2010-05-05 Alejandro G. Castro <alex@igalia.com>
[WebKit-https.git] / WebCore / ChangeLog
1 2010-05-05  Alejandro G. Castro  <alex@igalia.com>
2
3         Reviewed by Xan Lopez.
4
5         Fixed the gobject introspection compilation with the new DOM
6         bindings, we needed to add DOM objects.
7
8         * GNUmakefile.am:
9
10 2010-05-06  Adam Barth  <abarth@webkit.org>
11
12         Reviewed by Eric Seidel.
13
14         drop support for sessionStorage in sandbox iframes
15         https://bugs.webkit.org/show_bug.cgi?id=38151
16
17         This patch causes us to throw a security exception when a sandboxed
18         iframe attempts to access sessionStorage, matching our behavior for
19         localStorage.  The letter of the spec asks us to create a separate
20         storage area for each unique origin.  We might want to do that in a
21         future patch, but throwing a security error seems like a safe move now.
22
23         * page/DOMWindow.cpp:
24         (WebCore::DOMWindow::sessionStorage):
25         (WebCore::DOMWindow::localStorage):
26         * page/DOMWindow.h:
27         * page/DOMWindow.idl:
28         * page/SecurityOrigin.h:
29         (WebCore::SecurityOrigin::canAccessSessionStorage):
30         * storage/StorageEventDispatcher.cpp:
31         (WebCore::StorageEventDispatcher::dispatch):
32
33 2010-05-06  Andy Estes  <aestes@apple.com>
34
35         Reviewed by Maciej Stachowiak.
36
37         Allow forms submitting to target="_blank" to open popups if the submission
38         originated from a user gesture.
39
40         https://bugs.webkit.org/show_bug.cgi?id=37335
41         <rdar://problem/7884980>
42
43         Test: fast/events/popup-allowed-from-gesture-initiated-form-submit.html
44
45         * bindings/js/ScriptController.cpp:
46         (WebCore::ScriptController::processingUserGesture): If no DOM event is
47         being processed, consult UserGestureIndicator to determine return value.
48         * bindings/js/ScriptController.h: Moved the logic of processingUserGestureEvent()
49         into processingUserGesture().
50
51 2010-05-06  Xan Lopez  <xlopez@igalia.com>
52
53         Rubber-stamped by Eric Seidel.
54
55         [GTK] Refactor GenerateProperties in CodeGenerationGObject.pm
56         https://bugs.webkit.org/show_bug.cgi?id=38577
57
58         Thinko caught while refactoring, $custom variable was not
59         initialized.
60
61         * bindings/scripts/CodeGeneratorGObject.pm:
62         * bindings/scripts/test/GObject/WebKitDOMTestObj.cpp:
63         (webkit_dom_test_obj_set_property):
64         (webkit_dom_test_obj_get_property):
65         (webkit_dom_test_obj_class_init):
66
67 2010-05-06  Xan Lopez  <xlopez@igalia.com>
68
69         Rubber-stamped by Eric Seidel.
70
71         [GTK] Refactor GenerateProperties in CodeGenerationGObject.pm
72         https://bugs.webkit.org/show_bug.cgi?id=38577
73
74         Refactor GenerateProperty out of GenerateProperties. This is in
75         preparation for EventListeren attributes, which won't generate
76         normal GObject properties.
77
78         * bindings/scripts/CodeGeneratorGObject.pm:
79
80 2010-05-06  Pavel Feldman  <pfeldman@chromium.org>
81
82         Reviewed by Joseph Pecoraro.
83
84         Web Inspector: follow up to linkify event listeners.
85
86         https://bugs.webkit.org/show_bug.cgi?id=38257
87
88         * bindings/js/ScriptEventListener.cpp:
89         (WebCore::eventListenerHandlerLocation):
90         * bindings/v8/ScriptEventListener.cpp:
91         (WebCore::eventListenerHandlerLocation):
92         * inspector/front-end/ElementsPanel.js:
93         (WebInspector.ElementsPanel.prototype.linkifyNodeReference):
94         * inspector/front-end/EventListenersSidebarPane.js:
95
96 2010-05-05  Charles Wei  <charles.wei@torchmobile.com.cn>
97
98         Reviewed by George Staikos
99
100         https://bugs.webkit.org/show_bug.cgi?id=37848
101         This patch adds WCSS -wap-input-format and -wap-input-required support to WebKit
102
103         Tests: fast/wcss/wap-input-format.xhtml
104                fast/wcss/wap-input-required.xhtml
105
106         * css/CSSParser.cpp:
107         (WebCore::CSSParser::parseValue):
108         (WebCore::CSSParser::parseWCSSInputProperty):
109         * css/CSSParser.h:
110         * css/CSSStyleSelector.cpp:
111         (WebCore::CSSStyleSelector::applyProperty):
112         * css/WCSSPropertyNames.in:
113         * dom/InputElement.cpp:
114         (WebCore::InputElement::sanitizeValue):
115         (WebCore::InputElement::handleBeforeTextInsertedEvent):
116         (WebCore::InputElementData::InputElementData):
117         (WebCore::formatCodes):
118         (WebCore::cursorPositionToMaskIndex):
119         (WebCore::InputElement::isConformToInputMask):
120         (WebCore::InputElement::validateInputMask):
121         * dom/InputElement.h:
122         (WebCore::InputElementData::inputFormatMask):
123         (WebCore::InputElementData::setInputFormatMask):
124         (WebCore::InputElementData::maxInputCharsAllowed):
125         (WebCore::InputElementData::setMaxInputCharsAllowed):
126         * html/HTMLInputElement.cpp:
127         (WebCore::HTMLInputElement::setWapInputFormat):
128         * html/HTMLInputElement.h:
129         (WebCore::HTMLInputElement::data):
130         * wml/WMLInputElement.h:
131         (WebCore::WMLInputElement::data):
132
133 2010-05-05  MORITA Hajime  <morrita@google.com>
134
135         Reviewed by Darin Adler.
136
137         https://bugs.webkit.org/show_bug.cgi?id=38150
138         Refactoring: webkitEditableContentChangedEvent should be handled by the owner of appropriate the renderer.
139
140         Moved a part of code chunk in HTMLFormControlElementWithState::defaultEventHandler()
141         which accesses the renderer from foreign node, 
142         to TextControlInnerTextElement::defaultEventHandler() which owns the renderer.
143         
144         No new tests. No behavioral change.
145
146         * html/HTMLFormControlElement.cpp:
147         (WebCore::HTMLFormControlElementWithState::defaultEventHandler):
148         * html/HTMLFormControlElement.h:
149         * html/HTMLInputElement.h:
150         * html/HTMLTextAreaElement.h:
151         * rendering/TextControlInnerElements.cpp:
152         (WebCore::TextControlInnerTextElement::defaultEventHandler):
153
154 2010-05-05  Joseph Pecoraro  <joepeck@webkit.org>
155
156         Reviewed by Timothy Hatcher.
157
158         Web Inspector: Line Numbers should be Aligned with the Source Code Line
159         https://bugs.webkit.org/show_bug.cgi?id=38593
160
161         If there is a message bubble, such as an error message, in a source view
162         the line number should be in line with the source code line, not centered
163         between the source and bubble.
164
165         * inspector/front-end/textViewer.css:
166         (.webkit-line-number): added vertical-align: top
167
168 2010-05-05  Dan Bernstein  <mitz@apple.com>
169
170         Fix the decelerated compositing build.
171
172         * page/FrameView.cpp:
173         (WebCore::FrameView::isEnclosedInCompositingLayer):
174         * page/FrameView.h:
175
176 2010-05-05  Gavin Barraclough  <barraclough@apple.com>
177
178         Reviewed by Darin Adler.
179
180         Bug 38604 - workers-gc2 crashing on Qt
181
182         This appears to be due to a couple of issues.
183         (1) When the atomic string table is deleted it does not clear the 'inTable' bit
184         on AtomicStrings - it implicitly assumes that all AtomicStrings have already
185         been deleted at this point (otherwise they will crash in their destructor when
186         they try to remove themselves from the atomic string table).
187         (2) We don't fix the ordering in which WTF::WTFThreadData and
188         WebCore::ThreadGlobalData are destructed.
189
190         We should make sure that ThreadGlobalData is cleaned up before worker threads
191         terminate and WTF::WTFThreadData is destroyed, and we should clear the inTable
192         bit of members on atomic string table destruction.
193
194         WTF changes (fix issue 2, above) - clean up the thread data on worker termination.
195
196         * platform/ThreadGlobalData.cpp:
197         (WebCore::ThreadGlobalData::~ThreadGlobalData):
198         (WebCore::ThreadGlobalData::destroy):
199         * platform/ThreadGlobalData.h:
200         * workers/WorkerThread.cpp:
201         (WebCore::WorkerThread::workerThread):
202
203 2010-05-05  Dan Bernstein  <mitz@apple.com>
204
205         Reviewed by Simon Fraser.
206
207         <rdar://problem/7932072> Iframes in composited layers don’t repaint correctly (affects Yahoo! Mail with Flash Player 10.1)
208         https://bugs.webkit.org/show_bug.cgi?id=38427
209
210         Test: compositing/iframes/iframe-in-composited-layer.html
211
212         * WebCore.base.exp: Export FrameView::isEnclosedInCompositingLayer().
213         * page/FrameView.cpp:
214         (WebCore::FrameView::isEnclosedInCompositingLayer): Added this predicate.
215         * page/FrameView.h:
216
217 2010-05-05  Chris Marrin  <cmarrin@apple.com>
218
219         Reviewed by Simon Fraser.
220
221         Got composited iframes showing up on Mac again
222         https://bugs.webkit.org/show_bug.cgi?id=38565
223
224         This was broken by http://trac.webkit.org/changeset/58798. That
225         change connected iframes through the parent document to the root
226         of the layer tree. That is correct for Windows, but not for Mac.
227         So the places where the linkage is made were wrapped in 
228         shouldPropagateCompositingToIFrameParent() calls, which is 
229         always false for Mac.
230
231         Test: compositing/iframes/composited-iframe-alignment.html
232
233         * rendering/RenderLayerCompositor.cpp:Avoid doing composited iframe linkage on Mac
234         (WebCore::RenderLayerCompositor::rebuildCompositingLayerTree):
235         (WebCore::RenderLayerCompositor::didMoveOnscreen):
236         (WebCore::RenderLayerCompositor::willMoveOffscreen):
237         (WebCore::RenderLayerCompositor::ensureRootPlatformLayer):
238
239 2010-05-05  Alexey Proskuryakov  <ap@apple.com>
240
241         Reviewed by Darin Adler.
242
243         https://bugs.webkit.org/show_bug.cgi?id=38260
244         <rdar://problem/7917548> Fix whitespace removing in deprecatedParseURL().
245
246         Broken all the way since r4 (yes, that's a revision number).
247
248         Test: http/tests/security/xss-DENIED-javascript-with-spaces.html
249
250         * css/CSSHelper.cpp: (WebCore::deprecatedParseURL): Fixed loop conditions for remaining length.
251
252 2010-05-05  Chris Fleizach  <cfleizach@apple.com>
253
254         Reviewed by Beth Dakin.
255
256         Many AXUnknown showing up in the hierarchy
257         https://bugs.webkit.org/show_bug.cgi?id=38607
258
259         The RenderTableSection should not be an accessible element.
260
261         Test: platform/mac/accessibility/parent-of-table-row-is-table.html
262
263         * accessibility/AccessibilityRenderObject.cpp:
264         (WebCore::AccessibilityRenderObject::determineAccessibilityRole):
265
266 2010-05-05  Csaba Osztrogonác  <ossy@webkit.org>
267
268         Reviewed by Simon Hausmann.
269
270         [Qt]Disable Qt Multimedia backend for HTML 5 Audio and Video elements
271         Fall back to the Phonon backend until the release situation has been cleared up.
272
273         https://bugs.webkit.org/show_bug.cgi?id=38612
274
275         Buildfix for QT_VERSION >= 0x040700 after r58810.
276
277         * platform/graphics/MediaPlayer.cpp:
278
279 2010-05-05  Jian Li  <jianli@chromium.org>
280
281         Reviewed by David Levin.
282
283         XMLHttpRequestUpload events do not fire when sending a raw file or FormData object.
284         https://bugs.webkit.org/show_bug.cgi?id=37771
285
286         Test: http/tests/local/formdata/upload-events.html
287
288         * xml/XMLHttpRequest.cpp:
289         (WebCore::XMLHttpRequest::createRequest):
290
291 2010-05-05  Csaba Osztrogonác  <ossy@webkit.org>
292
293         Reviewed by Simon Hausmann.
294
295         [Qt] Qt port should use its own QUuid class to create UUID string.
296         https://bugs.webkit.org/show_bug.cgi?id=38581
297
298         * platform/UUID.cpp:
299         (WebCore::createCanonicalUUIDString): Added PLATFORM(QT) case.
300
301 2010-05-05  Alexey Proskuryakov  <ap@apple.com>
302
303         Roll out r58830 for breaking tests.
304
305         Was: https://bugs.webkit.org/show_bug.cgi?id=38546
306         Node.focus() fails to move focus from subframe properly
307
308         * html/HTMLFrameElementBase.cpp:
309         (WebCore::HTMLFrameElementBase::setFocus):
310
311 2010-05-05  Jian Li  <jianli@chromium.org>
312
313         Reviewed by Adam Barth.
314
315         Implement FileReader class.
316         https://bugs.webkit.org/show_bug.cgi?id=38157
317
318         This patch only contains the implementation of FileReader class as defined
319         in the File API spec: http://www.w3.org/TR/file-upload/#dfn-filereader.
320
321         New test will be added when a IDL is exposed and the FILE_READER is turned
322         on.
323
324         * GNUmakefile.am:
325         * WebCore.gypi:
326         * WebCore.pro:
327         * WebCore.vcproj/WebCore.vcproj:
328         * WebCore.xcodeproj/project.pbxproj:
329         * dom/EventNames.h:
330         * dom/EventTarget.cpp:
331         (WebCore::EventTarget::toFileReader):
332         * dom/EventTarget.h:
333         * html/FileError.h:
334         (WebCore::FileError::FileError):
335         * html/FileReader.cpp: Added.
336         * html/FileReader.h: Added.
337         * html/FileStream.cpp:
338         (WebCore::FileStream::FileStream):
339         (WebCore::FileStream::openForRead):
340         (WebCore::FileStream::close):
341         (WebCore::FileStream::read):
342         * html/FileStream.h:
343
344 2010-05-05  Steve Falkenburg  <sfalken@apple.com>
345
346         Reviewed by Maciej Stachowiak.
347
348         REGRESSION(r57969) Image decoder is repeatedly destroyed/created in CoreGraphics-based Windows WebKit
349         https://bugs.webkit.org/show_bug.cgi?id=38595
350
351         * platform/graphics/cg/ImageSourceCG.cpp:
352         (WebCore::ImageSource::setData):
353
354 2010-05-05  Alexey Proskuryakov  <ap@apple.com>
355
356         Reviewed by Adele Peterson.
357
358         https://bugs.webkit.org/show_bug.cgi?id=38546
359         Node.focus() fails to move focus from subframe properly
360
361         Test: fast/frames/take-focus-from-iframe.html
362
363         * html/HTMLFrameElementBase.cpp: (WebCore::HTMLFrameElementBase::setFocus): Don't clear
364         focus if this frame doesn't have it. This can happen if page's and HTMLFrameElement's ideas
365         of focused frame get out of sync temporarily.
366
367 2010-05-05  Alexey Proskuryakov  <ap@apple.com>
368
369         Reviewed by Adele Peterson.
370
371         https://bugs.webkit.org/show_bug.cgi?id=26824
372         <rdar://problem/7018610> EventHandler can operate on a wrong frame if focus changes during
373         keyboard event dispatch.
374
375         EventHandler object is tied to a frame, so it's wrong for it to continue processing a keyboard
376         event if focused frame changes between keydown and keypress.
377
378         * manual-tests/focus-change-between-key-events.html: Added.
379
380         * page/EventHandler.cpp: (WebCore::EventHandler::keyEvent): Bail out early if focused frame
381         changes while dispatching keydown. Also made similar changes for Windows to maintain matching
382         behavior, even though EventHandler was re-entered anyway due to WM_KEYDOWN and WM_CHAR being
383         separate events.
384
385 2010-05-05  Steve Block  <steveblock@google.com>
386
387         Reviewed by Adam Barth.
388
389         MediaError.h is missing PassRefPtr.h include
390         https://bugs.webkit.org/show_bug.cgi?id=38575
391
392         No new tests, build fix only.
393
394         * html/MediaError.h:
395
396 2010-05-05  Alexander Pavlov  <apavlov@chromium.org>
397
398         Reviewed by Pavel Feldman.
399
400         Web Inspector: Audits panel: Resource counters get reset when switching panels while reloading page
401         https://bugs.webkit.org/show_bug.cgi?id=38579
402
403         * inspector/front-end/AuditLauncherView.js:
404         (WebInspector.AuditLauncherView.prototype.updateResourceTrackingState):
405         (WebInspector.AuditLauncherView.prototype._updateResourceProgress):
406
407 2010-05-05  Pavel Feldman  <pfeldman@chromium.org>
408
409         Reviewed by Joseph Pecoraro.
410
411         Web Inspector: Doubleclick on line number should not trigger source editing.
412
413         https://bugs.webkit.org/show_bug.cgi?id=38588
414
415         * inspector/front-end/TextViewer.js:
416         (WebInspector.TextViewer.prototype._handleDoubleClick):
417
418 2010-05-05  Martin Robinson  <mrobinson@webkit.org>
419
420         Reviewed by Gustavo Noronha.
421
422         [GTK] Enable DOM clipboard and drag-and-drop access
423         https://bugs.webkit.org/show_bug.cgi?id=30623
424
425         Add support in the PasteboardHelper for images, in preparation for drag-and-drop data.
426
427         No new tests as functionality has not changed.
428
429         * platform/gtk/PasteboardHelper.cpp:
430         (WebCore::PasteboardHelper::fillSelectionData): Add support for images here.
431         (WebCore::PasteboardHelper::targetListForDataObject): Add support for images here.
432
433 2010-05-05  Stuart Morgan  <stuartmorgan@chromium.org>
434
435         Reviewed by Darin Fisher.
436
437         Add a "focused" parameter to Widget::setFocus, and updates Document
438         so that Widget is informed of focus loss as well as focus gain.
439         Changes all existing setFocus implementations so that they ignore
440         the 'false' case, so no behavior is changed until individual
441         platforms are updated to handle the new case (if they want to).
442
443         https://bugs.webkit.org/show_bug.cgi?id=37961
444
445         * dom/Document.cpp:
446         (WebCore::Document::setFocusedNode):
447         * platform/Widget.h:
448         * platform/android/WidgetAndroid.cpp:
449         (WebCore::Widget::setFocus):
450         * platform/chromium/WidgetChromium.cpp:
451         (WebCore::Widget::setFocus):
452         * platform/efl/WidgetEfl.cpp:
453         (WebCore::Widget::setFocus):
454         * platform/gtk/WidgetGtk.cpp:
455         (WebCore::Widget::setFocus):
456         * platform/haiku/WidgetHaiku.cpp:
457         (WebCore::Widget::setFocus):
458         * platform/mac/WidgetMac.mm:
459         (WebCore::Widget::setFocus):
460         * platform/qt/WidgetQt.cpp:
461         (WebCore::Widget::setFocus):
462         * platform/win/WidgetWin.cpp:
463         (WebCore::Widget::setFocus):
464         * platform/wx/WidgetWx.cpp:
465         (WebCore::Widget::setFocus):
466         * plugins/PluginView.h:
467         * plugins/PluginViewNone.cpp:
468         (WebCore::PluginView::setFocus):
469         * plugins/gtk/PluginViewGtk.cpp:
470         (WebCore::PluginView::setFocus):
471         * plugins/mac/PluginViewMac.cpp:
472         (WebCore::PluginView::setFocus):
473         * plugins/qt/PluginViewQt.cpp:
474         (WebCore::PluginView::setFocus):
475         * plugins/symbian/PluginViewSymbian.cpp:
476         (WebCore::PluginView::setFocus):
477         * plugins/win/PluginViewWin.cpp:
478         (WebCore::PluginView::setFocus):
479
480 2010-05-05  Steve Block  <steveblock@google.com>
481
482         Reviewed by Pavel Feldman.
483
484         Move V8 ScriptDebugServer::topStackFrame() to ScriptCallStack
485         https://bugs.webkit.org/show_bug.cgi?id=38531
486
487         ScriptCallStack::create() uses ScriptDebugServer::topStackFrame(), but ScriptDebugServer
488         is guarded with ENABLE(JAVASCRIPT_DEBUGGER). This prevents ScriptCallStack from being
489         built on platforms that do not define ENABLE(JAVASCRIPT_DEBUGGER).
490
491         No new tests, build fix only.
492
493         * bindings/v8/ScriptCallStack.cpp:
494         (WebCore::ScriptCallStack::callLocation):
495         (WebCore::ScriptCallStack::createUtilityContext): Helper for topStackFrame, moved from ScriptDebugServer
496         (WebCore::ScriptCallStack::topStackFrame): Moved from ScriptDebugServer
497         * bindings/v8/ScriptCallStack.h:
498         (WebCore::ScriptCallStack::utilityContext): Helper for topStackFrame, moved from ScriptDebugServer
499         * bindings/v8/ScriptDebugServer.cpp:
500         * bindings/v8/ScriptDebugServer.h:
501
502 2010-05-05  Csaba Osztrogonác  <ossy@webkit.org>
503
504         Rubber-stamped by Simon Hausmann.
505
506         [Qt] REGRESSION(r56869): WinCE build is broken
507         https://bugs.webkit.org/show_bug.cgi?id=36929
508
509         * WebCore.pro: LIBS += -lOle32 added.
