer::lastEffect):
1984         * rendering/RenderLayer.cpp:
1985         (WebCore::RenderLayer::paintLayer):
1986         (WebCore::RenderLayer::updateOrRemoveFilterEffect):
1987         (WebCore::RenderLayer::updateFilterBackingStore):
1988
1989 2011-11-16  Kentaro Hara  <haraken@chromium.org>
1990
1991         Remove all custom constructors of Events from JSC
1992         https://bugs.webkit.org/show_bug.cgi?id=72577
1993
1994         Reviewed by Adam Barth.
1995
1996         - Makes CodeGeneratorJS.pm generate Event constructors
1997         if [ConstructorTemplate=Event] IDL is specified.
1998         - Removes EventConstructors.h and JSEventConstructors.cpp.
1999         - Replaces all JSC custom constructors of Events
2000         with the generated code by [ConstructorTemplate=Event] IDL.
2001
2002         Tests: fast/events/constructors/before-load-event-constructor.html
2003                fast/events/constructors/close-event-constructor.html
2004                fast/events/constructors/custom-event-constructor.html
2005                fast/events/constructors/error-event-constructor.html
2006                fast/events/constructors/event-constructors.html
2007                fast/events/constructors/hash-change-event-constructor.html
2008                fast/events/constructors/message-event-constructor.html
2009                fast/events/constructors/overflow-event-constructor.html
2010                fast/events/constructors/page-transition-event-constructor.html
2011                fast/events/constructors/pop-state-event-constructor.html
2012                fast/events/constructors/progress-event-constructor.html
2013                fast/events/constructors/track-event-constructor.html
2014                fast/events/constructors/webkit-animation-event-constructor.html
2015                fast/events/constructors/webkit-transition-event-constructor.html
2016
2017         * bindings/generic/EventConstructors.h: Removed. This is what we wanted to do in this patch.
2018         * bindings/js/JSEventConstructors.cpp: Ditto.
2019         * GNUmakefile.list.am: Removed EventConstructors.h and JSEventConstructors.cpp.
2020         * Target.pri: Ditto.
2021         * UseJSC.cmake: Ditto.
2022         * WebCore.gypi: Ditto.
2023         * WebCore.vcproj/WebCore.vcproj: Ditto.
2024         * WebCore.xcodeproj/project.pbxproj: Ditto.
2025         * bindings/js/JSBindingsAllInOne.cpp: Ditto.
2026
2027         * bindings/scripts/CodeGeneratorJS.pm:
2028         (GenerateHeader): Added JSDictionary.h.
2029         (GenerateConstructorDeclaration): Added a header for fillXXXXEventInit(...).
2030         (GenerateConstructorDefinition): Generates an Event constructor. The generated code is the same as the code that had been written in JSEventConstructors.cpp.
2031         (IsConstructable): Judges if a given interface is constructable.
2032         (IsConstructorTemplate): Judges if a given interface has a given template, e.g. judges if an interface has [ConstructorTemplate=Event].
2033
2034         * bindings/scripts/test/TestEventConstructor.idl: Changed 'CustomConstructor=Event' to 'ConstructorTemplate=Event'. We should have changed this in r100108.
2035         * bindings/scripts/test/JS/JSTestEventConstructor.cpp: Updated a run-bindings-tests result.
2036         (WebCore::JSTestEventConstructorConstructor::constructJSTestEventConstructor):
2037         (WebCore::fillTestEventConstructorInit):
2038         * bindings/scripts/test/JS/JSTestEventConstructor.h: Ditto.
2039         * bindings/scripts/test/V8/V8TestEventConstructor.cpp: Ditto.
2040         (WebCore::V8TestEventConstructor::constructorCallback):
2041         (WebCore::fillTestEventConstructorInit):
2042         * bindings/scripts/test/V8/V8TestEventConstructor.h: Ditto.
2043
2044         * dom/BeforeLoadEvent.idl: In essence, replaced [JSCustomConstructor] IDL with [JSConstructorTemplate=Event] IDL.
2045         * dom/CustomEvent.idl: Ditto.
2046         * dom/ErrorEvent.idl: Ditto.
2047         * dom/Event.idl: Ditto.
2048         * dom/HashChangeEvent.idl: Ditto.
2049         * dom/MessageEvent.idl: Ditto.
2050         * dom/OverflowEvent.idl: Ditto.
2051         * dom/PageTransitionEvent.idl: Ditto.
2052         * dom/PopStateEvent.idl: Ditto.
2053         * dom/ProgressEvent.idl: Ditto.
2054         * dom/WebKitAnimationEvent.idl: Ditto.
2055         * dom/WebKitTransitionEvent.idl: Ditto.
2056         * html/track/TrackEvent.idl: Ditto.
2057         * websockets/CloseEvent.idl: Ditto.
2058
2059 2011-11-16  Dean Jackson  <dino@apple.com>
2060
2061         Build fix for Apple WebKit due to r100560.
2062         When I removed the violating header file it lost the
2063         reference to Color.
2064
2065         No review.
2066
2067         * platform/graphics/filters/FilterOperation.h:
2068
2069 2011-11-16  Dean Jackson  <dino@apple.com>
2070
2071         DropShadowFilterOperation violates platform isolation
2072         https://bugs.webkit.org/show_bug.cgi?id=72544
2073
2074         Reviewed by Simon Fraser.
2075
2076         Move ShadowData properties into the DropShadowFilterOperation
2077         to avoid depending on something outside platform.
2078
2079         * css/CSSComputedStyleDeclaration.cpp:
2080         (WebCore::CSSComputedStyleDeclaration::valueForFilter):
2081         * css/CSSStyleSelector.cpp:
2082         (WebCore::CSSStyleSelector::createFilterOperations):
2083         * platform/graphics/filters/FilterOperation.h:
2084         (WebCore::DropShadowFilterOperation::create):
2085         (WebCore::DropShadowFilterOperation::x):
2086         (WebCore::DropShadowFilterOperation::y):
2087         (WebCore::DropShadowFilterOperation::stdDeviation):
2088         (WebCore::DropShadowFilterOperation::color):
2089         (WebCore::DropShadowFilterOperation::operator==):
2090         (WebCore::DropShadowFilterOperation::DropShadowFilterOperation):
2091
2092 2011-11-16  Adam Bergkvist  <adam.bergkvist@ericsson.com>
2093
2094         Use a simple page client for user consent in getUserMedia()
2095         https://bugs.webkit.org/show_bug.cgi?id=70897
2096
2097         Reviewed by Adam Barth.
2098
2099         This is one in a series of patches that update the MediaStream feature
2100         to use WebCore platform interfaces.
2101
2102         Covered by existing tests.
2103
2104         * GNUmakefile.list.am:
2105         * WebCore.gypi:
2106         * mediastream/MediaStreamClient.h: Removed.
2107         * mediastream/MediaStreamController.cpp: Removed.
2108         * mediastream/MediaStreamController.h: Removed.
2109         * mediastream/MediaStreamFrameController.cpp: Removed.
2110         * mediastream/MediaStreamFrameController.h: Removed.
2111         * mediastream/UserMediaClient.h: Added.
2112         (WebCore::UserMediaClient::~UserMediaClient):
2113         * mediastream/UserMediaRequest.cpp: Added.
2114         (WebCore::UserMediaRequest::create):
2115         (WebCore::UserMediaRequest::UserMediaRequest):
2116         (WebCore::UserMediaRequest::~UserMediaRequest):
2117         (WebCore::UserMediaRequest::start):
2118         (WebCore::UserMediaRequest::mediaStreamSourcesQueryCompleted):
2119         (WebCore::UserMediaRequest::succeed):
2120         (WebCore::UserMediaRequest::fail):
2121         (WebCore::UserMediaRequest::contextDestroyed):
2122         (WebCore::UserMediaRequest::parseOptions):
2123         * mediastream/UserMediaRequest.h: Added.
2124         (WebCore::UserMediaRequest::audio):
2125         (WebCore::UserMediaRequest::video):
2126         (WebCore::UserMediaRequest::cameraPreferenceUser):
2127         (WebCore::UserMediaRequest::cameraPreferenceEnvironment):
2128         (WebCore::UserMediaRequest::successCallback):
2129         (WebCore::UserMediaRequest::errorCallback):
2130         * page/CallbackTask.h: Removed.
2131         * page/Frame.cpp:
2132         (WebCore::Frame::Frame):
2133         (WebCore::Frame::~Frame):
2134         (WebCore::Frame::pageDestroyed):
2135         (WebCore::Frame::transferChildFrameToNewDocument):
2136         * page/Frame.h:
2137         * page/Navigator.cpp:
2138         (WebCore::Navigator::webkitGetUserMedia):
2139         * page/NavigatorUserMediaErrorCallback.h:
2140         * page/Page.cpp:
2141         (WebCore::Page::Page):
2142         (WebCore::Page::~Page):
2143         (WebCore::Page::PageClients::PageClients):
2144         * page/Page.h:
2145         (WebCore::Page::userMediaClient):
2146
2147 2011-11-16  David Grogan  <dgrogan@chromium.org>
2148
2149         instantiate WorkerEventQueue in WorkerContext
2150         https://bugs.webkit.org/show_bug.cgi?id=72528
2151
2152         Reviewed by David Levin.
2153
2154         No new tests - nothing uses it yet.
2155
2156         * workers/WorkerContext.cpp:
2157         (WebCore::WorkerContext::WorkerContext):
2158         Instantiate WorkerEventQueue in InitializerList
2159
2160         * workers/WorkerEventQueue.h:
2161         Don't inherit from RefCounted, WorkerContext has an OwnPtr to it
2162
2163 2011-11-16  Kentaro Hara  <haraken@chromium.org>
2164
2165         Unreviewed. Rebaselined run-bindings-tests results.
2166
2167         * bindings/scripts/test/JS/JSTestEventConstructor.cpp:
2168         (WebCore::JSTestEventConstructor::JSTestEventConstructor):
2169         (WebCore::JSTestEventConstructorOwner::finalize):
2170         * bindings/scripts/test/JS/JSTestEventConstructor.h:
2171         (WebCore::JSTestEventConstructor::impl):
2172         (WebCore::JSTestEventConstructor::releaseImpl):
2173         * bindings/scripts/test/JS/JSTestInterface.cpp:
2174         (WebCore::JSTestInterface::JSTestInterface):
2175         (WebCore::JSTestInterfaceOwner::finalize):
2176         * bindings/scripts/test/JS/JSTestInterface.h:
2177         (WebCore::JSTestInterface::impl):
2178         (WebCore::JSTestInterface::releaseImpl):
2179         * bindings/scripts/test/JS/JSTestMediaQueryListListener.cpp:
2180         (WebCore::JSTestMediaQueryListListener::JSTestMediaQueryListListener):
2181         (WebCore::JSTestMediaQueryListListenerOwner::finalize):
2182         * bindings/scripts/test/JS/JSTestMediaQueryListListener.h:
2183         (WebCore::JSTestMediaQueryListListener::impl):
2184         (WebCore::JSTestMediaQueryListListener::releaseImpl):
2185         * bindings/scripts/test/JS/JSTestNamedConstructor.cpp:
2186         (WebCore::JSTestNamedConstructor::JSTestNamedConstructor):
2187         (WebCore::JSTestNamedConstructorOwner::finalize):
2188         * bindings/scripts/test/JS/JSTestNamedConstructor.h:
2189         (WebCore::JSTestNamedConstructor::impl):
2190         (WebCore::JSTestNamedConstructor::releaseImpl):
2191         * bindings/scripts/test/JS/JSTestObj.cpp:
2192         (WebCore::JSTestObj::JSTestObj):
2193         (WebCore::JSTestObjOwner::finalize):
2194         * bindings/scripts/test/JS/JSTestObj.h:
2195         (WebCore::JSTestObj::impl):
2196         (WebCore::JSTestObj::releaseImpl):
2197         * bindings/scripts/test/JS/JSTestSerializedScriptValueInterface.cpp:
2198         (WebCore::JSTestSerializedScriptValueInterface::JSTestSerializedScriptValueInterface):
2199         (WebCore::JSTestSerializedScriptValueInterfaceOwner::finalize):
2200         * bindings/scripts/test/JS/JSTestSerializedScriptValueInterface.h:
2201         (WebCore::JSTestSerializedScriptValueInterface::impl):
2202         (WebCore::JSTestSerializedScriptValueInterface::releaseImpl):
2203
2204 2011-11-16  Jay Civelli  <jcivelli@chromium.org>
2205
2206         Make sure MHTML documents use the domain of the MHTML file.
2207         https://bugs.webkit.org/show_bug.cgi?id=72445
2208
2209         Reviewed by Adam Barth.
2210
2211         * dom/Document.h:
2212         (WebCore::Document::setBaseURL):
2213         * loader/FrameLoader.cpp:
2214         (WebCore::FrameLoader::receivedFirstData):
2215
2216 2011-11-16  Julien Chaffraix  <jchaffraix@webkit.org>
2217
2218         Update supported display list after -webkit-grid and -webkit-inline-grid addition
2219         https://bugs.webkit.org/show_bug.cgi?id=72559
2220
2221         Reviewed by Tony Chang.
2222
2223         * css/CSSParser.cpp:
2224         (WebCore::CSSParser::parseValue): Added the 2 new supported
2225         values (forgotten in the previous change).
2226
2227 2011-11-11  Adrienne Walker  <enne@google.com>
2228
2229         [chromium] Expose mock scrollbars to window.internals
2230         https://bugs.webkit.org/show_bug.cgi?id=72195
2231
2232         Reviewed by James Robinson.
2233
2234         * testing/Internals.cpp:
2235         (WebCore::Internals::setMockScrollbarsEnabled):
2236         * testing/Internals.h:
2237         * testing/Internals.idl:
2238
2239 2011-11-16  Michael Nordman  <michaeln@google.com>
2240
2241         ApplicationCache manifest should work with any MIME type.
2242         https://bugs.webkit.org/show_bug.cgi?id=72082
2243
2244         Reviewed by Alexey Proskuryakov.
2245
2246         * loader/appcache/ApplicationCacheGroup.cpp:
2247         (WebCore::ApplicationCacheGroup::didReceiveManifestResponse): Remove the test for a particular type.
