1 2013-09-02 Anton Obzhirov <a.obzhirov@samsung.com>
3 <https://webkit.org/b/98350> [GTK] accessibility/aria-invalid.html times out
5 Reviewed by Mario Sanchez Prada.
7 The patch exposes aria-invalid attribute to ATK.
9 * accessibility/atk/AXObjectCacheAtk.cpp:
10 (WebCore::AXObjectCache::postPlatformNotification):
11 Added emitting state-change signal for aria-invalid event.
12 * accessibility/atk/WebKitAccessibleWrapperAtk.cpp:
13 (webkitAccessibleGetAttributes):
14 Added aria-invalid attribute.
16 2013-09-02 Commit Queue <commit-queue@webkit.org>
18 Unreviewed, rolling out r154955.
19 http://trac.webkit.org/changeset/154955
20 https://bugs.webkit.org/show_bug.cgi?id=120605
22 broke xcode4 build :| (Requested by kling on #webkit).
25 * GNUmakefile.list.am:
26 * WebCore.vcxproj/WebCore.vcxproj:
27 * WebCore.xcodeproj/project.pbxproj:
29 (defaultTagPropertyHash):
31 * html/HTMLAnchorElement.h:
32 (WebCore::isHTMLAnchorElement):
33 (WebCore::HTMLAnchorElement):
34 * html/HTMLAreaElement.h:
35 (WebCore::isHTMLAreaElement):
36 (WebCore::HTMLAreaElement):
37 * html/HTMLAudioElement.h:
38 (WebCore::isHTMLAudioElement):
39 * html/HTMLBaseElement.h:
40 (WebCore::isHTMLBaseElement):
41 (WebCore::HTMLBaseElement):
42 * html/HTMLCanvasElement.h:
43 (WebCore::isHTMLCanvasElement):
45 * html/HTMLFieldSetElement.h:
46 (WebCore::isHTMLFieldSetElement):
47 (WebCore::HTMLFieldSetElement):
48 * html/HTMLFormElement.h:
49 (WebCore::isHTMLFormElement):
50 * html/HTMLFrameSetElement.h:
51 (WebCore::isHTMLFrameSetElement):
52 (WebCore::HTMLFrameSetElement):
53 * html/HTMLImageElement.h:
54 (WebCore::isHTMLImageElement):
55 * html/HTMLInputElement.h:
56 (WebCore::isHTMLInputElement):
57 * html/HTMLLabelElement.h:
58 (WebCore::isHTMLLabelElement):
59 (WebCore::HTMLLabelElement):
60 * html/HTMLLegendElement.h:
61 (WebCore::isHTMLLegendElement):
62 (WebCore::HTMLLegendElement):
63 * html/HTMLMapElement.h:
64 (WebCore::isHTMLMapElement):
65 * html/HTMLMeterElement.h:
66 (WebCore::isHTMLMeterElement):
67 * html/HTMLOptGroupElement.h:
68 (WebCore::isHTMLOptGroupElement):
69 * html/HTMLOptionElement.h:
70 (WebCore::isHTMLOptionElement):
71 * html/HTMLParamElement.h:
72 (WebCore::isHTMLParamElement):
73 (WebCore::HTMLParamElement):
74 * html/HTMLProgressElement.h:
75 (WebCore::isHTMLProgressElement):
76 * html/HTMLScriptElement.h:
77 (WebCore::isHTMLScriptElement):
78 * html/HTMLSourceElement.h:
79 (WebCore::isHTMLSourceElement):
80 (WebCore::HTMLSourceElement):
81 * html/HTMLStyleElement.h:
82 (WebCore::isHTMLStyleElement):
83 (WebCore::HTMLStyleElement):
84 * html/HTMLTableElement.h:
85 (WebCore::isHTMLTableElement):
86 * html/HTMLTableRowElement.h:
87 (WebCore::isHTMLTableRowElement):
88 (WebCore::HTMLTableRowElement):
89 * html/HTMLTagNames.in:
90 * html/HTMLTextAreaElement.h:
91 (WebCore::isHTMLTextAreaElement):
92 * html/HTMLTitleElement.h:
93 (WebCore::isHTMLTitleElement):
94 (WebCore::HTMLTitleElement):
95 * html/HTMLTrackElement.h:
96 (WebCore::isHTMLTrackElement):
97 (WebCore::HTMLTrackElement):
99 * svg/SVGFontElement.h:
100 (WebCore::isSVGFontElement):
101 * svg/SVGFontFaceElement.h:
102 (WebCore::isSVGFontFaceElement):
103 (WebCore::SVGFontFaceElement):
104 * svg/SVGForeignObjectElement.h:
105 (WebCore::isSVGForeignObjectElement):
106 (WebCore::SVGForeignObjectElement):
107 * svg/SVGImageElement.h:
108 (WebCore::isSVGImageElement):
109 * svg/SVGScriptElement.h:
110 (WebCore::isSVGScriptElement):
113 2013-09-02 Antti Koivisto <antti@apple.com>
115 Clean up ContainerNode::childrenChanged
116 https://bugs.webkit.org/show_bug.cgi?id=120599
118 Reviewed by Andreas Kling.
120 - Make childrenChanged take a single struct argument instead of a long list of arguments.
121 - Use enum instead of childCountDelta. It was always -1, 0, 1 or the total number of children (in case of removing them all).
122 - Remove use of Node*, give the change range as Elements.
126 (WebCore::Attr::childrenChanged):
128 * dom/CharacterData.cpp:
129 (WebCore::CharacterData::parserAppendData):
130 (WebCore::CharacterData::dispatchModifiedEvent):
131 * dom/ContainerNode.cpp:
132 (WebCore::ContainerNode::insertBefore):
133 (WebCore::ContainerNode::notifyChildInserted):
134 (WebCore::ContainerNode::notifyChildRemoved):
136 Add private helpers for setting up the struct.
138 (WebCore::ContainerNode::parserInsertBefore):
139 (WebCore::ContainerNode::replaceChild):
140 (WebCore::ContainerNode::removeChild):
141 (WebCore::ContainerNode::parserRemoveChild):
142 (WebCore::ContainerNode::removeChildren):
143 (WebCore::ContainerNode::appendChild):
144 (WebCore::ContainerNode::parserAppendChild):
145 (WebCore::ContainerNode::childrenChanged):
146 (WebCore::ContainerNode::updateTreeAfterInsertion):
147 * dom/ContainerNode.h:
149 (WebCore::Document::childrenChanged):
152 (WebCore::checkForSiblingStyleChanges):
154 Clean up and simplify. Since we now get element range automatically we don't need to compute it.
156 (WebCore::Element::childrenChanged):
157 (WebCore::Element::finishParsingChildren):
159 * dom/ShadowRoot.cpp:
160 (WebCore::ShadowRoot::childrenChanged):
162 * html/HTMLElement.cpp:
163 (WebCore::HTMLElement::childrenChanged):
164 (WebCore::HTMLElement::adjustDirectionalityIfNeededAfterChildrenChanged):
166 Try to keep the existing behavior. This code needs more cleanup to be sane. It shouldn't operate on Nodes
167 as it only really cares about Elements.
169 * html/HTMLElement.h:
170 * html/HTMLFieldSetElement.cpp:
171 (WebCore::HTMLFieldSetElement::childrenChanged):
172 * html/HTMLFieldSetElement.h:
173 * html/HTMLObjectElement.cpp:
174 (WebCore::HTMLObjectElement::childrenChanged):
175 * html/HTMLObjectElement.h:
176 * html/HTMLOptGroupElement.cpp:
177 (WebCore::HTMLOptGroupElement::childrenChanged):
178 * html/HTMLOptGroupElement.h:
179 * html/HTMLOptionElement.cpp:
180 (WebCore::HTMLOptionElement::childrenChanged):
181 * html/HTMLOptionElement.h:
182 * html/HTMLOutputElement.cpp:
183 (WebCore::HTMLOutputElement::childrenChanged):
184 * html/HTMLOutputElement.h:
185 * html/HTMLScriptElement.cpp:
186 (WebCore::HTMLScriptElement::childrenChanged):
187 * html/HTMLScriptElement.h:
188 * html/HTMLSelectElement.cpp:
189 (WebCore::HTMLSelectElement::childrenChanged):
190 * html/HTMLSelectElement.h:
191 * html/HTMLStyleElement.cpp:
192 (WebCore::HTMLStyleElement::childrenChanged):
193 * html/HTMLStyleElement.h:
194 * html/HTMLTextAreaElement.cpp:
195 (WebCore::HTMLTextAreaElement::childrenChanged):
196 * html/HTMLTextAreaElement.h:
197 * html/HTMLTitleElement.cpp:
198 (WebCore::HTMLTitleElement::childrenChanged):
199 * html/HTMLTitleElement.h:
200 * html/shadow/InsertionPoint.cpp:
201 (WebCore::InsertionPoint::childrenChanged):
202 * html/shadow/InsertionPoint.h:
203 * svg/SVGClipPathElement.cpp:
204 (WebCore::SVGClipPathElement::childrenChanged):
205 * svg/SVGClipPathElement.h:
206 * svg/SVGElement.cpp:
207 (WebCore::SVGElement::childrenChanged):
209 * svg/SVGFELightElement.cpp:
210 (WebCore::SVGFELightElement::childrenChanged):
211 * svg/SVGFELightElement.h:
212 * svg/SVGFilterElement.cpp:
213 (WebCore::SVGFilterElement::childrenChanged):
214 * svg/SVGFilterElement.h:
215 * svg/SVGFilterPrimitiveStandardAttributes.cpp:
216 (WebCore::SVGFilterPrimitiveStandardAttributes::childrenChanged):
217 * svg/SVGFilterPrimitiveStandardAttributes.h:
218 * svg/SVGFontFaceElement.cpp:
219 (WebCore::SVGFontFaceElement::childrenChanged):
220 * svg/SVGFontFaceElement.h:
221 * svg/SVGFontFaceFormatElement.cpp:
222 (WebCore::SVGFontFaceFormatElement::childrenChanged):
223 * svg/SVGFontFaceFormatElement.h:
224 * svg/SVGFontFaceSrcElement.cpp:
225 (WebCore::SVGFontFaceSrcElement::childrenChanged):
226 * svg/SVGFontFaceSrcElement.h:
227 * svg/SVGFontFaceUriElement.cpp:
228 (WebCore::SVGFontFaceUriElement::childrenChanged):
229 * svg/SVGFontFaceUriElement.h:
230 * svg/SVGGradientElement.cpp:
231 (WebCore::SVGGradientElement::childrenChanged):
232 * svg/SVGGradientElement.h:
233 * svg/SVGMarkerElement.cpp:
234 (WebCore::SVGMarkerElement::childrenChanged):
235 * svg/SVGMarkerElement.h:
236 * svg/SVGMaskElement.cpp:
237 (WebCore::SVGMaskElement::childrenChanged):
238 * svg/SVGMaskElement.h:
239 * svg/SVGPatternElement.cpp:
240 (WebCore::SVGPatternElement::childrenChanged):
241 * svg/SVGPatternElement.h:
242 * svg/SVGScriptElement.cpp:
243 (WebCore::SVGScriptElement::childrenChanged):
244 * svg/SVGScriptElement.h:
245 * svg/SVGStyleElement.cpp:
246 (WebCore::SVGStyleElement::childrenChanged):
247 * svg/SVGStyleElement.h:
248 * svg/SVGTitleElement.cpp:
249 (WebCore::SVGTitleElement::childrenChanged):
250 * svg/SVGTitleElement.h:
252 2013-09-02 Andreas Kling <akling@apple.com>
254 Generate isFooElement() functions from tagname data.
255 <https://webkit.org/b/120584>
257 Reviewed by Antti Koivisto.
259 Add a "generateTypeChecks" attribute that can be used in HTMLTagNames.in & friends.
260 If present, isFooElement() style helpers will be added to HTMLElementTypeChecks.h.
261 This also outputs an isElementOfType<T> check for the Element iterators.
263 Removed all the hand-written isFooElement() functions that only checked tag name.
265 * html/HTMLTagNames.in:
268 Added "generateTypeChecks" attribute as appropriate.
271 * GNUmakefile.list.am:
272 * WebCore.vcxproj/WebCore.vcxproj:
273 * WebCore.xcodeproj/project.pbxproj:
275 Added to build systems based on how HTMLNames.h was done.
276 We're just outputting an additional header file in the generated code directory
277 so I suspect most ports will just pick this up automagically.
280 (defaultTagPropertyHash):
281 (printLicenseHeader):
283 (printTypeChecksHeaderFile):
285 Generate a separate file for each namespace with isFooElement() helpers for
286 elements with "generateTypeChecks" attribute set.
288 2013-09-02 Dirk Schulze <krit@webkit.org>
290 Use edgeMode=duplicate for blurring on filter() function
291 https://bugs.webkit.org/show_bug.cgi?id=120590
293 Reviewed by Antti Koivisto.
295 Filters on the CSS Image function filter() are not allowed to extend the
296 dimension of the input image. This causes weird results on blurring an image,
297 where the fading on the edges is clipped at the half of the fading.
298 We shouldn't fade edges at all and use the edgeMode=duplicate instead.
299 This will duplicate the pixel value on the nearest edge of the input image
300 instead of taking transparent black and results in nice blurred images with
303 Spec: http://dev.w3.org/fxtf/filters/#blurEquivalent
305 Test: fast/filter-image/filter-image-blur.html
307 * css/CSSFilterImageValue.cpp: Pass consumer information to the renderer.
308 (WebCore::CSSFilterImageValue::image):
309 * rendering/FilterEffectRenderer.cpp: Set edgeMode for feGaussianBlur to
310 'duplicate' or 'none' depending on the consumer.
311 (WebCore::FilterEffectRenderer::build):
312 * rendering/FilterEffectRenderer.h: Add enumeration to differ between the
313 different consumers of the renderer.
314 * rendering/RenderLayer.cpp:
315 (WebCore::RenderLayer::updateOrRemoveFilterEffectRenderer): Pass consumer
316 information to the renderer.
318 2013-09-01 Dirk Schulze <krit@webkit.org>
320 Add 'edgeMode' attribute to SVGFEGaussianBlur
321 https://bugs.webkit.org/show_bug.cgi?id=120582
323 Add 'edgeMode' attribute to the SVGFEGaussianBlur element. This attribute
324 allows users to define the behavior on edges with the values 'none' where
325 pixel values outside the input image are treated as transparent black. (The
326 current blurring behavior.) 'duplicate' which repeats the values on the
327 nearest edge and 'warp', which takes the pixel of the opposite site of
329 Beside the attribute, this patch implements the behavior of 'duplicate'.
331 http://dev.w3.org/fxtf/filters/#feGaussianBlurEdgeModeAttribute
333 Reviewed by Rob Buis.
335 Tests: svg/dynamic-updates/SVGFEGaussianBlurElement-dom-edgeMode-attr.html
336 svg/dynamic-updates/SVGFEGaussianBlurElement-svgdom-edgeMode-prop.html
337 svg/filters/svg-gaussianblur-edgeMode-duplicate-expected.svg
338 svg/filters/svg-gaussianblur-edgeMode-duplicate.svg
340 * platform/graphics/filters/FEGaussianBlur.cpp:
341 (WebCore::FEGaussianBlur::FEGaussianBlur):
342 (WebCore::FEGaussianBlur::create):
343 (WebCore::FEGaussianBlur::edgeMode):
344 (WebCore::FEGaussianBlur::setEdgeMode):
346 (WebCore::FEGaussianBlur::platformApplyGeneric):
347 (WebCore::FEGaussianBlur::determineAbsolutePaintRect):
348 * platform/graphics/filters/FEGaussianBlur.h:
349 * rendering/FilterEffectRenderer.cpp:
350 (WebCore::FilterEffectRenderer::build):
351 * svg/SVGFEGaussianBlurElement.cpp:
352 (WebCore::SVGFEGaussianBlurElement::SVGFEGaussianBlurElement):
353 (WebCore::SVGFEGaussianBlurElement::isSupportedAttribute):
354 (WebCore::SVGFEGaussianBlurElement::parseAttribute):
355 (WebCore::SVGFEGaussianBlurElement::svgAttributeChanged):
356 (WebCore::SVGFEGaussianBlurElement::build):
357 * svg/SVGFEGaussianBlurElement.h:
358 * svg/SVGFEGaussianBlurElement.idl:
360 2013-09-01 Andreas Kling <akling@apple.com>
362 EventHandler::m_frame should be a Frame&.
363 <https://webkit.org/b/120580>
365 Reviewed by Darin Adler.
367 EventHandler is tied to the lifetime of its frame, so let m_frame be a Frame&.
368 A handful of null checks and assertions removed.
370 2013-09-01 Darin Adler <darin@apple.com>
372 Fix a mistake in my recent pasteboard/editor refactoring that was causing tests to fail.
374 * editing/mac/EditorMac.mm:
375 (WebCore::getImage): One place this said cachedImage but it should have said tentativeCachedImage.
377 2013-09-01 Darin Adler <darin@apple.com>
379 HitTestResult should have innerNonSharedElement
380 https://bugs.webkit.org/show_bug.cgi?id=120579
382 Reviewed by Andreas Kling.
384 * editing/Editor.cpp:
385 (WebCore::Editor::copyImage): Call HitTestResult member function version of
386 innerNonSharedElement instead of a local function that does it.
389 (WebCore::Chrome::setToolTip): Use innerNonSharedElement instead of getting
390 the node and checking if it's an input element. Also added some missing braces.
392 * page/EventHandler.cpp:
393 (WebCore::EventHandler::selectClosestWordFromHitTestResult): Use targetNode for
394 local variables instead of innerNode to match the HitTestResult function name.
395 (WebCore::EventHandler::selectClosestWordOrLinkFromMouseEvent): Ditto.
396 (WebCore::EventHandler::handleMousePressEventTripleClick): Ditto.
397 (WebCore::EventHandler::handleMousePressEventSingleClick): Ditto.
398 (WebCore::EventHandler::handleMousePressEvent): Ditto.
400 * rendering/HitTestResult.cpp:
401 (WebCore::HitTestResult::innerElement): Rewrote so there there is no loop.
402 (WebCore::HitTestResult::innerNonSharedElement): Ditto.
404 * rendering/HitTestResult.h: Added innerNonSharedElement. Generally speaking,
405 we'd like to avoid using Node unless there is some real need.
407 2013-09-01 Xabier Rodriguez Calvar <calvaris@igalia.com>
409 Volume slider value should be 0 when audio is muted
410 https://bugs.webkit.org/show_bug.cgi?id=120553
412 Reviewed by Eric Carlson.
414 Fixed the problem of showing a non empty slider when audio is
417 Test: media/volume-bar-empty-when-muted.html.
419 * html/shadow/MediaControls.cpp:
420 (WebCore::MediaControls::reset): Use setSliderVolume.