510
511 2010-05-05  Simon Hausmann  <simon.hausmann@nokia.com>
512
513         Reviewed by Tor Arne Vestbø.
514
515         [Qt] Disable Qt Multimedia backend for HTML 5 Audio and Video elements
516
517         Fall back to the Phonon backend until the release situation has been cleared up.
518
519         * WebCore.pro:
520
521 2010-05-05  Ilya Tikhonovsky  <loislo@chromium.org>
522
523         Reviewed by Pavel Feldman.
524
525         WebInspector: Clickable links to resources should be used as 'details' part for the all suitable kinds of records.
526         https://bugs.webkit.org/show_bug.cgi?id=38542
527
528         * inspector/front-end/TimelinePanel.js:
529         (WebInspector.TimelinePanel.prototype._innerAddRecordToTimeline):
530         (WebInspector.TimelineRecordListRow):
531         (WebInspector.TimelineRecordListRow.prototype.update):
532         (WebInspector.TimelinePanel.FormattedRecord):
533         (WebInspector.TimelinePanel.FormattedRecord.prototype._generatePopupContent):
534         (WebInspector.TimelinePanel.FormattedRecord.prototype._getRecordDetails):
535         (WebInspector.TimelinePanel.PopupContentHelper):
536         (WebInspector.TimelinePanel.PopupContentHelper.prototype._createCell):
537         (WebInspector.TimelinePanel.PopupContentHelper.prototype._appendTextRow):
538         (WebInspector.TimelinePanel.PopupContentHelper.prototype._appendElementRow):
539         (WebInspector.TimelinePanel.PopupContentHelper.prototype._appendLinkRow):
540
541 2010-05-05  Nikolas Zimmermann  <nzimmermann@rim.com>
542
543         Not reviewed. Add missing files generated by "run-bindings-test", after the DOMTestCallback.idl addition.
544
545         * bindings/scripts/test/GObject/WebKitDOMTestCallback.cpp: Added.
546         (WebKit::kit):
547         (webkit_dom_test_callback_callback_with_class1param):
548         (webkit_dom_test_callback_callback_with_class2param):
549         (webkit_dom_test_callback_callback_with_non_bool_return_type):
550         (WebKit::wrapTestCallback):
551         (WebKit::core):
552         (webkit_dom_test_callback_finalize):
553         (webkit_dom_test_callback_set_property):
554         (webkit_dom_test_callback_get_property):
555         (webkit_dom_test_callback_class_init):
556         (webkit_dom_test_callback_init):
557         * bindings/scripts/test/GObject/WebKitDOMTestCallback.h: Added.
558         * bindings/scripts/test/GObject/WebKitDOMTestCallbackPrivate.h: Added.
559         * bindings/scripts/test/ObjC/DOMTestCallback.h: Added.
560         * bindings/scripts/test/ObjC/DOMTestCallback.mm: Added.
561         (-[DOMTestCallback dealloc]):
562         (-[DOMTestCallback finalize]):
563         (-[DOMTestCallback callbackWithClass1Param:]):
564         (-[DOMTestCallback callbackWithClass2Param:strArg:]):
565         (-[DOMTestCallback callbackWithNonBoolReturnType:]):
566         (-[DOMTestCallback customCallback:class6Param:]):
567         (core):
568         (kit):
569         * bindings/scripts/test/ObjC/DOMTestCallbackInternal.h: Added.
570
571 2010-05-05  Joseph Pecoraro  <joepeck@webkit.org>
572
573         Reviewed by Pavel Feldman.
574
575         Web Inspector: FontView needs Cleanup
576         https://bugs.webkit.org/show_bug.cgi?id=38567
577
578         FontView was missing some of the newer practices for initialization
579         that other Resource Views had. Also its width calculations could
580         result in a single character wrapping. This is general cleanup.
581
582         * inspector/front-end/FontView.js:
583         (WebInspector.FontView): move initialization into contentTabSelected
584         (WebInspector.FontView.prototype.contentTabSelected): handle initialization like other Resource Views
585         (WebInspector.FontView.prototype.updateFontPreviewSize): use a narrower width to prevent text from widowing
586
587 2010-05-05  Dirk Schulze  <krit@webkit.org>
588
589         Reviewed by Nikolas Zimmermann.
590
591         SVG hit testing is *way* too slow
592         https://bugs.webkit.org/show_bug.cgi?id=19312
593
594         Use the cached stroke and fill boundaries in RenderPath as a heuristik to
595         speed up SVG's hit testing.
596
597         No new tests added.
598
599         * rendering/RenderPath.cpp:
600         (WebCore::RenderPath::fillContains):
601         (WebCore::RenderPath::strokeContains):
602
603 2010-05-04  Dumitru Daniliuc  <dumi@chromium.org>
604
605         Reviewed by Adam Barth.
606
607         Auto-generate all Database callbacks.
608         https://bugs.webkit.org/show_bug.cgi?id=38503
609
610         * DerivedSources.cpp:
611         * DerivedSources.make:
612         * GNUmakefile.am:
613         * WebCore.gypi:
614         * WebCore.pri:
615         * WebCore.pro:
616         * WebCore.vcproj/WebCore.vcproj:
617         * WebCore.xcodeproj/project.pbxproj:
618         * bindings/js/JSBindingsAllInOne.cpp:
619         * bindings/js/JSCustomSQLStatementCallback.cpp: Removed.
620         * bindings/js/JSCustomSQLStatementCallback.h: Removed.
621         * bindings/js/JSCustomSQLStatementErrorCallback.cpp:
622         (WebCore::JSSQLStatementErrorCallback::handleEvent):
623         * bindings/js/JSCustomSQLStatementErrorCallback.h: Removed.
624         * bindings/js/JSCustomSQLTransactionCallback.cpp: Removed.
625         * bindings/js/JSCustomSQLTransactionCallback.h: Removed.
626         * bindings/js/JSCustomSQLTransactionErrorCallback.cpp: Removed.
627         * bindings/js/JSCustomSQLTransactionErrorCallback.h: Removed.
628         * bindings/js/JSDatabaseCallback.cpp: Removed.
629         * bindings/js/JSDatabaseCallback.h: Removed.
630         * bindings/js/JSDatabaseCustom.cpp:
631         (WebCore::JSDatabase::changeVersion):
632         (WebCore::createTransaction):
633         * bindings/js/JSSQLTransactionCustom.cpp:
634         (WebCore::JSSQLTransaction::executeSql):
635         * bindings/scripts/CodeGeneratorJS.pm:
636         * bindings/scripts/test/JS/JSTestCallback.cpp:
637         * bindings/v8/custom/V8CustomSQLStatementCallback.cpp: Removed.
638         * bindings/v8/custom/V8CustomSQLStatementCallback.h: Removed.
639         * bindings/v8/custom/V8CustomSQLStatementErrorCallback.cpp:
640         (WebCore::V8SQLStatementErrorCallback::handleEvent):
641         * bindings/v8/custom/V8CustomSQLStatementErrorCallback.h: Removed.
642         * bindings/v8/custom/V8CustomSQLTransactionCallback.cpp: Removed.
643         * bindings/v8/custom/V8CustomSQLTransactionCallback.h: Removed.
644         * bindings/v8/custom/V8CustomSQLTransactionErrorCallback.cpp: Removed.
645         * bindings/v8/custom/V8CustomSQLTransactionErrorCallback.h: Removed.
646         * bindings/v8/custom/V8DatabaseCallback.cpp: Removed.
647         * bindings/v8/custom/V8DatabaseCallback.h: Removed.
648         * bindings/v8/custom/V8DatabaseCustom.cpp:
649         (WebCore::V8Database::changeVersionCallback):
650         (WebCore::createTransaction):
651         * bindings/v8/custom/V8SQLTransactionCustom.cpp:
652         (WebCore::V8SQLTransaction::executeSqlCallback):
653         * storage/DatabaseCallback.h:
654         * storage/DatabaseCallback.idl: Added.
655         * storage/SQLStatement.cpp:
656         (WebCore::SQLStatement::performCallback):
657         * storage/SQLStatementCallback.h:
658         * storage/SQLStatementCallback.idl: Added.
659         * storage/SQLStatementErrorCallback.idl: Added.
660         * storage/SQLTransaction.cpp:
661         (WebCore::SQLTransaction::deliverTransactionCallback):
662         * storage/SQLTransactionCallback.h:
663         * storage/SQLTransactionCallback.idl: Added.
664         * storage/SQLTransactionErrorCallback.h:
665         * storage/SQLTransactionErrorCallback.idl: Added.
666
667 2010-05-04  Chris Marrin  <cmarrin@apple.com>
668
669         Reviewed by Simon Fraser.
670
671         Made composited iframes work on Windows
672         https://bugs.webkit.org/show_bug.cgi?id=32446
673         
674         This completes the work in http://trac.webkit.org/changeset/57919
675         to create compositing layers in the parent document when an iframe has 
676         a compositing layer. The parent document has a layer for the iframe 
677         element and builds a layer tree to the page root. The layer tree for the
678         iframe document is then parented to the iframe element's GraphicsLayer.
679         
680         The RenderLayerCompositor for the iframe document (which owns the
681         root of the layer tree) now has a clippingLayer which is the 
682         parent of the layer tree root so it can be clipped to the parent
683         iframe's bounds, taking into account borders, padding, etc. in
684         the parent iframe element.
685         
686         I also got rid of a no longer used function: RenderLayerCompositor::parentInRootLayer
687
688         Test: compositing/iframes/composited-parent-iframe.html
689
690         * rendering/RenderLayerBacking.cpp:Make calls to RenderLayerCompositor to set the clipping bounds for iframe content
691         * rendering/RenderLayerCompositor.cpp:Hook the iframe content to the parent iframe element
692         * rendering/RenderLayerCompositor.h:
693
694 2010-05-03  Alexey Proskuryakov  <ap@apple.com>
695
696         Reviewed by Adam Barth.
697
698         https://bugs.webkit.org/show_bug.cgi?id=38497
699         <rdar://problem/7759438> Make sure that http URLs always have a host in SecurityOrigin
700
701         This is a hardening fix, and behavior really depends on what an underlying networking layer
702         does. So, no test.
703
704         * page/SecurityOrigin.cpp:
705         (WebCore::schemeRequiresAuthority): List schemes that need an authority for successful loading.
706         (WebCore::SecurityOrigin::SecurityOrigin): Never let e.g. http origins with empty authorities
707         have the same security origin.
708
709 2010-05-04  Zhenyao Mo  <zmo@google.com>
710
711         Reviewed by Dimitri Glazkov.
712
713         getFramebufferAttachmentParameter should return the original WebGLTexture/WebGLRenderbuffer instead of creating new ones sharing names.
714         https://bugs.webkit.org/show_bug.cgi?id=38236
715
716         * html/canvas/CanvasObject.h: Add type check functions.
717         (WebCore::CanvasObject::isBuffer):
718         (WebCore::CanvasObject::isFramebuffer):
719         (WebCore::CanvasObject::isProgram):
720         (WebCore::CanvasObject::isRenderbuffer):
721         (WebCore::CanvasObject::isShader):
722         (WebCore::CanvasObject::isTexture):
723         * html/canvas/WebGLBuffer.h: Add type check functions.
724         (WebCore::WebGLBuffer::isBuffer):
725         * html/canvas/WebGLFramebuffer.h: Add type check functions.
726         (WebCore::WebGLFramebuffer::isFramebuffer):
727         * html/canvas/WebGLProgram.h: Add type check functions.
728         (WebCore::WebGLProgram::isProgram):
729         * html/canvas/WebGLRenderbuffer.cpp: remove constructor using existing name.
730         * html/canvas/WebGLRenderbuffer.h: Add type check functions; remove constructor using existing name.
731         (WebCore::WebGLRenderbuffer::isRenderbuffer):
732         * html/canvas/WebGLRenderingContext.cpp:
733         (WebCore::WebGLRenderingContext::getFramebufferAttachmentParameter): Return original Texture/Renderbuffer instead of creating new ones.
734         (WebCore::WebGLRenderingContext::findTexture): Find a WebGLTexture using a name.
735         (WebCore::WebGLRenderingContext::findRenderbuffer): Find a WebGLRenderbuffer using a name.
736         * html/canvas/WebGLRenderingContext.h: Add find* functions.
737         * html/canvas/WebGLShader.h: Add type check functions.
738         (WebCore::WebGLShader::isShader):
739         * html/canvas/WebGLTexture.cpp: remove constructor using existing name.
740         * html/canvas/WebGLTexture.h: Add type check functions; remove constructor using existing name.
741         (WebCore::WebGLTexture::isTexture):
742
743 2010-05-04  Luiz Agostini  <luiz.agostini@openbossa.org>
744
745         Reviewed by Simon Hausmann.
746
747         [Qt] QT_MOBILE_THEME compile time flag
748         https://bugs.webkit.org/show_bug.cgi?id=38439
749
750         Replacing preprocessor conditional used in RenderThemeQt from Q_WS_MAEMO_5 to
751         USE(QT_MOBILE_THEME).
752
753         * WebCore.pro:
754         * platform/qt/RenderThemeQt.cpp:
755         (WebCore::RenderThemeQt::RenderThemeQt):
756         (WebCore::RenderThemeQt::qStyle):
757         (WebCore::RenderThemeQt::extraDefaultStyleSheet):
758         (WebCore::RenderThemeQt::adjustMenuListButtonStyle):
759         (WebCore::RenderThemeQt::setPaletteFromPageClientIfExists):
760         * platform/qt/RenderThemeQt.h:
761
762 2010-05-04  Joseph Pecoraro  <joepeck@webkit.org>
763
764         Reviewed by Dan Bernstein.
765
766         SVG fonts trigger GlyphPage::fill with null font
767         https://bugs.webkit.org/show_bug.cgi?id=38530
768
769         SVG fonts do not use the glyph page cache. This change detects when
770         attempting to fill a GlyphPage with an SVG font and indicates that
771         the glyphs were not available.
772
773         * platform/graphics/GlyphPageTreeNode.cpp:
774         (WebCore::fill): helper method which handles attempts to fill a GlyphPage with SVG or non-SVG fonts
775         (WebCore::GlyphPageTreeNode::initializePage): use the helper where appropriate
776
777 2010-05-04  Kent Tamura  <tkent@chromium.org>
778
779         Reviewed by David Levin (and unofficially by Enrica Casucci).
780
781         REGRESSION(r54368): Text drag-and-drop from input/textarea doesn't work
782         if the text is like a URL
783         https://bugs.webkit.org/show_bug.cgi?id=38374
784
785         Since r54368, documentFragmentFromDragData() creates an <a> element
786         with no anchor text for URL-like text dragged from input/textarea
787         element. If such text is dropped to input/textarea element, the text
788         is not inserted.
789         To fix this problem, use the original text or the URL as an anchor
790         text.
791
792         Test: editing/pasteboard/drag-drop-url-text.html
793
794         * page/DragController.cpp:
795         (WebCore::documentFragmentFromDragData):
796
797 2010-05-04  Steve Block  <steveblock@google.com>
798
799         Reviewed by Darin Adler.
800
801         JavaInstanceJSC.cpp and JNIUtilityPrivate.cpp need to include jni_jsobject.h for jlong_to_pt() and ptr_to_jlong()
802         https://bugs.webkit.org/show_bug.cgi?id=38525
803
804         No new tests, build fix only.
805
806         * bridge/jni/jni_jsobject.h: Guard Mac-specific code with PLATFORM(MAC)
807         * bridge/jni/jsc/JNIUtilityPrivate.cpp: Include jni_jsobject.h
808         * bridge/jni/jsc/JavaInstanceJSC.cpp: Include jni_jsobject.h
809
810 2010-05-04  Steve Block  <steveblock@google.com>
811
812         Reviewed by Darin Adler.
813
814         New FileSystemPOSIX functions cause linker errors on Android
815         https://bugs.webkit.org/show_bug.cgi?id=38521
816
817         No new tests, build fix only.
818
819         * platform/android/FileSystemAndroid.cpp: Remove closeFile, which is provided in FileSystemPOSIX.cpp
820         * platform/posix/FileSystemPOSIX.cpp: Guard writeToFile with !PLATFORM(ANDROID)
821
822 2010-05-04  Pavel Feldman  <pfeldman@chromium.org>
823
824         Reviewed by Timothy Hatcher.
825
826         Web Inspector: restore main and sidebar scroller positions
827         when switching between panels.
828
829         https://bugs.webkit.org/show_bug.cgi?id=38522
830
831         * inspector/front-end/ElementsPanel.js:
832         (WebInspector.ElementsPanel.prototype.elementsToRestoreScrollPositionsFor):
833         * inspector/front-end/Panel.js:
834         (WebInspector.Panel.prototype.show):
835         (WebInspector.Panel.prototype.hide):
836         (WebInspector.Panel.prototype.elementsToRestoreScrollPositionsFor):
837         (WebInspector.Panel.prototype._storeScrollPositions):
838         (WebInspector.Panel.prototype._restoreScrollPositions):
839         * inspector/front-end/ResourcesPanel.js:
840         (WebInspector.ResourcesPanel.prototype.elementsToRestoreScrollPositionsFor):
841         * inspector/front-end/ScriptsPanel.js:
842         (WebInspector.ScriptsPanel.prototype.elementsToRestoreScrollPositionsFor):
843
844 2010-05-04  Steven Lai  <steven_lai@asia.apple.com>
845
846         Reviewed by Darin Adler.
847
848         https://bugs.webkit.org/show_bug.cgi?id=37900
849         <rdar://problem/7806164> cloneNode() does not preserve z-index with more than six digits 
850
851         Tests: fast/dom/HTMLInputElement/clone-input-checked.html
852                fast/dom/clone-node-z-index.html
853
854         * dom/Element.cpp: undid the order change of calling copyNonAttributeProperties() before setAttributes()
855         (WebCore::Element::cloneElementWithoutChildren): uncheck the previous radio button in the same radio button group only when the checked radio box is appended to the dom tree
856         * html/HTMLInputElement.cpp:
857         (WebCore::HTMLInputElement::updateCheckedRadioButtons):
858         (WebCore::HTMLInputElement::attach):
859         (WebCore::HTMLInputElement::setChecked):
860         * html/HTMLInputElement.h:
861
862 2010-04-29  John Gregg  <johnnyg@google.com>
863
864         Reviewed by Dmitry Titov.
865
866         notifications should have dir and replaceId attributes.
867         Note that tests are only enabled in Chromium, skipped elsewhere
868         since notifications aren't implemented.
869         https://bugs.webkit.org/show_bug.cgi?id=38336
870
871         Tests: fast/notifications/notifications-replace.html
872                fast/notifications/notifications-rtl.html
873
874         * notifications/Notification.h:
875         (WebCore::Notification::dir):
876         (WebCore::Notification::setDir):
877         (WebCore::Notification::replaceId):
878         (WebCore::Notification::setReplaceId):
879         * notifications/Notification.idl:
880
881 2010-05-04  Alejandro G. Castro  <alex@igalia.com>
882
883         Reviewed by Xan Lopez.
884
885         Fixed error defining the float/double minimum in the object
886         properties, the smallest value of a float/double is
887         -G_MAXDOUBLE/-G_MAXFLOAT.
888
889         * bindings/scripts/CodeGeneratorGObject.pm:
890
891 2010-05-03  Joseph Pecoraro  <joepeck@webkit.org>
892
893         Reviewed by Dan Bernstein.
894
895         REGRESSION: Text clipped in absolutely positioned search inputs
896         https://bugs.webkit.org/show_bug.cgi?id=38468
897
898         Previously I incorrectly added x() and y() to the tx and ty positioning.
899         This clips a search input like a text input, respecting the vertically
900         centered text.
901
902         * rendering/RenderTextControlSingleLine.cpp:
903         (WebCore::RenderTextControlSingleLine::layout): only the Y changes, so change setLocation to setY
904         (WebCore::RenderTextControlSingleLine::controlClipRect):
905
906 2010-05-04  Alexander Pavlov  <apavlov@chromium.org>
907
908         Reviewed by Timothy Hatcher.
909
910         Web Inspector: REGRESSION: Up/Down, PgUp/PgDown keys do not change numeric style properties
911         https://bugs.webkit.org/show_bug.cgi?id=38516
912
913         * inspector/front-end/StylesSidebarPane.js:
914         (WebInspector.StylePropertyTreeElement.prototype):
915
916 2010-05-04  Ben Murdoch  <benm@google.com>
917
918         Reviewed by Simon Hausmann.
919
920         Crash in handleTouchEvent: using dangling node ptrs in hashmap
921         https://bugs.webkit.org/show_bug.cgi?id=38514
922
923         When navigating away from a page, if you have your finger still
924         pressed and then lift it on the new page we see a crash if the
925         node got deleted as we still have a dangling pointer in the
926         m_originatingTouchPointTargets hashmap and try to use it as the
927         receiver to dispatch a touchend event.
928
929         Test: fast/events/touch/touch-stale-node-crash.html
930
931         * page/EventHandler.cpp:
932         (WebCore::EventHandler::clear): Clear the hashmap of touch targets.
933
934 2010-05-04  Joseph Pecoraro  <joepeck@webkit.org>
935
936         Reviewed by Pavel Feldman.
937
938         Web Inspector: Drawer Misbehaving when Docking / Undocking in Console Panel
939         https://bugs.webkit.org/show_bug.cgi?id=38510
940
941         * inspector/front-end/inspector.js:
942         (WebInspector.set attached): resize the drawer after docking/undocking
943
944 2010-05-04  Laszlo Gombos  <laszlo.1.gombos@nokia.com>
945
946         Unreviewed, build fix for Symbian.
947
948         [Symbian] Build fix after r58598. 
949
950         Use C99 integer types for the Symbian plugin
951         implementation.