2248
2249 2011-11-16  Daniel Sievers  <sievers@chromium.org>
2250
2251         [Chromium] Avoid color mask operations for root layers
2252         https://bugs.webkit.org/show_bug.cgi?id=72452
2253
2254         Instead of relying on the combination of clearing the surface and initializing
2255         the alpha channel to 1.0 followed by disabling alpha in the color mask when
2256         rendering the root layer tiles, add shaders to support writing out opaque layers
2257         (alpha channel values written as 1.0).
2258
2259         Reviewed by James Robinson.
2260
2261         No functional change made that requires new tests.
2262
2263         * platform/graphics/chromium/LayerRendererChromium.cpp:
2264         (WebCore::LayerRendererChromium::drawLayersInternal):
2265         (WebCore::LayerRendererChromium::initializeSharedObjects):
2266         (WebCore::LayerRendererChromium::tilerProgramOpaque):
2267         (WebCore::LayerRendererChromium::tilerProgramOpaqueAA):
2268         (WebCore::LayerRendererChromium::tilerProgramSwizzleOpaque):
2269         (WebCore::LayerRendererChromium::tilerProgramSwizzleOpaqueAA):
2270         (WebCore::LayerRendererChromium::cleanupSharedObjects):
2271         * platform/graphics/chromium/LayerRendererChromium.h:
2272         * platform/graphics/chromium/ShaderChromium.cpp:
2273         (WebCore::FragmentTexOpaqueBinding::FragmentTexOpaqueBinding):
2274         (WebCore::FragmentTexOpaqueBinding::init):
2275         (WebCore::FragmentShaderRGBATexOpaque::getShaderString):
2276         (WebCore::FragmentShaderRGBATexSwizzleOpaque::getShaderString):
2277         (WebCore::FragmentTexClampOpaqueAABinding::FragmentTexClampOpaqueAABinding):
2278         (WebCore::FragmentTexClampOpaqueAABinding::init):
2279         (WebCore::FragmentShaderRGBATexClampOpaqueAA::getShaderString):
2280         (WebCore::FragmentShaderRGBATexClampSwizzleOpaqueAA::getShaderString):
2281         * platform/graphics/chromium/ShaderChromium.h:
2282         (WebCore::FragmentTexOpaqueBinding::alphaLocation):
2283         (WebCore::FragmentTexOpaqueBinding::samplerLocation):
2284         (WebCore::FragmentTexClampOpaqueAABinding::alphaLocation):
2285         (WebCore::FragmentTexClampOpaqueAABinding::samplerLocation):
2286         (WebCore::FragmentTexClampOpaqueAABinding::fragmentTexTransformLocation):
2287         (WebCore::FragmentTexClampOpaqueAABinding::edgeLocation):
2288         * platform/graphics/chromium/cc/CCTiledLayerImpl.cpp:
2289         (WebCore::CCTiledLayerImpl::draw):
2290         * platform/graphics/chromium/cc/CCTiledLayerImpl.h:
2291
2292 2011-11-16  Tim Horton  <timothy_horton@apple.com>
2293
2294         Implement CSS3 Images cross-fade() image function
2295         https://bugs.webkit.org/show_bug.cgi?id=52162
2296         <rdar://problem/10209254>
2297
2298         Reviewed by Simon Fraser.
2299
2300         Render -webkit-cross-fade. Only cross-fades entirely composed of images will render for now,
2301         cross-fades involving generated images are not yet implemented.
2302
2303         Reorganize GeneratedImage to be the base class for GeneratorGeneratedImage and CrossfadeGeneratedImage.
2304
2305         Add a pending state to CSSImageGeneratorValue, which is used to enable the pending-images loading
2306         mechanism for -webkit-cross-fade's sub-images. Rework the logic in CSSStyleSelector to support pending
2307         generated images.
2308         
2309         Support parsing fractional values for the cross-fade amount (for example, 0.5 = 50%). Clamp cross-fade
2310         amount to 0-1 range.
2311
2312         Tests: css3/images/cross-fade-invalidation.html
2313                css3/images/cross-fade-simple.html
2314                css3/images/cross-fade-sizing.html
2315                css3/images/cross-fade-tiled.html
2316
2317         * CMakeLists.txt:
2318         * GNUmakefile.list.am:
2319         * Target.pri:
2320         * WebCore.gypi:
2321         * WebCore.xcodeproj/project.pbxproj:
2322         * css/CSSCanvasValue.h:
2323         (WebCore::CSSCanvasValue::isPending):
2324         (WebCore::CSSCanvasValue::loadSubimages):
2325         * css/CSSCrossfadeValue.cpp:
2326         (WebCore::CSSCrossfadeValue::isPending):
2327         (WebCore::CSSCrossfadeValue::loadSubimages):
2328         (WebCore::subimageIsPending):
2329         (WebCore::loadSubimage):
2330         (WebCore::cachedImageForCSSValue):
2331         (WebCore::CSSCrossfadeValue::image):
2332         (WebCore::CSSCrossfadeValue::crossfadeChanged):
2333         * css/CSSCrossfadeValue.h:
2334         (WebCore::CSSCrossfadeValue::create):
2335         (WebCore::CSSCrossfadeValue::~CSSCrossfadeValue):
2336         (WebCore::CSSCrossfadeValue::fixedSize):
2337         (WebCore::CSSCrossfadeValue::CSSCrossfadeValue):
2338         (WebCore::CSSCrossfadeValue::CrossfadeObserverProxy::CrossfadeObserverProxy):
2339         * css/CSSGradientValue.cpp:
2340         (WebCore::CSSGradientValue::image):
2341         * css/CSSGradientValue.h:
2342         (WebCore::CSSGradientValue::isPending):
2343         (WebCore::CSSGradientValue::loadSubimages):
2344         * css/CSSImageGeneratorValue.cpp:
2345         (WebCore::CSSImageGeneratorValue::generatedOrPendingImage):
2346         (WebCore::CSSImageGeneratorValue::generatedImage):
2347         (WebCore::CSSImageGeneratorValue::isPending):
2348         (WebCore::CSSImageGeneratorValue::loadSubimages):
2349         * css/CSSImageGeneratorValue.h:
2350         * css/CSSParser.cpp:
2351         (WebCore::CSSParser::parseCrossfade):
2352         * css/CSSStyleSelector.cpp:
2353         (WebCore::CSSStyleSelector::styleImage):
2354         (WebCore::CSSStyleSelector::generatedOrPendingFromValue):
2355         (WebCore::CSSStyleSelector::loadPendingImage):
2356         (WebCore::CSSStyleSelector::loadPendingImages):
2357         * css/CSSStyleSelector.h:
2358         * platform/graphics/BitmapImage.h:
2359         * platform/graphics/CrossfadeGeneratedImage.cpp: Added.
2360         (WebCore::CrossfadeGeneratedImage::CrossfadeGeneratedImage):
2361         (WebCore::CrossfadeGeneratedImage::~CrossfadeGeneratedImage):
2362         (WebCore::CrossfadeGeneratedImage::drawCrossfade):
2363         (WebCore::CrossfadeGeneratedImage::draw):
2364         (WebCore::CrossfadeGeneratedImage::drawPattern):
2365         (WebCore::CrossfadeGeneratedImage::imageChanged):
2366         * platform/graphics/CrossfadeGeneratedImage.h: Added.
2367         (WebCore::CrossfadeGeneratedImage::create):
2368         (WebCore::CrossfadeSubimageObserverProxy::CrossfadeSubimageObserverProxy):
2369         (WebCore::CrossfadeSubimageObserverProxy::setReady):
2370         * platform/graphics/GeneratedImage.h:
2371         (WebCore::GeneratedImage::GeneratedImage):
2372         * platform/graphics/GeneratorGeneratedImage.cpp: Renamed from Source/WebCore/platform/graphics/GeneratedImage.cpp.
2373         (WebCore::GeneratorGeneratedImage::draw):
2374         (WebCore::GeneratorGeneratedImage::drawPattern):
2375         (WebCore::GeneratedImage::computeIntrinsicDimensions):
2376         * platform/graphics/GeneratorGeneratedImage.h: Copied from Source/WebCore/platform/graphics/GeneratedImage.h.
2377         (WebCore::GeneratorGeneratedImage::create):
2378         (WebCore::GeneratorGeneratedImage::~GeneratorGeneratedImage):
2379         (WebCore::GeneratorGeneratedImage::GeneratorGeneratedImage):
2380         * platform/graphics/Image.h:
2381         * platform/graphics/ImageBuffer.h:
2382         * rendering/style/StylePendingImage.h:
2383         (WebCore::StylePendingImage::create):
2384         (WebCore::StylePendingImage::data):
2385         (WebCore::StylePendingImage::cssImageValue):
2386         (WebCore::StylePendingImage::cssImageGeneratorValue):
2387         (WebCore::StylePendingImage::StylePendingImage):
2388
2389
2390 2011-11-16  Dan Bernstein  <mitz@apple.com>
2391
2392         WebCore part of <rdar://problem/10262242> Add API for paginated display
2393         https://bugs.webkit.org/show_bug.cgi?id=72537
2394
2395         Reviewed by Anders Carlsson.
2396
2397         * WebCore.exp.in: Exported Page::pageCount().
2398         * page/Page.cpp:
2399         (WebCore::Page::pageCount): Added this getter.
2400         * page/Page.h:
2401
2402 2011-11-16  Shawn Singh  <shawnsingh@chromium.org>
2403
2404         [chromium] Track property changes for render surfaces.
2405         https://bugs.webkit.org/show_bug.cgi?id=72521
2406
2407         Reviewed by James Robinson.
2408
2409         Created CCRenderSurfaceTest for testing.
2410
2411         * platform/graphics/chromium/LayerRendererChromium.cpp:
2412         (WebCore::LayerRendererChromium::drawLayersOntoRenderSurfaces):
2413         * platform/graphics/chromium/cc/CCLayerImpl.cpp:
2414         (WebCore::CCLayerImpl::resetPropertyChangedFlagForSubtree):
2415         * platform/graphics/chromium/cc/CCLayerImpl.h:
2416         * platform/graphics/chromium/cc/CCRenderSurface.cpp:
2417         (WebCore::CCRenderSurface::CCRenderSurface):
2418         (WebCore::CCRenderSurface::setClipRect):
2419         (WebCore::CCRenderSurface::setContentRect):
2420         (WebCore::CCRenderSurface::surfacePropertyChanged):
2421         * platform/graphics/chromium/cc/CCRenderSurface.h:
2422         (WebCore::CCRenderSurface::resetPropertyChangedFlag):
2423
2424 2011-11-16  Ben Wells  <benwells@chromium.org>
2425
2426         Seaming on border corners with mixed colour alpha borders
2427         https://bugs.webkit.org/show_bug.cgi?id=70471
2428
2429         Reviewed by Simon Fraser.
2430
2431         Seaming is fixed by antialiasing mitred corners for edges that have alpha and are joining
2432         a side that is of a different color.
2433
2434         Test: fast/borders/border-mixed-alpha.html
2435
2436         * rendering/RenderBoxModelObject.cpp:
2437         (WebCore::colorNeedsAntiAliasAtCorner):
2438         (WebCore::RenderBoxModelObject::paintOneBorderSide):
2439
2440 2011-11-16  Sam Weinig  <sam@webkit.org>
2441
2442         JS wrappers of DOM objects should have no-op constructors
2443         https://bugs.webkit.org/show_bug.cgi?id=72556
2444
2445         Reviewed by Geoffrey Garen.
2446
2447         Stop using a RefPtr to hold DOM objects contained by JavaScript
2448         wrappers and instead use a raw pointer. We were already releasing
2449         the underlying object before the destructor ran (via the finalizer)
2450         so the default behavior of destroying the RefPtr is always unnecessary
2451         busy work. 
2452
2453         * bindings/js/JSCSSValueCustom.cpp:
2454         (WebCore::JSCSSValueOwner::finalize):
2455         * bindings/js/JSNodeCustom.cpp:
2456         (WebCore::JSNodeOwner::finalize):
2457         (WebCore::JSNode::visitChildren):
2458         Call releaseImpl() instead of clearImpl().
2459
2460         * bindings/scripts/CodeGeneratorJS.pm:
2461         (GenerateHeader):
2462         Stop storing m_impl in a RefPtr and instead use a raw pointer. Switch
2463         clearImpl() to releaseImpl(), which explicitly derefs the pointer and
2464         clear it.
2465
2466         (GenerateImplementation):
2467         Use leakPtr() to explicitly adopt the PassRefPtr into the raw pointer.
2468         Change default finalize to call releaseImpl() instead of clearImpl().
2469
2470 2011-11-16  Michael Saboff  <msaboff@apple.com>
2471
2472         Enable 8 Bit Strings in JavaScriptCore
2473         https://bugs.webkit.org/show_bug.cgi?id=71337
2474
2475         This patch turns on 8 bit strings in StringImpl and enables
2476         their use in JavaScriptCore. Some of the changes are to
2477         turn on code that had been staged (Lexer.cpp, Identifier.cpp,
2478         SmallStrings.cpp and some of StringImpl.{h,cpp}).
2479         Other changes are minor fixes to make 8 bit strings work
2480         (UString.h, StringImpl::getData16SlowCase()).
2481         Changed StringBuffer to be a templated class based on character
2482         type.  This change riplled into WebCore code as well.
2483
2484         Reviewed by Geoffrey Garen.
2485
2486         No new tests - Changes in response to refactoring StringBuffer to
2487         be a template on character type.