421 (WebCore::MediaControls::changedVolume): Use setSliderVolume.
422 (WebCore::MediaControls::setSliderVolume): Added to set the volume
423 to 0 when muted and to its value otherwise.
424 * html/shadow/MediaControls.h: Added setSliderVolume.
425 * html/shadow/MediaControlsApple.cpp:
426 (WebCore::MediaControlsApple::reset): Used setSliderVolume and
427 setFullscreenSliderVolume.
428 (WebCore::MediaControlsApple::changedVolume): Used
429 setFullscreenSliderVolume.
430 (WebCore::MediaControlsApple::setFullscreenSliderVolume): Added to
431 set the volume to 0 when muted and to its value otherwise.
432 * html/shadow/MediaControlsApple.h: Added setFullscreenSliderVolume
433 * html/shadow/MediaControlsBlackBerry.cpp:
434 (WebCore::MediaControlsBlackBerry::reset): Used setSliderVolume.
436 2013-09-01 Andreas Kling <akling@apple.com>
438 Give EditCommand a protected Frame& getter.
439 <https://webkit.org/b/120574>
441 Reviewed by Darin Adler.
443 EditCommand is only created for documents that are attached to a Frame,
444 we already ASSERTed as much in the EditCommand constructor.
446 This patch adds a "Frame& EditCommand::frame()" helper, so EditCommand
447 and its subclasses don't have to fumble around with pointers.
449 2013-08-31 Antti Koivisto <antti@apple.com>
451 Add element ancestor iterator
452 https://bugs.webkit.org/show_bug.cgi?id=120563
454 Reviewed by Andreas Kling.
456 This patch adds ancestor iterators. They iterate over elements parent chain up to the root.
458 To iterate over Element ancestors:
460 auto ancestors = elementAncestors(this);
461 for (auto it = ancestors.begin(), end = ancestors.end(); it != end; ++it) {
462 Element& element = *it;
465 To iterate over Element ancestors including the current element:
467 auto lineage = elementLineage(this);
468 for (auto it = lineage.begin(), end = lineage.end(); it != end; ++it) {
469 Element& element = *it;
472 To iterate over ancestors of a specific Element subclass:
474 auto htmlAncestors = ancestorsOfType<HTMLElement>(this);
475 for (auto it = htmlAncestors.begin(), end = htmlAncestors.end(); it != end; ++it) {
476 HTMLElement& htmlElement = *it;
479 To iterate over ancestors of a specific Element subclass including the current element:
481 auto htmlLineage = lineageOfType<HTMLElement>(this);
482 for (auto it = htmlLineage.begin(), end = htmlLineage.end(); it != end; ++it) {
483 HTMLElement& htmlElement = *it;
486 The patch also uses the new types in a few places.
488 * WebCore.xcodeproj/project.pbxproj:
489 * accessibility/AccessibilityNodeObject.cpp:
490 (WebCore::AccessibilityNodeObject::mouseButtonListener):
491 (WebCore::AccessibilityNodeObject::labelForElement):
492 * dom/ElementAncestorIterator.h: Added.
493 (WebCore::::ElementAncestorIterator):
494 (WebCore::::operator):
495 (WebCore::::ElementAncestorConstIterator):
496 (WebCore::::ElementAncestorIteratorAdapter):
499 (WebCore::::ElementAncestorConstIteratorAdapter):
500 (WebCore::elementLineage):
501 (WebCore::elementAncestors):
502 (WebCore::ancestorsOfType):
503 * dom/ElementIterator.h:
504 (WebCore::findElementAncestorOfType):
505 (WebCore::::traverseAncestor):
507 * html/HTMLElement.cpp:
508 (WebCore::HTMLElement::adjustDirectionalityIfNeededAfterChildAttributeChanged):
509 * html/HTMLFieldSetElement.h:
510 (WebCore::isHTMLFieldSetElement):
511 (WebCore::HTMLFieldSetElement):
512 * html/HTMLFrameSetElement.cpp:
513 (WebCore::HTMLFrameSetElement::findContaining):
514 * html/HTMLFrameSetElement.h:
515 (WebCore::HTMLFrameSetElement):
516 * html/HTMLInputElement.h:
517 (WebCore::isHTMLInputElement):
518 (WebCore::toHTMLInputElement):
519 * html/HTMLLegendElement.cpp:
520 (WebCore::HTMLLegendElement::associatedControl):
522 2013-08-31 Darin Adler <darin@apple.com>
524 Refactor URL and image writing so layer-violating parts are in Editor, not Pasteboard (Mac-only at first)
525 https://bugs.webkit.org/show_bug.cgi?id=120573
527 Reviewed by Andreas Kling.
529 * editing/Editor.cpp:
530 (WebCore::Editor::copy): Use writeImageToPasteboard instead of Pasteboard::writeImage on Mac.
531 (WebCore::Editor::copyURL): Use writeURLToPasteboard instead of Pasteboard::writeURL on Mac.
532 (WebCore::innerNonSharedElement): Added. Used in copyImage.
533 (WebCore::Editor::copyImage): Use innerNonSharedElement instead of innerNonSharedNode.
534 Use writeImageToPasteboard instead of Pasteboard::writeImage on Mac.
536 * editing/Editor.h: Add new functions, writeURLToPasteboard and writeImageToPasteboard,
537 both Mac-only for now.
539 * editing/mac/EditorMac.mm:
540 (WebCore::Editor::pasteWithPasteboard): Removed unneeded "m_frame.editor()" round trip that
541 was left behind in this function.
542 (WebCore::getImage): Added. Helper used by writeImageToPasteboard.
543 (WebCore::Editor::writeURLToPasteboard): Added. Sets up PasteboardURL and then calls
544 Pasteboard::write with it.
545 (WebCore::Editor::writeImageToPasteboard): Added. Sets up PasteboardImage and then calls
546 Pasteboard::write with it.
547 * page/DragController.cpp:
548 (WebCore::DragController::startDrag): Use Editor::writeURLToPasteboard instead of
549 Pasteboard::writeURL on Mac.
551 * platform/Pasteboard.h: Added PasteboardURL and PasteboardImage structures.
552 Declare write functions for PasteboardWebContent, PasteboardURL, and PasteboardImage.
553 Guard writeURL and writeImage with !PLATFORM(MAC).
555 * platform/mac/PasteboardMac.mm:
556 (WebCore::writeURLForTypes): Changed this to take a PasteboardURL.
557 (WebCore::Pasteboard::write): Renamed writeURL and writeImage to this. Refactor both
558 to take PasteboardURL and PasteboardImage arguments.
559 (WebCore::fileWrapper): Renamed from fileWrapperForImage since the fact that this is
560 for an image is now clear from its argument, a PasteboardImage.
561 (WebCore::writeFileWrapperAsRTFDAttachment): Changed this function to use early return.
563 2013-08-31 Andreas Kling <akling@apple.com>
565 EditCommand constructors should take Document&.
566 <https://webkit.org/b/120566>
568 Reviewed by Darin Adler.
570 Let EditCommand's and all subclass constructors take Document& instead of Document*.
571 EditCommand::document() now returns a Document&.
573 Added Editor::document() which returns Editor::frame().document() after asserting
574 that it's non-null, to make passing a Document& from Editor functions easier.
576 2013-08-31 Andreas Kling <akling@apple.com>
578 Make more use of toRenderFoo(RenderObject&) cast helpers.
579 <https://webkit.org/b/120565>
581 Reviewed by Darin Adler.
583 To make amends for putting ugly toRenderFoo(&renderer) casts everywhere,
584 here's a patch that switches a bunch of code over to using reference-based casts.
586 I removed pointer-based casts altogether for these renderers:
594 2013-08-31 Santosh Mahto <santosh.ma@samsung.com>
596 warning: unused parameter ‘renderingMode' in ImageBufferCairo.cpp:94
597 https://bugs.webkit.org/show_bug.cgi?id=120543
599 Reviewed by Darin Adler.
601 Fixing warning by adding ASSERT_UNUSED.
603 * platform/graphics/cairo/ImageBufferCairo.cpp:
604 (WebCore::ImageBuffer::ImageBuffer):
606 2013-08-31 Andreas Kling <akling@apple.com>
608 Fix typo in r154931 that caused assertions in continuation tests.
610 * rendering/InlineFlowBox.cpp:
611 (WebCore::InlineFlowBox::paint):
613 2013-08-31 Rob Buis <rwlbuis@webkit.org>
615 XMLSerializer-attribute-namespace-prefix-conflicts can't produce reliable results
616 https://bugs.webkit.org/show_bug.cgi?id=120490
618 Reviewed by Ryosuke Niwa.
620 Retain the key of the namespaces map by using AtomicString as key instead of AtomicStringImpl*.
622 * editing/MarkupAccumulator.h:
624 2013-08-31 Andreas Kling <akling@apple.com>
626 InlineBox::renderer() and pals should return references.
627 <https://webkit.org/b/120562>
629 Reviewed by Antti Koivisto.
631 Make InlineBox::renderer() and its friends return references instead of pointers.
632 Every box always has a renderer, so this clears up any ambiguity, and a number of
633 weird assertions in SVG code.
635 Made InlineBox::m_renderer private so subclasses have to go through renderer().
636 The next step will be to replace renderer() with tightly-typed accessors where
637 applicable (e.g InlineTextBox's renderer is always a RenderText.)