952
953         No new tests, as there is no new functionality.
954
955         * plugins/symbian/PluginPackageSymbian.cpp:
956         (WebCore::PluginPackage::NPVersion):
957         * plugins/symbian/PluginViewSymbian.cpp:
958         (WebCore::PluginView::handlePostReadFile):
959
960 2010-05-04  Nikolas Zimmermann  <nzimmermann@rim.com>
961
962         Reviewed by Dirk Schulze.
963
964         Split SVGCharacterLayoutInfo in smaller pieces
965         https://bugs.webkit.org/show_bug.cgi?id=38513
966
967         Split SVGCharacterLayoutInfo into SVGCharacterLayoutInfo/SVGCharacterData and SVGTextChunkLayoutInfo.
968         This is a preparation for more work in the text area.
969
970         * Android.mk: Add SVGCharacterData.(cpp|h) and SVGTextChunkLayoutInfo.h to build.
971         * GNUmakefile.am: Ditto.
972         * WebCore.gypi: Ditto.
973         * WebCore.pro: Ditto
974         * WebCore.vcproj/WebCore.vcproj: Ditto.
975         * WebCore.xcodeproj/project.pbxproj: Ditto.
976         * rendering/SVGCharacterData.cpp: Copied from rendering/SVGCharacterLayoutInfo.cpp.
977         * rendering/SVGCharacterData.h: Copied from rendering/SVGCharacterLayoutInfo.h.
978         (WebCore::SVGChar::SVGChar):
979         * rendering/SVGCharacterLayoutInfo.cpp:
980         (WebCore::SVGCharacterLayoutInfo::isInitialLayout): Introduced new helper function to share code between addLayoutInformation/addStackContent.
981         (WebCore::SVGCharacterLayoutInfo::addLayoutInformation): Use new helper function.
982         (WebCore::SVGCharacterLayoutInfo::addStackContent): Ditto
983         * rendering/SVGCharacterLayoutInfo.h:
984         * rendering/SVGRootInlineBox.h: Include new files.
985         * rendering/SVGTextChunkLayoutInfo.h: Copied from rendering/SVGCharacterLayoutInfo.h.
986
987 2010-05-04  Xan Lopez  <xlopez@igalia.com>
988
989         Reviewed by Holger Freyther.
990
991         [GTK] GObject DOM bindings
992         https://bugs.webkit.org/show_bug.cgi?id=33590
993
994         Use helper functions from CodeGenerator.pm to figure out whether a
995         type is "fundamental" or not (basically whether it's anything
996         other than a string or a non-pointer type).
997
998         * bindings/scripts/CodeGeneratorGObject.pm:
999
1000 2010-04-30  Alexander Pavlov  <apavlov@chromium.org>
1001
1002         Reviewed by Yury Semikhatsky.
1003
1004         Web Inspector: REGRESSION: Disabled style properties are absent in Styles sidebar after WebInspector is re-opened
1005         https://bugs.webkit.org/show_bug.cgi?id=38255
1006
1007         Moved stylesheet-related mappings into a separate object stored
1008         in InspectorController rather than InspectorDOMAgent (which gets reset
1009         on every frontend [dis]connect).
1010
1011         * GNUmakefile.am:
1012         * WebCore.gypi:
1013         * WebCore.pro:
1014         * WebCore.vcproj/WebCore.vcproj:
1015         * WebCore.xcodeproj/project.pbxproj:
1016         * inspector/InspectorCSSStore.cpp: Added.
1017         (WebCore::InspectorCSSStore::InspectorCSSStore):
1018         (WebCore::InspectorCSSStore::~InspectorCSSStore):
1019         (WebCore::InspectorCSSStore::reset):
1020         * inspector/InspectorCSSStore.h: Added.
1021         * inspector/InspectorController.cpp:
1022         (WebCore::InspectorController::InspectorController):
1023         (WebCore::InspectorController::setFrontend):
1024         (WebCore::InspectorController::didCommitLoad):
1025         * inspector/InspectorController.h:
1026         * inspector/InspectorDOMAgent.cpp:
1027         (WebCore::InspectorDOMAgent::InspectorDOMAgent):
1028         (WebCore::InspectorDOMAgent::discardBindings):
1029         (WebCore::InspectorDOMAgent::applyStyleText):
1030         (WebCore::InspectorDOMAgent::setStyleText):
1031         (WebCore::InspectorDOMAgent::setStyleProperty):
1032         (WebCore::InspectorDOMAgent::toggleStyleEnabled):
1033         (WebCore::InspectorDOMAgent::setRuleSelector):
1034         (WebCore::InspectorDOMAgent::addRule):
1035         (WebCore::InspectorDOMAgent::bindStyle):
1036         (WebCore::InspectorDOMAgent::bindRule):
1037         (WebCore::InspectorDOMAgent::buildObjectForStyle):
1038         (WebCore::InspectorDOMAgent::buildObjectForRule):
1039         * inspector/InspectorDOMAgent.h:
1040         (WebCore::InspectorDOMAgent::create):
1041         (WebCore::InspectorDOMAgent::cssStore):
1042
1043 2010-05-04  Mikhail Naganov  <mnaganov@chromium.org>
1044
1045         Reviewed by Pavel Feldman.
1046
1047         Display "Recording..." item when recording an user-initiated CPU profile.
1048
1049         https://bugs.webkit.org/show_bug.cgi?id=38043
1050
1051         * English.lproj/localizedStrings.js:
1052         * inspector/front-end/ProfilesPanel.js:
1053         (WebInspector.ProfilesPanel.prototype.addProfileHeader):
1054         (WebInspector.ProfilesPanel.prototype.removeProfileHeader):
1055         (WebInspector.ProfilesPanel.prototype.showProfile):
1056         * inspector/front-end/inspector.js:
1057         (WebInspector.setRecordingProfile):
1058
1059 2010-05-04  Tucker Jay  <jay.tucker@nokia.com>
1060
1061         Reviewed by Holger Freyther.
1062
1063         Animated GIF images does not animate 10x as expected by default.
1064         https://bugs.webkit.org/show_bug.cgi?id=36818
1065
1066         Added test case to existing manual test to test the
1067         fixed functionality.
1068
1069         * manual-tests/qt/qt-10loop-anim.gif: Added.
1070         * manual-tests/qt/qt-gif-test.html:
1071         * platform/graphics/qt/ImageDecoderQt.cpp:
1072         (WebCore::ImageDecoderQt::repetitionCount):
1073
1074 2010-05-04  Dirk Schulze  <krit@webkit.org>
1075
1076         Unreviewed sort of XCodes project file.
1077
1078         * WebCore.xcodeproj/project.pbxproj:
1079
1080 2010-05-03  Steven Lai  <steven_lai@asia.apple.com>
1081
1082         Reviewed by Brady Eidson.
1083
1084         Reverted hashchange() event back to async.
1085         (This change does not update HashChangeEvent to its new proposed interface)
1086         https://bugs.webkit.org/show_bug.cgi?id=36201
1087         rdar://problem/7780794
1088         rdar://problem/7761278 (partial fix)
1089
1090         Tests: fast/loader/hashchange-event-async.html
1091
1092         * dom/Document.cpp: reverted hashchange() event back to async
1093         (WebCore::Document::enqueueHashchangeEvent):
1094
1095 2010-05-03  Holger Hans Peter Freyther  <zecke@selfish.org>
1096
1097         Rubber-stamped by Xan Lopez.
1098
1099         [Cairo,WX] Stop leaking a FontPlatformData.
1100         https://bugs.webkit.org/show_bug.cgi?id=37500
1101
1102         Stephan Aßmus pointed out that the pango font backend
1103         is leaking memory and fixed it. The WX font backend
1104         and the Cairo/Fontconfig backend have the same snippet
1105         of code and are leaking memory as well. This commit is
1106         fixing that.
1107
1108         * platform/graphics/cairo/SimpleFontDataCairo.cpp:
1109         (WebCore::SimpleFontData::smallCapsFontData):
1110         * platform/graphics/wx/SimpleFontDataWx.cpp:
1111         (WebCore::SimpleFontData::smallCapsFontData):
1112
1113 2010-05-03  James Robinson  <jamesr@chromium.org>
1114
1115         Reviewed by Eric Seidel.
1116
1117         Clean up a few compiler warnings
1118         https://bugs.webkit.org/show_bug.cgi?id=38073
1119
1120         * html/TextMetrics.h:
1121         (WebCore::TextMetrics::width):
1122         * rendering/style/StyleRareInheritedData.h:
1123         * rendering/style/StyleRareNonInheritedData.h:
1124
1125 2010-05-02  Dumitru Daniliuc  <dumi@chromium.org>
1126
1127         Reviewed by Adam Barth.
1128
1129         Add the ability to auto-generate callbacks to all code generators.
1130         https://bugs.webkit.org/show_bug.cgi?id=38414
1131
1132         * bindings/scripts/CodeGeneratorJS.pm:
1133         * bindings/scripts/CodeGeneratorV8.pm:
1134         * bindings/scripts/test/JS/JSTestCallback.cpp: Added.
1135         (WebCore::JSTestCallback::JSTestCallback):
1136         (WebCore::JSTestCallback::~JSTestCallback):
1137         (WebCore::JSTestCallback::callbackWithClass1Param):
1138         (WebCore::JSTestCallback::callbackWithClass2Param):
1139         * bindings/scripts/test/JS/JSTestCallback.h: Added.
1140         (WebCore::JSTestCallback::create):
1141         * bindings/scripts/test/V8/JSTestCallback.cpp: Added.
1142         (WebCore::V8TestCallback::V8TestCallback):
1143         (WebCore::V8TestCallback::~V8TestCallback):
1144         (WebCore::V8TestCallback::callbackWithClass1Param):
1145         (WebCore::V8TestCallback::callbackWithClass2Param):
1146         * bindings/scripts/test/V8/V8TestCallback.h: Added.
1147         (WebCore::V8TestCallback::create):
1148         * bindings/scripts/test/TestCallback.idl: Added.
1149
1150 2010-05-03  Kevin Watters  <kevinwatters@gmail.com>
1151
1152         Reviewed by Kevin Ollivier.
1153
1154         [wx] Build and use Mac's ComplexTextController to support complex text in wx.
1155         https://bugs.webkit.org/show_bug.cgi?id=38482
1156
1157         * platform/graphics/FloatSize.h:
1158         * platform/graphics/GlyphBuffer.h:
1159         (WebCore::GlyphBuffer::advanceAt):
1160         (WebCore::GlyphBuffer::add):
1161         * platform/graphics/SimpleFontData.h:
1162         (WebCore::SimpleFontData::getNSFont):
1163         * platform/graphics/mac/ComplexTextController.cpp:
1164         * platform/graphics/mac/ComplexTextController.h:
1165         * platform/graphics/wx/FontCacheWx.cpp:
1166         (WebCore::FontCache::getFontDataForCharacters):
1167         (WebCore::FontCache::getLastResortFallbackFont):
1168         * platform/graphics/wx/FontPlatformData.h:
1169         (toCTFontRef):
1170         (WebCore::FontPlatformData::FontPlatformData):
1171         (WebCore::FontPlatformData::allowsLigatures):
1172         * platform/graphics/wx/FontPlatformDataWx.cpp:
1173         (WebCore::FontPlatformData::FontPlatformData):
1174         (WebCore::FontPlatformData::cgFont):
1175         * platform/graphics/wx/FontPlatformDataWxMac.mm: Added.
1176         (WebCore::FontPlatformData::nsFont):
1177         (WebCore::FontPlatformData::cacheNSFont):
1178         * platform/graphics/wx/FontWx.cpp:
1179         (WebCore::Font::canReturnFallbackFontsForComplexText):
1180         (WebCore::Font::selectionRectForComplexText):
1181         (WebCore::Font::drawComplexText):
1182         (WebCore::Font::floatWidthForComplexText):
1183         (WebCore::Font::offsetForPositionForComplexText):
1184         * platform/graphics/wx/SimpleFontDataWx.cpp:
1185         (WebCore::SimpleFontData::platformInit):
1186         (WebCore::SimpleFontData::containsCharacters):
1187         (WebCore::SimpleFontData::platformWidthForGlyph):
1188         * platform/wx/wxcode/fontprops.h:
1189         * platform/wx/wxcode/mac/carbon/fontprops.mm:
1190         (wxFontContainsCharacters):
1191         (GetTextExtent):
1192         * platform/wx/wxcode/mac/carbon/non-kerned-drawing.cpp:
1193         (WebCore::drawTextWithSpacing):
1194         * platform/wx/wxcode/win/fontprops.cpp:
1195         (wxFontContainsCharacters):
1196         * wscript:
1197
1198 2010-05-03  Abhishek Arya  <inferno@chromium.org>
1199
1200         Reviewed by Adam Barth.
1201
1202         Add support for controlling clipboard access from javascript.
1203         Clipboard access from javascript is disabled by default.
1204         https://bugs.webkit.org/show_bug.cgi?id=27751
1205
1206         Test: editing/execCommand/clipboard-access.html
1207
1208         * WebCore.base.exp:
1209         * editing/EditorCommand.cpp:
1210         (WebCore::supportedCopyCut):
1211         (WebCore::supportedPaste):
1212         (WebCore::createCommandMap):
1213         * page/Settings.cpp:
1214         (WebCore::Settings::Settings):
1215         (WebCore::Settings::setJavaScriptCanAccessClipboard):
1216         * page/Settings.h:
1217         (WebCore::Settings::javaScriptCanAccessClipboard):
1218
1219 2010-05-03  Alexey Proskuryakov  <ap@apple.com>
1220
1221         Reviewed by Adam Barth.
1222
1223         https://bugs.webkit.org/show_bug.cgi?id=38285
1224         <rdar://problem/7903453> REGRESSION: Javascript command window.open does not work in empty tab
1225
1226         Cannot be tested, because new windows created in DRT always have an opener, and thus inherit
1227         its security origin. Only new windows and tabs created by browser chrome had this problem.
1228
1229         * loader/FrameLoader.cpp: (WebCore::FrameLoader::init): Moved updateSandboxFlags() call to
1230         the beginning, so that an initial document would get correct flags.
1231
1232 2010-05-03  Noam Rosenthal  <noam.rosenthal@nokia.com>
1233
1234         Reviewed by Darin Adler.
1235
1236         WebGL compile issue.
1237         Added ExceptionCode.h to JSWebGLArrayBufferConstructor.cpp, for some reason it was missing.
1238         https://bugs.webkit.org/show_bug.cgi?id=38453
1239
1240         No new tests: compile fix.
1241
1242         * bindings/js/JSWebGLArrayBufferConstructor.cpp:
1243
1244 2010-05-03  Eric Seidel  <eric@webkit.org>
1245
1246         Unreviewed, rolling out r58685.
1247         http://trac.webkit.org/changeset/58685
1248         https://bugs.webkit.org/show_bug.cgi?id=38461
1249
1250         Broke a test on Gtk
1251
1252         * platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp:
1253         (WebCore::MediaPlayerPrivateGStreamer::updateStates):
1254
1255 2010-05-03  Yael Aharon  <yael.aharon@nokia.com>
1256
1257         Reviewed by Darin Adler.
1258
1259         Use HTML5 number parsing in HTMLProgressElement
1260         https://bugs.webkit.org/show_bug.cgi?id=38434
1261
1262         Use parseToDoubleForNumberType instead of toDouble.
1263         Throw an exception when the number is NaN or Infinity.
1264
1265         * html/HTMLProgressElement.cpp:
1266         (WebCore::HTMLProgressElement::value):
1267         (WebCore::HTMLProgressElement::setValue):
1268         (WebCore::HTMLProgressElement::max):
1269         (WebCore::HTMLProgressElement::setMax):
1270         * html/HTMLProgressElement.h:
1271         * html/HTMLProgressElement.idl:
1272
1273 2010-05-03  Jens Alfke  <snej@chromium.org>
1274
1275         Reviewed by Darin Fisher.
1276
1277         [chromium] Add "willSendSubmitEvent" hook to WebFrameClient and FrameLoaderClient
1278         https://bugs.webkit.org/show_bug.cgi?id=38397
1279
1280         No tests (functionality is exposed only through native WebKit API.)
1281
1282         * html/HTMLFormElement.cpp:
1283         (WebCore::HTMLFormElement::prepareSubmit):  Call frame loader's dispatchWillSendSubmitEvent
1284         * loader/EmptyClients.h:
1285         * loader/FrameLoaderClient.h:
1286         (WebCore::FrameLoaderClient::dispatchWillSendSubmitEvent):  New empty method
1287
1288 2010-05-03  Philippe Normand  <pnormand@igalia.com>
1289
1290         Reviewed by Eric Carlson.
1291
1292         [GStreamer] forgotten call to durationChanged in updateStates()
1293         https://bugs.webkit.org/show_bug.cgi?id=38461
1294
1295         Notify MediaPlayer if duration is known after playback started.
1296
1297         * platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp:
1298         (WebCore::MediaPlayerPrivateGStreamer::updateStates):
1299
1300 2010-05-03  Ryuan Choi  <ryuan.choi@gmail.com>
1301
1302         Reviewed by Darin Adler.
1303
1304         fixing build break due to clearWatch() when Geolocation feature is
1305         disabled.
1306
1307         https://bugs.webkit.org/show_bug.cgi?id=38091
1308
1309         no test because this is a build fix only
1310
1311         * page/Geolocation.cpp:
1312         (WebCore::Geolocation::clearWatch):
1313
1314 2010-05-03  Stephan Aßmus  <superstippi@gmx.de>
1315
1316         Reviewed by Holger Freyther.
1317
1318         [Gtk] Fix leaking the FontPlatformData instance used to create the the
1319         small caps font data.
1320         https://bugs.webkit.org/show_bug.cgi?id=37500
1321
1322         No new tests needed.
1323
1324         * platform/graphics/gtk/SimpleFontDataPango.cpp:
1325         (WebCore::SimpleFontData::smallCapsFontData):
1326             - Use a stack allocated FontPlatformData instead of a heap allocated
1327               one that is never freed.
1328
1329 2010-05-03  Jarkko Sakkinen  <jarkko.j.sakkinen@gmail.com>
1330
1331         Reviewed by Simon Hausmann.
1332
1333         [Qt] GraphicsLayer: support webGL
1334         https://bugs.webkit.org/show_bug.cgi?id=35388
1335
1336         Added support GraphicsContext3D to GraphicsLayer.
1337         Added paint method to GraphicsContext3D for Qt platform that
1338         uses drawTexture() when QGLWidget is used as viewport of
1339         QGraphicsWebView.
1340         Fine-tuned texture and handling and image to texture conversion to
1341         work also when drawTexture() blitting is used.
1342
1343         * platform/graphics/GraphicsContext3D.h:
1344         * platform/graphics/qt/GraphicsContext3DQt.cpp:
1345         (WebCore::GraphicsContext3DInternal::GraphicsContext3DInternal):
1346         (WebCore::GraphicsContext3D::beginPaint):
1347         (WebCore::GraphicsContext3D::paint):
1348         (WebCore::GraphicsContext3D::texImage2D):
1349         (WebCore::GraphicsContext3D::texSubImage2D):
1350         (WebCore::GraphicsContext3D::getImageData):
1351         * platform/graphics/qt/GraphicsLayerQt.cpp:
1352         (WebCore::GraphicsLayerQtImpl::):
1353         (WebCore::GraphicsLayerQtImpl::GraphicsLayerQtImpl):
1354         (WebCore::GraphicsLayerQtImpl::paint):
1355         (WebCore::GraphicsLayerQtImpl::flushChanges):
1356         (WebCore::GraphicsLayerQt::setContentsToGraphicsContext3D):
1357         (WebCore::GraphicsLayerQt::setGraphicsContext3DNeedsDisplay):
1358         * platform/graphics/qt/GraphicsLayerQt.h:
1359
1360 2010-05-03  Janne Koskinen  <janne.p.koskinen@digia.com>
1361
1362         Reviewed by Simon Hausmann.
1363
1364         [Qt] Fix qtlibraryinfix not to contain space
1365
1366         List catenation with += adds whitespace cutting the infix
1367         from the final target.
1368
1369         * WebCore.pro:
1370
1371 2010-05-03  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
1372
1373         Reviewed by Simon Hausmann.
1374
1375         [Qt] Fix rendering of <button> elements on Mac OS X
1376
1377         The <button> element has ButtonPart appearance, not PushButton part,
1378         so we have to include ButtonPart when we decide if we should draw the
1379         button as raised on Mac OS X.
1380
1381         https://bugs.webkit.org/show_bug.cgi?id=38458
1382
1383         * platform/qt/RenderThemeQt.cpp:
1384         (WebCore::RenderThemeQt::initializeCommonQStyleOptions):
1385
1386 2010-04-30  Philippe Normand  <pnormand@igalia.com>
1387
1388         Reviewed by Eric Seidel.
1389
1390         [GStreamer] endless loop after playback ended
1391         https://bugs.webkit.org/show_bug.cgi?id=38384
1392
1393         At playback end ensure duration() will return a valid duration if
1394         we managed to calculate it based on current position.
1395
1396         Test: media/video-duration-known-after-eos.html
1397
1398         * platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp:
1399         (WebCore::MediaPlayerPrivateGStreamer::didEnd):
1400
1401 2010-05-03  Tor Arne Vestbø  <tor.arne.vestbo@nokia.com>
1402
1403         [Qt] Fix build break on Mac OS X
1404
1405         * plugins/mac/PluginPackageMac.cpp: Use correct type
1406         * WebCore.pro: Remove duplicate symbol, we now have a Qt implementation
1407
1408 2010-05-03  Thomas Zander <t.zander@nokia.com>
1409
1410         Reviewed by Simon Hausmann.
1411
1412         [Qt] Fix library infix usage when compiling inside of Qt
1413
1414         Don't apply the infix when building inside Qt, as that's done through the
1415         inclusion of qbase.pri.
1416
1417         * WebCore.pro:
1418
1419 2010-05-02  Dan Bernstein  <mitz@apple.com>
1420
1421         Reviewed by Simon Fraser.
1422
1423         Another case of <rdar://problem/7552959> REGRESSION: Infinite recursion in Position::getInlineBoxAndOffset()
1424         https://bugs.webkit.org/show_bug.cgi?id=38445
1425
1426         Test: editing/selection/mixed-editability-11.html
1427
1428         * dom/Position.cpp:
1429         (WebCore::downstreamIgnoringEditingBoundaries): Added. Returns the furthest visually equivalent
1430         position downstream, crossing any editability boundaries.