2488
2489         * css/CSSParser.cpp:
2490         (WebCore::quoteCSSString):
2491         * css/CSSPrimitiveValue.cpp:
2492         (WebCore::formatNumber):
2493         * dom/Document.cpp:
2494         (WebCore::canonicalizedTitle):
2495         * platform/Length.cpp:
2496         (WebCore::newCoordsArray):
2497         * platform/sql/SQLiteStatement.cpp:
2498         (WebCore::SQLiteStatement::prepare):
2499         * platform/text/TextCodecUTF16.cpp:
2500         (WebCore::TextCodecUTF16::decode):
2501         * platform/text/TextCodecUTF8.cpp:
2502         (WebCore::TextCodecUTF8::decode):
2503         * rendering/RenderText.cpp:
2504         (WebCore::makeCapitalized):
2505         * xml/XSLTProcessorLibxslt.cpp:
2506         (WebCore::writeToStringBuilder):
2507
2508 2011-11-16  Alexandre Elias  <aelias@google.com>
2509
2510         [chromium] Improvements for page scale delta during commit
2511         https://bugs.webkit.org/show_bug.cgi?id=72471
2512
2513         Reviewed by James Robinson.
2514
2515         Page scale now follows the same commit flow as scroll position:
2516         the delta is folded into m_pageScale at BFAC time, and a "sent" value
2517         is preserved for temporary use until the commit finishes.
2518
2519         I also merged setPageScaleFactor and setPageScaleFactorLimits into one
2520         function on the impl side.  The reason is that setPageFactor must
2521         be applied after the limits are updated, but on the other hand setting
2522         the limits first may cause an unnecessary clamp of the scale delta.
2523         Merging the methods avoids this bind.
2524
2525         No new tests. (planning to add later: https://bugs.webkit.org/show_bug.cgi?id=71529)
2526
2527         * platform/graphics/chromium/cc/CCLayerTreeHost.cpp:
2528         (WebCore::CCLayerTreeHost::finishCommitOnImplThread):
2529         (WebCore::CCLayerTreeHost::applyScrollAndScale):
2530         * platform/graphics/chromium/cc/CCLayerTreeHostCommon.h:
2531         * platform/graphics/chromium/cc/CCLayerTreeHostImpl.cpp:
2532         (WebCore::CCLayerTreeHostImpl::CCLayerTreeHostImpl):
2533         (WebCore::CCLayerTreeHostImpl::setPageScaleFactorAndLimits):
2534         (WebCore::CCLayerTreeHostImpl::adjustScrollsForPageScaleChange):
2535         (WebCore::CCLayerTreeHostImpl::setScaleDelta):
2536         (WebCore::CCLayerTreeHostImpl::applyScaleDeltaToScrollLayer):
2537         (WebCore::CCLayerTreeHostImpl::scrollRootLayer):
2538         (WebCore::CCLayerTreeHostImpl::processScrollDeltas):
2539         * platform/graphics/chromium/cc/CCLayerTreeHostImpl.h:
2540
2541 2011-11-16  Joshua Bell  <jsbell@chromium.org>
2542
2543         IndexedDB: Reduce nested key depth threshold, re-enable flaky test
2544         https://bugs.webkit.org/show_bug.cgi?id=72529
2545
2546         Reviewed by Adam Barth.
2547
2548         Drop maximum array key depth from 20k to 2k.
2549
2550         * bindings/v8/IDBBindingUtilities.cpp:
2551
2552 2011-11-16  Nate Chapin  <japhet@chromium.org>
2553
2554         Fix incorrect multipart handling in r100311.
2555         SubresourceLoader::didReceiveData() is getting called
2556         twice, which has unintended side effects.
2557         https://bugs.webkit.org/show_bug.cgi?id=72436
2558
2559         Reviewed by Adam Barth.
2560
2561         http/tests/multipart/invalid-image-data.html should stop
2562         asserting after this.
2563
2564         * loader/SubresourceLoader.cpp:
2565         (WebCore::SubresourceLoader::didReceiveResponse):
2566         (WebCore::SubresourceLoader::didReceiveData):
2567         (WebCore::SubresourceLoader::sendDataToResource):
2568         * loader/SubresourceLoader.h:
2569
2570 2011-11-16  Justin Schuh  <jschuh@chromium.org>
2571
2572         Clear SVG filter client when its node is detached
2573         https://bugs.webkit.org/show_bug.cgi?id=71741
2574
2575         Reviewed by Eric Seidel.
2576
2577         Test: svg/filters/reparent-animated-filter-target.html
2578
2579         * rendering/svg/SVGResourcesCache.cpp:
2580         (WebCore::SVGResourcesCache::clientDestroyed):
2581
2582 2011-11-16  John Bates  <jbates@google.com>
2583
2584         Page/layer flashes after GPU-accelerated CSS transition
2585         https://bugs.webkit.org/show_bug.cgi?id=72343
2586
2587         LayerRendererChromium was resizing the window to 1x1 at initialization.
2588         In some cases, there is no drawLayers before switching back to
2589         software rendering. This left the window resized to 1x1 and the
2590         following software paints would therefore not be visible. This change
2591         moves the reshape call into drawLayers so that it will only be called
2592         if rendering will occur.
2593
2594         Reviewed by James Robinson.
2595
2596         New test: CCLayerTreeHostImplTest.reshapeNotCalledUntilDraw.
2597
2598         * platform/graphics/chromium/LayerRendererChromium.cpp:
2599         (WebCore::LayerRendererChromium::viewportChanged):
2600         (WebCore::LayerRendererChromium::doViewportChanged):
2601         (WebCore::LayerRendererChromium::drawLayersInternal):
2602         * platform/graphics/chromium/LayerRendererChromium.h:
2603
2604 2011-11-16  Alexandre Elias  <aelias@google.com>
2605
2606         [chromium] Add null pointer check in setDeviceScaleFactor
2607         https://bugs.webkit.org/show_bug.cgi?id=72464
2608
2609         Reviewed by James Robinson.
2610
2611         No new tests. (Tiny fix.)
2612
2613         * page/Page.cpp:
2614         (WebCore::Page::setDeviceScaleFactor):
2615
2616 2011-11-16  Sheriff Bot  <webkit.review.bot@gmail.com>
2617
2618         Unreviewed, rolling out r100438.
2619         http://trac.webkit.org/changeset/100438
2620         https://bugs.webkit.org/show_bug.cgi?id=72536
2621
2622         Broke unit tests (Requested by jamesr_ on #webkit).
2623
2624         * platform/PlatformScreen.h:
2625         * platform/chromium/PlatformScreenChromium.cpp:
2626         * platform/chromium/PlatformSupport.h:
2627         * platform/graphics/chromium/cc/CCLayerTreeHost.h:
2628         (WebCore::CCSettings::CCSettings):
2629         * platform/graphics/chromium/cc/CCThreadProxy.cpp:
2630         (WebCore::CCThreadProxy::initializeImplOnImplThread):
2631
2632 2011-11-16  Andy Estes  <aestes@apple.com>
2633
2634         Assertion failure in LayerFlushScheduler::resume() when running some layout tests in WebKitTestRunner
2635         https://bugs.webkit.org/show_bug.cgi?id=72535
2636
2637         Reviewed by Anders Carlsson.
2638
2639         LayerFlushScheduler attempted to use a counter strategy for calls to
2640         suspend() and resume(), which allowed us to assert that these calls
2641         were balanced. Unfortunately it is hard to guarantee this in WebKit2,
2642         where we sometimes try to call suspend() before we've entered
2643         compositing mode (hence before we have a LayerTreeHost and a
2644         LayerFlushScheduler). When we later call resume(), this call ends up
2645         being unbalanced and asserts.
2646
2647         For now, remove the assertions and allow unbalanced calls to suspend()
2648         and resume().
2649
2650         * platform/graphics/ca/LayerFlushScheduler.cpp:
2651         (WebCore::LayerFlushScheduler::suspend):
2652         (WebCore::LayerFlushScheduler::resume):
2653         * platform/graphics/ca/LayerFlushScheduler.h:
2654         * platform/graphics/ca/mac/LayerFlushSchedulerMac.cpp:
2655         (WebCore::LayerFlushScheduler::LayerFlushScheduler):
2656         (WebCore::LayerFlushScheduler::runLoopObserverCallback):
2657         (WebCore::LayerFlushScheduler::schedule):
2658
2659 2011-11-16  Sheriff Bot  <webkit.review.bot@gmail.com>
2660
2661         Unreviewed, rolling out r100473.
2662         http://trac.webkit.org/changeset/100473
2663         https://bugs.webkit.org/show_bug.cgi?id=72534
2664
2665         "Broke the Mac Build" (Requested by mwenge2 on #webkit).
2666
2667         * CMakeLists.txt:
2668         * GNUmakefile.list.am:
2669         * Target.pri:
2670         * WebCore.gypi:
2671         * rendering/RenderBlock.cpp:
2672         (WebCore::RenderBlock::addChildIgnoringAnonymousColumnBlocks):
2673         * rendering/RenderObject.cpp:
2674         (WebCore::RenderObject::createObject):
2675         (WebCore::RenderObject::addChild):
2676         * rendering/RenderObject.h:
2677         * rendering/RenderTable.cpp:
2678         (WebCore::RenderTable::addChild):
2679         * rendering/RenderTable.h:
2680         * rendering/RenderTableCaption.cpp: Removed.
2681         * rendering/RenderTableCaption.h: Removed.
2682
2683 2011-11-16  Sheriff Bot  <webkit.review.bot@gmail.com>
2684
2685         Unreviewed, rolling out r100479.
2686         http://trac.webkit.org/changeset/100479
2687         https://bugs.webkit.org/show_bug.cgi?id=72533
2688
2689         "Broke the Mac Build" (Requested by mwenge2 on #webkit).
2690
2691         * WebCore.vcproj/WebCore.vcproj:
2692         * WebCore.xcodeproj/project.pbxproj:
2693         * rendering/RenderingAllInOne.cpp:
2694
2695 2011-11-16  Chris Fleizach  <cfleizach@apple.com>
2696
2697         WebKitTestRunner needs to support accessibility-related DRT APIs
2698         https://bugs.webkit.org/show_bug.cgi?id=42131
2699
2700         Minor changes needed to support AX testing in WKTestRunner.
2701
2702         Reviewed by Beth Dakin.
2703
2704         * WebCore.exp.in:
2705              Expose focusedUIElementForPage so that the WK2 injected bundle can retrieve it.
2706         * accessibility/mac/WebAccessibilityObjectWrapper.mm:
2707         (accessibilitySearchKeyForString):
2708             Remove an unncessary assert that was causing issues with the WK2 test run.
2709
2710 2011-11-16  Sergio Villar Senin  <svillar@igalia.com>
2711
2712         [Soup] Somet tests fail with FAIL Unexpected response data received: Wrong method: GET
2713         https://bugs.webkit.org/show_bug.cgi?id=69219
2714
2715         Reviewed by Martin Robinson.
2716
2717         Do not stop appending data to the request body if any of the blob
2718         items to upload is not accesible.
2719
2720         * platform/network/soup/ResourceHandleSoup.cpp:
2721         (WebCore::addEncodedBlobToSoupMessageBody):
2722
2723 2011-11-16  Beth Dakin  <bdakin@apple.com>
2724
2725         https://bugs.webkit.org/show_bug.cgi?id=72400
2726         Scrollbar uiStateTransitionProgress requires tracking the mouse all the time
2727         -and corresponding-
2728         <rdar://problem/10409328>
2729
2730         Reviewed by Darin Adler.
2731
2732         This patch makes it so we track the mouse all the time when we have legacy 
2733         scrollbars (rather than only tracking the mouse when the window is key). When 
2734         we're in that mode, we want to do as little work as possible when handling the 
2735         mouseMoved event so that this extra tracking has little to no performance impact. 
2736         Also, we don't want to change basic behaviors by having normal web content hover 
2737         effects start happening when a window is in the background. So this patch also 
2738         introduces a way to handle a mouseMoved event that will only affect scrollbars.
2739
2740         EventHandler::mouseMoved() and EventHandler::handleMouseEvent() both now take a 
2741         boolean parameter that indicates if we are only updating scrollbars. If that is 
2742         the case, then we make our HitTestRequest ReadOnly, and we return early once 
2743         updateLastScrollbarUnderMouse() is called.
2744         * WebCore.exp.in:
2745         * page/EventHandler.cpp:
2746         (WebCore::EventHandler::mouseMoved):
2747         (WebCore::EventHandler::handleMouseMoveEvent):
2748
2749         In addition to calling Scrollbar::mouseExited() when appropriate, this function 
2750         now calls a new function, Scrollbar::mouseEntered() when appropriate. 
2751         (WebCore::EventHandler::updateLastScrollbarUnderMouse):
2752         * page/EventHandler.h:
2753
2754         Scrollbar::mouseMoved() used to be responsible for calling 
2755         ScrollAnimator::mouseEnteredScrollbar(), but now Scrollbar::mouseEntered() takes 
2756         care of that instead, much like how Scrollbar::mouseExited() takes care of calling 
2757         the animator's exit function.
2758         * platform/Scrollbar.cpp:
2759         (WebCore::Scrollbar::mouseMoved):
2760         (WebCore::Scrollbar::mouseEntered):
2761         * platform/Scrollbar.h:
2762
2763 2011-11-16  Andreas Kling  <kling@webkit.org>
2764
2765         CSSValue: isInheritedValue() doesn't need a dedicated bit.
2766         <http://webkit.org/b/72514>
2767
2768         Reviewed by Antti Koivisto.
2769
2770         Remove CSSValue::m_isInherited and have isInheritedValue() check the
2771         class type instead. There's no compelling reason for CSSInheritedValue
2772         to have a dedicated bit, since nobody subclasses it anyway.