639 Also added reference versions of these RenderObject casts:
644 - toRenderSVGInlineText()
646 2013-08-31 Andreas Kling <akling@apple.com>
648 PostAttachCallbackDisabler should take a ContainerNode&.
649 <https://webkit.org/b/120560>
651 Reviewed by Antti Koivisto.
653 Let PostAttachCallbackDisabler take a reference and remove an ugly assertion.
655 * dom/ContainerNode.h:
656 (WebCore::PostAttachCallbackDisabler::PostAttachCallbackDisabler):
657 (WebCore::PostAttachCallbackDisabler::~PostAttachCallbackDisabler):
659 (WebCore::Document::recalcStyle):
660 * style/StyleResolveTree.cpp:
661 (WebCore::Style::attachRenderTree):
663 2013-08-31 Antti Koivisto <antti@apple.com>
665 Rename ChildIterator -> ElementChildIterator, DescendantIterator -> ElementDescendantIterator
666 https://bugs.webkit.org/show_bug.cgi?id=120561
668 Reviewed by Andreas Kling.
670 New names are more descriptive and consistent. They also put related iterator files close by in file lists.
671 The types are always used with auto so longer names don't make them any clumsier to use.
673 For simplicity included all Element iterator types from ElementIterator.h so it is the only file that has
674 to be included by the clients.
676 * GNUmakefile.list.am:
678 * WebCore.vcxproj/WebCore.vcxproj:
679 * WebCore.xcodeproj/project.pbxproj:
680 * accessibility/AccessibilityNodeObject.cpp:
681 * accessibility/AccessibilityRenderObject.cpp:
682 * accessibility/AccessibilityTable.cpp:
683 * css/CSSFontFaceSource.cpp:
684 * dom/ChildIterator.h: Removed.
685 * dom/DescendantIterator.h: Removed.
687 * dom/ElementChildIterator.h: Copied from Source/WebCore/dom/ChildIterator.h.
688 (WebCore::::ElementChildIterator):
689 (WebCore::::operator):
690 (WebCore::::ElementChildConstIterator):
691 (WebCore::::ElementChildIteratorAdapter):
694 (WebCore::::ElementChildConstIteratorAdapter):
695 (WebCore::elementChildren):
696 (WebCore::childrenOfType):
697 * dom/ElementDescendantIterator.h: Copied from Source/WebCore/dom/DescendantIterator.h.
698 (WebCore::::ElementDescendantIterator):
699 (WebCore::::operator):
700 (WebCore::::ElementDescendantConstIterator):
701 (WebCore::::ElementDescendantIteratorAdapter):
704 (WebCore::::ElementDescendantConstIteratorAdapter):
705 (WebCore::elementDescendants):
706 (WebCore::descendantsOfType):
707 * dom/ElementIterator.h:
710 * editing/ApplyStyleCommand.cpp:
711 * editing/markup.cpp:
712 * html/HTMLAppletElement.cpp:
713 * html/HTMLFieldSetElement.cpp:
714 * html/HTMLLabelElement.cpp:
715 * html/HTMLMapElement.cpp:
716 * html/HTMLMediaElement.cpp:
717 * html/HTMLObjectElement.cpp:
718 * loader/PlaceholderDocument.cpp:
719 * rendering/FilterEffectRenderer.cpp:
720 * style/StyleResolveTree.cpp:
721 * svg/SVGElement.cpp:
722 * svg/SVGSVGElement.cpp:
723 * svg/animation/SMILTimeContainer.cpp:
724 * svg/graphics/SVGImage.cpp:
726 2013-08-31 Andreas Kling <akling@apple.com>
728 Don't do document style recalc unless there's a RenderView.
729 <https://webkit.org/b/120558>
731 Reviewed by Antti Koivisto.
733 There's no sense in computing style for a Document that has no RenderView.
734 Checking this before continuing also lets us know that there's a Frame & FrameView
735 present, simplifying some things later on.
738 (WebCore::Document::recalcStyle):
739 * style/StyleResolveForDocument.cpp:
740 (WebCore::Style::resolveForDocument):
742 2013-08-31 Antti Koivisto <antti@apple.com>
744 Add common base for element iterators
745 https://bugs.webkit.org/show_bug.cgi?id=120557
747 Reviewed by Andreas Kling.
749 Add ElementIterator and use it as the base for child and descendant iterators. The only difference is the definition of operator++.
751 Also renamed DescendantIteratorAssertions to ElementIteratorAssertions.
753 * GNUmakefile.list.am:
755 * WebCore.vcxproj/WebCore.vcxproj:
756 * WebCore.xcodeproj/project.pbxproj:
757 * dom/ChildIterator.h:
758 (WebCore::::ChildIterator):
759 (WebCore::::operator):
760 (WebCore::::ChildConstIterator):
763 * dom/DescendantIterator.h:
764 (WebCore::::DescendantIterator):
765 (WebCore::::operator):
766 (WebCore::::DescendantConstIterator):
767 * dom/DescendantIteratorAssertions.h: Removed.
768 * dom/ElementIterator.h: Added.
769 (WebCore::::ElementIterator):
770 (WebCore::::traverseNext):
771 (WebCore::::traversePrevious):
772 (WebCore::::traverseNextSibling):
773 (WebCore::::traversePreviousSibling):
774 (WebCore::::operator):
776 (WebCore::::ElementConstIterator):
777 * dom/ElementIteratorAssertions.h: Copied from Source/WebCore/dom/DescendantIteratorAssertions.h.
778 (WebCore::ElementIteratorAssertions::ElementIteratorAssertions):
779 (WebCore::ElementIteratorAssertions::domTreeHasMutated):
780 (WebCore::ElementIteratorAssertions::dropEventDispatchAssertion):
782 2013-08-30 Brent Fulgham <bfulgham@apple.com>
784 [Windows] Build fix after r154921.
786 AVFoundationCF uses an internal wrapper object around the C-API. The call point
787 in the original patch was inside the wrapper, which did not (yet) provide
788 a method implementation.
790 * platform/graphics/avfoundation/cf/MediaPlayerPrivateAVFoundationCF.cpp:
791 (WebCore::MediaPlayerPrivateAVFoundationCF::updateVideoLayerGravity):
792 (WebCore::AVFWrapper::updateVideoLayerGravity):
794 2013-08-30 Simon Fraser <simon.fraser@apple.com>
796 Video with object-fit: cover can spill outside the box
797 https://bugs.webkit.org/show_bug.cgi?id=52103
799 Reviewed by Dean Jackson.
801 object-fit on renderers which use accelerated compositing needs special
804 For directly composited images, and video, GraphicsLayer needs to know
805 both the size of the content layer, and also a rectangle at which this
806 should be clipped (because, for the first time, that content layer can be
807 larger than the renderer's content box).
809 AVFoundation would always aspect-ratio fit video by default, so plumb
810 through MediaPlayer a way to override that when object-fit requires it.
812 Added a LAYER_TREE_INCLUDES_CONTENT_LAYERS enum to the layerTreeAsText()
813 flags so we can dump content layers for testing.
815 Tests: compositing/images/direct-image-object-fit.html
816 compositing/reflections/direct-image-object-fit-reflected.html
817 compositing/video/video-object-fit.html
819 * page/Frame.h: New LayerTreeFlagsIncludeContentLayers flag.
820 * platform/graphics/GraphicsLayer.h: New flag.
821 * platform/graphics/MediaPlayer.cpp:
822 (WebCore::MediaPlayer::shouldMaintainAspectRatio):
823 (WebCore::MediaPlayer::setShouldMaintainAspectRatio):
824 * platform/graphics/MediaPlayer.h:
825 * platform/graphics/MediaPlayerPrivate.h:
826 (WebCore::MediaPlayerPrivateInterface::shouldMaintainAspectRatio):
827 (WebCore::MediaPlayerPrivateInterface::setShouldMaintainAspectRatio):
828 * platform/graphics/avfoundation/MediaPlayerPrivateAVFoundation.cpp:
829 (WebCore::MediaPlayerPrivateAVFoundation::MediaPlayerPrivateAVFoundation):
830 (WebCore::MediaPlayerPrivateAVFoundation::setShouldMaintainAspectRatio):
831 * platform/graphics/avfoundation/MediaPlayerPrivateAVFoundation.h:
832 * platform/graphics/avfoundation/cf/MediaPlayerPrivateAVFoundationCF.cpp:
833 (WebCore::MediaPlayerPrivateAVFoundationCF::updateVideoLayerGravity):
834 (WebCore::AVFWrapper::platformLayer):
835 * platform/graphics/avfoundation/cf/MediaPlayerPrivateAVFoundationCF.h:
836 * platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.h:
837 * platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.mm:
838 (WebCore::MediaPlayerPrivateAVFoundationObjC::createVideoLayer):
839 (WebCore::MediaPlayerPrivateAVFoundationObjC::updateVideoLayerGravity):
840 * platform/graphics/ca/GraphicsLayerCA.cpp: We need a new m_contentsClippingLayer to
841 clip the contents layer, which only gets created when necessary. It has to be cloned
843 (WebCore::GraphicsLayerCA::willBeDestroyed):
844 (WebCore::GraphicsLayerCA::setContentsRect):
845 (WebCore::GraphicsLayerCA::setContentsClippingRect):
846 (WebCore::GraphicsLayerCA::commitLayerChangesBeforeSublayers):
847 (WebCore::GraphicsLayerCA::updateSublayerList):
848 (WebCore::GraphicsLayerCA::updateContentsImage):
849 (WebCore::GraphicsLayerCA::updateContentsMediaLayer):
850 (WebCore::GraphicsLayerCA::updateContentsCanvasLayer):
851 (WebCore::GraphicsLayerCA::updateContentsColorLayer):
852 (WebCore::GraphicsLayerCA::updateContentsRects):
853 (WebCore::GraphicsLayerCA::dumpAdditionalProperties):
854 (WebCore::GraphicsLayerCA::ensureCloneLayers):
855 (WebCore::GraphicsLayerCA::removeCloneLayers):
856 (WebCore::GraphicsLayerCA::fetchCloneLayers):
857 * platform/graphics/ca/GraphicsLayerCA.h:
858 * rendering/RenderLayerBacking.cpp: Need to push both the contentsRect and
859 the contentsClippingRect down to the GraphicsLayers. Most of the time they
860 are the same, unless object-fit makes them different.