1431         (WebCore::upstreamIgnoringEditingBoundaries): Similarly for upstream.
1432         (WebCore::Position::getInlineBoxAndOffset): Changed the logic for finding an inline box for positions
1433         whose node is a block flow. Instead of traversing the DOM, advance downstream or upstream as far as
1434         possible, crossing any editability boudaries. Infinite recursion is avoided by advancing all the way
1435         and checking that the new position is different from the starting position. Also replaced the specific
1436         test for buttons with the generic and more comprehensive canHaveChildrenForEditing().
1437
1438 2010-05-02  Tasuku Suzuki  <tasuku.suzuki@nokia.com>
1439
1440         Reviewed by Simon Hausmann.
1441
1442         [Qt] Fix compilation with QT_NO_BEARERMANAGEMENT
1443         https://bugs.webkit.org/show_bug.cgi?id=38324
1444
1445         * platform/network/NetworkStateNotifier.h:
1446         * platform/network/qt/NetworkStateNotifierQt.cpp:
1447
1448 2010-04-29  Janne Koskinen  <janne.p.koskinen@digia.com>
1449
1450         Reviewed by Simon Hausmann.
1451
1452         [Qt] QtWebKit versioning added
1453         https://bugs.webkit.org/show_bug.cgi?id=37207
1454
1455         QtWebkit releases separated from Qt release cycle.
1456
1457         * WebCore.pro:
1458
1459 2010-05-02  Joseph Pecoraro  <joepeck@webkit.org>
1460
1461         Reviewed by Timothy Hatcher.
1462
1463         Web Inspector: Old Style trimWhitespace() should be trim()
1464         https://bugs.webkit.org/show_bug.cgi?id=38441
1465
1466         * inspector/front-end/SourceFrame.js:
1467         (WebInspector.SourceFrame.prototype._evalSelectionInCallFrame):
1468
1469 2010-05-02  Tasuku Suzuki  <tasuku.suzuki@nokia.com>
1470
1471         Reviewed by Simon Hausmann.
1472
1473         [Qt] Fix compilation with QT_NO_LINEEDIT
1474         https://bugs.webkit.org/show_bug.cgi?id=38324
1475
1476         * platform/qt/RenderThemeQt.cpp:
1477         (WebCore::RenderThemeQt::~RenderThemeQt):
1478         (WebCore::RenderThemeQt::findFrameLineWidth):
1479
1480 2010-05-02  Pavel Feldman  <pfeldman@chromium.org>
1481
1482         Not reviewed: Touch inspector controller to kick windows tests.
1483
1484         * inspector/InspectorController.cpp:
1485
1486 2010-05-02  Pavel Feldman  <pfeldman@chromium.org>
1487
1488         Reviewed by Timothy Hatcher.
1489
1490         Web Inspector: IMG nodes are being added to the DOM tree late, sometimes hiding the revealed element.
1491
1492         https://bugs.webkit.org/show_bug.cgi?id=38432
1493
1494         * inspector/front-end/ElementsTreeOutline.js:
1495         (WebInspector.ElementsTreeOutline.prototype._onmousemove):
1496         (WebInspector.ElementsTreeElement.prototype._createTooltipForNode.setTooltip):
1497         (WebInspector.ElementsTreeElement.prototype._createTooltipForNode):
1498         (WebInspector.ElementsTreeElement.prototype.updateTitle):
1499         (WebInspector.ElementsTreeElement.prototype._attributeHTML):
1500         ():
1501
1502 2010-05-02  Dirk Schulze  <krit@webkit.org>
1503
1504         Reviewed by Nikolas Zimmermann.
1505
1506         SVG hkern implementation incomplete
1507         https://bugs.webkit.org/show_bug.cgi?id=38407
1508
1509         Test: svg/text/text-hkern.svg
1510
1511         The current SVG hkern implementation is incomplete and partly wrong. We pass the ACID3 test
1512         by accident.
1513         The new implementation supports all glyph and unicode combinations that are allowed by the Spec
1514         and fixes various of bugs. The parser moved from SVGFontElement to the general parsing code in
1515         SVGParserUtilities.
1516         Some clean-up makes the code more readable and reuseable for the upcoming vkern implementation.
1517         hkern support for text on path is missing and will be added by a following patch.
1518         Unicode strings of hkern elements are just parsed once and not on every glyph again anymore.
1519
1520         * rendering/SVGRootInlineBox.cpp:
1521         (WebCore::calculateCSSKerning):
1522         (WebCore::applySVGKerning):
1523         (WebCore::SVGRootInlineBox::buildLayoutInformationForTextBox):
1524         * svg/SVGFontElement.cpp:
1525         (WebCore::stringMatchesUnicodeRange):
1526         (WebCore::stringMatchesGlyphName):
1527         (WebCore::matches):
1528         (WebCore::SVGFontElement::getHorizontalKerningPairForStringsAndGlyphs):
1529         * svg/SVGFontElement.h:
1530         * svg/SVGHKernElement.cpp:
1531         (WebCore::SVGHKernElement::buildHorizontalKerningPair):
1532         * svg/SVGHKernElement.h:
1533         * svg/SVGParserUtilities.cpp:
1534         (WebCore::parseGlyphName):
1535         (WebCore::parseUnicodeRange):
1536         (WebCore::parseKerningUnicodeString):
1537         * svg/SVGParserUtilities.h:
1538
1539 2010-05-02  Pavel Feldman  <pfeldman@chromium.org>
1540
1541         Reviewed by Timothy Hatcher.
1542
1543         Web Inspector: debugger shortcuts are processed twice if source frame has focus.
1544
1545         https://bugs.webkit.org/show_bug.cgi?id=38431
1546
1547         * inspector/front-end/SourceFrame.js:
1548         (WebInspector.SourceFrame):
1549         (WebInspector.SourceFrame.prototype._createViewerIfNeeded):
1550
1551 2010-05-02  Jarkko Sakkinen  <jarkko.j.sakkinen@gmail.com>
1552
1553         Reviewed by Eric Seidel.
1554
1555         [Qt] Build error in GraphicsContext3DQt.cpp
1556         https://bugs.webkit.org/show_bug.cgi?id=38382
1557
1558         Removed duplicate implementation of isGLES2Compliant from
1559         GraphicsContext3DQt.cpp. Removed deprecated API stuff for
1560         texImage2D/texSubImage2D.
1561         * platform/graphics/qt/GraphicsContext3DQt.cpp:
1562
1563 2010-05-02  Garret Kelly  <gdk@chromium.org>
1564
1565         Reviewed by David Levin.
1566
1567         Make the Touch RuntimeEnabledFeature disabled by default.
1568         https://bugs.webkit.org/show_bug.cgi?id=38392
1569
1570         * bindings/generic/RuntimeEnabledFeatures.cpp: Disable the Touch feature by default.
1571
1572 2010-05-02  Michael Nordman  <michaeln@google.com>
1573
1574         Reviewed by Dmitry Titov.
1575
1576         Define two new ResourceRequestBase TargetTypes for worker and shared worker
1577         main resources. Use the new target types where appropiate. Add logic to marshal
1578         the target type specified by requests initiated on a background worker thread.
1579
1580         https://bugs.webkit.org/show_bug.cgi?id=38295
1581
1582         No new tests. This doesn't have script visible artifacts.
1583
1584         * platform/network/ResourceRequestBase.cpp: marshal the values
1585         (WebCore::ResourceRequestBase::adopt):
1586         (WebCore::ResourceRequestBase::copyData):
1587         * platform/network/ResourceRequestBase.h:  define the types
1588         (WebCore::ResourceRequestBase::):
1589         * workers/DefaultSharedWorkerRepository.cpp: use TargetIsSharedWorker
1590         (WebCore::SharedWorkerScriptLoader::load):
1591         * workers/Worker.cpp: use TargetIsWorker
1592         (WebCore::Worker::Worker):
1593         * workers/WorkerContext.cpp: use TargetIsScript for importScripts
1594         (WebCore::WorkerContext::importScripts):
1595         * workers/WorkerScriptLoader.cpp: add a data member for the target type
1596         (WebCore::WorkerScriptLoader::WorkerScriptLoader):
1597         (WebCore::WorkerScriptLoader::createResourceRequest):
1598         * workers/WorkerScriptLoader.h:
1599
1600 2010-05-02  Noam Rosenthal  <noam.rosenthal@nokia.com>
1601
1602         Reviewed by Adele Peterson.
1603
1604         Webkit doesn't compile with 3D-canvas enabled and video disabled
1605         https://bugs.webkit.org/show_bug.cgi?id=38297
1606
1607         Added a #ifdef ENABLE(VIDEO) to WebGL code
1608
1609         No new tests: compile fix.
1610
1611         * bindings/js/JSWebGLRenderingContextCustom.cpp:
1612         (WebCore::JSWebGLRenderingContext::texImage2D):
1613         (WebCore::JSWebGLRenderingContext::texSubImage2D):
1614
1615 2010-05-01  Evan Stade  <estade@chromium.org>
1616
1617         Reviewed by David Levin.
1618
1619         [chromium] Skia needs to fade DragImages
1620         https://bugs.webkit.org/show_bug.cgi?id=38008
1621
1622         tested by DragImageTest
1623
1624         * platform/chromium/DragImageChromiumSkia.cpp:
1625         (WebCore::dissolveDragImageToFraction):implement
1626         (WebCore::createDragImageFromImage):deep copy instead of shallow
1627
1628 2010-05-01  Maciej Stachowiak  <mjs@apple.com>
1629
1630         Reviewed by Sam Weinig.
1631
1632         REGRESSION (r58273): Visited links do not change color immediately when Cmd-clicked
1633         https://bugs.webkit.org/show_bug.cgi?id=38422
1634         <rdar://problem/7921778>
1635
1636         Tests:
1637             manual-tests/visited-link-new-window.html
1638
1639         * css/CSSStyleSelector.cpp:
1640         (WebCore::CSSStyleSelector::initElement): Only cache the visited link state
1641         when invoked as part of a helper call to styleForElement or pseudoStyleForElement,
1642         to avoid caching the visited link state beyond the scope of a single style lookup.
1643         (WebCore::CSSStyleSelector::styleForElement): Adjust for above change.
1644         (WebCore::CSSStyleSelector::pseudoStyleForElement): Adjust for above change.
1645         * css/CSSStyleSelector.h:
1646         (WebCore::CSSStyleSelector::styleForElement): Change so "visited link helper mode"
1647         can't accidentally be called from outside CSSStyleSelector itself.
1648         (WebCore::CSSStyleSelector::pseudoStyleForElement): ditto
1649         * manual-tests/visited-link-new-window.html: Added. I could not figure out any way
1650         to make an automated test that supports visited link coloring.
1651
1652 2010-05-01  Yael Aharon  <yael.aharon@nokia.com>
1653
1654         Reviewed by Darin Adler.
1655
1656         Move number parsing code out of HTMLInputElement.
1657         https://bugs.webkit.org/show_bug.cgi?id=38203
1658
1659         The numebr parsing code follows HTML5 parsing rules and should be available outside of HTMLInputElement.
1660         No new tests as no new functionality was introduced.
1661
1662         * html/HTMLInputElement.cpp:
1663         * html/HTMLInputElement.h:
1664         * html/HTMLParser.cpp:
1665         (WebCore::serializeForNumberType):
1666         (WebCore::parseToDoubleForNumberType):
1667         * html/HTMLParser.h:
1668         * html/StepRange.cpp:
1669         (WebCore::StepRange::clampValue):
1670         (WebCore::StepRange::valueFromElement):
1671         * html/ValidityState.cpp:
1672         (WebCore::ValidityState::typeMismatch):
1673         * rendering/RenderSlider.cpp:
1674         (WebCore::RenderSlider::setValueForPosition):
1675
1676 2010-05-01  Xan Lopez  <xlopez@igalia.com>
1677
1678         Reviewed by Oliver Hunt.
1679
1680         [GTK] GObject DOM bindings
1681         https://bugs.webkit.org/show_bug.cgi?id=33590
1682
1683         Unify more logic to skip functions into SkipFunction, and
1684         whitelist the two [Custom] methods in HTMLCollection.idl.
1685
1686         * bindings/scripts/CodeGeneratorGObject.pm:
1687
1688 2010-05-01  Robert Hogan  <robert@webkit.org>
1689
1690         Reviewed by Simon Hausmann.
1691
1692         [Qt] Add smart paste support
1693
1694         https://bugs.webkit.org/show_bug.cgi?id=38136
1695
1696         * WebCore.pro:
1697         * editing/qt/SmartReplaceQt.cpp: Added.
1698         (WebCore::isCharacterSmartReplaceExempt):
1699         * platform/qt/PasteboardQt.cpp:
1700         (WebCore::Pasteboard::writeSelection):
1701         (WebCore::Pasteboard::canSmartReplace):
1702
1703 2010-04-30  Yoshiki Hayashi  <yhayashi@google.com>
1704
1705         Reviewed by Shinichiro Hamaji.
1706
1707         https://bugs.webkit.org/show_bug.cgi?id=38249
1708         
1709         Fixes an issue where border height and padding height are ignored when computing vertically shrinking flexbox's height.
1710
1711         Test: fast/flexbox/child-flexing.html
1712
1713         * rendering/RenderFlexibleBox.cpp:
1714         (WebCore::RenderFlexibleBox::allowedChildFlex):
1715
1716 2010-04-30  Shinichiro Hamaji  <hamaji@chromium.org>
1717
1718         Reviewed by Darin Adler.
1719
1720         Add layoutTestController.setPrinting()
1721         https://bugs.webkit.org/show_bug.cgi?id=37203
1722
1723         Use the renderer's width insteead of screen's width as the width of
1724         a screen depends on machines.
1725
1726         * rendering/RenderTreeAsText.cpp:
1727         (WebCore::externalRepresentation):
1728
1729 2010-04-30  Joseph Pecoraro  <joepeck@webkit.org>
1730
1731         Reviewed by Timothy Hatcher.
1732
1733         Web Inspector: Missing INSPECTOR Guard in Console::lastWMLErrorMessage
1734         https://bugs.webkit.org/show_bug.cgi?id=38366
1735
1736         Console::lastWMLErrorMessage is only available if WML is enabled, however
1737         its implementation only makes sense as long as INSPECTOR is enabled
1738         as well. So this adds the ENABLE(INSPECTOR) guard in the function. A
1739         browser without ENABLE(INSPECTOR) will always get an empty result.
1740
1741         * page/Console.cpp:
1742
1743 2010-04-28  Joseph Pecoraro  <joepeck@webkit.org>
1744
1745         Reviewed by Oliver Hunt.
1746
1747         REGRESSION(r58313): Regression evident in pixel tests: the search icon is always clipped at the bottom.
1748         https://bugs.webkit.org/show_bug.cgi?id=38253
1749
1750         Test: fast/css/input-search-padding.html
1751
1752         An <input type="search"> contains an inner block, which is explicitly
1753         centered in RenderTextControlSingleLine based on the height of the element.
1754         However, the clipping rect was not using the set location, and instead
1755         calculated off of the top border and padding alone. This also vertically
1756         centers the Caps Lock indicator.
1757
1758         * rendering/RenderTextControl.cpp: moved controlClipRect implementation to RenderTextControlSingleLine
1759         * rendering/RenderTextControl.h: allow a subclass implementation of controlClipRect, removed redundant hasControlClip implementation, and moved controlClipRect
1760         * rendering/RenderTextControlSingleLine.cpp:
1761         (WebCore::RenderTextControlSingleLine::paint): vertically center the Caps Lock indicator
1762         (WebCore::RenderTextControlSingleLine::controlClipRect): use the set location of the anonymous inner block instead
1763         * rendering/RenderTextControlSingleLine.h: allow for an implementation of controlClipRect for <input type="search">
1764
1765 2010-04-30  Jon Honeycutt  <jhoneycutt@apple.com>
1766
1767         Caret may fail to blink if a focus handler brings up a modal dialog 
1768         https://bugs.webkit.org/show_bug.cgi?id=38372
1769
1770         Reviewed by Darin Adler.
1771
1772         * manual-tests/onfocus-alert-blinking-caret.html: Added.
1773
1774         * page/EventHandler.cpp:
1775         (WebCore::EventHandler::handleMousePressEvent):
1776         Moved the call to setCaretBlinkingSuspended() from here...
1777         (WebCore::EventHandler::handleMousePressEvent):
1778         ... to here. This makes us suspend caret blinking before dispatching the
1779         mouse down event. If dispatching the mouse down event allows the message
1780         loop to run, we want mouse up events received in that message loop to be
1781         able to resume caret blinking.
1782         (WebCore::EventHandler::lostMouseCapture):
1783         We've lost mouse capture and won't be notified of mouse up events;
1784         resume caret blinking.
1785
1786         * page/EventHandler.h:
1787         Declare lostMouseCapture().
1788
1789 2010-04-30  Dimitri Glazkov  <dglazkov@chromium.org>
1790
1791         Unreviewed, build fix.
1792
1793         [Chromium] Added a simple IdentifierRep struct and removed dependency on
1794         IdentifierRep.h.
1795
1796         * bindings/v8/V8NPObject.cpp: Added simple IdentifierRep struct.
1797
1798 2010-04-30  Dimitri Glazkov  <dglazkov@chromium.org>
1799
1800         Unreviewed, build fix.
1801
1802         [Chromium] Remove bridge/ from include paths.
1803
1804         * WebCore.gyp/WebCore.gyp: Removed bridge/ from include path.
1805
1806 2010-04-30  Abhishek Arya  <inferno@chromium.org>
1807
1808         Reviewed by David Kilzer.
1809
1810         Convert m_documentUnderMouse, m_dragInitiator to RefPtr.
1811         Eliminated unused m_dragInitiator accessor to prevent dereferencing.
1812         https://bugs.webkit.org/show_bug.cgi?id=37618
1813
1814         Test: editing/pasteboard/drag-drop-iframe-refresh-crash.html
1815
1816         * page/DragController.cpp:
1817         (WebCore::DragController::tryDocumentDrag):
1818         (WebCore::DragController::concludeEditDrag):
1819         * page/DragController.h:
1820         (WebCore::DragController::draggingImageURL):
1821         (WebCore::DragController::documentUnderMouse):
1822
1823 2010-04-29  James Robinson  <jamesr@chromium.org>
1824
1825         Reviewed by Simon Fraser.
1826
1827         Calls FrameView::scrollPositionChanged whenever a ScrollView is scrolled
1828         https://bugs.webkit.org/show_bug.cgi?id=38286
1829
1830         When a ScrollView's  scroll position is changed, we have to call
1831         FrameView::scrollPositionChanged to generate repaint invalidation for
1832         fixed position elements.  This ends up getting called indirectly when
1833         the ScrollView has a platformWidget through the port layer
1834         (see WebHTMLView.mm's _frameOrBoundsChanged method for how the mac
1835         port does it) but not when there is no platformWidget.
1836
1837         This is tested by the fast/repaint/fixed-* tests when run in pixel
1838         mode.
1839
1840         Test: fast/repaint/fixed-move-after-keyboard-scroll.html
1841
1842         * page/FrameView.h:
1843         * platform/ScrollView.cpp:
1844         (WebCore::ScrollView::valueChanged):
1845         * platform/ScrollView.h:
1846         (WebCore::ScrollView::scrollPositionChanged):
1847
1848 2010-04-30  Anders Carlsson  <andersca@apple.com>
1849
1850         Reviewed by Darin Adler.
1851
1852         Use C99 integer types in more places.
1853
1854         * manual-tests/NPN_Invoke/main.c:
1855         (NPP_New):
1856         (NPP_NewStream):
1857         (NPP_WriteReady):
1858         (NPP_Write):
1859         (NPP_HandleEvent):
1860         (functionPointerForTVector):
1861         * plugins/mac/PluginViewMac.cpp:
1862         (WebCore::PluginView::platformGetValueStatic):
1863         (WebCore::PluginView::handlePostReadFile):
1864
1865 2010-04-30  Darin Adler  <darin@apple.com>
1866
1867         Reviewed by Oliver Hunt.
1868
1869         Remove unused scrollRectIntoViewRecursively function
1870         https://bugs.webkit.org/show_bug.cgi?id=38403
1871
1872         * page/Chrome.cpp:
1873         (WebCore::Chrome::scrollRectIntoView): Moved comment here that was previously
1874         in the scrollRectIntoViewRecursively function.
1875
1876         * platform/ScrollView.cpp: Get rid scrollRectIntoViewRecursively.
1877         * platform/ScrollView.h: Ditto. Fix comment that refers to the two functions.
1878         Also correct all uses of the term "method" to use the C++ term "function" and
1879         got rid of double spaces after periods.
1880
1881 2010-04-30  Anders Carlsson  <andersca@apple.com>
1882
1883         Another Qt build fix.
1884
1885         * plugins/qt/PluginViewQt.cpp:
1886         (WebCore::PluginView::platformGetValueStatic):
1887
1888 2010-04-30  Anders Carlsson  <andersca@apple.com>
1889
1890         Add back TRUE, FALSE and NULL macros. They were not meant to be removed!
1891
1892         * bridge/npapi.h:
1893
1894 2010-04-30  Anders Carlsson  <andersca@apple.com>
1895
1896         Try to fix the Qt build this time.
1897
1898         * plugins/qt/PluginPackageQt.cpp:
1899         (WebCore::staticPluginQuirkRequiresGtkToolKit_NPN_GetValue):
1900         (WebCore::PluginPackage::NPVersion):
1901
1902 2010-04-30  Anders Carlsson  <andersca@apple.com>
1903
1904         Yet another build fix.
1905
1906         * plugins/gtk/PluginPackageGtk.cpp:
1907         (WebCore::PluginPackage::NPVersion):
1908
1909 2010-04-30  Anders Carlsson  <andersca@apple.com>
1910
1911         Fix build.
1912
1913         * bridge/npapi.h:
1914
1915 2010-04-30  Jian Li  <jianli@chromium.org>
1916
1917         Reviewed by Darin Fisher.
1918
1919         [chromium] Add WebFileSystem interface and hook up with all FileSystem methods.