2773
2774         * css/CSSValue.h:
2775         (WebCore::CSSValue::isInheritedValue):
2776         (WebCore::CSSValue::CSSValue):
2777
2778 2011-11-16  Robert Hogan  <robert@webkit.org>
2779
2780         Fix build on Windows and Mac after r100473
2781
2782         Unreviewed, fix build.
2783
2784         * WebCore.vcproj/WebCore.vcproj:
2785         * WebCore.xcodeproj/project.pbxproj:
2786         * rendering/RenderingAllInOne.cpp:
2787
2788 2011-11-16  Philip Rogers  <pdr@google.com>
2789
2790         Remove extra GraphicsContextStateSaver restore() call.
2791         https://bugs.webkit.org/show_bug.cgi?id=72497
2792
2793         Reviewed by Andreas Kling.
2794
2795         * html/canvas/CanvasRenderingContext2D.cpp:
2796         (WebCore::CanvasRenderingContext2D::drawTextInternal):
2797
2798 2011-11-08  Robert Hogan  <robert@webkit.org>
2799
2800         CSS 2.1 failure: border-collapse-offset-002.htm fails
2801         https://bugs.webkit.org/show_bug.cgi?id=71705
2802
2803         Table captions are implemented as children of the table but have a special
2804         requirement to expand to the full width of the table rather than just the 'available'
2805         width, i.e. the full width minus padding and borders.
2806
2807         To accomodate this create a RenderTableCaption object that reimplements containingBlockLogicalWidthForContent()
2808         to return the full width of the containing block (i.e. the table) rather than the available width.
2809
2810         Reviewed by Antti Koivisto.
2811
2812         * CMakeLists.txt: Add RenderTableCaption.[cpp|h]
2813         * GNUmakefile.list.am: Add RenderTableCaption.[cpp|h]
2814         * Target.pri: Add RenderTableCaption.[cpp|h]
2815         * WebCore.gypi: Add RenderTableCaption.[cpp|h]
2816         * rendering/RenderBlock.cpp:
2817         (WebCore::RenderBlock::addChildIgnoringAnonymousColumnBlocks): Use RenderTableCaption
2818         * rendering/RenderObject.cpp:
2819         (WebCore::RenderObject::createObject): Add RenderTableCaption.[cpp|h]
2820         (WebCore::RenderObject::addChild): ditto
2821         * rendering/RenderObject.h:
2822         (WebCore::RenderObject::isTableCaption):
2823         * rendering/RenderTable.cpp:
2824         (WebCore::RenderTable::addChild):
2825         * rendering/RenderTable.h:
2826         * rendering/RenderTableCaption.cpp: Added.
2827         (WebCore::RenderTableCaption::RenderTableCaption): Implement RenderTableCaption
2828         (WebCore::RenderTableCaption::~RenderTableCaption):
2829         (WebCore::RenderTableCaption::containingBlockLogicalWidthForContent): Return the containing block's full width rather than it's available width.
2830         * rendering/RenderTableCaption.h: Added.
2831         (WebCore::RenderTableCaption::isTableCaption):
2832         (WebCore::toRenderTableCaption):
2833
2834 2011-11-16  Antaryami Pandia  <antaryami.pandia@motorola.com>
2835
2836         [Gtk] display:none has no effect on <option> element.
2837         https://bugs.webkit.org/show_bug.cgi?id=72370
2838
2839         Reviewed by Martin Robinson.
2840
2841         * platform/gtk/GtkPopupMenu.cpp:
2842         (WebCore::GtkPopupMenu::appendItem):
2843         * platform/gtk/PopupMenuGtk.cpp:
2844         (WebCore::PopupMenuGtk::createGtkActionForMenuItem):
2845
2846 2011-11-16  Dan Winship  <danw@gnome.org>
2847
2848         [GTK] Fix platformDefersLoading to handle non-http requests, and
2849         to not use broken-ish libsoup APIs.
2850         https://bugs.webkit.org/show_bug.cgi?id=72227
2851
2852         Reviewed by Martin Robinson.
2853
2854         * platform/network/ResourceHandleInternal.h:
2855         * platform/network/soup/ResourceHandleSoup.cpp:
2856         (WebCore::sendRequestCallback):
2857         (WebCore::startHTTPRequest):
2858         (WebCore::hasBeenSent):
2859         (WebCore::ResourceHandle::platformSetDefersLoading):
2860         (WebCore::readCallback): rather than deferring by using
2861         soup_session_pause_message(), let the read complete, but just don't
2862         process the result until we're no longer deferred.
2863         (WebCore::startNonHTTPRequest): Don't start the request if
2864         it's deferred.
2865
2866 2011-11-16  Vsevolod Vlasov  <vsevik@chromium.org>
2867
2868         Web Inspector: Application cache status should be updated after swapCache().
2869         https://bugs.webkit.org/show_bug.cgi?id=72123
2870
2871         Reviewed by Pavel Feldman.
2872
2873         Application cache view resources and status are now updated after swapCache() call.
2874         Refresh button removed from application cache view.
2875         Application cache inspector tests are moved to their own folder.
2876
2877         Tests: http/tests/inspector/appcache/appcache-iframe-manifests.html
2878                http/tests/inspector/appcache/appcache-manifest-with-non-existing-file.html
2879                http/tests/inspector/appcache/appcache-swap.html
2880
2881         * inspector/front-end/ApplicationCacheItemsView.js:
2882         (WebInspector.ApplicationCacheItemsView):
2883         (WebInspector.ApplicationCacheItemsView.prototype.get statusBarItems):
2884         (WebInspector.ApplicationCacheItemsView.prototype.wasShown):
2885         (WebInspector.ApplicationCacheItemsView.prototype._maybeUpdate):
2886         (WebInspector.ApplicationCacheItemsView.prototype._markDirty):
2887         (WebInspector.ApplicationCacheItemsView.prototype.updateStatus):
2888         (WebInspector.ApplicationCacheItemsView.prototype._updateCallback):
2889         (WebInspector.ApplicationCacheItemsView.prototype._deleteCallback):
2890         * loader/appcache/ApplicationCacheGroup.cpp:
2891         (WebCore::ApplicationCacheGroup::setNewestCache):
2892         (WebCore::ApplicationCacheGroup::makeObsolete):
2893         (WebCore::ApplicationCacheGroup::setUpdateStatus):
2894         * loader/appcache/ApplicationCacheHost.cpp:
2895         (WebCore::ApplicationCacheHost::notifyDOMApplicationCache):
2896         (WebCore::ApplicationCacheHost::swapCache):
2897
2898 2011-11-16  Eric Carlson  <eric.carlson@apple.com>
2899
2900         addTrack() must throw an exception if 'kind' is unknown
2901         https://bugs.webkit.org/show_bug.cgi?id=71915
2902
2903         Reviewed by Philippe Normand.
2904
2905         Tests: media/track/track-addtrack-kind.html
2906                media/track/track-kind.html
2907
2908         * html/HTMLMediaElement.cpp:
2909         (WebCore::HTMLMediaElement::textTrackKindChanged): New. Will be implemented for 62885.
2910         (WebCore::HTMLMediaElement::addTrack): Throw if  'kind' is not a known value.
2911         (WebCore::HTMLMediaElement::addTextTrack): Call textTracks(), it will allocate the track
2912             list object if necessary.
2913         (WebCore::HTMLMediaElement::textTracks): Never return NULL, a TextTrackList with no tracks
2914             is allowed.
2915         * html/HTMLMediaElement.h:
2916         (WebCore::HTMLMediaElement::addTrack): Add variants to deal with optional parameters plus
2917             a mandatory ExceptionCode parameter.
2918         * html/HTMLMediaElement.idl: addTrack can generate an exception.
2919
2920         * html/HTMLTrackElement.cpp:
2921         (WebCore::HTMLTrackElement::attributeChanged): kind, label, and srclang attribute changes should
2922             percolate down the the TextTrack.
2923         (WebCore::HTMLTrackElement::kind): Return the TextTrack kind because it is not necessarily the
2924             same as the attribute value.
2925         (WebCore::HTMLTrackElement::ensureTrack): Only pass legal 'kind' keywords to create a TextTrack.
2926         (WebCore::HTMLTrackElement::textTrackKindChanged): Notify parent element, if any.
2927         * html/HTMLTrackElement.h:
2928
2929         * html/TextTrack.cpp:
2930         (WebCore::TextTrack::subtitlesKeyword): New, return legal kind attribute value.
2931         (WebCore::TextTrack::captionsKeyword): Ditto.
2932         (WebCore::TextTrack::descriptionsKeyword): Ditto.
2933         (WebCore::TextTrack::chaptersKeyword): Ditto.
2934         (WebCore::TextTrack::metadataKeyword): Ditto.
2935         (WebCore::TextTrack::TextTrack): Call setKind to make sure m_kind is always set to legal value.
2936         (WebCore::TextTrack::isValidKindKeyword): New, validate 'kind' value.
2937         (WebCore::TextTrack::setKind): Only allow legal values.
2938         * html/TextTrack.h:
2939         (WebCore::TextTrack::kind):
2940         (WebCore::TextTrack::label):
2941         (WebCore::TextTrack::setLabel):
2942         (WebCore::TextTrack::language):
2943         (WebCore::TextTrack::setLanguage):
2944         (WebCore::TextTrack::readyState):
2945         (WebCore::TextTrack::mode):
2946
2947 2011-11-16  Andreas Kling  <kling@webkit.org>
2948
2949         CSSValue: Clean up initial value construction.
2950         <http://webkit.org/b/72502>
2951
2952         Reviewed by Antti Koivisto.
2953
2954         Instead of determining whether a given CSSInitialValue is 'implicit' or not by
2955         querying the CSSValue::ClassType (InitialClass vs. ImplicitInitialClass),
2956         add a protected CSSValue member and set it from the CSSInitialValue constructor.
2957
2958         Also get rid of the CSSValue::m_isInitial bit since we can now replace the
2959         checks by classType() == InitialClass.
2960
2961         No new tests, this is a cleanup.
2962
2963         * css/CSSInitialValue.h:
2964         (WebCore::CSSInitialValue::CSSInitialValue):
2965
2966             Poke 'implicit' constructor argument into CSSValue::m_isImplicit.
2967
2968         * css/CSSValue.cpp:
2969         (WebCore::CSSValue::cssText):
2970         (WebCore::CSSValue::destroy):
2971
2972             Remove ImplicitInitialClass cases.
2973
2974         * css/CSSValue.h:
2975         (WebCore::CSSValue::isImplicitInitialValue):
2976         (WebCore::CSSValue::isInitialValue):
2977         (WebCore::CSSValue::CSSValue):
2978
2979 2011-11-16  Antaryami Pandia  <antaryami.pandia@motorola.com>
2980
2981         Remove unnecessary if check from RenderListBox::paintItemForeground.
2982         https://bugs.webkit.org/show_bug.cgi?id=72488
2983
2984         Reviewed by Andreas Kling.
2985
2986         * rendering/RenderListBox.cpp:
2987         (WebCore::RenderListBox::paintItemForeground):
2988
2989 2011-11-15  Simon Hausmann  <simon.hausmann@nokia.com>
2990
2991         [Qt] Centralize hide_symbols and ensure all libs are built with symbol visibility & bsymbolic_functions
2992
2993         Reviewed by Tor Arne Vestbø.
2994
2995         * Target.pri: Eliminate duplicated symbol stuff that lives now in default_post.prf.
2996
2997 2011-11-16  Iain Merrick  <husky@google.com>
2998
2999         [chromium] Pass screen refresh rate into compositor.
3000         https://bugs.webkit.org/show_bug.cgi?id=71040
3001
3002         Reviewed by Tony Gentilcore.
3003
3004         * platform/PlatformScreen.h:
3005         * platform/chromium/PlatformScreenChromium.cpp:
3006         (WebCore::screenRefreshRate):
3007         * platform/chromium/PlatformSupport.h:
3008         * platform/graphics/chromium/cc/CCLayerTreeHost.h:
3009         (WebCore::CCSettings::CCSettings):
3010         * platform/graphics/chromium/cc/CCThreadProxy.cpp:
3011         (WebCore::CCThreadProxy::initializeImplOnImplThread):
3012
3013 2011-11-15  Andrey Kosyakov  <caseq@chromium.org>
3014
3015         Web Inspector: [Extensions API] drop ExtensionSidebarPane.onUpdated, use callbacks instead
3016         https://bugs.webkit.org/show_bug.cgi?id=72388
3017
3018         Reviewed by Pavel Feldman.
3019
3020         Tests: inspector/extensions/extensions-panel.html
3021                inspector/extensions/extensions-sidebar.html
3022
3023         - fire ExtensionSidebarPane.on{Hidden,Shown} for non-iframe content (experssions/objects);
3024         - drop ExtensionsSidebarPane.onUpdated, provide callback for setObject()/setExpression() instead;
3025         - fix an exception when a page is replaced with object/expression.
3026
3027          inspector/front-end/ExtensionAPI.js:
3028         (injectedExtensionAPI.ExtensionSidebarPaneImpl):
3029         (injectedExtensionAPI.ExtensionSidebarPaneImpl.prototype.setExpression):
3030         (injectedExtensionAPI.ExtensionSidebarPaneImpl.prototype.setObject):
3031         * inspector/front-end/ExtensionPanel.js:
3032         (WebInspector.ExtensionNotifierView):
3033         (WebInspector.ExtensionNotifierView.prototype.wasShown):
3034         (WebInspector.ExtensionNotifierView.prototype.willHide):
3035         (WebInspector.ExtensionSidebarPane.prototype.setObject):
3036         (WebInspector.ExtensionSidebarPane.prototype.setExpression):
3037         (WebInspector.ExtensionSidebarPane.prototype.setPage):
3038         (WebInspector.ExtensionSidebarPane.prototype._onEvaluate):
3039         (WebInspector.ExtensionSidebarPane.prototype._makeObjectPropertiesView):
3040         (WebInspector.ExtensionSidebarPane.prototype._setObject):
3041         * inspector/front-end/ExtensionServer.js:
3042         (WebInspector.ExtensionServer.prototype._onSetSidebarContent.callback):
3043         (WebInspector.ExtensionServer.prototype._onSetSidebarContent):
3044         (WebInspector.ExtensionServer.prototype._dispatchCallback):
3045         * inspector/front-end/View.js:
3046         (WebInspector.View.prototype.detach):
3047
3048 2011-11-16  Simon Hausmann  <simon.hausmann@nokia.com>
3049
3050         Unreviewed, rolling out r100266.
3051         http://trac.webkit.org/changeset/100266
3052
3053         Broke WTR.
3054
3055         * Target.pri:
3056
3057 2011-11-16  Per-Erik Brodin  <per-erik.brodin@ericsson.com>
3058
3059         [GTK] fast/events/event-creation.html fails creating MediaStreamEvent
3060         https://bugs.webkit.org/show_bug.cgi?id=70720
3061
3062         Reviewed by Philippe Normand.
3063
3064         Added missing overriding of Event::interfaceName()
3065
3066         * mediastream/MediaStreamEvent.cpp:
3067         (WebCore::MediaStreamEvent::stream): Changed return value to raw pointer.