861 (WebCore::RenderLayerBacking::resetContentsRect):
862 (WebCore::RenderLayerBacking::positionOverflowControlsLayers):
863 (WebCore::RenderLayerBacking::updateDirectlyCompositedBackgroundColor):
864 (WebCore::RenderLayerBacking::updateDirectlyCompositedBackgroundImage):
865 (WebCore::RenderLayerBacking::updateImageContents):
866 (WebCore::RenderLayerBacking::contentsBox):
867 * rendering/RenderLayerCompositor.cpp:
868 (WebCore::RenderLayerCompositor::layerTreeAsText):
869 * rendering/RenderVideo.cpp:
870 (WebCore::RenderVideo::updatePlayer):
871 * testing/Internals.cpp:
872 (WebCore::Internals::layerTreeAsText):
873 * testing/Internals.h:
874 * testing/Internals.idl:
876 2013-08-30 Brent Fulgham <bfulgham@apple.com>
878 [Windows] Update to incorporate additional suggestions
879 https://bugs.webkit.org/show_bug.cgi?id=120448
881 Reviewed by Darin Adler
883 * platform/graphics/avfoundation/cf/MediaPlayerPrivateAVFoundationCF.cpp: Add
884 notes as to why AVFoundationCF needs this extra method call.
885 * platform/graphics/avfoundation/cf/MediaPlayerPrivateAVFoundationCF.h: Use OVERRIDE
886 * rendering/RenderVideo.cpp: Replace ternary operator with logical &&.
888 2013-08-30 Brent Fulgham <bfulgham@apple.com>
890 [Windows] Vide element in page always uses non-hw accelerated mode.
891 https://bugs.webkit.org/show_bug.cgi?id=120448
893 Reviewed by Darin Adler
895 This patch is unfortunately larger than my original idea, but seems to make the
896 layout system happier. Instead of switching into composited mode when building
897 the media player, we now build the original layout tree with compositing active
898 if the underlying media element requires it. The AVFoundationCF player needs to
899 have the compositor available at construction time so it can attach to the
900 rendering device. Otherwise it falls back to CPU-only mode.
902 * platform/graphics/MediaPlayer.cpp:
903 (WebCore::MediaPlayer::requiresImmediateCompositing): Added
904 * platform/graphics/MediaPlayer.h:
905 * platform/graphics/MediaPlayerPrivate.h:
906 (WebCore::MediaPlayerPrivateInterface::requiresImmediateCompositing): Added
907 * platform/graphics/avfoundation/cf/MediaPlayerPrivateAVFoundationCF.cpp:
908 (WebCore::MediaPlayerPrivateAVFoundationCF::createAVPlayer): Added
909 * platform/graphics/avfoundation/cf/MediaPlayerPrivateAVFoundationCF.h:
910 (WebCore::MediaPlayerPrivateAVFoundationCF::requiresImmediateCompositing):
911 * rendering/RenderLayerCompositor.cpp:
912 (WebCore::RenderLayerCompositor::requiresCompositingForVideo): Uses new
913 'requiresImmediateCompositing' to short-circuit check for whether a
914 compositor is required.
915 * rendering/RenderVideo.cpp:
916 (WebCore::RenderVideo::requiresImmediateCompositing):
917 * rendering/RenderVideo.h:
919 2013-08-30 Joseph Pecoraro <pecoraro@apple.com>
921 Web Inspector: Breakpoints should have Automatically Continue Option
922 https://bugs.webkit.org/show_bug.cgi?id=120187
924 Reviewed by Timothy Hatcher.
926 Tests: inspector-protocol/debugger/removeBreakpoint.html
927 inspector-protocol/debugger/setBreakpoint-autoContinue.html
928 inspector-protocol/debugger/setBreakpoint-column.html
929 inspector-protocol/debugger/setBreakpoint-condition.html
930 inspector-protocol/debugger/setBreakpoint.html
932 * inspector/Inspector.json:
933 Convert the "condition" argument of setBreakpoint and setBreakpointByUrl
934 to a BreakpointOptions object with optional properties. One of which
935 is "condition" and a new option "autoContinue".
937 * bindings/js/ScriptDebugServer.h:
938 * bindings/js/ScriptDebugServer.cpp:
939 (WebCore::ScriptDebugServer::hasBreakpoint):
940 (WebCore::ScriptDebugServer::pauseIfNeeded):
941 Automatically continue if the breakpoint was hit and has autoContinue.
943 * inspector/InspectorDebuggerAgent.h:
944 * inspector/InspectorDebuggerAgent.cpp:
945 (WebCore::buildObjectForBreakpointCookie):
946 (WebCore::InspectorDebuggerAgent::setBreakpointByUrl):
947 (WebCore::InspectorDebuggerAgent::setBreakpoint):
948 (WebCore::InspectorDebuggerAgent::continueToLocation):
949 (WebCore::InspectorDebuggerAgent::didParseSource):
950 * inspector/ScriptBreakpoint.h:
951 (WebCore::ScriptBreakpoint::ScriptBreakpoint):
952 Handle the new incoming BreakpointOptions type, and set the
953 autoContinue state on ScriptBreakpoints.
955 * inspector/front-end/DebuggerModel.js:
956 (WebInspector.DebuggerModel.prototype.setBreakpointByURL):
957 (WebInspector.DebuggerModel.prototype.setBreakpointBySourceId):
958 Update old front-end to be compatible with the new API.
960 2013-08-30 Dean Jackson <dino@apple.com>
962 Animations should use double for key values, not floats
963 https://bugs.webkit.org/show_bug.cgi?id=120547
965 Reviewed by Simon Fraser.
967 Merge https://chromium.googlesource.com/chromium/blink/+/71de046541c77120874b9bff82958ee9e0e20c7c
969 Some files have been renamed in the Blink port, and they have made some
970 improvements, but I took what applied to us.
972 All our existing tests passed.
974 * css/StyleResolver.cpp:
975 (WebCore::StyleResolver::keyframeStylesForAnimation):
976 * css/WebKitCSSKeyframeRule.cpp:
977 (WebCore::StyleKeyframe::parseKeyString):
978 * css/WebKitCSSKeyframeRule.h:
979 (WebCore::StyleKeyframe::getKeys):
980 * platform/graphics/GraphicsLayer.h:
981 (WebCore::AnimationValue::keyTime):
982 (WebCore::AnimationValue::AnimationValue):
983 (WebCore::FloatAnimationValue::create):
984 (WebCore::FloatAnimationValue::FloatAnimationValue):
985 (WebCore::TransformAnimationValue::create):
986 (WebCore::TransformAnimationValue::TransformAnimationValue):
987 (WebCore::FilterAnimationValue::create):
988 (WebCore::FilterAnimationValue::FilterAnimationValue):
989 * rendering/RenderLayerBacking.cpp:
990 (WebCore::RenderLayerBacking::startAnimation):
991 * rendering/style/KeyframeList.h:
992 (WebCore::KeyframeValue::KeyframeValue):
993 (WebCore::KeyframeValue::key):
994 (WebCore::KeyframeValue::setKey):
996 2013-08-30 Brendan Long <b.long@cablelabs.com>
998 [GStreamer] support in-band text tracks
999 https://bugs.webkit.org/show_bug.cgi?id=103771
1001 Reviewed by Eric Carlson.
1003 Tests: New tests added because existing tests were too specific.
1004 media/track/in-band/track-in-band-kate-ogg-cues-added-once.html
1005 media/track/in-band/track-in-band-kate-ogg-kind.html
1006 media/track/in-band/track-in-band-kate-ogg-language.html
1007 media/track/in-band/track-in-band-kate-ogg-mode.html
1008 media/track/in-band/track-in-band-kate-ogg-style.html
1009 media/track/in-band/track-in-band-kate-ogg-track-order.html
1010 media/track/in-band/track-in-band-srt-mkv-cues-added-once.html
1011 media/track/in-band/track-in-band-srt-mkv-kind.html
1012 media/track/in-band/track-in-band-srt-mkv-language.html
1013 media/track/in-band/track-in-band-srt-mkv-mode.html
1014 media/track/in-band/track-in-band-srt-mkv-style.html
1015 media/track/in-band/track-in-band-srt-mkv-track-order.html
1017 * CMakeLists.txt: Add InbandTextTrackPrivateGStreamer, InbandGenericTextTrack, InbandWebVTTTextTrack, and TextCombinerGStreamer files.
1018 * GNUmakefile.list.am: Same.