1920         https://bugs.webkit.org/show_bug.cgi?id=38228
1921
1922         * platform/chromium/ChromiumBridge.h:
1923         * platform/chromium/FileSystemChromium.cpp:
1924         (WebCore::openFile):
1925         (WebCore::closeFile):
1926         (WebCore::seekFile):
1927         (WebCore::truncateFile):
1928         (WebCore::readFromFile):
1929         (WebCore::writeToFile):
1930
1931 2010-04-30  Anders Carlsson  <andersca@apple.com>
1932
1933         Reviewed by Dan Bernstein.
1934
1935         Final part of 
1936         
1937         https://bugs.webkit.org/show_bug.cgi?id=20784
1938         move npapi.h to C99 integer types
1939
1940         * bridge/npapi.h:
1941         Remove the old types.
1942
1943 2010-04-30  Sheriff Bot  <webkit.review.bot@gmail.com>
1944
1945         Unreviewed, rolling out r58569.
1946         http://trac.webkit.org/changeset/58569
1947         https://bugs.webkit.org/show_bug.cgi?id=38399
1948
1949         This broke the GTK bots due to bad GC behavior (Requested by
1950         ericu on #webkit).
1951
1952         * bindings/js/JSWorkerContextCustom.cpp:
1953         * bindings/v8/custom/V8WorkerContextCustom.cpp:
1954         * storage/Database.idl:
1955         * storage/SQLError.idl:
1956         * storage/SQLResultSet.idl:
1957         * storage/SQLResultSetRowList.idl:
1958         * storage/SQLTransaction.idl:
1959         * workers/WorkerContext.cpp:
1960         (WebCore::WorkerContext::openDatabase):
1961         * workers/WorkerContext.h:
1962         (WebCore::WorkerContext::databaseExceededQuota):
1963         * workers/WorkerContext.idl:
1964
1965 2010-04-30  Anders Carlsson  <andersca@apple.com>
1966
1967         Fix GTK+ build.
1968
1969         * plugins/gtk/PluginViewGtk.cpp:
1970         (WebCore::PluginView::handlePostReadFile):
1971         (WebCore::PluginView::platformGetValueStatic):
1972
1973 2010-04-30  Anders Carlsson  <andersca@apple.com>
1974
1975         Fix Qt build.
1976
1977         * plugins/qt/PluginViewQt.cpp:
1978         (WebCore::PluginView::handlePostReadFile):
1979         (WebCore::PluginView::platformGetValue):
1980
1981 2010-04-30  Anders Carlsson  <andersca@apple.com>
1982
1983         Reviewed by Timothy Hatcher.
1984
1985         Next step towards fixing
1986         
1987         https://bugs.webkit.org/show_bug.cgi?id=20784
1988         move npapi.h to C99 integer types
1989
1990         Use the C99 types everywhere. The "old" types are still around but will be removed
1991         in a subsequent commit.
1992
1993         * bridge/npapi.h:
1994         (_NPCocoaEvent::):
1995         * plugins/PluginPackage.h:
1996         * plugins/PluginStream.cpp:
1997         (WebCore::PluginStream::deliverData):
1998         * plugins/PluginStream.h:
1999         * plugins/PluginView.cpp:
2000         (WebCore::PluginView::postURLNotify):
2001         (WebCore::PluginView::postURL):
2002         (WebCore::PluginView::write):
2003         (WebCore::PluginView::handlePost):
2004         * plugins/PluginView.h:
2005         * plugins/PluginViewNone.cpp:
2006         (WebCore::PluginView::handlePostReadFile):
2007         * plugins/npapi.cpp:
2008         (NPN_MemAlloc):
2009         (NPN_MemFlush):
2010         (NPN_PostURLNotify):
2011         (NPN_PostURL):
2012         (NPN_Write):
2013         * plugins/npfunctions.h:
2014         * plugins/win/PluginPackageWin.cpp:
2015         (WebCore::PluginPackage::NPVersion):
2016         * plugins/win/PluginViewWin.cpp:
2017         (WebCore::PluginView::handlePostReadFile):
2018
2019 2010-04-30  Peter Kasting  <pkasting@google.com>
2020
2021         Reviewed by David Levin.
2022
2023         Make all image decoders set the "failed" bit if an image could not be
2024         completely decoded, but no more data is coming.  The ICO and BMP
2025         decoders already did this.
2026         https://bugs.webkit.org/show_bug.cgi?id=35411
2027
2028         "Failed" does not cause the image to not be displayed, it simply causes
2029         us to not bother to try to decode again if future requests are made, and
2030         for some decoders, lets the decoder clean up some of its temporary
2031         objects.
2032
2033         No layout tests because this does not change the visible output of decoding in any way.
2034
2035         * platform/image-decoders/gif/GIFImageDecoder.cpp:
2036         (WebCore::GIFImageDecoder::frameComplete): Return whether the frame could be marked as complete.
2037         (WebCore::GIFImageDecoder::decode): Fail if read() needs more data (and thus returns false) and no more is coming.
2038         * platform/image-decoders/gif/GIFImageDecoder.h:
2039         * platform/image-decoders/gif/GIFImageReader.cpp:
2040         (GIFImageReader::do_lzw): Instead of returning true for buffer underrun and false for failure, return false for both and set the failure flag on failure.
2041         (GIFImageReader::read): Ditto.
2042         * platform/image-decoders/gif/GIFImageReader.h:
2043         * platform/image-decoders/jpeg/JPEGImageDecoder.cpp:
2044         (WebCore::):
2045         (WebCore::JPEGImageReader::decode): See do_lzw() comment above.
2046         (WebCore::JPEGImageDecoder::decode): Fail if decode() needs more data (and thus returns false) and no more is coming.
2047         * platform/image-decoders/jpeg/JPEGImageDecoder.h:
2048         * platform/image-decoders/png/PNGImageDecoder.cpp:
2049         (WebCore::PNGImageReader::decode): Return true for decode success, false for buffer underrun or decode failure, and set the failure flag on decode failure.
2050         (WebCore::PNGImageDecoder::decode): See JPEGImageDecoder::decode() comment above.
2051         * platform/image-decoders/png/PNGImageDecoder.h:
2052
2053 2010-04-30  Brady Eidson  <beidson@apple.com>
2054
2055         Reviewed by Eric Carlson.
2056
2057         <rdar://problem/7902467> - Audio plays upon loading of npr.org but shouldn't
2058
2059         No new tests. (Currently no way to test such site specific hack behavior)
2060
2061         * html/HTMLMediaElement.cpp:
2062         (WebCore::HTMLMediaElement::HTMLMediaElement):
2063         (WebCore::HTMLMediaElement::asyncEventTimerFired): If the event to be dispatched is the canplay
2064           event, wrap the dispatch with m_dispatchingCanPlayEvent set.
2065         (WebCore::HTMLMediaElement::play): If m_dispatchingCanPlayEvent is true and the site is npr.org, 
2066           don't perform the play().
2067         * html/HTMLMediaElement.h: Add m_dispatchingCanPlayEvent member.
2068
2069 2010-04-30  Dan Bernstein  <mitz@apple.com>
2070
2071         Reviewed by Adele Peterson.
2072
2073         Part of <rdar://problem/6649734> Text repainting does not account for glyphs which draw outside the typographic bounds of the font
2074         https://bugs.webkit.org/show_bug.cgi?id=6274
2075
2076         Account for glyph overflow of characters in the range U+1E00..U+2000, but without sending them
2077         through the complex text code path. Instead, introduce a variant of the fast path that tracks
2078         glyph overflow.
2079
2080         * platform/graphics/Font.cpp:
2081         (WebCore::Font::drawText): Use codePath().
2082         (WebCore::Font::floatWidth): Use codePath(). Pass the GlyphOverflow pointer through to
2083         floatWidthForSimpleText() if the code path is SimpleWithGlyphOverflow.
2084         (WebCore::Font::selectionRectForText): Use codePath().
2085         (WebCore::Font::offsetForPosition): Ditto.
2086         * platform/graphics/Font.h: Replaced canUseGlyphCache() with codePath(). Added a GlyphOverflow
2087         parameter to floatWidthForSimpleText().
2088         * platform/graphics/FontFastPath.cpp:
2089         Removed ROMAN_AND_GREEK_DIACRITICS_CAN_USE_GLYPH_CACHE.
2090         (WebCore::Font::codePath): Renamed canUseGlyphCache() to this. Where it used to return false,
2091         it now returns Complex. Where it used to return true, it now returns Simple, except for
2092         the range U+1E00..U+2000, where it now returns SimpleWithGlyphOverflow.
2093         (WebCore::Font::floatWidthForSimpleText): Added a GlyphOverflow parameter. If not 0, have the
2094         width iterator account for glyph bounds, then update the GlyphOverflow accordingly.
2095         * platform/graphics/WidthIterator.cpp:
2096         (WebCore::WidthIterator::WidthIterator): Added boolean parameter telling the width iterator
2097         whether to account for glyph bounds. Initialize m_accountForGlyphBounds accordingly. Initialize
2098         m_maxGlyphBoundingBoxY, m_minGlyphBoundingBoxY, m_firstGlyphOverflow and m_lastGlyphOverflow.
2099         (WebCore::WidthIterator::advance): If accounting for glyph bounds, update the above member variables.
2100         * platform/graphics/WidthIterator.h:
2101         (WebCore::WidthIterator::maxGlyphBoundingBoxY): Added this accessor.
2102         (WebCore::WidthIterator::minGlyphBoundingBoxY): Ditto.
2103         (WebCore::WidthIterator::firstGlyphOverflow): Ditto.
2104         (WebCore::WidthIterator::lastGlyphOverflow): Ditto.
2105
2106 2010-04-30  Chris Marrin  <cmarrin@apple.com>
2107
2108         Reviewed by Simon Fraser.
2109
2110         Reversed the order of the CSSMatrix.multiply method
2111         https://bugs.webkit.org/show_bug.cgi?id=38337
2112
2113         Test: transforms/svg-vs-css.xhtml
2114
2115         * css/WebKitCSSMatrix.cpp:
2116         (WebCore::WebKitCSSMatrix::multiply):
2117         * platform/graphics/transforms/TransformationMatrix.cpp:
2118
2119 2010-04-30  Kevin Ollivier  <kevino@theolliviers.com>
2120
2121         Unreviewed. Attempt to fix the Chromium Mac build after the last commit.
2122
2123         * WebCore.gypi:
2124
2125 2010-04-30  Kevin Ollivier  <kevino@theolliviers.com>
2126
2127         Reviewed by Dan Bernstein.
2128
2129         Allow other ports to compile ATSUI and CoreText functions in SimpleFontData for Mac.
2130         https://bugs.webkit.org/show_bug.cgi?id=38334
2131
2132         * WebCore.xcodeproj/project.pbxproj:
2133         * platform/graphics/mac/SimpleFontDataATSUI.mm: Copied from WebCore/platform/graphics/mac/SimpleFontDataMac.mm.
2134         * platform/graphics/mac/SimpleFontDataCoreText.cpp: Copied from WebCore/platform/graphics/mac/SimpleFontDataMac.mm.
2135         * platform/graphics/mac/SimpleFontDataMac.mm:
2136
2137 2010-04-30  Shinichiro Hamaji  <hamaji@chromium.org>
2138
2139         Reviewed by Simon Fraser.
2140
2141         SHOULD NEVER BE REACHED assertion loading forbes.com
2142         https://bugs.webkit.org/show_bug.cgi?id=38272
2143
2144         Ignore page media related pseudo classes.
2145
2146         Test: printing/pseudo-class-outside-page.html
2147
2148         * css/CSSStyleSelector.cpp:
2149         (WebCore::CSSStyleSelector::SelectorChecker::checkOneSelector):
2150
2151 2010-04-30  Laszlo Gombos  <laszlo.1.gombos@nokia.com>
2152
2153         Unreviewed, build fix.
2154
2155         Fix compiler warning "suggest parentheses around"
2156
2157         No new tests as there is no new functionality.
2158
2159         * svg/SVGAnimateElement.cpp:
2160         (WebCore::SVGAnimateElement::calculateFromAndToValues):
2161
2162 2010-04-30  Kent Tamura  <tkent@chromium.org>
2163
2164         Unreviewed. Regression fix.
2165
2166         Revert a part of r58564 to be compatible with prior behavior
2167         https://bugs.webkit.org/show_bug.cgi?id=38383
2168
2169         * rendering/RenderTextControlSingleLine.cpp:
2170         (WebCore::RenderTextControlSingleLine::forwardEvent):
2171          r58564 made a region check for the cancel button stricter, but it
2172          made some tests failing on Chromium. So, relax the check again.
2173
2174 2010-04-29  Jeremy Orlow  <jorlow@chromium.org>
2175
2176         Reviewed by Darin Fisher.
2177
2178         Change StorageEvent.uri to StorageEvent.url to match the spec
2179         https://bugs.webkit.org/show_bug.cgi?id=38331
2180
2181         As I mentioned in http://www.mail-archive.com/public-webapps@w3.org/msg08495.html
2182         WebKit is the only one who places the document's URL in a 'uri' property
2183         rather than a 'url' property.  Even though we've shipped several versions of
2184         browsers with the old name, we probably should change this to comply with the
2185         spec.
2186
2187         This stuff is covered by existing tests.
2188
2189         * storage/StorageEvent.cpp:
2190         (WebCore::StorageEvent::create):
2191         (WebCore::StorageEvent::StorageEvent):
2192         (WebCore::StorageEvent::initStorageEvent):
2193         * storage/StorageEvent.h:
2194         (WebCore::StorageEvent::url):
2195         * storage/StorageEvent.idl:
2196
2197 2010-04-30  Nikolas Zimmermann  <nzimmermann@rim.com>
2198
2199         Reviewed by Dirk Schulze.
2200
2201         REGRESSION: RenderPath does not handle repaints correctly anymore if bounds changed
2202         https://bugs.webkit.org/show_bug.cgi?id=38385
2203
2204         The last SVG performance patch broke repainting if bounds of a RenderPath get smaller.
2205         It would only repaint the smaller part, not the original larger bounds.
2206
2207         Remove all lazy calculation of the repaint rects, instead calculate object/strokeBoundingBox and repaintRectInLocalCoordinates
2208         once in layout - after LayoutRepainter grabbed the initial bounds, before calling repaintAfterLayout(). We can now inline
2209         all these functions, and save a lot of m_path.isEmpty() checks, which are expensive. No need to store a seperated markerBoundingBox(),
2210         combine with strokeBoundingBox() -> save one FloatRect per RenderPath. Move strokeBoundingBox() from SVGRenderBase to RenderObject,
2211         right next to objectBoundingBox() - to save unnecessary toSVGRenderBase() calls. Completly remove this method.
2212
2213         Overall this is a regression fix, a performance improvement and saves memory. Something for everyone.
2214
2215         Tests: svg/custom/repaint-stroke-width-changes.svg
2216
2217         * rendering/RenderObject.cpp: Added strokeBoundingBox() here, to avoid the toSVGRenderBase() dance.
2218         (WebCore::RenderObject::strokeBoundingBox):
2219         * rendering/RenderObject.h: Ditto.
2220         * rendering/RenderPath.cpp:
2221         (WebCore::RenderPath::RenderPath):
2222         (WebCore::RenderPath::layout): Fix regression, do repainting correctly, by recalculating the boundaries, if needed, instead of nulling them.
2223         (WebCore::RenderPath::paint): Cache SVGRenderStyle in local variable, remove no longer valid FIXME.
2224         (WebCore::RenderPath::calculateMarkerBoundsIfNeeded): Return a FloatRect, to avoid having to store the marker bounding box seperated.
2225         (WebCore::RenderPath::styleWillChange): Mark boundaries as dirty.
2226         (WebCore::RenderPath::updateCachedBoundaries): New function to (re-)calculate all cached boundaries, only called from layout().
2227         * rendering/RenderPath.h: Rename cached rect variables to have more sensible names.
2228         (WebCore::RenderPath::objectBoundingBox): Inlined, just returns the cached value - no more lazy creation. Huge speedup as this is hot code.
2229         (WebCore::RenderPath::strokeBoundingBox): Ditto.
2230         (WebCore::RenderPath::repaintRectInLocalCoordinates): Ditto.
2231         * rendering/RenderSVGBlock.h: Remove toSVGRenderBase() method.
2232         * rendering/RenderSVGImage.h: Ditto.
2233         * rendering/RenderSVGInline.cpp: No need to call toSVGRenderBase() just to get the strokeBoundingBox(). Unifies code to retrieve bounding boxes.
2234         (WebCore::RenderSVGInline::strokeBoundingBox):
2235         * rendering/RenderSVGInline.h: Remove toSVGRenderBase() method.
2236         * rendering/RenderSVGModelObject.h: Ditto.
2237         * rendering/RenderSVGResourceFilter.cpp: No need to call toSVGRenderBase() anymore, just grab the strokeBoundingBox() from the RenderObject.
2238         (WebCore::RenderSVGResourceFilter::applyResource):
2239         * rendering/RenderSVGText.h: Remove toSVGRenderBase() method.
2240         * rendering/SVGRenderSupport.h: Ditto. Remove markerBoundingBox() method, now combined with strokeBoundingBox().
2241         (WebCore::SVGRenderBase::strokeBoundingBox):
2242         * rendering/SVGRootInlineBox.h: Remove toSVGRenderBase() method.
2243         * rendering/style/SVGRenderStyle.h: Add hasMarkers() helper method, to avoid doing unnecessary work in RenderPath.
2244         (WebCore::SVGRenderStyle::hasMarkers):
2245
2246 2010-04-30  Eric Uhrhane  <ericu@chromium.org>
2247
2248         Reviewed by Dmitry Titov.
2249
2250         Add bindings for async DB API in Workers.
2251         https://bugs.webkit.org/show_bug.cgi?id=34992
2252
2253         Tests: storage/change-version-handle-reuse-worker.html
2254                storage/execute-sql-args-worker.html
2255
2256         * bindings/js/JSWorkerContextCustom.cpp: Add openDatabase binding.
2257         (WebCore::JSWorkerContext::openDatabase):
2258
2259         * bindings/v8/custom/V8WorkerContextCustom.cpp: Add openDatabase stub; Chromium will need work both in V8 and in the browser process before we can turn this on there.
2260         (WebCore::V8WorkerContext::openDatabaseCallback):
2261
2262         Add NoStaticTables flags to all objects now shared with workers.
2263         * storage/Database.idl:
2264         * storage/SQLError.idl:
2265         * storage/SQLResultSet.idl:
2266         * storage/SQLResultSetRowList.idl:
2267         * storage/SQLTransaction.idl:
2268         
2269         * workers/WorkerContext.h: Add databaseExceededQuota.
2270         * workers/WorkerContext.cpp:
2271         (WebCore::WorkerContext::databaseExceededQuota): Add stub implementation for testing; you just get 5MB for now.
2272         (WebCore::WorkerContext::openDatabase): Remove invalid assertion.
2273
2274         Add the IDL for the call to openDatabase.
2275         * workers/WorkerContext.idl:
2276
2277 2010-04-30  Shinichiro Hamaji  <hamaji@chromium.org>
2278
2279         Reviewed by Darin Adler.
2280
2281         Unnecessary PrintContext::end() calls
2282         https://bugs.webkit.org/show_bug.cgi?id=38247
2283
2284         Refactoring only, so no new tests.
2285
2286         * page/PrintContext.cpp:
2287         (WebCore::PrintContext::pageNumberForElement):
2288         (WebCore::PrintContext::numberOfPages):
2289
2290 2010-04-30  Kent Tamura  <tkent@chromium.org>
2291
2292         Reviewed by Adele Peterson.
2293
2294         Implement interactive behavior of spin buttons.
2295         https://bugs.webkit.org/show_bug.cgi?id=35686
2296
2297         Introduce SpinButtonElement. It is a shadow element class for
2298         spin buttons. If the upper side of the element is clicked, calls
2299         HTMLInputElement::stepUpFromRenderer(1). If the lower button is
2300         clicked, calls HTMLInputElement::stepUpFromRenderer(-1).
2301
2302         SpinButtonElement tracks the mouse pointer position, and
2303         RenderTheme sets ControlStates::SpinUpState if the pointer is on
2304         the upper side.
2305
2306         Test: platform/mac/fast/forms/input-number-click.html
2307
2308         * dom/Element.h:
2309         (WebCore::Element::isSpinButtonElement):
2310         * editing/VisibleSelection.cpp:
2311         (WebCore::VisibleSelection::adjustSelectionToAvoidCrossingEditingBoundaries):
2312          Remove an assertion. lastEditablePositionBeforePositionInRoot() can
2313          return null in a case that m_end is at a shadow element (a spin button)
2314          and baseRoot is another shadow element (inner text block) in the same
2315          node (an INPUT element).