3068         (WebCore::MediaStreamEvent::interfaceName): Added back from r98044.
3069         * mediastream/MediaStreamEvent.h:
3070         * mediastream/MediaStreamEvent.idl: Changed module name to "events".
3071
3072 2011-11-16  Mario Sanchez Prada  <msanchez@igalia.com>
3073
3074         [GTK] Use GQuark's in the ATK wrapper to get and set arbitrary data
3075         https://bugs.webkit.org/show_bug.cgi?id=72394
3076
3077         Reviewed by Martin Robinson.
3078
3079         No new functionality, no new tests needed.
3080
3081         * accessibility/gtk/AccessibilityObjectWrapperAtk.cpp:
3082         (webkit_accessible_class_init): Initialize GQuarks.
3083         (getGailTextUtilForAtk): Use gailTextUtilQuark.
3084         (getPangoLayoutForAtk): Remove unused call to
3085         g_object_set_data_full, since that data is no used anywhere.
3086         (webkitAccessibleHyperlinkImplGetHyperlink): Use hyperlinkObjectQuark.
3087
3088 2011-11-15  Alexandru Chiculita  <achicu@adobe.com>
3089
3090         [CSSShaders] Implement the style cached resources and computed style for the shader urls
3091         https://bugs.webkit.org/show_bug.cgi?id=72378
3092
3093         Reviewed by Dean Jackson.
3094
3095         Test: css3/filters/custom-filter-property-computed-style.html
3096
3097         * GNUmakefile.list.am:
3098         * Target.pri:
3099         * WebCore.gypi:
3100         * WebCore.vcproj/WebCore.vcproj:
3101         * WebCore.xcodeproj/project.pbxproj:
3102         * css/CSSComputedStyleDeclaration.cpp:
3103         (WebCore::CSSComputedStyleDeclaration::valueForFilter):
3104         * css/CSSStyleSelector.cpp:
3105         (WebCore::CSSStyleSelector::CSSStyleSelector):
3106         (WebCore::CSSStyleSelector::styleForKeyframe):
3107         (WebCore::CSSStyleSelector::pseudoStyleForElement):
3108         (WebCore::CSSStyleSelector::styleForPage):
3109         (WebCore::CSSStyleSelector::applyMatchedDeclarations):
3110         (WebCore::CSSStyleSelector::styleShader):
3111         (WebCore::CSSStyleSelector::cachedOrPendingStyleShaderFromValue):
3112         (WebCore::CSSStyleSelector::loadPendingShaders):
3113         (WebCore::CSSStyleSelector::createCustomFilterOperation):
3114         (WebCore::CSSStyleSelector::createFilterOperations):
3115         * css/CSSStyleSelector.h:
3116         * css/WebKitCSSShaderValue.cpp:
3117         (WebCore::WebKitCSSShaderValue::WebKitCSSShaderValue):
3118         (WebCore::WebKitCSSShaderValue::~WebKitCSSShaderValue):
3119         (WebCore::WebKitCSSShaderValue::cachedShader):
3120         (WebCore::WebKitCSSShaderValue::cachedOrPendingShader):
3121         * css/WebKitCSSShaderValue.h:
3122         * loader/cache/CachedResource.cpp:
3123         (WebCore::defaultPriorityForResourceType):
3124         * loader/cache/CachedResource.h:
3125         * loader/cache/CachedResourceLoader.cpp:
3126         (WebCore::createResource):
3127         (WebCore::CachedResourceLoader::requestShader):
3128         (WebCore::CachedResourceLoader::checkInsecureContent):
3129         (WebCore::CachedResourceLoader::canRequest):
3130         * loader/cache/CachedResourceLoader.h:
3131         * loader/cache/CachedShader.cpp:
3132         (WebCore::CachedShader::CachedShader):
3133         (WebCore::CachedShader::~CachedShader):
3134         * loader/cache/CachedShader.h:
3135         * platform/graphics/filters/CustomFilterOperation.h:
3136         * rendering/style/RenderStyle.h:
3137         (WebCore::InheritedFlags::filter):
3138         * rendering/style/StyleCachedShader.cpp:
3139         (WebCore::StyleCachedShader::StyleCachedShader):
3140         (WebCore::StyleCachedShader::cssValue):
3141         * rendering/style/StyleCachedShader.h:
3142         (WebCore::StyleCachedShader::create):
3143         * rendering/style/StylePendingShader.h:
3144         (WebCore::StylePendingShader::create):
3145         (WebCore::StylePendingShader::cssValue):
3146         (WebCore::StylePendingShader::cssShaderValue):
3147         (WebCore::StylePendingShader::StylePendingShader):
3148         * rendering/style/StyleShader.h:
3149         (WebCore::StyleShader::~StyleShader):
3150         (WebCore::StyleShader::isCachedShader):
3151         (WebCore::StyleShader::isPendingShader):
3152         (WebCore::StyleShader::StyleShader):
3153
3154 2011-11-15  Sergio Villar Senin  <svillar@igalia.com>
3155
3156         [WK2] [GTK] fast/css/webkit-mask-crash-fieldset-legend.html asserts WebKitWebProcess
3157         https://bugs.webkit.org/show_bug.cgi?id=69510
3158
3159         Reviewed by Simon Fraser.
3160
3161         End the current transparency layer before early returning from
3162         paintMask() when there is a maskBoxImage which is still being
3163         loaded. This will balance the previous call to
3164         beginTransparencyLayer().
3165
3166         * rendering/InlineFlowBox.cpp:
3167         (WebCore::InlineFlowBox::paintMask):
3168
3169 2011-11-15  Darin Adler  <darin@apple.com>
3170
3171         Incorrect type checks in RenderTheme media code
3172         https://bugs.webkit.org/show_bug.cgi?id=72184
3173
3174         Reviewed by Eric Carlson.
3175
3176         No tests added. Ideally this patch should be revised to add tests!
3177
3178         * accessibility/AccessibilityMediaControls.cpp:
3179         (WebCore::AccessibilityMediaControl::create): Use mediaControlElementType.
3180         (WebCore::AccessibilityMediaControl::controlType): Ditto.
3181         (WebCore::AccessibilityMediaTimeline::valueDescription): Use early return
3182         rather than an assertion to check type of input element.
3183
3184         * html/shadow/MediaControlElements.cpp:
3185         (WebCore::mediaControlElementType): Added. A type-safe way to get the
3186         media control element type after checking isMediaControlElement but with
3187         no other assumptions.
3188         * html/shadow/MediaControlElements.h: Added mediaControlElementType.
3189
3190         * platform/efl/RenderThemeEfl.cpp:
3191         (WebCore::RenderThemeEfl::paintMediaPlayButton): Use mediaControlElementType.
3192         (WebCore::RenderThemeEfl::paintMediaSeekBackButton): Use mediaControlElementType.
3193         (WebCore::RenderThemeEfl::paintMediaSeekForwardButton): Use mediaControlElementType.
3194         * platform/gtk/RenderThemeGtk.cpp:
3195         (WebCore::RenderThemeGtk::paintMediaPlayButton): Check isMediaControlElement and
3196         use mediaControlElementType.
3197         * rendering/RenderThemeMac.mm:
3198         (WebCore::RenderThemeMac::paintMediaMuteButton): Ditto. Also remove uneeded
3199         redundant null check.
3200         (WebCore::RenderThemeMac::paintMediaPlayButton): Ditto.
3201         (WebCore::RenderThemeMac::paintMediaToggleClosedCaptionsButton): Ditto.
3202
3203 2011-11-15  Jeff Timanus  <twiz@chromium.org>
3204
3205         [chromium] During tear down, prevent the WebGLLayerChromium instance from attempting to stop a timer for a NULL context.
3206         https://bugs.webkit.org/show_bug.cgi?id=72423
3207
3208         Reviewed by Kenneth Russell.
3209
3210         * platform/graphics/chromium/WebGLLayerChromium.cpp:
3211         (WebCore::WebGLLayerChromium::setDrawingBuffer):
3212
3213 2011-11-15  Mark Hahnenberg  <mhahnenberg@apple.com>
3214
3215         Rebaseline generated WebCore bindings
3216
3217         Unreviewed build fix
3218
3219         No new tests.
3220
3221         * bindings/scripts/test/JS/JSTestEventConstructor.cpp:
3222         (WebCore::isObservable):
3223         (WebCore::JSTestEventConstructorOwner::isReachableFromOpaqueRoots):
3224         (WebCore::JSTestEventConstructorOwner::finalize):
3225         * bindings/scripts/test/JS/JSTestEventConstructor.h:
3226         (WebCore::JSTestEventConstructor::clearImpl):
3227         (WebCore::wrapperOwner):
3228         (WebCore::wrapperContext):
3229         * bindings/scripts/test/JS/JSTestInterface.cpp:
3230         (WebCore::JSTestInterfaceOwner::finalize):
3231         * bindings/scripts/test/JS/JSTestInterface.h:
3232         (WebCore::JSTestInterface::clearImpl):
3233         * bindings/scripts/test/JS/JSTestMediaQueryListListener.cpp:
3234         (WebCore::isObservable):
3235         (WebCore::JSTestMediaQueryListListenerOwner::isReachableFromOpaqueRoots):
3236         (WebCore::JSTestMediaQueryListListenerOwner::finalize):
3237         * bindings/scripts/test/JS/JSTestMediaQueryListListener.h:
3238         (WebCore::JSTestMediaQueryListListener::clearImpl):
3239         (WebCore::wrapperOwner):
3240         (WebCore::wrapperContext):
3241         * bindings/scripts/test/JS/JSTestNamedConstructor.cpp:
3242         (WebCore::JSTestNamedConstructorOwner::finalize):
3243         * bindings/scripts/test/JS/JSTestNamedConstructor.h:
3244         (WebCore::JSTestNamedConstructor::clearImpl):
3245         * bindings/scripts/test/JS/JSTestObj.cpp:
3246         (WebCore::isObservable):
3247         (WebCore::JSTestObjOwner::isReachableFromOpaqueRoots):
3248         (WebCore::JSTestObjOwner::finalize):
3249         * bindings/scripts/test/JS/JSTestObj.h:
3250         (WebCore::JSTestObj::clearImpl):
3251         (WebCore::wrapperOwner):
3252         (WebCore::wrapperContext):
3253         * bindings/scripts/test/JS/JSTestSerializedScriptValueInterface.cpp:
3254         (WebCore::isObservable):
3255         (WebCore::JSTestSerializedScriptValueInterfaceOwner::isReachableFromOpaqueRoots):
3256         (WebCore::JSTestSerializedScriptValueInterfaceOwner::finalize):
3257         * bindings/scripts/test/JS/JSTestSerializedScriptValueInterface.h:
3258         (WebCore::JSTestSerializedScriptValueInterface::clearImpl):
3259         (WebCore::wrapperOwner):
3260         (WebCore::wrapperContext):
3261
3262 2011-11-15  Jeff Timanus  <twiz@chromium.org>
3263
3264         Patch removing duplicated code in the setup of the DrawingBuffer used
3265         to host the back-buffer for WebGL contents.
3266         https://bugs.webkit.org/show_bug.cgi?id=72327
3267
3268         Reviewed by Kenneth Russell.
3269
3270         * html/canvas/WebGLRenderingContext.cpp:
3271         (WebCore::WebGLRenderingContext::copyTexImage2D):
3272         (WebCore::WebGLRenderingContext::copyTexSubImage2D):
3273         (WebCore::WebGLRenderingContext::readPixels):
3274
3275 2011-11-15  Mark Hahnenberg  <mhahnenberg@apple.com>
3276
3277         JS DOM wrappers depend on destructor to deref impl RefPtrs
3278         https://bugs.webkit.org/show_bug.cgi?id=72341
3279
3280         Reviewed by Sam Weinig.
3281
3282         No new tests.
3283
3284         Added clearing of impl RefPtrs to JS DOM wrapper nodes and removed the default 
3285         wrapperOwner function in favor of generating all WeakHandleOwners and wrapperOwner functions.
3286
3287         * bindings/js/JSCSSValueCustom.cpp:
3288         (WebCore::JSCSSValueOwner::finalize):
3289         * bindings/js/JSDOMBinding.h:
3290         * bindings/js/JSNodeCustom.cpp:
3291         (WebCore::JSNodeOwner::finalize):
3292         * bindings/scripts/CodeGeneratorJS.pm:
3293         (GenerateHeader):
3294         (GenerateImplementation):
3295
3296 2011-11-15  Joseph Pecoraro  <pecoraro@apple.com>
3297
3298         Web Inspector: Share Highlight Code for Drawing Outlined Quad
3299         https://bugs.webkit.org/show_bug.cgi?id=72451
3300
3301         Reviewed by Timothy Hatcher.
3302
3303         * inspector/DOMNodeHighlighter.cpp:
3304
3305 2011-11-15  Sheriff Bot  <webkit.review.bot@gmail.com>
3306
3307         Unreviewed, rolling out r100308.
3308         http://trac.webkit.org/changeset/100308
3309         https://bugs.webkit.org/show_bug.cgi?id=72450
3310
3311         Introduces WebGL conformance test regressions. (Requested by
3312         twiz on #webkit).