1019 * PlatformEfl.cmake: Same.
1021 * WebCore.vcxproj/WebCore.vcxproj: Same.
1022 * WebCore.vcxproj/WebCore.vcxproj.filters: Same.
1023 * WebCore.xcodeproj/project.pbxproj: Same.
1024 * html/track/InbandGenericTextTrack.cpp: Split out code for handling generic cues.
1025 (WebCore::GenericTextTrackCueMap::GenericTextTrackCueMap): Move from InbandTextTrack.
1026 (WebCore::GenericTextTrackCueMap::~GenericTextTrackCueMap): Same.
1027 (WebCore::GenericTextTrackCueMap::add): Same.
1028 (WebCore::GenericTextTrackCueMap::find): Same.
1029 (WebCore::GenericTextTrackCueMap::remove): Same.
1030 (WebCore::InbandGenericTextTrack::create): Same.
1031 (WebCore::InbandGenericTextTrack::updateCueFromCueData): Same.
1032 (WebCore::InbandGenericTextTrack::addGenericCue): Same.
1033 (WebCore::InbandGenericTextTrack::updateGenericCue): Same.
1034 (WebCore::InbandGenericTextTrack::removeGenericCue): Same.
1035 (WebCore::InbandGenericTextTrack::removeCue): Same.
1036 (WebCore::InbandGenericTextTrack::InbandGenericTextTrack): Empty.
1037 (WebCore::InbandGenericTextTrack::~InbandGenericTextTrack): Empty.
1038 * html/track/InbandGenericTextTrack.h: Copied from Source/WebCore/html/track/InbandTextTrack.h.
1039 The only addition is the ASSERT_NOT_REACHED() for WebVTT cues.
1040 * html/track/InbandTextTrack.cpp: Add label and language changed callbacks.
1041 (WebCore::InbandTextTrack::create): Return a generic or WebVTT text track based on the private CueFormat.
1042 (WebCore::InbandTextTrack::labelChanged): Added.
1043 (WebCore::InbandTextTrack::languageChanged): Added.
1044 * html/track/InbandTextTrack.h: Add label and language changed callbacks.
1045 * html/track/InbandWebVTTTextTrack.cpp: Added, based on InbandTextTrack.
1046 (WebCore::InbandWebVTTTextTrack::create): Same.
1047 (WebCore::InbandWebVTTTextTrack::InbandWebVTTTextTrack): Empty.
1048 (WebCore::InbandWebVTTTextTrack::~InbandWebVTTTextTrack): Empty.
1049 (WebCore::InbandWebVTTTextTrack::parseWebVTTCueData): Sends data to a WebVTTParser.
1050 (WebCore::InbandWebVTTTextTrack::newCuesParsed): Adds cues when WebVTTParser parses them.
1051 (WebCore::InbandWebVTTTextTrack::fileFailedToParse): Prints a warning when WebVTTParser has a problem.
1052 * html/track/InbandWebVTTTextTrack.h: Added.
1053 * platform/graphics/InbandTextTrackPrivate.h:
1054 (WebCore::InbandTextTrackPrivate::cueFormat): For determining if the track will have generic or WebVTT cues.
1055 (WebCore::InbandTextTrackPrivate::InbandTextTrackPrivate): Pass CueFormat in the constructor.
1056 * platform/graphics/InbandTextTrackPrivateClient.h: Same.
1057 * platform/graphics/avfoundation/InbandTextTrackPrivateAVF.cpp:
1058 (WebCore::InbandTextTrackPrivateAVF::InbandTextTrackPrivateAVF): Pass CueFormat (Generic) to InbandTextTrackPrivate.
1059 * platform/graphics/gstreamer/GRefPtrGStreamer.cpp: Add GRefPtr specializations for GstSample and GstEvent.
1060 * platform/graphics/gstreamer/GRefPtrGStreamer.h: Same.
1061 * platform/graphics/gstreamer/GStreamerUtilities.h: Add WARN_MEDIA_MESSAGE.
1062 * platform/graphics/gstreamer/GStreamerVersioning.h: Add a function to check GStreamer version at runtime.
1063 * platform/graphics/gstreamer/InbandTextTrackPrivateGStreamer.cpp: Added.
1064 (WebCore::textTrackPrivateEventCallback): Watches for tag and stream start events.
1065 (WebCore::textTrackPrivateSampleTimeoutCallback): See notifyTrackOfSample();
1066 (WebCore::textTrackPrivateStreamTimeoutCallback): See notifyTrackOfStreamChanged();
1067 (WebCore::textTrackPrivateTagsChangeTimeoutCallback): See notifyTrackOfTagsChanged();
1068 (WebCore::InbandTextTrackPrivateGStreamer::InbandTextTrackPrivateGStreamer): Initializes tags and stream and sets up event callback.
1069 (WebCore::InbandTextTrackPrivateGStreamer::~InbandTextTrackPrivateGStreamer): Calls disconnect:
1070 (WebCore::InbandTextTrackPrivateGStreamer::disconnect): Removes signal handlers and frees some memory.
1071 (WebCore::InbandTextTrackPrivateGStreamer::handleSample): Adds samples to a list and sets up callback.
1072 (WebCore::InbandTextTrackPrivateGStreamer::streamChanged): Sets up callback.
1073 (WebCore::InbandTextTrackPrivateGStreamer::tagsChanged): Same.
1074 (WebCore::InbandTextTrackPrivateGStreamer::notifyTrackOfSample): Parses all queued samples with WebVTTParser.
1075 (WebCore::InbandTextTrackPrivateGStreamer::notifyTrackOfStreamChanged): Keeps track of current stream.
1076 (WebCore::InbandTextTrackPrivateGStreamer::notifyTrackOfTagsChanged): Sets label and language from tags.
1077 * platform/graphics/gstreamer/InbandTextTrackPrivateGStreamer.h: Added.
1078 (WebCore::InbandTextTrackPrivateGStreamer::create): Basic RefPtr create function.
1079 (WebCore::InbandTextTrackPrivateGStreamer::pad): Returns the pad this track is associated with (used
1080 to determine if a playbin text stream has already been associated with a text track).
1081 (WebCore::InbandTextTrackPrivateGStreamer::setIndex): Sets the track index (used for sorting).
1082 (WebCore::InbandTextTrackPrivateGStreamer::streamId): Returns the stream ID (used to handle new samples).
1083 * platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp:
1084 (WebCore::mediaPlayerPrivateTextChangedCallback): Called for playbin "text-changed" event. See textChanged().
1085 (WebCore::mediaPlayerPrivateTextChangeTimeoutCallback): See notifyPlayerOfText().
1086 (WebCore::mediaPlayerPrivateNewTextSampleCallback): See newTextSample().
1087 (WebCore::MediaPlayerPrivateGStreamer::MediaPlayerPrivateGStreamer): Initialize m_textTimerHandler.
1088 (WebCore::MediaPlayerPrivateGStreamer::~MediaPlayerPrivateGStreamer): Disconnect text tracks (they don't necessarily
1089 get freed here, since a script could hold a reference).
1090 (WebCore::MediaPlayerPrivateGStreamer::textChanged): Setup callback for notifyPlayerOfText.
1091 (WebCore::MediaPlayerPrivateGStreamer::notifyPlayerOfText): Create text tracks.
1092 (WebCore::MediaPlayerPrivateGStreamer::newTextSample): Handle new samples by giving them to a text track
1093 with a matching stream. This method is syncryonous because we need to get the stream start sticky event
1095 * platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.h:
1096 * platform/graphics/gstreamer/TextCombinerGStreamer.cpp: Added. This element forwards buffers from all
1097 of its input pads, but also converts plain text to WebVTT as needed.
1098 (webkit_text_combiner_init): Setup internal funnel.
1099 (webkitTextCombinerPadEvent): If the caps are plain text make sure we have a webvttenv, otherwise connect
1100 directly to the funnel.
1101 (webkitTextCombinerRequestNewPad): Setup ghostpad and event callback.
1102 (webkitTextCombinerReleasePad): Release pad and optional associated webvttenc.
1103 (webkit_text_combiner_class_init): Setup pad templates and request/release pad functions.
1104 (webkitTextCombinerNew): Returns a new WebKitTextCombiner.
1105 * platform/graphics/gstreamer/TextCombinerGStreamer.h: Added.
1106 * platform/graphics/gstreamer/TextSinkGStreamer.cpp: Added.
1107 (webkit_text_sink_init): Set sync=false.
1108 (webkitTextSinkGetProperty): Ignore sync property.
1109 (webkitTextSinkSetProperty): Same.
1110 (webkitTextSinkQuery): Ignore position and duration queries, forward everything else to appsink.
1111 (webkit_text_sink_class_init): Setup property and query functions.
1112 (webkitTextSinkNew): Return a new WebKitTextSink.
1113 * platform/graphics/gstreamer/TextSinkGStreamer.h: Added.
1115 2013-08-30 Dirk Schulze <krit@webkit.org>
1117 Animate CSS Image filter() function
1118 https://bugs.webkit.org/show_bug.cgi?id=119938
1120 Reviewed by Simon Fraser.
1122 With this patch, the new introduced CSS Image function filter() can be
1123 animated. According to the spec, just filter functions can be
1126 The patch also prepares StyleImage blending for interpolation of other
1127 generated images like gradients or cross-fade().