2316         * html/HTMLInputElement.cpp:
2317         (WebCore::HTMLInputElement::stepUpFromRenderer):
2318         * html/HTMLInputElement.h:
2319         (WebCore::HTMLInputElement::hasSpinButton):
2320          Add types supporting step attribute except RANGE.
2321         * rendering/RenderTextControlSingleLine.cpp:
2322         (WebCore::RenderTextControlSingleLine::nodeAtPoint):
2323         (WebCore::RenderTextControlSingleLine::forwardEvent):
2324         (WebCore::RenderTextControlSingleLine::preferredContentWidth):
2325         (WebCore::RenderTextControlSingleLine::createSubtreeIfNeeded):
2326         * rendering/RenderTextControlSingleLine.h:
2327         * rendering/RenderTheme.cpp:
2328         (WebCore::RenderTheme::controlStatesForRenderer):
2329         (WebCore::RenderTheme::isSpinUpButtonPartPressed):
2330         (WebCore::RenderTheme::isSpinUpButtonPartHovered):
2331         * rendering/RenderTheme.h:
2332         * rendering/TextControlInnerElements.cpp:
2333         (WebCore::SpinButtonElement::SpinButtonElement):
2334         (WebCore::SpinButtonElement::defaultEventHandler):
2335         * rendering/TextControlInnerElements.h:
2336         (WebCore::SpinButtonElement::isSpinButtonElement):
2337         (WebCore::SpinButtonElement::isEnabledFormControl):
2338         (WebCore::SpinButtonElement::onUpButton):
2339
2340 2010-04-30  Yael Aharon  <yael.aharon@nokia.com>
2341
2342         Reviewed by Kenneth Rohde Christiansen.
2343
2344         [Qt] Enable DOMWindow constructor for HTMLProgressElement
2345         https://bugs.webkit.org/show_bug.cgi?id=38333
2346
2347         Add ENABLE_PROGRESS_TAG to FEATURES_DEFINES_JAVASCRIPT.
2348
2349         * WebCore.pri:
2350
2351 2010-04-30  Kent Tamura  <tkent@chromium.org>
2352
2353         Reviewed by Adele Peterson.
2354
2355         Mac implementation of outer-spin-button appearance, and anonymous
2356         element generation for <input type=number>.
2357         https://bugs.webkit.org/show_bug.cgi?id=32813
2358
2359         The implementation uses NSStepperCell. Like the other Mac
2360         controls, it has only three candidates for sizes.
2361
2362         The editable block of an input element is shrunk, and the
2363         anonymous block for a spin button is put on the right of the
2364         editable block.
2365
2366         Tests: platform/mac/fast/forms/input-appearance-spinbutton-size.html
2367                platform/mac/fast/forms/input-appearance-spinbutton.html
2368
2369         * dom/Element.cpp:
2370         (WebCore::Element::pseudoStyleCacheIsInvalid):
2371         * dom/InputElement.h:
2372         (WebCore::InputElement::hasSpinButton):
2373         * html/HTMLInputElement.h:
2374         (WebCore::HTMLInputElement::hasSpinButton): Return true for NUMBER type.
2375         * platform/mac/ThemeMac.mm:
2376         (WebCore::sizeFromNSControlSize): Split the main part of sizeFromFont()
2377           to this in order to use stepperControlSizeForFont() instead of
2378           controlSizeForFont().
2379         (WebCore::sizeFromFont): Just calls sizeFromNSControlSize() with
2380           sizeFromFont(). No behavior changes.
2381         (WebCore::stepperSizes): Returns sizes for mini, small, and regular.
2382         (WebCore::stepperControlSizeForFont):
2383           Dedicated version of controlSizeForFont().
2384         (WebCore::stepper): Returns NSStepperCell object with specified settings.
2385         (WebCore::paintStepper):
2386         (WebCore::ThemeMac::controlSize): Support for OuterSpinButton.
2387         (WebCore::ThemeMac::minimumControlSize): ditto.
2388         (WebCore::ThemeMac::inflateControlPaintRect): ditto.
2389         (WebCore::ThemeMac::paint): ditto.
2390         * rendering/RenderBox.cpp:
2391         (WebCore::RenderBox::paintBoxDecorations):
2392           Move the content to paintBoxDecorationsWithSize().
2393         (WebCore::RenderBox::paintBoxDecorationsWithSize):
2394         * rendering/RenderBox.h: Declare paintBoxDecorationsWithSize().
2395         * rendering/RenderTextControlSingleLine.cpp:
2396         (WebCore::RenderTextControlSingleLine::~RenderTextControlSingleLine):
2397         (WebCore::RenderTextControlSingleLine::paintBoxDecorations):
2398           Call paintBoxDecorationsWithSize() with smaller width by decorationWidthRight().
2399         (WebCore::RenderTextControlSingleLine::addFocusRingRects):
2400           Add a rectangle of which width is smaller by decorationWidthRight().
2401         (WebCore::RenderTextControlSingleLine::layout):
2402           Adjust m_outerSpinButton position.
2403         (WebCore::RenderTextControlSingleLine::styleDidChange):
2404         (WebCore::RenderTextControlSingleLine::textBlockWidth):
2405         (WebCore::RenderTextControlSingleLine::decorationWidthRight):
2406         (WebCore::RenderTextControlSingleLine::preferredDecorationWidthRight):
2407         (WebCore::RenderTextControlSingleLine::createSubtreeIfNeeded):
2408           Creates an element for m_outerSpinButton if it is needed.
2409         (WebCore::RenderTextControlSingleLine::createInnerTextStyle):
2410         (WebCore::RenderTextControlSingleLine::createOuterSpinButtonStyle):
2411         * rendering/RenderTextControlSingleLine.h: Declare new methods and m_outerSpinButton.
2412         * rendering/RenderThemeMac.mm:
2413         (WebCore::RenderThemeMac::adjustRepaintRect): Support for OuterSpinButton.
2414
2415 2010-04-29  Adam Barth  <abarth@webkit.org>
2416
2417         Unreviewed.  Update JSC CodeGenerator baseline.  Not sure how I missed
2418         this one earlier.
2419
2420         * bindings/scripts/test/JS/JSTestObj.cpp:
2421         (WebCore::jsTestObjPrototypeFunctionSerializedValue):
2422
2423 2010-04-29  Justin Garcia  <justin.garcia@apple.com>
2424
2425         Reviewed by Adele Peterson.
2426
2427         Need to updateLayout after typing commands too
2428         https://bugs.webkit.org/show_bug.cgi?id=38352
2429
2430         Replaced !m_parent checks in EditCommand.cpp by the more descriptive isTopLevelCommand().
2431         Move the post editing operation updateLayout() call to {un,re}appliedEditing so that text insertions,
2432         which don't go through EditCommand::{un,re}apply() can benefit from it too.  No test case possible
2433         since most platforms have a layout performed as a side effect of post operation selection code.
2434
2435         * editing/EditCommand.cpp:
2436         (WebCore::EditCommand::apply):
2437         (WebCore::EditCommand::unapply):
2438         (WebCore::EditCommand::reapply):
2439         * editing/EditCommand.h:
2440         (WebCore::EditCommand::isTopLevelCommand):
2441         * editing/Editor.cpp:
2442         (WebCore::Editor::appliedEditing):
2443         (WebCore::Editor::unappliedEditing):
2444         (WebCore::Editor::reappliedEditing):
2445
2446 2010-04-29  Dan Bernstein  <mitz@apple.com>
2447
2448         Reviewed by Simon Fraser.
2449
2450         <rdar://problem/7918086> REGRESSION (r57820): Controller is not displayed in window when opening a MP3 file in browser window
2451         https://bugs.webkit.org/show_bug.cgi?id=38350
2452
2453         Tests: media/audio-only-video-intrinsic-size.html
2454                media/media-document-audio-size.html
2455
2456         * rendering/RenderVideo.cpp:
2457         (WebCore::RenderVideo::RenderVideo): Until metadata is available, ignore the natural size
2458         reported by the player.
2459         (WebCore::RenderVideo::videoSizeChanged): Respect a natural size of zero if reported by the
2460         player, except in standalone media documents.
2461
2462 2010-04-29  Anders Carlsson  <andersca@apple.com>
2463
2464         Reviewed by Dan Bernstein.
2465
2466         First part of
2467         https://bugs.webkit.org/show_bug.cgi?id=20784
2468         move npapi.h to C99 integer types.
2469
2470         Add nptypes.h to the build.
2471
2472         * WebCore.xcodeproj/project.pbxproj:
2473         * bridge/npapi.h:
2474         * bridge/npruntime.h:
2475         * bridge/nptypes.h: Added.
2476
2477 2010-04-29  Xan Lopez  <xlopez@igalia.com>
2478
2479         Rubber-stamped by Adam Barth.
2480
2481         Update GObject bindings test results. We are actually moving
2482         backwards here, but we'll update them again when we figure out
2483         what broke.
2484
2485         * bindings/scripts/test/GObject/WebKitDOMTestObj.cpp:
2486         (webkit_dom_test_obj_set_property):
2487         (webkit_dom_test_obj_get_property):
2488
2489 2010-04-29  Sam Weinig  <sam@webkit.org>
2490
2491         Reviewed by Anders Carlsson.
2492
2493         Fix for https://bugs.webkit.org/show_bug.cgi?id=38277
2494         MiniBrowser: -[WebCoreFlippedView currentEditor]: unrecognized selector
2495
2496         * platform/mac/ThemeMac.mm:
2497         (-[WebCoreFlippedView currentEditor]): Added currentEditor nil implementation.
2498
2499 2010-04-29  Adam Barth  <abarth@webkit.org>
2500
2501         Reviewed by Eric Seidel.
2502
2503         CodeGeneratorJS.pm should be consistent about castedThis versus castedThisObj
2504         https://bugs.webkit.org/show_bug.cgi?id=38338
2505
2506         Currently CodeGeneratorJS.pm uses castThis for methods and
2507         castedThisObj for attributes.  This inconsistency makes it difficult to
2508         factor common code genereration code into methods shared by both kinds
2509         of bindings.  This match aligns the names so that a future patch (e.g.,
2510         in https://bugs.webkit.org/show_bug.cgi?id=38313) can reduce copy/paste
2511         code.
2512
2513         * bindings/scripts/CodeGeneratorJS.pm:
2514         * bindings/scripts/test/JS/JSTestObj.cpp:
2515         (WebCore::setJSTestObjIntAttr):
2516         (WebCore::setJSTestObjLongLongAttr):
2517         (WebCore::setJSTestObjUnsignedLongLongAttr):
2518         (WebCore::setJSTestObjStringAttr):
2519         (WebCore::setJSTestObjTestObjAttr):
2520         (WebCore::setJSTestObjAttrWithException):
2521         (WebCore::setJSTestObjAttrWithSetterException):
2522         (WebCore::setJSTestObjAttrWithGetterException):
2523         (WebCore::jsTestObjPrototypeFunctionVoidMethod):
2524         (WebCore::jsTestObjPrototypeFunctionVoidMethodWithArgs):
2525         (WebCore::jsTestObjPrototypeFunctionIntMethod):
2526         (WebCore::jsTestObjPrototypeFunctionIntMethodWithArgs):
2527         (WebCore::jsTestObjPrototypeFunctionObjMethod):
2528         (WebCore::jsTestObjPrototypeFunctionObjMethodWithArgs):
2529         (WebCore::jsTestObjPrototypeFunctionMethodWithException):
2530         (WebCore::jsTestObjPrototypeFunctionCustomMethod):
2531         (WebCore::jsTestObjPrototypeFunctionCustomMethodWithArgs):
2532         (WebCore::jsTestObjPrototypeFunctionCustomArgsAndException):
2533         (WebCore::jsTestObjPrototypeFunctionAddEventListener):
2534         (WebCore::jsTestObjPrototypeFunctionRemoveEventListener):
2535         (WebCore::jsTestObjPrototypeFunctionWithDynamicFrame):
2536         (WebCore::jsTestObjPrototypeFunctionWithDynamicFrameAndArg):
2537         (WebCore::jsTestObjPrototypeFunctionWithDynamicFrameAndOptionalArg):
2538         (WebCore::jsTestObjPrototypeFunctionWithDynamicFrameAndUserGesture):
2539         (WebCore::jsTestObjPrototypeFunctionWithDynamicFrameAndUserGestureASAD):
2540         (WebCore::jsTestObjPrototypeFunctionWithScriptStateVoid):
2541         (WebCore::jsTestObjPrototypeFunctionWithScriptStateObj):
2542         (WebCore::jsTestObjPrototypeFunctionWithScriptStateVoidException):
2543         (WebCore::jsTestObjPrototypeFunctionWithScriptStateObjException):
2544         (WebCore::jsTestObjPrototypeFunctionMethodWithOptionalArg):
2545         (WebCore::jsTestObjPrototypeFunctionMethodWithNonOptionalArgAndOptionalArg):
2546         (WebCore::jsTestObjPrototypeFunctionMethodWithNonOptionalArgAndTwoOptionalArgs):
2547
2548 2010-04-29  Gustavo Noronha Silva  <gustavo.noronhaollabora.co.uk>
2549
2550         Reviewed by Xan Lopez.
2551
2552         [GTK] pointerCursor should use the default cursor set for the window, not GDK_LEFT_PTR
2553         https://bugs.webkit.org/show_bug.cgi?id=36963
2554
2555         Use the default cursor instead of hard-coding left pointer.
2556
2557         * platform/gtk/CursorGtk.cpp:
2558         (WebCore::Cursor::Cursor):
2559         (WebCore::pointerCursor):
2560
2561 2010-04-29  Ilya Tikhonovsky  <loislo@chromium.org>
2562
2563         Reviewed by Yury Semikhatsky.
2564
2565         WebInspector: If Timeline panel is in recording mode and is not visible and has received
2566         new events then these events do not appear in the panel when the panel becomes visible.
2567         Timeline popup may appear in the upper left window corner when you switch to another panel.
2568         https://bugs.webkit.org/show_bug.cgi?id=38322
2569
2570         * inspector/front-end/Popover.js:
2571         (WebInspector.PopoverHelper.prototype._mouseMove.doHide):
2572         (WebInspector.PopoverHelper.prototype._mouseMove):
2573         (WebInspector.PopoverHelper.prototype.hidePopup):
2574         (WebInspector.PopoverHelper.prototype._hidePopup):
2575         * inspector/front-end/TimelinePanel.js:
2576         (WebInspector.TimelinePanel.prototype.show):
2577         (WebInspector.TimelinePanel.prototype._scheduleRefresh):
2578         (WebInspector.TimelinePanel.prototype._refresh):
2579
2580 2010-04-29  Anton Muhin  <antonm@chromium.org>
2581
2582         Reviewed by Darin Adler.
2583
2584         Let's cache nodelists instead of DynamicNodeList::Caches
2585         https://bugs.webkit.org/show_bug.cgi?id=33696
2586
2587         Test: fast/dom/Element/node-list-identity.html
2588
2589         * bindings/js/JSNodeCustom.cpp:
2590         (WebCore::JSNode::markChildren): Mark all cached node lists as well
2591         * dom/ClassNodeList.cpp:
2592         (WebCore::ClassNodeList::ClassNodeList): Don't need DynamicNodeList::Caches argument any more
2593         (WebCore::ClassNodeList::~ClassNodeList): Remove from the cache
2594         * dom/ClassNodeList.h: Added a field with original class names to be used as a key for removal from the cache
2595         (WebCore::ClassNodeList::create): Don't need DynamicNodeList::Caches argument any more
2596         * dom/NameNodeList.cpp:
2597         (WebCore::NameNodeList::NameNodeList): Don't need DynamicNodeList::Caches argument any more
2598         (WebCore::NameNodeList::~NameNodeList): Remove from the cache
2599         * dom/NameNodeList.h:
2600         (WebCore::NameNodeList::create): Don't need DynamicNodeList::Caches argument any more
2601         * dom/Node.cpp:
2602         (WebCore::Node::removeCachedClassNodeList): Remove ClassNodeList from the cache
2603         (WebCore::Node::removeCachedNameNodeList): Remove NameNodeList from the cache
2604         (WebCore::Node::removeCachedTagNodeList): Remove TagNodeList from the cache
2605         (WebCore::Node::getElementsByTagNameNS): Switch to caching node lists themselves, not the data
2606         (WebCore::Node::getElementsByName): Switch to caching node lists themselves, not the data
2607         (WebCore::Node::getElementsByClassName): Switch to caching node lists themselves, not the data
2608         (WebCore::NodeListsNodeData::invalidateCaches): Switch to caching node lists themselves, not the data
2609         (WebCore::NodeListsNodeData::invalidateCachesThatDependOnAttributes): Switch to caching node lists themselves, not the data
2610         (WebCore::NodeListsNodeData::isEmpty): Switch to caching node lists themselves, not the data
2611         (WebCore::markNodeLists): Helper to mark all the node lists in the cache
2612         (WebCore::Node::markCachedNodeListsSlow): Mark all the cached node lists if any could be present
2613         * dom/Node.h:
2614         (WebCore::Node::markCachedNodeLists): Fast-path marking of cached node lists---bails out if there is no rare data
2615         * dom/NodeRareData.h: Changed type of caches to hold raw pointers to node lists, not RefPtr's to data
2616         * dom/TagNodeList.cpp:
2617         (WebCore::TagNodeList::TagNodeList): Don't need DynamicNodeList::Caches argument any more
2618         (WebCore::TagNodeList::~TagNodeList): Remove from the cache
2619         * dom/TagNodeList.h:
2620         (WebCore::TagNodeList::create): Don't need DynamicNodeList::Caches argument any more
2621
2622 2010-04-29  Mikhail Naganov  <mnaganov@chromium.org>
2623
2624         Reviewed by Yury Semikhatsky.
2625
2626         [Chromium] Update ScriptProfileNode to retrieve execution time in milliseconds.
2627
2628         https://bugs.webkit.org/show_bug.cgi?id=38330
2629
2630         * bindings/v8/ScriptProfileNode.cpp:
2631         (WebCore::ScriptProfileNode::totalTime):
2632         (WebCore::ScriptProfileNode::selfTime):
2633
2634 2010-04-28  Dimitri Glazkov  <dglazkov@chromium.org>
2635
2636         Reviewed by Darin Adler.
2637
2638         A form without a submit button unexpectedly performs its action when Return is pressed
2639         https://bugs.webkit.org/show_bug.cgi?id=9756
2640
2641         Implemented implicit form submission algorithm as defined in HTML5 spec:
2642         http://www.whatwg.org/specs/web-apps/current-work/multipage/association-of-controls-and-forms.html#implicit-submission,
2643         falling back to match IE's behavior in the edge cases.
2644
2645         The underlying rules are:
2646
2647         * If the form has no enabled submit buttons, submit if Enter/Return is pressed on the only single-line text field.
2648
2649         * Otherwise, submit form using first enabled submit button if Enter/Return is pressed on a field
2650           that's not a textarea or a select.
2651
2652         Test: fast/forms/implicit-submission.html
2653
2654         * dom/SelectElement.cpp:
2655         (WebCore::SelectElement::menuListDefaultEventHandler): Ripped out implicit submission for select elements.
2656         (WebCore::SelectElement::listBoxDefaultEventHandler): Ditto.
2657         (WebCore::SelectElement::defaultEventHandler): Ditto.
2658         * dom/SelectElement.h: Ditto.
2659         (WebCore::HTMLFormElement::submitImplicitly): Renamed submitClick to submitImplicitly to better match HTML5 spec
2660             language, changed the logic to match the rules above.
2661         * html/HTMLInputElement.cpp:
2662         (WebCore::HTMLInputElement::defaultEventHandler): Reamed clickDefaultButton to implicitSubmission to better match
2663             HTML5 spec language, made radio element to trigger implicit submission.
2664         * html/HTMLSelectElement.cpp:
2665         (WebCore::HTMLSelectElement::defaultEventHandler): Removed passing of form() as parameter, because it's no longer
2666             necessary.
2667
2668 2010-04-29  Paweł Hajdan, Jr.  <phajdan.jr@chromium.org>
2669
2670         Reviewed by Jeremy Orlow.
2671
2672         Fix building with libpng-1.4.
2673         https://bugs.webkit.org/show_bug.cgi?id=33287
2674
2675         No new tests (no behavior change).
2676
2677         Original patch by John Bowler <jbowler@acm.org>
2678
2679         * platform/image-encoders/skia/PNGImageEncoder.cpp:
2680         (WebCore::PNGImageEncoder::encode):
2681
2682 2010-04-29  Adam Langley  <agl@chromium.org>
2683
2684         Reviewed by David Levin.
2685
2686         This patch adds support for WOFF in Chromium. Since Chromium
2687         already transcodes all OpenType files for security reasons we
2688         are adding WOFF support into the transcoder.
2689
2690         https://bugs.webkit.org/show_bug.cgi?id=38217
2691
2692         * css/CSSFontFaceSrcValue.cpp:
2693         (WebCore::CSSFontFaceSrcValue::isSupportedFormat):
2694           Recognise "woff" as a font-face format value (guarded by
2695           ENABLE(OPENTYPE_SANITIZER) at this point)
2696         * platform/graphics/opentype/OpenTypeSanitizer.cpp:
2697         (WebCore::OpenTypeSanitizer::sanitize):
2698           Change so that the transcoded font can be larger than the original.
2699           (WOFF files are compressed, so the transcoded TTF is typically
2700           larger.)
2701
2702 2010-04-29  Alex Milowski  <alex@milowski.com>
2703
2704         Reviewed by Kenneth Rohde Christiansen.
2705
2706         Updates to the Qt build to enable building MathML support.
2707
2708         * WebCore.pri:
2709         * WebCore.pro:
2710
2711 2010-04-29  Xan Lopez  <xlopez@igalia.com>
2712
2713         Reviewed by Gustavo Noronha.
2714
2715         [GTK] GObject DOM bindings
2716         https://bugs.webkit.org/show_bug.cgi?id=33590
2717
2718         Do not generate unneeded boilerplate in {get,set}_property methods
2719         when there are no properties to generate code for. This gets rid
2720         of lots of compiler warnings.
2721
2722         * bindings/scripts/CodeGeneratorGObject.pm:
2723
2724 2010-04-29  Simon Hausmann  <simon.hausmann@nokia.com>
2725
2726         Reviewed by Kenneth Rohde Christiansen.
2727
2728         [Qt] Speed up text layouting
2729         https://bugs.webkit.org/show_bug.cgi?id=31719
2730
2731         Use QFontMetrics::width() for the text width calculation instead
2732         of QTextLayout. This avoids expensive bearing calculations and the
2733         line breaking code.