3313
3314         * html/canvas/WebGLRenderingContext.cpp:
3315         (WebCore::WebGLRenderingContext::copyTexImage2D):
3316         (WebCore::WebGLRenderingContext::copyTexSubImage2D):
3317         (WebCore::WebGLRenderingContext::readPixels):
3318
3319 2011-11-15  James Robinson  <jamesr@chromium.org>
3320
3321         Rollout http://trac.webkit.org/changeset/99813, caused some crashes in
3322         TiledLayerChromium::updateCompositorResources()
3323
3324         * WebCore.gypi:
3325         * platform/graphics/chromium/BitmapCanvasLayerTextureUpdater.cpp: Removed.
3326         * platform/graphics/chromium/BitmapCanvasLayerTextureUpdater.h: Removed.
3327         * platform/graphics/chromium/CanvasLayerTextureUpdater.cpp: Removed.
3328         * platform/graphics/chromium/CanvasLayerTextureUpdater.h: Removed.
3329         * platform/graphics/chromium/ContentLayerChromium.cpp:
3330         (WebCore::ContentLayerChromium::createTextureUpdater):
3331         * platform/graphics/chromium/FrameBufferSkPictureCanvasLayerTextureUpdater.h: Removed.
3332         * platform/graphics/chromium/ImageLayerChromium.cpp:
3333         (WebCore::ImageLayerTextureUpdater::prepareToUpdate):
3334         * platform/graphics/chromium/LayerRendererChromium.cpp:
3335         * platform/graphics/chromium/LayerTextureUpdater.h:
3336         * platform/graphics/chromium/LayerTextureUpdaterCanvas.cpp: Renamed from Source/WebCore/platform/graphics/chromium/FrameBufferSkPictureCanvasLayerTextureUpdater.cpp.
3337         (WebCore::FrameBuffer::FrameBuffer::FrameBuffer):
3338         (WebCore::FrameBuffer::FrameBuffer::~FrameBuffer):
3339         (WebCore::FrameBuffer::FrameBuffer::initialize):
3340         (WebCore::LayerTextureUpdaterCanvas::LayerTextureUpdaterCanvas):
3341         (WebCore::LayerTextureUpdaterCanvas::paintContents):
3342         (WebCore::LayerTextureUpdaterBitmap::create):
3343         (WebCore::LayerTextureUpdaterBitmap::LayerTextureUpdaterBitmap):
3344         (WebCore::LayerTextureUpdaterBitmap::sampledTexelFormat):
3345         (WebCore::LayerTextureUpdaterBitmap::prepareToUpdate):
3346         (WebCore::LayerTextureUpdaterBitmap::updateTextureRect):
3347         (WebCore::LayerTextureUpdaterSkPicture::create):
3348         (WebCore::LayerTextureUpdaterSkPicture::LayerTextureUpdaterSkPicture):
3349         (WebCore::LayerTextureUpdaterSkPicture::~LayerTextureUpdaterSkPicture):
3350         (WebCore::LayerTextureUpdaterSkPicture::sampledTexelFormat):
3351         (WebCore::LayerTextureUpdaterSkPicture::prepareToUpdate):
3352         (WebCore::LayerTextureUpdaterSkPicture::updateTextureRect):
3353         * platform/graphics/chromium/LayerTextureUpdaterCanvas.h: Added.
3354         (WebCore::LayerTextureUpdaterCanvas::contentRect):
3355         (WebCore::LayerTextureUpdaterBitmap::orientation):
3356         (WebCore::LayerTextureUpdaterSkPicture::orientation):
3357         * platform/graphics/chromium/SkPictureCanvasLayerTextureUpdater.cpp: Removed.
3358         * platform/graphics/chromium/SkPictureCanvasLayerTextureUpdater.h: Removed.
3359         * platform/graphics/chromium/TiledLayerChromium.cpp:
3360         (WebCore::UpdatableTile::UpdatableTile):
3361         (WebCore::UpdatableTile::texture):
3362         (WebCore::TiledLayerChromium::updateCompositorResources):
3363         (WebCore::TiledLayerChromium::pushPropertiesTo):
3364         (WebCore::TiledLayerChromium::createTile):
3365         (WebCore::TiledLayerChromium::protectTileTextures):
3366         (WebCore::TiledLayerChromium::prepareToUpdate):
3367         * platform/graphics/chromium/cc/CCTextureUpdater.cpp:
3368         (WebCore::CCTextureUpdater::append):
3369         (WebCore::CCTextureUpdater::update):
3370         * platform/graphics/chromium/cc/CCTextureUpdater.h:
3371
3372 2011-11-15  Julien Chaffraix  <jchaffraix@webkit.org>
3373
3374         Add the needed plumbing to parse display: -webkit-inline-grid
3375         https://bugs.webkit.org/show_bug.cgi?id=72438
3376
3377         Reviewed by Tony Chang.
3378
3379         Test: fast/css-grid-layout/display-grid-set-get.html
3380
3381         Added the needed constants and plugged everything together.
3382         Again we treat display: -webkit-inline-grid like display: none
3383         for the moment.
3384
3385         * css/CSSParser.cpp:
3386         (WebCore::CSSParser::parseValue):
3387         * css/CSSPrimitiveValueMappings.h:
3388         (WebCore::CSSPrimitiveValue::CSSPrimitiveValue):
3389         * css/CSSValueKeywords.in:
3390         * rendering/RenderObject.cpp:
3391         (WebCore::RenderObject::createObject):
3392         * rendering/style/RenderStyleConstants.h:
3393
3394 2011-11-15  W. James MacLean  <wjmaclean@chromium.org>
3395
3396         [chromium] Move setVisibleRect() calls into calculateDrawTransformAndVisibility()
3397         https://bugs.webkit.org/show_bug.cgi?id=72162
3398
3399         Reviewed by Kenneth Russell.
3400
3401         Refactoring of existing functionality, so uses existing tests.
3402
3403         * platform/graphics/chromium/LayerRendererChromium.cpp:
3404         (WebCore::LayerRendererChromium::drawLayer):
3405         * platform/graphics/chromium/cc/CCLayerTreeHost.cpp:
3406         (WebCore::paintContentsIfDirty):
3407         (WebCore::CCLayerTreeHost::paintMaskAndReplicaForRenderSurface):
3408         (WebCore::CCLayerTreeHost::paintLayerContents):
3409         * platform/graphics/chromium/cc/CCLayerTreeHostCommon.cpp:
3410         (WebCore::walkLayersAndCalculateVisibleLayerRects):
3411         (WebCore::CCLayerTreeHostCommon::calculateDrawTransformsAndVisibility):
3412
3413 2011-11-15  Sheriff Bot  <webkit.review.bot@gmail.com>
3414
3415         Unreviewed, rolling out r100340.
3416         http://trac.webkit.org/changeset/100340
3417         https://bugs.webkit.org/show_bug.cgi?id=72448
3418
3419         Caused assertion failure in Win dbg canary. (Requested by
3420         pkasting on #webkit).
3421
3422         * platform/graphics/chromium/LayerRendererChromium.cpp:
3423         (WebCore::LayerRendererChromium::LayerRendererChromium):
3424         (WebCore::LayerRendererChromium::viewportChanged):
3425         (WebCore::LayerRendererChromium::drawLayersInternal):
3426         * platform/graphics/chromium/LayerRendererChromium.h:
3427
3428 2011-11-15  Julien Chaffraix  <jchaffraix@webkit.org>
3429
3430         Switch table indexing to unsigned
3431         https://bugs.webkit.org/show_bug.cgi?id=72083
3432
3433         Reviewed by Darin Adler.
3434
3435         No expected change in behavior.
3436
3437         All of the code is now using unsigned for indexing!
3438
3439         * rendering/FixedTableLayout.cpp:
3440         (WebCore::FixedTableLayout::layout):
3441         * rendering/RenderTable.cpp:
3442         (WebCore::RenderTable::colElement):
3443         (WebCore::RenderTable::cellAbove):
3444         * rendering/RenderTableSection.cpp:
3445         (WebCore::RenderTableSection::splitColumn):
3446         Added some ASSERTs to make sure we don't underflow. Looking at how
3447         the different variables are populated, they should not be reached.
3448
3449         * rendering/RenderTableCell.cpp:
3450         (WebCore::RenderTableCell::colSpan):
3451         (WebCore::RenderTableCell::rowSpan):
3452         Those 2 functions promotes HTMLTableCellElement's int to unsigned
3453         which should be fine as we make sure their are positive. Also HTML5
3454         makes those 2 fields "unsigned long" which goes in the same direction.
3455
3456         * rendering/AutoTableLayout.cpp:
3457         (WebCore::AutoTableLayout::layout):
3458         * rendering/RenderTableSection.cpp:
3459         (WebCore::RenderTableSection::nodeAtPoint):
3460         Rewrote a couple of reverse iterating to be able to use unsigned
3461         without overflowing.
3462
3463         * rendering/AutoTableLayout.cpp:
3464         (WebCore::AutoTableLayout::recalcColumn):
3465         (WebCore::AutoTableLayout::fullRecalc):
3466         (WebCore::AutoTableLayout::calcEffectiveLogicalWidth):
3467         (WebCore::AutoTableLayout::insertSpanCell):
3468         * rendering/AutoTableLayout.h:
3469         * rendering/FixedTableLayout.cpp:
3470         (WebCore::FixedTableLayout::calcWidthArray):
3471         * rendering/RenderTable.cpp:
3472         (WebCore::RenderTable::splitColumn):
3473         (WebCore::RenderTable::appendColumn):
3474         (WebCore::RenderTable::recalcSections):
3475         * rendering/RenderTable.h:
3476         (WebCore::RenderTable::getColumnPos):
3477         (WebCore::RenderTable::spanOfEffCol):
3478         (WebCore::RenderTable::effColToCol):
3479         * rendering/RenderTableCell.cpp:
3480         (WebCore::RenderTableCell::styleOrColLogicalWidth):
3481         (WebCore::CollapsedBorders::nextBorder):
3482         * rendering/RenderTableCell.h:
3483         * rendering/RenderTableCol.cpp:
3484         (WebCore::RenderTableCol::updateFromElement):
3485         * rendering/RenderTableCol.h:
3486         (WebCore::RenderTableCol::span):
3487         (WebCore::RenderTableCol::setSpan):
3488         * rendering/RenderTableSection.cpp:
3489         (WebCore::RenderTableSection::addCell):
3490         (WebCore::RenderTableSection::setCellLogicalWidths):
3491         (WebCore::RenderTableSection::layoutRows):
3492         (WebCore::RenderTableSection::calcOuterBorderBefore):
3493         (WebCore::RenderTableSection::calcOuterBorderAfter):
3494         (WebCore::RenderTableSection::calcOuterBorderStart):
3495         (WebCore::RenderTableSection::calcOuterBorderEnd):
3496         (WebCore::RenderTableSection::paintObject):
3497         (WebCore::RenderTableSection::appendColumn):
3498         * rendering/RenderTableSection.h:
3499         (WebCore::RenderTableSection::cellAt):
3500         (WebCore::RenderTableSection::primaryCellAt):
3501         (WebCore::RenderTableSection::getBaseline):
3502         Mechanical change int -> unsigned.
3503
3504 2011-11-15  Andy Estes  <aestes@apple.com>
3505
3506         Consolidate the logic that creates run loop observers for flushing layer tree changes to CoreAnimation
3507         https://bugs.webkit.org/show_bug.cgi?id=72106
3508
3509         Reviewed by Anders Carlsson.
3510
3511         Add a class that encapsulates the logic of scheduling, enabling and
3512         invalidating a run loop observer that fires before Core Animation's
3513         commit observer. Clients can subclass LayerFlushSchedulerClient and
3514         implement flushLayers(), which will be called by the observer.
3515
3516         * WebCore.exp.in:
3517         * WebCore.xcodeproj/project.pbxproj:
3518         * platform/graphics/ca/LayerFlushScheduler.cpp: Added.
3519         (WebCore::LayerFlushScheduler::suspend): Suspend scheduling by
3520         invalidating the run loop observer. Keep a count of calls to suspend()
3521         in m_suspendCount.
3522         (WebCore::LayerFlushScheduler::resume): Decrement m_suspendCount.
3523         Install the run loop observer when it reaches 0.
3524         * platform/graphics/ca/LayerFlushSchedulerClient.h: Added.
3525         (WebCore::LayerFlushSchedulerClient::~LayerFlushSchedulerClient):
3526         * platform/graphics/ca/LayerFlushScheduler.h: Added.
3527         * platform/graphics/ca/mac/LayerFlushSchedulerMac.cpp: Added.
3528         (LayerFlushScheduler::LayerFlushScheduler):
3529         (LayerFlushScheduler::~LayerFlushScheduler):
3530         (LayerFlushScheduler::runLoopObserverCallback): Call flushLayers() on
3531         the LayerFlushSchedulerClient.
3532         (LayerFlushScheduler::schedule): Install the run loop observer.
3533         (LayerFlushScheduler::invalidate): Remove the run loop
3534         observer if it is installed.