1129 http://dev.w3.org/fxtf/filters/#interpolating-filter-image
1131 Test: fast/filter-image/filter-image-animation.html
1133 * css/CSSComputedStyleDeclaration.cpp: Reuse the code that creates a
1134 CSSValueList from ComputeStyle logic.
1135 (WebCore::valueForPixel):
1136 For StyleRules we want to have not-adjusted length values.
1137 (WebCore::ComputedStyleExtractor::valueForShadow):
1138 Add argument to switch between adjusted and not-adjusted length.
1139 (WebCore::ComputedStyleExtractor::valueForFilter):
1141 (WebCore::ComputedStyleExtractor::propertyValue):
1142 * css/CSSComputedStyleDeclaration.h:
1143 * css/CSSFilterImageValue.h: Add helper functions
1144 for animating filters. We need to pass the FilterOperations for
1145 the image generation and the CSSValueList for StyleRule.
1146 (WebCore::CSSFilterImageValue::filterOperations):
1147 (WebCore::CSSFilterImageValue::setFilterOperations):
1148 (WebCore::CSSFilterImageValue::cachedImage):
1149 * page/animation/CSSPropertyAnimation.cpp:
1150 Add animation code to support animations between two filter()
1152 (WebCore::blendFilterOperations):
1153 (WebCore::blendFunc):
1154 (WebCore::filterBlend):
1155 * rendering/style/StyleGeneratedImage.h: Add helper functions.
1156 (WebCore::CSSFilterImageValue::imageValue):
1159 2013-08-30 Leo Yang <leoyang@blackberry.com>
1161 Make sure remove CachedResourceClient when destructing IconLoader
1162 https://bugs.webkit.org/show_bug.cgi?id=120541
1164 Reviewed by Darin Adler.
1166 It's a good practice to call CachedResource::removeClient(client)
1167 when the client is being destructed. We need to do this for InconLoader
1168 to prevent m_resource from keeping dangling client in case m_resource
1169 is referenced by someone else in the future.
1171 Found by code inspection. Just a defensive enhancement no new tests.
1173 * loader/icon/IconLoader.cpp:
1174 (WebCore::IconLoader::~IconLoader):
1176 2013-08-30 Hans Muller <hmuller@adobe.com>
1178 [CSS Shapes] Redefine the ShapeIntervals class as a template
1179 https://bugs.webkit.org/show_bug.cgi?id=120381
1181 Reviewed by Alexandru Chiculita.
1183 No new tests, this was just an internal refactoring.
1185 The existing ShapeIntervals class has been converted into a template whose
1186 type specifies the type of the interval's x1 and x2 horizontal endpoints
1187 (formerly float). There were several other minor changes, all in the realm
1190 - The original type was a struct with public x1 and x2 fields. It's now a class
1191 with x1 and x2 accessors. ASSERTS are now used to maintain the x2 >= x1 invariant.
1192 In the original code the invariant was not checked.
1194 - The logical comparison operators have been overloaded for ShapeInterval.
1195 This obviates the IntervalX1Comparator class which has been removed.
1197 - The names of the global ShapeInterval Vector set operation methods have been
1198 changed to reflect the fact that they're now members of the template class,
1199 rather than globals.
1201 PolygonShape.cpp depended on the ShapeInterval class. In the one or two places
1202 where an interval's x1 or x1 fields had been set explicitly, ShapeInterval::set()
1203 is now used to set both fields. This also enables the invariant check mentioned
1204 earlier. The other changes to this class are syntatic, to account for the ShapeInterval
1208 * GNUmakefile.list.am:
1209 * WebCore.vcxproj/WebCore.vcxproj:
1210 * WebCore.xcodeproj/project.pbxproj:
1211 * rendering/shapes/PolygonShape.cpp:
1212 (WebCore::appendIntervalX):
1213 (WebCore::computeXIntersections):
1214 (WebCore::computeOverlappingEdgeXProjections):
1215 (WebCore::PolygonShape::getExcludedIntervals):
1216 (WebCore::PolygonShape::getIncludedIntervals):
1217 * rendering/shapes/PolygonShape.h:
1218 * rendering/shapes/ShapeInterval.cpp: Removed.
1219 * rendering/shapes/ShapeInterval.h: The entire implementation is now here.
1220 (WebCore::ShapeInterval::ShapeInterval):
1221 (WebCore::ShapeInterval::x1):
1222 (WebCore::ShapeInterval::x2):
1223 (WebCore::ShapeInterval::set):
1224 (WebCore::ShapeInterval::overlaps):
1225 (WebCore::ShapeInterval::intersect):
1226 (WebCore::ShapeInterval::sortVector):
1227 (WebCore::ShapeInterval::uniteVectors):
1228 (WebCore::ShapeInterval::intersectVectors):
1229 (WebCore::ShapeInterval::subtractVectors):
1230 (WebCore::operator==):
1231 (WebCore::operator!=):
1232 (WebCore::operator< ):
1233 (WebCore::operator> ):
1234 (WebCore::operator<=):
1235 (WebCore::operator>=):
1237 2013-08-30 Antti Koivisto <antti@apple.com>
1239 Use Element& in StyleResolveTree
1240 https://bugs.webkit.org/show_bug.cgi?id=120540
1242 Reviewed by Andreas Kling.
1244 * dom/ContainerNode.cpp:
1245 (WebCore::attachChild):
1246 (WebCore::detachChild):
1248 (WebCore::Document::recalcStyle):
1249 (WebCore::Document::createRenderTree):
1250 (WebCore::Document::detach):
1252 (WebCore::Element::lazyReattach):
1253 (WebCore::Element::updateExistingPseudoElement):
1254 (WebCore::Element::createPseudoElementIfNeeded):
1255 (WebCore::disconnectPseudoElement):
1256 * dom/ShadowRoot.cpp:
1257 (WebCore::ShadowRoot::setResetStyleInheritance):
1258 * html/HTMLDetailsElement.cpp:
1259 (WebCore::HTMLDetailsElement::parseAttribute):
1260 * html/HTMLInputElement.cpp:
1261 (WebCore::HTMLInputElement::updateType):
1262 (WebCore::HTMLInputElement::parseAttribute):
1263 * html/HTMLObjectElement.cpp:
1264 (WebCore::HTMLObjectElement::renderFallbackContent):
1265 * html/HTMLPlugInImageElement.cpp:
1266 (WebCore::HTMLPlugInImageElement::willRecalcStyle):
1267 (WebCore::HTMLPlugInImageElement::documentWillSuspendForPageCache):
1268 (WebCore::HTMLPlugInImageElement::documentDidResumeFromPageCache):
1269 (WebCore::HTMLPlugInImageElement::restartSnapshottedPlugIn):
1270 * html/HTMLSelectElement.cpp:
1271 (WebCore::HTMLSelectElement::parseAttribute):
1272 (WebCore::HTMLSelectElement::parseMultipleAttribute):
1273 * html/HTMLViewSourceDocument.cpp:
1274 (WebCore::HTMLViewSourceDocument::createContainingTable):
1275 (WebCore::HTMLViewSourceDocument::addSpanWithClassName):
1276 (WebCore::HTMLViewSourceDocument::addLine):
1277 (WebCore::HTMLViewSourceDocument::finishLine):
1278 (WebCore::HTMLViewSourceDocument::addBase):
1279 (WebCore::HTMLViewSourceDocument::addLink):
1280 * html/parser/HTMLConstructionSite.cpp:
1281 (WebCore::executeTask):
1282 * html/parser/HTMLTreeBuilder.cpp:
1283 (WebCore::HTMLTreeBuilder::callTheAdoptionAgency):
1284 * html/shadow/InsertionPoint.cpp:
1285 (WebCore::InsertionPoint::willAttachRenderers):
1286 (WebCore::InsertionPoint::willDetachRenderers):
1287 * loader/PlaceholderDocument.cpp:
1288 (WebCore::PlaceholderDocument::createRenderTree):
1289 * style/StyleResolveTree.cpp:
1290 (WebCore::Style::attachChildren):
1291 (WebCore::Style::attachRenderTree):
1292 (WebCore::Style::detachChildren):
1293 (WebCore::Style::detachRenderTree):
1294 (WebCore::Style::resolveLocal):
1295 (WebCore::Style::updateTextStyle):
1296 (WebCore::Style::resolveShadowTree):
1297 (WebCore::Style::resolveTree):
1299 Documents only ever have one child element. Remove the loop.
1301 (WebCore::Style::detachRenderTreeInReattachMode):
1302 (WebCore::Style::reattachRenderTree):
1303 * style/StyleResolveTree.h:
1305 (WebCore::SVGTests::handleAttributeChange):
1306 * xml/XMLErrors.cpp:
1307 (WebCore::XMLErrors::insertErrorMessageBlock):
1308 * xml/parser/XMLDocumentParserLibxml2.cpp:
1309 (WebCore::XMLDocumentParser::startElementNs):
1311 2013-08-30 Andreas Kling <akling@apple.com>
1313 Document style resolve should take Document&.
1314 <https://webkit.org/b/120534>
1316 Reviewed by Darin Adler.
1318 Tweak Style::resolveTree(Document*) and Style::resolveForDocument(Document*) to take Document& instead.
1321 (WebCore::Document::recalcStyle):
1322 * html/HTMLLinkElement.cpp:
1323 (WebCore::HTMLLinkElement::process):
1324 * style/StyleResolveForDocument.cpp:
1325 (WebCore::Style::resolveForDocument):
1326 * style/StyleResolveForDocument.h:
1327 * style/StyleResolveTree.cpp:
1328 (WebCore::Style::resolveTree):
1329 * style/StyleResolveTree.h:
1331 2013-08-30 Darin Adler <darin@apple.com>
1333 [Mac] No need for Pasteboard::getDataSelection
1334 https://bugs.webkit.org/show_bug.cgi?id=120536
1336 Reviewed by Anders Carlsson.
1338 * editing/Editor.h: Added some Mac-only private member functions.