2734
2735         * platform/graphics/qt/FontQt.cpp:
2736         (WebCore::Font::floatWidthForComplexText):
2737
2738 2010-04-29  Kwang Yul Seo  <skyul@company100.net>
2739
2740         Reviewed by Simon Hausmann.
2741
2742         [WINCE] Export g_stackBase with JS_EXPORTDATA
2743         https://bugs.webkit.org/show_bug.cgi?id=37437
2744
2745         Declare g_stackBase with JS_EXPORTDATA as it is imported from JavaScriptCore.
2746
2747         * platform/wince/SharedTimerWince.cpp:
2748
2749 2010-04-29  Adam Barth  <abarth@webkit.org>
2750
2751         Reviewed by Maciej Stachowiak.
2752
2753         Remove custom bindings for PopStateEvent.initPopStateEvent
2754         https://bugs.webkit.org/show_bug.cgi?id=38311
2755
2756         Our code generation of SerializedScriptValue was slightly buggy, but
2757         it's easy to fix.  Notice that the conversion to an atomic string is
2758         handled by the C++ type system and doesn't require logic in the code
2759         generator.
2760
2761         * bindings/js/JSPopStateEventCustom.cpp:
2762         * bindings/scripts/CodeGeneratorJS.pm:
2763         * bindings/scripts/CodeGeneratorV8.pm:
2764         * bindings/scripts/test/GObject/WebKitDOMTestObj.cpp:
2765         (webkit_dom_test_obj_serialized_value):
2766         * bindings/scripts/test/GObject/WebKitDOMTestObj.h:
2767         * bindings/scripts/test/JS/JSTestObj.cpp:
2768         (WebCore::):
2769         (WebCore::jsTestObjPrototypeFunctionSerializedValue):
2770         * bindings/scripts/test/JS/JSTestObj.h:
2771         * bindings/scripts/test/ObjC/DOMTestObj.h:
2772         * bindings/scripts/test/ObjC/DOMTestObj.mm:
2773         (-[DOMTestObj serializedValue:]):
2774         * bindings/scripts/test/TestObj.idl:
2775         * bindings/scripts/test/V8/V8TestObj.cpp:
2776         (WebCore::TestObjInternal::serializedValueCallback):
2777         (WebCore::ConfigureV8TestObjTemplate):
2778         * bindings/v8/custom/V8PopStateEventCustom.cpp:
2779         * dom/PopStateEvent.idl:
2780
2781 2010-04-29  Kent Tamura  <tkent@chromium.org>
2782
2783         Reviewed by Darin Adler.
2784
2785         Fix a bug that selection drag-and-drop doesn't work for input/textarea.
2786         https://bugs.webkit.org/show_bug.cgi?id=38175
2787
2788         The code supposed the selected region was in the destination text
2789         field. It is not true in a case of drag-and-drop.
2790
2791         Test: editing/pasteboard/drag-drop-input-textarea.html
2792
2793         * dom/InputElement.cpp:
2794         (WebCore::InputElement::handleBeforeTextInsertedEvent):
2795         * html/HTMLTextAreaElement.cpp:
2796         (WebCore::HTMLTextAreaElement::handleBeforeTextInsertedEvent):
2797
2798 2010-04-29  Adam Barth  <abarth@webkit.org>
2799
2800         Reviewed by Maciej Stachowiak.
2801
2802         Remove custom bindings for NodeFilter.acceptNode
2803         https://bugs.webkit.org/show_bug.cgi?id=38309
2804
2805         This "custom" code was just an instance of the CallWith=ScriptState
2806         pattern.  Also, it looks like V8 just had a garbage implemenation that
2807         did nothing.
2808
2809         * WebCore.gypi:
2810         * bindings/js/JSNodeFilterCustom.cpp:
2811         * bindings/v8/custom/V8NodeFilterCustom.cpp: Removed.
2812         * dom/NodeFilter.idl:
2813
2814 2010-04-29  Noam Rosenthal  <noam.rosenthal@nokia.com>
2815
2816         Reviewed by Laszlo Gombos.
2817
2818         [Qt] QtWebkit doesn't link with 3D canvas
2819         https://bugs.webkit.org/show_bug.cgi?id=38299
2820
2821         Added implementation for GraphicsContext3D::isGLES2Compliant, which returns the correct value based on a Qt #define.
2822
2823         No new tests: build fix
2824
2825         * platform/graphics/qt/GraphicsContext3DQt.cpp:
2826         (WebCore::GraphicsContext3D::isGLES2Compliant):
2827
2828 2010-04-29  Simon Hausmann  <simon.hausmann@nokia.com>
2829
2830         Reviewed by Tor Arne Vestbø.
2831
2832         [Qt] REGRESSION(r57638): tst_qwebframe::objectDeleted() fails
2833         https://bugs.webkit.org/show_bug.cgi?id=38316
2834
2835         Accessing properties of a deleted objects doesn't throw an exception
2836         anymore.
2837
2838         Continue to expose the QObject class wrapper for objects that
2839         previously existed but don't exist anymore. QtClass is safe to
2840         use with a deleted QObject.
2841
2842         * bridge/qt/qt_instance.cpp:
2843         (JSC::Bindings::QtInstance::getClass): Return null only if m_class
2844         doesn't exist yet and there's no m_object.
2845
2846 2010-04-29  Noam Rosenthal  <noam.rosenthal@nokia.com>
2847
2848         Reviewed by Simon Hausmann.
2849
2850         Reverse animations don't work in some use cases        
2851         https://bugs.webkit.org/show_bug.cgi?id=38075
2852
2853         This was due to a code path special-casing reverse animations, that became obselete when we aligned our animation code
2854         with the CA implementation. That special case code path is now a bug - and this patch removes it.
2855
2856         http://staff.washington.edu/fmf/2009/03/25/iphone-3d-css-transformations/ now runs the reverse
2857         animation correctly.
2858
2859         * platform/graphics/qt/GraphicsLayerQt.cpp:
2860         (WebCore::TransformAnimationQt::applyFrame):
2861
2862 2010-04-29  Jan Hanssen  <jhanssen@codeaurora.org>
2863
2864         Reviewed by Adam Barth.
2865
2866         HTMLOptionElement::ownerSelectElement() needs to consider keygen elements
2867         https://bugs.webkit.org/show_bug.cgi?id=26016
2868
2869         Patch written by Grace Kloba <klobag@gmail.com>, test fixed by me.
2870
2871         Test: fast/dom/HTMLKeygenElement/keygen-option-select.html
2872
2873         * html/HTMLOptionElement.cpp:
2874         (WebCore::HTMLOptionElement::ownerSelectElement):
2875         Make HTMLOptionElement::ownerSelectElement() consider the keygen element in addition to the current select element.
2876
2877 2010-04-29  Gustavo Sverzut Barbieri  <barbieri@profusion.mobi>
2878
2879         Reviewed by Eric Seidel.
2880
2881         Add EFL-specific code to Widget.h and move the empty
2882         frameRectsChanged() definition to Widget.cpp, since the EFL port
2883         needs to override that.
2884         http://webkit.org/b/36317
2885
2886         No new tests required.
2887
2888         * WebCore.base.exp:
2889         * platform/Widget.cpp:
2890         (WebCore::Widget::frameRectsChanged):
2891         * platform/Widget.h:
2892
2893 2010-04-29  Jarkko Sakkinen  <jarkko.j.sakkinen@gmail.com>
2894
2895         Reviewed by Simon Hausmann.
2896
2897         [Qt] GraphicsContext3DQt.cpp does not implement isGLES2Compliant() 
2898         https://bugs.webkit.org/show_bug.cgi?id=38216
2899
2900         * platform/graphics/qt/GraphicsContext3DQt.cpp:
2901         (WebCore::GraphicsContext3D::isGLES2Compliant):
2902
2903 2010-04-29  Zhenyao Mo  <zmo@google.com>
2904
2905         Reviewed by Dimitri Glazkov.
2906
2907         Remove the unnecessary texImage2D function with Image as input in GraphicsContext3D
2908         https://bugs.webkit.org/show_bug.cgi?id=38235
2909
2910         * html/canvas/WebGLRenderingContext.cpp:
2911         (WebCore::WebGLRenderingContext::texImage2D): Add extractImageData; add a common entry point for texImage2D with image input.
2912         (WebCore::WebGLRenderingContext::texImage2DBase): Add this function as the common entry point for texImage2D.
2913         (WebCore::WebGLRenderingContext::texSubImage2D): Add extractImageData; add a common entry point for texSubImage2D with image input.
2914         (WebCore::WebGLRenderingContext::texSubImage2DBase): Add this function as the common entry point for texSubImage2D.
2915         * html/canvas/WebGLRenderingContext.h: Add tex*Image{Base/Image} function declaration.
2916         * platform/graphics/GraphicsContext3D.h: Remove tex*Image declaration with Image input.
2917         * platform/graphics/mac/GraphicsContext3DMac.cpp: Remove tex*Image implementation with Image input.
2918         (WebCore::GraphicsContext3D::texImage2D):
2919         (WebCore::GraphicsContext3D::texSubImage2D):
2920
2921 2010-04-29  Noam Rosenthal  <noam.rosenthal@nokia.com>
2922
2923         Reviewed by Simon Hausmann.
2924
2925         [Qt] GraphicsLayer: flicker when starting an animation before the previous one ended.
2926         https://bugs.webkit.org/show_bug.cgi?id=38076
2927
2928         This was due to the cude in the removeAnimations functions, which called deleteLater() without stopping the
2929         animation synchronously. The delay between the call to that function and the actual call to the animation's destructor
2930         is when the flicker occured. We fix this by calling stop() synchronously, and making sure that the value is reverted
2931         upon stop (updateState) and not upon the object's destruction.
2932
2933         http://staff.washington.edu/fmf/2009/03/25/iphone-3d-css-transformations/ now doesn't flicker when
2934         the animation is toggled frequently.
2935
2936         * platform/graphics/qt/GraphicsLayerQt.cpp:
2937         (WebCore::TransformAnimationQt::~TransformAnimationQt):
2938         (WebCore::TransformAnimationQt::applyFrame):
2939         (WebCore::TransformAnimationQt::updateState):
2940         (WebCore::OpacityAnimationQt::~OpacityAnimationQt):
2941         (WebCore::OpacityAnimationQt::updateState):
2942         (WebCore::GraphicsLayerQt::removeAnimationsForProperty):
2943         (WebCore::GraphicsLayerQt::removeAnimationsForKeyframes):
2944
2945 2010-04-28  Luiz Agostini  <luiz.agostini@openbossa.org>
2946
2947         Reviewed by Simon Fraser.
2948
2949         Media queries empty values
2950         https://bugs.webkit.org/show_bug.cgi?id=38116
2951
2952         Adding isValid() method to MediaQueryExp to make it possible to differentiate
2953         between queries with empty values and queries with invalid values.
2954
2955         Test: fast/media/media-query-invalid-value.html
2956
2957         * css/MediaQueryEvaluator.cpp:
2958         (WebCore::MediaQueryEvaluator::eval):
2959         * css/MediaQueryExp.cpp:
2960         (WebCore::MediaQueryExp::MediaQueryExp):
2961         * css/MediaQueryExp.h:
2962         (WebCore::MediaQueryExp::isValid):
2963
2964 2010-04-28  Pavel Feldman  <pfeldman@chromium.org>
2965
2966         Reviewed by Yury Semikhatsky.
2967
2968         Web Inspector: Allow editing script resources when resource tracking is enabled.
2969
2970         https://bugs.webkit.org/show_bug.cgi?id=38269
2971
2972         * inspector/front-end/ScriptView.js:
2973         * inspector/front-end/ScriptsPanel.js:
2974         (WebInspector.ScriptsPanel.prototype.canEditScripts):
2975         (WebInspector.ScriptsPanel.prototype.editScriptSource):
2976         * inspector/front-end/SourceFrame.js:
2977         (WebInspector.SourceFrame.prototype.get textModel):
2978         * inspector/front-end/SourceView.js:
2979         (WebInspector.SourceView):
2980         (WebInspector.SourceView.prototype._addBreakpoint):
2981         (WebInspector.SourceView.prototype._editLine):
2982         (WebInspector.SourceView.prototype._editLineComplete):
2983         (WebInspector.SourceView.prototype._sourceIDForLine):
2984
2985 2010-04-25  Antonio Gomes  <tonikitoo@webkit.org>
2986
2987         Reviewed by Darin Adler.
2988
2989         Clicking a scrollbar blurs the currently focused element
2990         https://bugs.webkit.org/show_bug.cgi?id=16809
2991
2992         WebKit ports that do not use native (platform) widgets for rendering the scrollbars
2993         are currently mismatching a common behaviour of other browser vendors (including
2994         Firefox and Opera): clicking on a frame scrollbar *should not* steal focus from content.
2995         WebKit browsers based ports that do use WebCore for handling scrollbars, like QtWebKit for
2996         instance, can observe the opposite behaviour.
2997
2998         Patch fixes this behaviour by checking if current MouseDown event targets a frame scrollbar.
2999         If that is the case, it bails out and do not change focus behavior at all.
3000
3001         It is important to note that the given way this is implemented by this patch, non-frame
3002         scrollbars (e.g. a vertical scrollbar of an overflowed textareas or divs) will keep working
3003         in the same way as it works currently.
3004
3005         Tests: scrollbars/scrollbar-click-does-not-blur-content.html
3006                scrollbars/scrollbar-iframe-click-does-not-blur-content.html
3007
3008         * page/EventHandler.cpp:
3009         (WebCore::EventHandler::dispatchMouseEvent):
3010
3011 2010-04-28  Dan Bernstein  <mitz@apple.com>
3012
3013         Rubber-stamped by Maciej Stachowiak.
3014
3015         Try to fix test failures seen on the Leopard build bot after r58467
3016
3017         * platform/graphics/SimpleFontData.cpp:
3018         (WebCore::SimpleFontData::platformGlyphInit): Initialize m_zeroWidthSpaceGlyph before
3019         calling widthForGlyph(), as the latter uses the former.
3020
3021 2010-04-28  Martin Robinson  <mrobinson@webkit.org>
3022
3023         Reviewed by Gustavo Noronha.
3024
3025         [GTK] Enable DOM clipboard and drag-and-drop access
3026         https://bugs.webkit.org/show_bug.cgi?id=30623
3027
3028         Make ClipboardGtk a "live" DataTransfer object, able to modify
3029         the clipboard when setData(...) is called.
3030
3031         * platform/gtk/ClipboardGtk.cpp:
3032         (WebCore::Editor::newGeneralClipboard): Pass the GtkClipboard into the factory method.
3033         (WebCore::ClipboardGtk::ClipboardGtk): Create two separate constructors, one for DnD data and one for clipboard data.
3034         (WebCore::dataObjectTypeFromHTMLClipboardType): Added.
3035         (WebCore::ClipboardGtk::clearData): Clear the member DataObject, optionally write the clipboard.
3036         (WebCore::ClipboardGtk::clearAllData): Ditto.
3037         (WebCore::joinURIList): Added.
3038         (WebCore::ClipboardGtk::getData): Get the data from the clipboard, if possible.
3039         (WebCore::ClipboardGtk::setData): Write data to the DataObject and maybe the clipboard.
3040         (WebCore::ClipboardGtk::types): Read the clipboard/DataObject to find applicable types.
3041         (WebCore::ClipboardGtk::files): Read the clipboard/DataObject to find the files.
3042         (WebCore::ClipboardGtk::writeURL): Write to the DataObject and maybe the clipboard.
3043         (WebCore::ClipboardGtk::writeRange): Ditto.
3044         (WebCore::ClipboardGtk::writePlainText): Ditto.
3045         (WebCore::ClipboardGtk::hasData): Query the clipboard/DataObject.
3046         * platform/gtk/ClipboardGtk.h: 
3047         (WebCore::ClipboardGtk::create): Added one factory for pasteboard-backed DataObjects and one for DnD-backed objects.
3048         * platform/gtk/DataObjectGtk.cpp:
3049         (WebCore::replaceNonBreakingSpaceWithSpace): Added this helper function to clean &nbsp; from plain text.
3050         (WebCore::DataObjectGtk::markup): Actually return the range if it's set.
3051         (WebCore::DataObjectGtk::setText): Clean &nbsp; from plain text.
3052         (WebCore::DataObjectGtk::clearText): Added.
3053         (WebCore::DataObjectGtk::clearMarkup): Added.
3054         * platform/gtk/DataObjectGtk.h:
3055         (WebCore::DataObjectGtk::clearURIList): Added.
3056         (WebCore::DataObjectGtk::clearImage): Added.
3057         * platform/gtk/PasteboardHelper.cpp:
3058         (WebCore::PasteboardHelper::initializeTargetList): Added target ID's for URI list and Netscape URL targets.
3059         (WebCore::urisToKURLVector): Added.
3060         (WebCore::PasteboardHelper::getClipboardContents): Added.
3061         (WebCore::PasteboardHelper::fillSelectionData): Added logic for URI lists and Netscape URLs.
3062         (WebCore::PasteboardHelper::targetListForDataObject): Ditto.
3063         * platform/gtk/PasteboardHelper.h: Added default argument to writeClipboardContents and new method.
3064
3065 2010-04-28  Martin Robinson  <mrobinson@webkit.org>
3066
3067         Reviewed by Gustavo Noronha.
3068
3069         [GTK] Enable DOM clipboard and drag-and-drop access
3070         https://bugs.webkit.org/show_bug.cgi?id=30623
3071
3072         Use the length of the UTF-8 markup data in bytes when filling GtkSelectionData.
3073
3074         No new tests, because pasteboard DataTransfer tests will be enabled
3075         with the completion of the GTK+ DataTransfer object.
3076
3077         * platform/gtk/PasteboardGtk.cpp:
3078         (WebCore::clipboard_get_contents_cb): Use strlen here instead of g_utf8_strlen.
3079
3080 2010-04-28  Dan Bernstein  <mitz@apple.com>
3081
3082         Reviewed by Adele Peterson.
3083
3084         More of <rdar://problem/7855777> REGRESSION: Memory usage increase caused by storing glyph bounds in GlyphMetricsMap
3085         https://bugs.webkit.org/show_bug.cgi?id=37936
3086
3087         Assigning zero width and empty bounds to the ZERO WIDTH SPACE glyph often allocates a width map
3088         page and a bounds map page for the glyph, each of which pages contains 255 unused entries. Save
3089         this space by not storing the zero width and empty bounds in the metrics maps.
3090
3091         * platform/graphics/SimpleFontData.cpp:
3092         (WebCore::SimpleFontData::SimpleFontData): Initialize m_zeroWidthSpaceGlyph.
3093         (WebCore::SimpleFontData::platformGlyphInit): Set m_zeroWidthSpaceGlyph. Don’t create entries
3094         for the ZERO WIDTH SPACE glyph in the width map and in the bounds map.
3095         * platform/graphics/SimpleFontData.h:
3096         (WebCore::SimpleFontData::boundsForGlyph): Return empty bounds for the ZERO WIDTH SPACE glyph
3097         without consulting the bounds map.
3098         (WebCore::SimpleFontData::widthForGlyph): Return zero for the ZERO WIDTH SPACE glyph without
3099         consulting the width map.
3100
3101 2010-04-28  Mark Rowe  <mrowe@apple.com>
3102
3103         Reviewed by Adele Peterson.
3104
3105         <rdar://problem/7847573> Safari welcome page logs JavaScript exception during page load
3106
3107         The Safari welcome page runs afoul of the same-origin restriction on access to stylesheet rules
3108         that was added to address <https://bugs.webkit.org/show_bug.cgi?id=20527>. To work around this we're
3109         adding a temporary site-specific quirk that relaxes this restriction for the particular cross-origin
3110         access that occurs on the Safari welcome page.
3111
3112         * css/CSSStyleSheet.cpp:
3113         (WebCore::CSSStyleSheet::cssRules):
3114
3115 2010-04-28  Mike Thole  <mthole@apple.com>
3116
3117         Reviewed by Mark Rowe.
3118
3119         Add separate exports file for symbols dependent on WTF_USE_PROTECTION_SPACE_AUTH_CALLBACK.
3120
3121         * DerivedSources.make: Updated for WebCore.ProtectionSpaceAuthCallback.exp.
3122         * WebCore.ProtectionSpaceAuthCallback.exp: Added.
3123         * WebCore.xcodeproj/project.pbxproj:
3124
3125 2010-04-28  Evan Martin  <evan@chromium.org>
3126
3127         Reviewed by David Levin.
3128
3129         [chromium] revert getFontDataForCharacters change in r58341
3130         https://bugs.webkit.org/show_bug.cgi?id=38288
3131
3132         It caused a performance regression.
3133
3134         * platform/chromium/ChromiumBridge.h:
3135         * platform/graphics/chromium/FontCacheLinux.cpp:
3136         (WebCore::FontCache::getFontDataForCharacters):
3137
3138 2010-04-28  Eric Seidel  <eric@webkit.org>
3139
3140         Unreviewed, rolling out r58441.
3141         http://trac.webkit.org/changeset/58441
3142         https://bugs.webkit.org/show_bug.cgi?id=37618
3143
3144         Broke 3 test on Qt.
3145
3146         * page/DragController.cpp:
3147         (WebCore::DragController::tryDocumentDrag):
3148         (WebCore::DragController::concludeEditDrag):
3149         * page/DragController.h:
3150         (WebCore::DragController::setDragInitiator):
3151         (WebCore::DragController::dragInitiator):
3152         (WebCore::DragController::documentUnderMouse):
3153
3154 2010-04-28  Darin Adler  <darin@apple.com>
3155
3156         Reviewed by Dan Bernstein.
3157
3158         Remove some obsolete scrolling code
3159         https://bugs.webkit.org/show_bug.cgi?id=38293
3160
3161         * page/Chrome.cpp:
3162         (WebCore::Chrome::scrollRectIntoView): Removed ScrollView* argument.
3163         * page/Chrome.h: Made scrollRectIntoView non-virtual, and removed ScrollView* argument.
3164
3165         * page/FrameView.cpp: Removed scrollRectIntoViewRecursively.
3166         * page/FrameView.h: Removed scrollRectIntoViewRecursively and made
3167         setScrollPosition non-virtual, since there is no class derived from
3168         this class, and ScrollView's setScrollPosition is non-virtual.