3535
3536 2011-11-15  Adam Klein  <adamk@chromium.org>
3537
3538         [v8] Use throwError instead of compiling and running script in handleMaxRecursionDepthExceeded
3539         https://bugs.webkit.org/show_bug.cgi?id=72432
3540
3541         Reviewed by Adam Barth.
3542
3543         * bindings/v8/V8Proxy.cpp:
3544         (WebCore::handleMaxRecursionDepthExceeded):
3545
3546 2011-11-15  Vincent Scheib  <scheib@chromium.org>
3547
3548         Pointer Lock: Refactoring: PointerLock.idl: Dropping webkit prefix
3549         https://bugs.webkit.org/show_bug.cgi?id=72431
3550
3551         Reviewed by Adam Barth.
3552
3553         * page/PointerLock.cpp:
3554         (WebCore::PointerLock::lock):
3555         (WebCore::PointerLock::unlock):
3556         (WebCore::PointerLock::isLocked):
3557         * page/PointerLock.h:
3558         * page/PointerLock.idl:
3559
3560 2011-11-15  Nat Duca  <nduca@chromium.org>
3561
3562         [chromium] Fuse MainThread and CCThread
3563         https://bugs.webkit.org/show_bug.cgi?id=72426
3564
3565         Reviewed by James Robinson.
3566
3567         * WebCore.gypi:
3568         * platform/graphics/chromium/LayerRendererChromium.cpp:
3569         * platform/graphics/chromium/cc/CCLayerTreeHost.cpp:
3570         (WebCore::CCLayerTreeHost::CCLayerTreeHost):
3571         * platform/graphics/chromium/cc/CCLayerTreeHostImpl.cpp:
3572         * platform/graphics/chromium/cc/CCMainThread.cpp: Removed.
3573         * platform/graphics/chromium/cc/CCMainThread.h: Removed.
3574         * platform/graphics/chromium/cc/CCMainThreadTask.h: Removed.
3575         * platform/graphics/chromium/cc/CCProxy.cpp:
3576         (WebCore::CCProxy::setMainThread):
3577         (WebCore::CCProxy::mainThread):
3578         (WebCore::CCProxy::setImplThread):
3579         (WebCore::CCProxy::implThread):
3580         (WebCore::CCProxy::isMainThread):
3581         (WebCore::CCProxy::isImplThread):
3582         (WebCore::CCProxy::~CCProxy):
3583         * platform/graphics/chromium/cc/CCProxy.h:
3584         * platform/graphics/chromium/cc/CCScopedThreadProxy.h: Renamed from Source/WebCore/platform/graphics/chromium/cc/CCScopedMainThreadProxy.h.
3585         (WebCore::CCScopedThreadProxy::create):
3586         (WebCore::CCScopedThreadProxy::postTask):
3587         (WebCore::CCScopedThreadProxy::shutdown):
3588         (WebCore::CCScopedThreadProxy::CCScopedThreadProxy):
3589         (WebCore::CCScopedThreadProxy::runTaskIfNotShutdown):
3590         * platform/graphics/chromium/cc/CCSingleThreadProxy.cpp:
3591         * platform/graphics/chromium/cc/CCThreadProxy.cpp:
3592         (WebCore::CCThreadProxy::CCThreadProxy):
3593         (WebCore::CCThreadProxy::compositeAndReadback):
3594         (WebCore::CCThreadProxy::finishAllRendering):
3595         (WebCore::CCThreadProxy::initializeLayerRenderer):
3596         (WebCore::CCThreadProxy::setNeedsAnimate):
3597         (WebCore::CCThreadProxy::setNeedsCommit):
3598         (WebCore::CCThreadProxy::onSwapBuffersCompleteOnImplThread):
3599         (WebCore::CCThreadProxy::setNeedsRedraw):
3600         (WebCore::CCThreadProxy::setVisible):
3601         (WebCore::CCThreadProxy::start):
3602         (WebCore::CCThreadProxy::stop):
3603         (WebCore::CCThreadProxy::obtainBeginFrameAndCommitTaskFromCCThread):
3604         (WebCore::CCThreadProxy::createBeginFrameAndCommitTaskOnImplThread):
3605         (WebCore::CCThreadProxy::beginFrameAndCommit):
3606         (WebCore::CCThreadProxy::scheduledActionDrawAndSwap):
3607         (WebCore::CCThreadProxy::initializeImplOnImplThread):
3608         * platform/graphics/chromium/cc/CCThreadProxy.h:
3609
3610 2011-11-15  Vincent Scheib  <scheib@chromium.org>
3611
3612         Pointer Lock: Refactor: MouseEvent.idl movementX/Y
3613         https://bugs.webkit.org/show_bug.cgi?id=72427
3614
3615         - [Conditional...] vs #if defined
3616         - Runtime enabled
3617         - .movementX/Y prefixed with 'webkit'
3618
3619         Reviewed by Adam Barth.
3620
3621         No new tests.
3622
3623         * bindings/generic/RuntimeEnabledFeatures.h:
3624         (WebCore::RuntimeEnabledFeatures::webkitMovementXEnabled):
3625         (WebCore::RuntimeEnabledFeatures::webkitMovementYEnabled):
3626         * dom/MouseEvent.idl:
3627         * dom/MouseRelatedEvent.h:
3628         (WebCore::MouseRelatedEvent::webkitMovementX):
3629         (WebCore::MouseRelatedEvent::webkitMovementY):
3630
3631 2011-10-28  Ojan Vafai  <ojan@chromium.org>
3632
3633         implement flex-align for flex-flow: column
3634         https://bugs.webkit.org/show_bug.cgi?id=70754
3635
3636         Reviewed by David Hyatt.
3637
3638         Tests: css3/flexbox/flex-align-column.html
3639                css3/flexbox/line-wrapping.html
3640
3641         * rendering/RenderBox.cpp:
3642         (WebCore::RenderBox::sizesToIntrinsicLogicalWidth):
3643         When flexitems are column, they should size to the intrinsic width unless flex-flow is stretch.
3644
3645         * rendering/RenderFlexibleBox.cpp:
3646         (WebCore::RenderFlexibleBox::alignChildrenBlockDirection):
3647         This was just using the wrong, non-flow-aware method.
3648
3649 2011-11-15  Jochen Eisinger  <jochen@chromium.org>
3650
3651         Rename ReferrerPolicy to clarify its meaning
3652         https://bugs.webkit.org/show_bug.cgi?id=72420
3653
3654         Reviewed by Nate Chapin.
3655
3656         On the one hand, even if the ReferrerPolicy was set to SendReferrer, the
3657         referrer wasn't necessarily send. On the other hand, I want to use the
3658         name ReferrerPolicy when implementing the meta referrer tag.
3659
3660         No change in behavior expected so no test.
3661
3662         * html/HTMLAnchorElement.cpp:
3663         (WebCore::HTMLAnchorElement::handleClick):
3664         (WebCore::handleLinkClick):
3665         * loader/FrameLoader.cpp:
3666         (WebCore::FrameLoader::changeLocation):
3667         (WebCore::FrameLoader::urlSelected):
3668         (WebCore::FrameLoader::loadFrameRequest):
3669         * loader/FrameLoader.h:
3670         * loader/FrameLoaderTypes.h:
3671         * loader/NavigationScheduler.cpp:
3672         (WebCore::ScheduledHistoryNavigation::fire):
3673         (WebCore::ScheduledFormSubmission::fire):
3674         * page/ContextMenuController.cpp:
3675         (WebCore::openNewWindow):
3676         (WebCore::ContextMenuController::contextMenuItemSelected):
3677         * WebCore.exp.in: updated.
3678
3679 2011-11-15  Adam Klein  <adamk@chromium.org>
3680
3681         Factor out V8Proxy's max recursion depth handling code
3682         https://bugs.webkit.org/show_bug.cgi?id=72422
3683
3684         Reviewed by Nate Chapin.
3685
3686         Previously, V8Proxy used slightly different code to handle stack limit
3687         violations depending on whether they occured in runScript or
3688         callFunction. As described in http://webkit.org/b/72063, I intend to
3689         expand the usage of m_recursion when calling into script. This patch
3690         is intended to unify the existing handling code, making it easier to
3691         move elsewhere without causing unintended side-effects.
3692
3693         No tests changed, as the only change in behavior is the string passed
3694         to RangeError in the runScript case, and it's not mentioned anywhere
3695         in the LayoutTests.
3696
3697         * bindings/v8/V8Proxy.cpp:
3698         (WebCore::handleMaxRecursionDepthExceeded):
3699         (WebCore::V8Proxy::runScript): Use callFunction's factored-out code.
3700         (WebCore::V8Proxy::callFunction): Simplify and factor out code into handleMaxRecursionDepthExceeded.
3701
3702 2011-11-15  Jessie Berlin  <jberlin@apple.com>
3703
3704         NSURLRequest leak beneath ResourceRequest::setStorageSession seen on Leaks bot.
3705         https://bugs.webkit.org/show_bug.cgi?id=72419
3706
3707         Reviewed by Adam Roben.
3708
3709         Adopt the copied NSURLRequest.
3710
3711         * platform/network/mac/ResourceRequestMac.mm:
3712         (WebCore::ResourceRequest::setStorageSession):
3713
3714 2011-11-15  John Bates  <jbates@google.com>
3715
3716         Page/layer flashes after GPU-accelerated CSS transition
3717         https://bugs.webkit.org/show_bug.cgi?id=72343
3718
3719         LayerRendererChromium was resizing the window to 1x1 at initialization.
3720         In some cases, there is no drawLayers before switching back to
3721         software rendering. This left the window resized to 1x1 and the
3722         following software paints would therefore not be visible. This change
3723         moves the reshape call into drawLayers so that it will only be called
3724         if rendering will occur.
3725
3726         Reviewed by James Robinson.
3727
3728         New test: CCLayerTreeHostImplTest.reshapeNotCalledUntilDraw.
3729
3730         * platform/graphics/chromium/LayerRendererChromium.cpp:
3731         (WebCore::LayerRendererChromium::viewportChanged):
3732         (WebCore::LayerRendererChromium::doViewportChanged):
3733         (WebCore::LayerRendererChromium::drawLayersInternal):
3734         * platform/graphics/chromium/LayerRendererChromium.h:
3735
3736 2011-11-15  Julien Chaffraix  <jchaffraix@webkit.org>
3737
3738         Add the needed plumbing to parse display: -webkit-grid
3739         https://bugs.webkit.org/show_bug.cgi?id=72331
3740
3741         Reviewed by Tony Chang.
3742
3743         Test: fast/css-grid-layout/display-grid-set-get.html
3744
3745         Added parsing support for display: -webkit-grid. From a rendering perspective,
3746         the value is equivalent to display: none until we properly implement it.
3747
3748         * css/CSSParser.cpp:
3749         (WebCore::CSSParser::parseValue):
3750         * css/CSSPrimitiveValueMappings.h:
3751         (WebCore::CSSPrimitiveValue::CSSPrimitiveValue):
3752         * css/CSSValueKeywords.in:
3753         * rendering/style/RenderStyleConstants.h:
3754         Added the new CSS value and plumbed the parsing and style application of
3755         the new value.
3756
3757         * css/CSSPrimitiveValueMappings.h:
3758         (WebCore::CSSPrimitiveValue::operator EDisplay):
3759         Added an ASSERT here as I bumped into some non-trivial issues due to bug 72296.
3760         -wap-marquee was offsetting the new value and was wrongly casted by the CSSPrimitiveValueMapping
3761         logic outside the EDisplay range which would lead to crashes.
3762
3763         * rendering/RenderObject.cpp:
3764         (WebCore::RenderObject::createObject):
3765         Fixed the indentation to follow our coding rules.
3766
3767 2011-11-15  Vincent Scheib  <scheib@chromium.org>
3768
3769         Mouse Lock: Renaming to 'Pointer Lock': MouseLockable to PointerLock
3770         https://bugs.webkit.org/show_bug.cgi?id=72315
3771
3772         Reviewed by Dimitri Glazkov.
3773
3774         * WebCore.gypi:
3775         * page/Navigator.cpp:
3776         (WebCore::Navigator::webkitPointer):
3777         * page/Navigator.h:
3778         * page/Navigator.idl:
3779         * page/PointerLock.cpp: Renamed from Source/WebCore/page/MouseLockable.cpp.
3780         (WebCore::PointerLock::PointerLock):
3781         (WebCore::PointerLock::~PointerLock):
3782         (WebCore::PointerLock::webkitLock):
3783         (WebCore::PointerLock::webkitUnlock):
3784         (WebCore::PointerLock::webkitIsLocked):
3785         * page/PointerLock.h: Renamed from Source/WebCore/page/MouseLockable.h.
3786         (WebCore::PointerLock::create):
3787         * page/PointerLock.idl: Renamed from Source/WebCore/page/MouseLockable.idl.
3788
3789 2011-10-28  Ojan Vafai  <ojan@chromium.org>
3790
3791         Overflow and relayout are broken in the new flexboxes
3792         https://bugs.webkit.org/show_bug.cgi?id=71161
3793
3794         Reviewed by David Hyatt.
3795
3796         Tests: css3/flexbox/auto-height-dynamic.html
3797                css3/flexbox/flex-item-child-overflow-expected.html
3798                css3/flexbox/flex-item-child-overflow.html
3799
3800         * rendering/RenderFlexibleBox.cpp:
3801         (WebCore::RenderFlexibleBox::layoutBlock):
3802         -Always set the logical height to 0 to start with to ensure we don't
3803         use the height from the previous layout when we are computing the
3804         intrinsic size of the flexbox.
3805         -Call computeOverflow after computeLogicalHeight so that flex-item's children's
3806         overflow is properly rendered.
3807
3808         (WebCore::RenderFlexibleBox::layoutAndPlaceChildrenInlineDirection):
3809         -Now that we setLogicalHeight in layoutBlock, we no longer need to do it here.
3810         -Refactor flipping code. The behavior is the same, but the variable names are just
3811         more correct.
3812
3813 2011-10-28  Ojan Vafai  <ojan@chromium.org>
3814
3815         Overflow and relayout are broken in the new flexboxes
3816         https://bugs.webkit.org/show_bug.cgi?id=71161
3817
3818         Reviewed by David Hyatt.
3819
3820         Tests: css3/flexbox/auto-height-dynamic.html
3821                css3/flexbox/flex-item-child-overflow-expected.html
3822                css3/flexbox/flex-item-child-overflow.html
3823
3824         * rendering/RenderFlexibleBox.cpp:
3825         (WebCore::RenderFlexibleBox::layoutBlock):
3826         -Always set the logical height to 0 to start with to ensure we don't
3827         use the height from the previous layout when we are computing the
3828         intrinsic size of the flexbox.
3829         -Call computeOverflow after computeLogicalHeight so that flex-item's children's
3830         overflow is properly rendered.
3831
3832         (WebCore::RenderFlexibleBox::layoutAndPlaceChildrenInlineDirection):
3833         -Now that we setLogicalHeight in layoutBlock, we no longer need to do it here.
3834         -Refactor flipping code. The behavior is the same, but the variable names are just
3835         more correct.
3836
3837 2011-11-15  Nate Chapin  <japhet@chromium.org>
3838
3839         CachedResourceRequest is now the only SubresourceLoaderClient
3840         Merge CachedResourceRequest into SubresourceLoader and delete
3841         the SubresourceLoaderClient interface. A few items were moved
3842         to CachedResource instead of SubresourceLoader.
3843         https://bugs.webkit.org/show_bug.cgi?id=71149
3844
3845         Reviewed by Adam Barth.
3846
3847         * CMakeLists.txt:
3848         * GNUmakefile.list.am:
3849         * WebCore.gypi:
3850         * WebCore.pro:
3851         * WebCore.vcproj/WebCore.vcproj:
3852         * WebCore.xcodeproj/project.pbxproj:
3853         * loader/ResourceLoadScheduler.cpp:
3854         * loader/ResourceLoadScheduler.h:
3855         * loader/ResourceLoader.cpp:
3856         * loader/SubresourceLoader.cpp:
3857         (WebCore::SubresourceLoader::SubresourceLoader):
3858         (WebCore::SubresourceLoader::create):
3859         (WebCore::SubresourceLoader::init): Do work that had previously been
3860             done in SubresourceLoader::create() after the constructor.
3861         (WebCore::SubresourceLoader::willSendRequest):
3862         (WebCore::SubresourceLoader::didSendData):
3863         (WebCore::SubresourceLoader::didReceiveResponse):
3864         (WebCore::SubresourceLoader::didReceiveData):
3865         (WebCore::SubresourceLoader::didReceiveCachedMetadata):
3866         (WebCore::SubresourceLoader::didFinishLoading):
3867         (WebCore::SubresourceLoader::didFail):
3868         (WebCore::SubresourceLoader::willCancel):
3869         (WebCore::SubresourceLoader::releaseResources): Do the cleanup work that was
3870             duplicated throughout the various terminal CachedResourceRequest callbacks.
3871         * loader/SubresourceLoader.h: Fix indentation style issues.
3872         * loader/SubresourceLoaderClient.h: Removed.
3873         * loader/cache/CachedImage.cpp:
3874         * loader/cache/CachedRawResource.cpp:
3875         * loader/cache/CachedResource.cpp:
3876         (WebCore::cachedResourceTypeToTargetType):
3877         (WebCore::CachedResource::load): Do the work that had been done in
3878             CachedResourceRequest::load().
3879         (WebCore::CachedResource::finish):
3880         (WebCore::CachedResource::setResponse):
3881         (WebCore::CachedResource::stopLoading):
3882         * loader/cache/CachedResource.h:
3883         (WebCore::CachedResource::canDelete):
3884         * loader/cache/CachedResourceLoader.cpp:
3885         * loader/cache/CachedResourceRequest.cpp: Removed.
3886         * loader/cache/CachedResourceRequest.h: Removed.
3887         * loader/cf/SubresourceLoaderCF.cpp:
3888         * loader/chromium/CachedResourceRequestChromium.cpp: Removed.
3889         * loader/chromium/SubresourceLoaderChromium.cpp:
3890
3891 2011-11-15  Anders Carlsson  <andersca@apple.com>
3892
3893         HostWindow screenToWindow/windowToScreen should be screenToRootView/rootViewToScreen
3894         https://bugs.webkit.org/show_bug.cgi?id=72397
3895
3896         Reviewed by Dan Bernstein.
3897
3898         screenToWindow and windowToScreen already use root view coordinates everywhere, with the
3899         exception of Mac WebKit1 which doesn't even implement the functions.
3900
3901         * accessibility/mac/WebAccessibilityObjectWrapper.mm:
3902         (-[WebAccessibilityObjectWrapper position]):
3903         * loader/EmptyClients.h:
3904         * page/Chrome.cpp:
3905         (WebCore::Chrome::screenToRootView):
3906         (WebCore::Chrome::rootViewToScreen):
3907         * page/Chrome.h:
3908         * page/ChromeClient.h:
3909         * platform/HostWindow.h:
3910         * platform/ScrollView.cpp:
3911         (WebCore::ScrollView::contentsToScreen):
3912         (WebCore::ScrollView::screenToContents):
3913         * platform/chromium/PopupContainer.cpp:
3914         (WebCore::PopupContainer::layoutAndCalculateWidgetRect):
3915         (WebCore::PopupContainer::refresh):
3916
3917 2011-11-15  Jeff Timanus  <twiz@chromium.org>
3918
3919         Patch removing duplicated code in the setup of the DrawingBuffer used
3920         to host the back-buffer for WebGL contents.
3921         https://bugs.webkit.org/show_bug.cgi?id=72327
3922
3923         Reviewed by Julien Chaffraix.
3924
3925         * html/canvas/WebGLRenderingContext.cpp:
3926         (WebCore::WebGLRenderingContext::copyTexImage2D):
3927         (WebCore::WebGLRenderingContext::copyTexSubImage2D):
3928         (WebCore::WebGLRenderingContext::readPixels):
3929
3930 2011-11-15  Eugene Nalimov  <enal@google.com>
3931
3932         Event listener for active DOM object that is also DOM node can be garbage collected prematurely.
3933         https://bugs.webkit.org/show_bug.cgi?id=70421 
3934
3935         Reviewed by Adam Barth.
3936
3937         Problem demonstrated itself when HTMLAudioElement was changed to become active DOM object.
3938         Before that there were no DOM objects that simultaneously were nodes and active objects.
3939         DOM object could be held in one of 3 maps -- node map, active objects map, and all other
3940         objects map, and HTMLAudioElement should be in 2 maps simultaneously. When it was in the
3941         active DOM objects map only, its event listener could be garbage collected, because special
3942         code that groups listeners with wrappers could handle only wrappers for objects in the node
3943         map. If we put HTMLAudioElement into nodes map, it would not be active DOM node, and can be
3944         garbage collected prematurely itself (see https://bugs.webkit.org/show_bug.cgi?id=66878).
3945         Fix is to introduce 4th map -- active nodes map, and change the code accordingly.
3946
3947         Test: media/audio-garbage-collect.html
3948
3949         * bindings/scripts/CodeGeneratorV8.pm:
3950         (GenerateNamedConstructorCallback):
3951         (GetDomMapFunction):
3952         * bindings/v8/DOMDataStore.cpp:
3953         (WebCore::DOMDataStore::DOMDataStore):
3954         (WebCore::DOMDataStore::getDOMWrapperMap):
3955         (WebCore::DOMDataStore::weakNodeCallback):
3956         * bindings/v8/DOMDataStore.h:
3957         (WebCore::DOMDataStore::activeDomNodeMap):
3958         * bindings/v8/ScopedDOMDataStore.cpp:
3959         (WebCore::ScopedDOMDataStore::ScopedDOMDataStore):
3960         (WebCore::ScopedDOMDataStore::~ScopedDOMDataStore):
3961         * bindings/v8/StaticDOMDataStore.cpp:
3962         (WebCore::StaticDOMDataStore::StaticDOMDataStore):
3963         * bindings/v8/StaticDOMDataStore.h:
3964         * bindings/v8/V8DOMMap.cpp:
3965         (WebCore::getActiveDOMNodeMap):
3966         (WebCore::removeAllDOMObjects):
3967         (WebCore::visitActiveDOMNodes):
3968         * bindings/v8/V8DOMMap.h:
3969         * bindings/v8/V8DOMWrapper.cpp:
3970         (WebCore::V8DOMWrapper::setJSWrapperForDOMNode):
3971         (WebCore::V8DOMWrapper::getWrapperSlow):
3972         * bindings/v8/V8GCController.cpp:
3973         (WebCore::GCPrologueSpecialCase):
3974         (WebCore::void):
3975         (WebCore::Node):
3976         (WebCore::GCPrologueVisitor::visitDOMWrapper):
3977         (WebCore::V8GCController::gcPrologue):
3978         (WebCore::GCEpilogueHelper::GCEpilogueSpecialCase):
3979         (WebCore::GCEpilogueVisitor::visitDOMWrapper):
3980         (WebCore::V8GCController::gcEpilogue):
3981         * dom/Node.h:
3982         (WebCore::Node::isActiveNode):
3983         * html/HTMLAudioElement.h:
3984         (WebCore::HTMLAudioElement::isActiveNode):
3985
3986 2011-11-15  David Kilzer  <ddkilzer@apple.com>
3987
3988         Remove useless const modifier from KURL::init
3989         <http://webkit.org/b/72387>
3990
3991         Reviewed by Darin Adler.
3992
3993         * platform/KURL.cpp:
3994         (WebCore::KURL::init): Remove useless const.
3995
3996 2011-11-14  Anders Carlsson  <andersca@apple.com>
3997
3998         HostWindow invalidation functions should use root view coordinates
3999         https://bugs.webkit.org/show_bug.cgi?id=72338
4000
4001         Reviewed by Dan Bernstein.
4002
4003         Rename invalidateWindow to invalidateRootView, and invalidateContentsAndWindow
4004         to invalidateContentsAndRootView. Make sure that the rects passed to the renamed functions
4005         are in root view coordinates by changing contentsToWindow calls to contentsToRootView.
4006         
4007         In practice this doesn't matter because for all platforms except Mac WebKit1, root view coordinates
4008         and window coordinates are equivalent, and Mac WebKit1 doesn't use these invalidation functions.
4009
4010         * loader/EmptyClients.h:
4011         * page/Chrome.cpp:
4012         (WebCore::Chrome::invalidateRootView):
4013         (WebCore::Chrome::invalidateContentsAndRootView):
4014         * page/Chrome.h:
4015         * page/ChromeClient.h:
4016         * page/Frame.cpp:
4017         (WebCore::Frame::tiledBackingStorePaintEnd):
4018         * page/FrameView.cpp:
4019         (WebCore::FrameView::invalidateRect):
4020         (WebCore::FrameView::scrollContentsFastPath):
4021         * platform/HostWindow.h:
4022         * platform/ScrollView.cpp:
4023         (WebCore::ScrollView::rectToCopyOnScroll):
4024         (WebCore::ScrollView::scrollContents):
4025         (WebCore::ScrollView::wheelEvent):
4026         * platform/chromium/FramelessScrollView.cpp:
4027         (WebCore::FramelessScrollView::invalidateRect):
4028         * svg/graphics/SVGImage.cpp:
4029         (WebCore::SVGImageChromeClient::invalidateContentsAndRootView):
4030
4031 2011-11-15  Philip Rogers  <pdr@google.com>
4032
4033         Implement maxWidth for fillText and strokeText, fixing the canvas/philip/tests/2d.text.draw.fill.maxWidth.fontface.html test.
4034         https://bugs.webkit.org/show_bug.cgi?id=61528
4035
4036         Reviewed by Stephen White.
4037
4038         Tests: fast/canvas/2d.text.draw.fill.maxWidth.gradient.html
4039                fast/canvas/2d.text.draw.fill.maxWidth.negative.html
4040                fast/canvas/2d.text.draw.fill.maxWidth.veryLarge.html
4041                fast/canvas/2d.text.draw.fill.maxWidth.verySmall.html
4042
4043         * html/canvas/CanvasRenderingContext2D.cpp:
4044         (WebCore::CanvasRenderingContext2D::drawTextInternal):
4045
4046 2011-11-15  Carlos Garcia Campos  <cgarcia@igalia.com>
4047
4048         [GTK] Implement download support in WebKit2
4049         https://bugs.webkit.org/show_bug.cgi?id=72258
4050
4051         Reviewed by Martin Robinson.
4052
4053         Add common download errors to ErrorsGtk.
4054
4055         * platform/gtk/ErrorsGtk.cpp:
4056         (WebCore::downloadNetworkError):
4057         (WebCore::downloadCancelledByUserError):
4058         (WebCore::downloadDestinationError):
4059         * platform/gtk/ErrorsGtk.h:
4060
4061 2011-11-15  Cary Clark  <caryclark@google.com>
4062
4063         [chromium-mac] Enable vertical text using Skia
4064         https://bugs.webkit.org/show_bug.cgi?id=72137
4065
4066         Use Skia to draw vertical text. This is much
4067         faster and has higher fidelity than the old method
4068         of drawing text on a path.
4069
4070         The graphics context passed to Skia has been
4071         rotated 90 degrees but the character advances
4072         have not, so it is necessary to unrotate the canvas,
4073         and re-rotate the positions.
4074
4075         This generates correct output (or, at least,
4076         consistent with Chromium CG on Mac) for all vertical
4077         text tests, one of which is mentioned below.
4078         
4079         Reviewed by Stephen White.
4080
4081         Test: fast/writing-mode/text-orientation-basic.html
4082
4083         * platform/graphics/skia/FontSkia.cpp:
4084         (WebCore::setupPaint):
4085         (WebCore::Font::drawGlyphs):
4086
4087 2011-11-15  Philip Rogers  <pdr@google.com>
4088
4089         Fix SVG hit testing when padding is present
4090         https://bugs.webkit.org/show_bug.cgi?id=37325
4091
4092         Reviewed by Nikolas Zimmermann.
4093
4094         Test: svg/hittest/svg-padding.xhtml
4095
4096         * rendering/svg/RenderSVGRoot.cpp:
4097         (WebCore::RenderSVGRoot::nodeAtPoint):
4098
4099 2011-11-15  Pavel Feldman  <pfeldman@google.com>
4100
4101         Web Inspector: move generic code from DevTools.js into the WebCore.
4102         https://bugs.webkit.org/show_bug.cgi?id=72377