3169
3170         * platform/HostWindow.h: Removed scrollRectIntoView.
3171
3172         * platform/ScrollView.cpp:
3173         (WebCore::ScrollView::scrollRectIntoViewRecursively): Updated comment
3174         since I was able to do most of the tasks listed here.
3175
3176         * rendering/RenderLayer.cpp:
3177         (WebCore::RenderLayer::scrollRectToVisible): Removed 0 passed for
3178         ScrollView* argument to Chrome::scrollRectIntoView.
3179
3180 2010-04-28  Mike Thole  <mthole@apple.com>
3181
3182         Reviewed by David Kilzer.
3183
3184         Add canAuthenticateAgainstProtectionSpace() to frame loader so that a protection space 
3185         can be inspected before attempting to authenticate against it
3186         https://bugs.webkit.org/show_bug.cgi?id=38271
3187
3188         * loader/EmptyClients.h:
3189         (WebCore::EmptyFrameLoaderClient::canAuthenticateAgainstProtectionSpace): Added.
3190         * loader/FrameLoader.cpp:
3191         (WebCore::FrameLoader::canAuthenticateAgainstProtectionSpace): Added.
3192         * loader/FrameLoader.h:
3193         * loader/FrameLoaderClient.h:
3194         * loader/ResourceLoader.cpp:
3195         (WebCore::ResourceLoader::canAuthenticateAgainstProtectionSpace): Added.
3196         * loader/ResourceLoader.h:
3197         (WebCore::ResourceLoader::canAuthenticateAgainstProtectionSpace): Added.
3198         * platform/network/ProtectionSpace.h:
3199         (WebCore::):
3200         * platform/network/ResourceHandle.h:
3201         * platform/network/ResourceHandleClient.h:
3202         (WebCore::ResourceHandleClient::canAuthenticateAgainstProtectionSpace): Added.
3203         * platform/network/mac/AuthenticationMac.mm:
3204         (WebCore::mac):
3205         Add cases for case ProtectionSpaceAuthenticationSchemeServerTrustEvaluationRequested and
3206         ProtectionSpaceAuthenticationSchemeClientCertificateRequested.  Guarded with the
3207         #if USE(PROTECTION_SPACE_AUTH_CALLBACK) as the new NSURLProtectionSpace constants don't
3208         exist prior to Mac OS X 10.6.
3209         (WebCore::core): Ditto.
3210         * platform/network/mac/ResourceHandleMac.mm:
3211         (WebCore::ResourceHandle::canAuthenticateAgainstProtectionSpace): Added.
3212         (-[WebCoreResourceHandleAsDelegate connection:canAuthenticateAgainstProtectionSpace:]): Added.
3213
3214 2010-04-28  Abhishek Arya  <inferno@chromium.org>
3215
3216         Reviewed by Eric Seidel.
3217
3218         Convert m_documentUnderMouse, m_dragInitiator to RefPtr.
3219         Eliminated unused m_dragInitiator accessor to prevent dereferencing.
3220         https://bugs.webkit.org/show_bug.cgi?id=37618
3221
3222         Test: editing/pasteboard/drag-drop-iframe-refresh-crash.html
3223
3224         * page/DragController.cpp:
3225         (WebCore::DragController::tryDocumentDrag):
3226         (WebCore::DragController::concludeEditDrag):
3227         * page/DragController.h:
3228         (WebCore::DragController::draggingImageURL):
3229         (WebCore::DragController::documentUnderMouse):
3230
3231 2010-04-28  Dumitru Daniliuc  <dumi@chromium.org>
3232
3233         Unreviewed, fixing a build problem introduced by the previous patch.
3234
3235         * storage/DatabaseSync.cpp:
3236         (WebCore::DatabaseSync::openDatabaseSync):
3237
3238 2010-04-23  Dumitru Daniliuc  <dumi@chromium.org>
3239
3240         Reviewed by Jeremy Orlow.
3241
3242         Adding some required classes for the sync WebSQLDatabases API.
3243         https://bugs.webkit.org/show_bug.cgi?id=34994
3244
3245         * GNUmakefile.am:
3246         * WebCore.gypi:
3247         * WebCore.pro:
3248         * WebCore.vcproj/WebCore.vcproj:
3249         * WebCore.xcodeproj/project.pbxproj:
3250         * storage/DatabaseSync.cpp: Added.
3251         (WebCore::DatabaseSync::databaseInfoTableName):
3252         (WebCore::DatabaseSync::setIsAvailable):
3253         (WebCore::DatabaseSync::isAvailable):
3254         (WebCore::DatabaseSync::openDatabaseSync):
3255         (WebCore::DatabaseSync::DatabaseSync):
3256         (WebCore::DatabaseSync::~DatabaseSync):
3257         (WebCore::DatabaseSync::version):
3258         (WebCore::DatabaseSync::changeVersion):
3259         (WebCore::DatabaseSync::transaction):
3260         * storage/DatabaseSync.h: Added.
3261         (WebCore::DatabaseSync::databaseDebugName):
3262         * storage/SQLTransactionSync.cpp: Added.
3263         (WebCore::SQLTransactionSync::create):
3264         (WebCore::SQLTransactionSync::SQLTransactionSync):
3265         (WebCore::SQLTransactionSync::~SQLTransactionSync):
3266         (WebCore::SQLTransactionSync::executeSQL):
3267         * storage/SQLTransactionSync.h: Added.
3268         (WebCore::SQLTransactionSync::database):
3269         (WebCore::SQLTransactionSync::isReadOnly):
3270         * storage/SQLTransactionSyncCallback.h: Added.
3271         (WebCore::SQLTransactionSyncCallback::~SQLTransactionSyncCallback):
3272
3273 2010-04-28  İsmail Dönmez  <ismail@namtrac.org>
3274
3275         Reviewed by Simon Hausmann.
3276
3277         Fix compilation with QT_NO_CURSOR defined.
3278
3279         * platform/qt/QWebPageClient.h:
3280         (QWebPageClient::resetCursor):
3281         (QWebPageClient::setCursor):
3282
3283 2010-04-28  Abhishek Arya  <inferno@chromium.org>
3284
3285         Reviewed by Alexey Proskuryakov.
3286
3287         Added a check to make sure that resources from a different https origin are not cached.
3288         https://bugs.webkit.org/show_bug.cgi?id=33456
3289
3290         Test: http/tests/appcache/different-https-origin-resource-main.html
3291
3292         * loader/appcache/ManifestParser.cpp:
3293         (WebCore::parseManifest):
3294
3295 2010-04-28  Sam Weinig  <sam@webkit.org>
3296
3297         Reviewed by Mark Rowe.
3298
3299         Only set -allowable_client WebKit2 for engineering builds on SnowLeopard
3300         and later.
3301
3302         * Configurations/WebCore.xcconfig:
3303
3304 2010-04-28  Darin Adler  <darin@apple.com>
3305
3306         Reviewed by Adele Peterson.
3307
3308         REGRESSION: Autoscroll does not work in Mail messages
3309         https://bugs.webkit.org/show_bug.cgi?id=38267
3310         rdar://problem/7559799
3311
3312         Still haven't figured out a good way to test this with DumpRenderTree
3313         or with Safari. Testing has to be done with Mail for now.
3314
3315         The machinery to make autoscrolling work on Mac OS X when a WebView is embedded in another
3316         view had gotten broken in multiple ways. For some reason, a combination of bugs made it
3317         partly work until around r48064. This brings it back.
3318
3319         There were three problems:
3320
3321             1) Code in EventHandler decided there was nothing to scroll, so didn't start
3322                the autoscroll timer.
3323             2) The wrong rectangle was passed to Chrome::scrollRectIntoView.
3324             3) The Mac WebKit implementation of ChromeClient::scrollRectIntoView did incorrect
3325                coordinate conversion.
3326
3327         I verified that none of these have any effect on regression tests, or behavior in
3328         web browsers, or behavior on platforms other than Mac.
3329
3330         * page/EventHandler.cpp:
3331         (WebCore::canAutoscroll): Added. Returns true for boxes that can scroll directly
3332         and for the top level box of the top frame.
3333         (WebCore::EventHandler::handleMouseDraggedEvent): Use canAutoscroll.
3334         (WebCore::EventHandler::updateAutoscrollRenderer): Ditto.
3335
3336         * page/FrameView.cpp:
3337         (WebCore::FrameView::scrollToAnchor): Fixed comment.
3338
3339         * platform/ScrollView.cpp:
3340         (WebCore::ScrollView::scrollRectIntoViewRecursively): Put ASSERT_NOT_REACHED into this
3341         now-unused function along with some comments about removing some obsolete code.
3342
3343         * rendering/RenderLayer.cpp:
3344         (WebCore::RenderLayer::scrollRectToVisible): Removed call to scrollRectIntoViewRecursively
3345         since from the WebKit point of view this is the topmost scroll view anyway. Instead call
3346         setScrollPosition. Moved the code to call Chrome::scrollRectIntoView here since it needs
3347         to use a different rectangle anyway.
3348
3349 2010-04-21  Ojan Vafai  <ojan@chromium.org>
3350
3351         Reviewed by Adele Peterson.
3352
3353         http://trac.webkit.org/changeset/57215 caused perf/memory regressions
3354         https://bugs.webkit.org/show_bug.cgi?id=37292
3355
3356         #if out the canUseGlyphCache Changes from r57215 as they caused a 
3357         8% perf regression on Chromium's international page load tests so that
3358         the perf regression can be fixed properly without being left in the tree.
3359
3360         * platform/graphics/FontFastPath.cpp:
3361         (WebCore::Font::canUseGlyphCache):
3362
3363 2010-04-28  Beth Dakin  <bdakin@apple.com>
3364
3365         Reviewed by Darin Adler.
3366
3367         Fix for <rdar://problem/7474349>
3368
3369         Add a synchronous display mechanism for WKCACFLayerRenderer. 
3370
3371         * platform/graphics/win/WKCACFLayerRenderer.cpp:
3372         (WebCore::WKCACFLayerRenderer::setRootContentsAndDisplay): This is 
3373         just like setRootContents(), but it calls paint() instead of 
3374         renderSoon().
3375         * platform/graphics/win/WKCACFLayerRenderer.h:
3376
3377 2010-04-28  Dmitry Titov  <dimich@chromium.org>
3378
3379         Reviewed by Geoffrey Garen.
3380
3381         REGRESSION: fast/workers/wrapper-map-gc.html crashes on Snow Leopard Release Bot
3382         https://bugs.webkit.org/show_bug.cgi?id=37554
3383
3384         The flaky fast/workers/wrapper-map-gc.html will stop being flaky.
3385
3386         * bindings/js/JSEventListener.cpp:
3387         (WebCore::JSEventListener::handleEvent):
3388         check if JS execution was terminated, as in cases of Worker.terminate() or WorkerGlobalScope.close().
3389         * bindings/js/JSWorkerContextBase.cpp:
3390         (WebCore::toJS): ASSERT the value of workerContextWrapper, it should never be 0.
3391         * bindings/js/WorkerScriptController.h:
3392         (WebCore::WorkerScriptController::workerContextWrapper): remove returning 0 if JS execution was forbidden.
3393         (WebCore::WorkerScriptController::isExecutionForbidden):
3394         * bindings/v8/WorkerScriptController.h:
3395         (WebCore::WorkerScriptController::isExecutionForbidden):
3396
3397         Add ScriptExecutionContext::isJSExecutionTerminated(), it is always 'false' for Document
3398         and 'true' for WorkerContext when script is terminated.
3399         * dom/ScriptExecutionContext.h:
3400         * dom/Document.h:
3401         (WebCore::Document::isJSExecutionTerminated):
3402         * workers/WorkerContext.cpp:
3403         (WebCore::WorkerContext::isJSExecutionTerminated):
3404         * workers/WorkerContext.h:
3405
3406 2010-04-28  Ilya Tikhonovsky  <loislo@chromium.org>
3407
3408         Reviewed by Yury Semikhatsky.
3409
3410         WebInspector: Multiple Main Resource Content Loaded marks appear in Resource panel.
3411         https://bugs.webkit.org/show_bug.cgi?id=38270
3412
3413         * inspector/front-end/ResourcesPanel.js:
3414         (WebInspector.ResourcesPanel.prototype.updateGraphDividersIfNeeded):
3415
3416 2010-04-28  Ilya Tikhonovsky  <loislo@chromium.org>
3417
3418         Reviewed by Yury Semikhatsky.
3419
3420         WebInspector: Timeline: We can have precise urls in the EvaluateScript records.
3421         https://bugs.webkit.org/show_bug.cgi?id=38264
3422
3423         * inspector/front-end/TimelinePanel.js:
3424         (WebInspector.TimelinePanel.FormattedRecord.prototype._generatePopupContent):
3425         (WebInspector.TimelinePanel.FormattedRecord.prototype._getRecordDetails):
3426
3427 2010-04-28  Pavel Feldman  <pfeldman@chromium.org>
3428
3429         Reviewed by Timothy Hatcher.
3430
3431         Web Inspector: Linkify node and function in the event listeners panel.
3432
3433         https://bugs.webkit.org/show_bug.cgi?id=38251
3434
3435         * bindings/js/ScriptEventListener.cpp:
3436         (WebCore::eventListenerHandlerBody):
3437         (WebCore::eventListenerHandlerLocation):
3438         * bindings/js/ScriptEventListener.h:
3439         * bindings/v8/ScriptEventListener.cpp:
3440         (WebCore::eventListenerHandlerBody):
3441         (WebCore::eventListenerHandlerLocation):
3442         * bindings/v8/ScriptEventListener.h:
3443         * inspector/InspectorDOMAgent.cpp:
3444         (WebCore::InspectorDOMAgent::buildObjectForEventListener):
3445         * inspector/front-end/ElementsPanel.js:
3446         (WebInspector.ElementsPanel.prototype.linkifyNodeReference):
3447         (WebInspector.ElementsPanel.prototype.linkifyNodeReference.preventDefault):
3448         * inspector/front-end/EventListenersSidebarPane.js:
3449         (WebInspector.EventListenersSidebarPane.prototype.update.callback):
3450         (WebInspector.EventListenersSidebarPane.prototype.update):
3451         ():
3452         * inspector/front-end/StylesSidebarPane.js:
3453         (WebInspector.StylesSidebarPane.prototype._rebuildSectionsForStyleRules):
3454         * inspector/front-end/inspector.css:
3455         (.node-link):
3456
3457 2010-04-28  Julien Chaffraix  <jchaffraix@webkit.org>
3458
3459         Reviewed by Alexey Proskuryakov.
3460
3461         [XHR] Cross-Origin synchronous request with credential raises NETWORK_ERR
3462         https://bugs.webkit.org/show_bug.cgi?id=37781
3463         <rdar://problem/7905150>
3464
3465         Tests: http/tests/xmlhttprequest/access-control-preflight-credential-async.html
3466                http/tests/xmlhttprequest/access-control-preflight-credential-sync.html
3467
3468         Rolling the patch in as I could not reproduce Qt results locally.
3469
3470         * loader/DocumentThreadableLoader.cpp:
3471         (WebCore::DocumentThreadableLoader::DocumentThreadableLoader): Now we remove the
3472         credential from the request here to avoid forgetting to do so in the different code path.
3473         (WebCore::DocumentThreadableLoader::makeSimpleCrossOriginAccessRequest): Just add the
3474         "Origin" header.
3475         (WebCore::DocumentThreadableLoader::loadRequest): Check here the the credential have
3476         been removed so that we don't leak them. Also tweaked a comment to make it clear that
3477         the URL check has issue when credential is involved.
3478
3479 2010-04-28  Noam Rosenthal  <noam.rosenthal@nokia.com>
3480
3481         Reviewed by Kenneth Rohde Christiansen.
3482
3483         [Qt] GraphicsLayer: preserves-3d and backface visibility
3484         https://bugs.webkit.org/show_bug.cgi?id=35312
3485
3486         Implement preserves-3d by maintaining the 3D transformation heirarchy inside GraphicsLayerQt, and extrapolating
3487         the relative QTransform. When the extrapolation fails (un-invertible matrix) we ignore the transformation change.
3488
3489         WebKitSite/blog-files/3d-transforms test now work with Qt.
3490
3491         * platform/graphics/qt/GraphicsLayerQt.cpp:
3492         (WebCore::GraphicsLayerQtImpl::updateTransform):
3493         (WebCore::GraphicsLayerQtImpl::opaqueArea):
3494         (WebCore::GraphicsLayerQtImpl::boundingRect):
3495         (WebCore::GraphicsLayerQtImpl::paint):
3496         (WebCore::GraphicsLayerQtImpl::flushChanges):
3497
3498 2010-04-28  Ilya Tikhonovsky  <loislo@chromium.org>
3499
3500         Reviewed by Yury Semikhatsky.
3501
3502         Web Inspector: View frame selector for Timeline overview panel is redesigned a bit.
3503         Now it is possible to adjust view frame by resizer bars and by click-n-drag in overview pane.
3504         https://bugs.webkit.org/show_bug.cgi?id=38251
3505
3506         * inspector/front-end/TimelineOverviewPane.js:
3507         (WebInspector.TimelineOverviewPane):
3508         (WebInspector.TimelineOverviewPane.prototype.reset):
3509         (WebInspector.TimelineOverviewPane.prototype._dragWindow):
3510         (WebInspector.TimelineOverviewPane.prototype._windowSelectorDragging):
3511         (WebInspector.TimelineOverviewPane.prototype._endWindowSelectorDragging):
3512         (WebInspector.TimelineOverviewPane.prototype._resizeWindowLeft):
3513         (WebInspector.TimelineOverviewPane.prototype._resizeWindowRight):
3514         (WebInspector.TimelineOverviewPane.prototype._setWindowPosition):
3515         (WebInspector.TimelinePanel.WindowSelector):
3516         (WebInspector.TimelinePanel.WindowSelector.prototype._createSelectorElement):
3517         (WebInspector.TimelinePanel.WindowSelector.prototype._close):
3518         (WebInspector.TimelinePanel.WindowSelector.prototype._updatePosition):
3519         * inspector/front-end/inspector.css:
3520         (#timeline-overview-grid):
3521         (.timeline-window-selector):
3522         (#timeline-overview-window):
3523         (.timeline-overview-dividers-background):
3524         (.timeline-overview-window-rulers):
3525         (.timeline-window-resizer):
3526
3527 2010-04-28  Marcus Bulach  <bulach@chromium.org>
3528
3529         Reviewed by Jeremy Orlow.
3530
3531         Reverts 58340 (https://bugs.webkit.org/show_bug.cgi?id=38158) due to regression on fast/text/international/khmer-selection.html
3532         https://bugs.webkit.org/show_bug.cgi?id=38254
3533
3534         * platform/graphics/chromium/FontLinux.cpp:
3535         (WebCore::adjustTextRenderMode):
3536         (WebCore::TextRunWalker::getTextRun):
3537         (WebCore::TextRunWalker::getNormalizedTextRun):
3538         (WebCore::TextRunWalker::setGlyphXPositions):
3539         (WebCore::glyphIndexForXPositionInScriptRun):
3540         (WebCore::Font::offsetForPositionForComplexText):
3541         (WebCore::Font::selectionRectForComplexText):
3542         * platform/graphics/chromium/FontPlatformDataLinux.cpp:
3543         (WebCore::FontPlatformData::setupPaint):
3544         * platform/graphics/chromium/HarfbuzzSkia.cpp:
3545         (WebCore::getOutlinePoint):
3546         * platform/graphics/skia/GraphicsContext3DSkia.cpp:
3547         (WebCore::GraphicsContext3D::getImageData):
3548         * platform/graphics/skia/GraphicsContextSkia.cpp:
3549         (WebCore::isCoordinateSkiaSafe):
3550         (WebCore::GraphicsContext::fillRect):
3551         (WebCore::GraphicsContext::strokePath):
3552         (WebCore::GraphicsContext::strokeRect):
3553
3554 2010-04-28  Andrey Kosyakov  <caseq@chromium.org>
3555
3556         Reviewed by Yury Semikhatsky.
3557
3558         Log error message to inspector console if a resource fails to load.
3559         Disable checking of mime-type consistency for failed resources.
3560         https://bugs.webkit.org/show_bug.cgi?id=37215
3561
3562         Test: http/tests/inspector/console-resource-errors.html
3563
3564         * inspector/InspectorController.cpp:
3565         (WebCore::InspectorController::didReceiveResponse):
3566         (WebCore::InspectorController::didFailLoading):
3567         * inspector/front-end/Resource.js:
3568         (WebInspector.Resource.prototype._mimeTypeIsConsistentWithType):
3569         * inspector/front-end/ResourcesPanel.js:
3570         (WebInspector.ResourcesPanel.prototype.recreateViewForResourceIfNeeded):
3571
3572 2010-04-28  Yury Semikhatsky  <yurys@chromium.org>
3573
3574         Reviewed by Pavel Feldman.
3575
3576         Support pause on exceptions in v8 implementation of ScriptDebugServer.
3577
3578         https://bugs.webkit.org/show_bug.cgi?id=38205        
3579
3580         * bindings/v8/ScriptDebugServer.cpp:
3581         (WebCore::ScriptDebugServer::addListener):
3582         (WebCore::ScriptDebugServer::pauseOnExceptionsState):
3583         (WebCore::ScriptDebugServer::setPauseOnExceptionsState):
3584         (WebCore::ScriptDebugServer::currentCallFrame):
3585         (WebCore::ScriptDebugServer::handleV8DebugMessage):
3586         (WebCore::ScriptDebugServer::dispatchDidParseSource):
3587         * bindings/v8/ScriptDebugServer.h:
3588
3589 2010-04-28  Sheriff Bot  <webkit.review.bot@gmail.com>
3590
3591         Unreviewed, rolling out r58313.
3592         http://trac.webkit.org/changeset/58313
3593         https://bugs.webkit.org/show_bug.cgi?id=38253
3594
3595         Regression evident in pixel tests: the search icon is always
3596         clipped at the bottom. (Requested by jorlow on #webkit).
3597