1340 * editing/mac/EditorMac.mm:
1341 (WebCore::Editor::selectionInWebArchiveFormat): Added.
1342 (WebCore::Editor::adjustedSelectionRange): Added.
1343 (WebCore::attributedStringForRange): Added.
1344 (WebCore::dataInRTFDFormat): Added.
1345 (WebCore::dataInRTFFormat): Added.
1346 (WebCore::Editor::dataSelectionForPasteboard): Moved the implementation
1347 of Pasteboard::getDataSelection here, refactoring to share code with the
1348 writeSelectionToPasteboard function.
1349 (WebCore::Editor::writeSelectionToPasteboard): Refactored to share code
1350 with the new function above.
1352 * platform/Pasteboard.h: Removed getDataSelection. One less layering violation.
1353 * platform/mac/PasteboardMac.mm: Ditto.
1355 2013-08-30 Rob Buis <rwlbuis@webkit.org>
1357 SVG error parsing empty path
1358 https://bugs.webkit.org/show_bug.cgi?id=78980
1360 Reviewed by Darin Adler.
1362 According to the spec (http://www.w3.org/TR/SVG/paths.html#PathData), path 'd' attribute can be empty.
1363 No error should be reported in this case.
1365 * svg/SVGPathParser.cpp:
1366 (WebCore::SVGPathParser::parsePathDataFromSource):
1367 * svg/SVGPathUtilities.cpp:
1368 (WebCore::buildPathFromString):
1369 (WebCore::buildSVGPathByteStreamFromSVGPathSegList):
1370 (WebCore::buildPathFromByteStream):
1371 (WebCore::buildSVGPathSegListFromByteStream):
1372 (WebCore::buildStringFromByteStream):
1373 (WebCore::buildStringFromSVGPathSegList):
1374 (WebCore::buildSVGPathByteStreamFromString):
1375 (WebCore::buildAnimatedSVGPathByteStream):
1376 (WebCore::addToSVGPathByteStream):
1378 2013-08-30 Andreas Kling <akling@apple.com>
1380 Try to fix CSS_VARIABLES and CSS_DEVICE_ADAPTATION builds.
1382 * css/StyleResolver.cpp:
1383 (WebCore::StyleResolver::StyleResolver):
1384 (WebCore::StyleResolver::resolveVariables):
1386 2013-08-30 Commit Queue <commit-queue@webkit.org>
1388 Unreviewed, rolling out r154891.
1389 http://trac.webkit.org/changeset/154891
1390 https://bugs.webkit.org/show_bug.cgi?id=120539
1392 broke the mac build (Requested by Ossy on #webkit).
1394 * platform/network/SynchronousLoaderClient.h:
1396 2013-08-30 Csaba Osztrogonác <ossy@webkit.org>
1398 Add USE(CFNETWORK) guard to SynchronousLoaderClient::didReceiveAuthenticationChallengedidReceiveAuthenticationChallenge()
1399 https://bugs.webkit.org/show_bug.cgi?id=120532
1401 Reviewed by Darin Adler.
1403 * platform/network/SynchronousLoaderClient.h:
1405 2013-08-30 Brent Fulgham <bfulgham@apple.com>
1407 [Windows] Video inside page always uses non-hardware accelerated playback
1408 https://bugs.webkit.org/show_bug.cgi?id=120448
1410 Reviewed by Eric Carlson.
1412 * platform/graphics/avfoundation/cf/MediaPlayerPrivateAVFoundationCF.cpp:
1413 (WebCore::MediaPlayerPrivateAVFoundationCF::createAVPlayer): Check for
1414 D3D device. If it doesn't exist, switch to compositing mode and try again.
1416 2013-08-30 Leo Yang <leoyang@blackberry.com>
1418 Possible dangling CachedResourceClient of StyleRuleImport and XSLImportRule
1419 https://bugs.webkit.org/show_bug.cgi?id=120479
1421 Reviewed by Darin Adler.
1423 In StyleRuleImport::requestStyleSheet() and XSLImportRule::loadSheet() we
1424 didn't call removeClient() for m_cachedSheet before assigning m_cachedSheet
1425 a new value. This could leave the client as a client of the old cached
1426 sheet and dangling after the client is deleted. Fix them by calling removeClient()
1427 before assigning m_cacheSheet a new value.
1429 Found by code inspection. Seems no way to test it automatically.
1431 * css/StyleRuleImport.cpp:
1432 (WebCore::StyleRuleImport::requestStyleSheet):
1433 * xml/XSLImportRule.cpp:
1434 (WebCore::XSLImportRule::loadSheet):
1436 2013-08-30 Gabor Abraham <abrhm@inf.u-szeged.hu>
1438 Fix unused parameter warning in graphics/cairo/GraphicsContextCairo.cpp file.
1439 https://bugs.webkit.org/show_bug.cgi?id=120524
1441 Reviewed by Darin Adler.
1443 * platform/graphics/cairo/GraphicsContextCairo.cpp:
1444 (WebCore::GraphicsContext::fillRectWithRoundedHole):
1446 2013-08-30 Andreas Kling <akling@apple.com>
1448 StyleResolver & friends should use Document&.
1449 <https://webkit.org/b/120527>
1451 Reviewed by Antti Koivisto.
1453 Document&-ify StyleResolver, ElementRuleCollector and SelectorChecker.
1455 2013-08-30 Brent Fulgham <bfulgham@apple.com>
1457 [Windows] Unreviewed build correction after r154835. Only seen when doing a
1458 full (clean) rebuild.
1460 * DerivedSources.cpp: Remove reference to deleted JSHTMLDialogElement.cpp
1462 2013-08-30 Tamas Czene <tczene@inf.u-szeged.hu>
1464 Resolve unused parameter warning in FindController.cpp
1465 https://bugs.webkit.org/show_bug.cgi?id=120407
1467 Reviewed by Darin Adler.
1469 * accessibility/atk/WebKitAccessibleInterfaceText.cpp:
1470 (wordAtPositionForAtkBoundary):
1472 2013-08-30 Arpita Bahuguna <a.bah@samsung.com>
1474 setAttributeNode() does not set the new value to an existing attribute if specified attribute is in a different case.
1475 https://bugs.webkit.org/show_bug.cgi?id=120293
1477 Reviewed by Darin Adler.
1479 setAttributeNode() performs a case-insensitive search for an existing
1480 attribute. If an existing attribute is found, it retrieves the index of
1481 such an attribute. For setting the attribute's new value, we call upon
1482 setAttributeInternal() to which both the index as well as the name of
1483 the attribute is passed.
1484 The name passed to this method is the same as the one passed to the
1485 setAttributeNode() API from the webpage and thus can be in any case.
1487 However, setAttributeInternal() uses this name to get the corresponding
1488 existing attribute node. Since this retrieval is not case-insensitive,
1489 the existing node is not returned and thus the new value is not set on
1491 We should instead use the passed index and use that to retrieve the
1494 Note that obtaining the attribute's value using getAttributeNode() would
1495 still return the correct value, i.e. the new one.
1497 Also, this change shall make our behavior similar to that of FF and IE.
1499 Test: fast/dom/Element/setAttributeNode-for-existing-attribute.html
1502 (WebCore::Element::setAttributeInternal):
1503 If the passed index is not equal to attributeNotFound, we use that index
1504 to retrieve the existing attribute.
1506 2013-08-30 Mikhail Pozdnyakov <mikhail.pozdnyakov@intel.com>
1508 SimpleClassVectorTraits shall be used for RuleData
1509 https://bugs.webkit.org/show_bug.cgi?id=120526
1511 Reviewed by Andreas Kling.
1513 Whereas RuleData is POD type and whereas there are quite a lot of Vector<RuleData> occurrences in the code,
1514 it makes sense to use SimpleClassVectorTraits for RuleData and hence to allow using more efficient mem functions
1519 2013-08-30 Andreas Kling <akling@apple.com>
1521 Node::document() should return a reference.
1522 <https://webkit.org/b/120496>
1524 Reviewed by Antti Koivisto.
1526 Now that orphan DocumentType nodes also have document pointers, it's no longer
1527 possible to have a null Node::document().
1529 Cement this by making document() return a reference, and remove the various
1530 null checks exposed by this.
1532 2013-08-30 Andrei Parvu <parvu@adobe.com>
1534 [CSS Masking] -webkit-mask-repeat: space does not work
1535 Added the space option to background-repeat and -webkit-mask-repeat.
1536 With the property value 'space', the background or mask image gets repeated as often as it fits within the background positioning
1537 area. The repeated images are spaced equally to fill the unused area.
1538 https://bugs.webkit.org/show_bug.cgi?id=119324
1540 Reviewed by Dirk Schulze.
1542 Tests: css3/background/background-repeat-space-border.html
1543 css3/background/background-repeat-space-content.html
1544 css3/background/background-repeat-space-padding.html
1545 css3/masking/mask-repeat-space-border.html
1546 css3/masking/mask-repeat-space-content.html
1547 css3/masking/mask-repeat-space-padding.html
1549 * platform/graphics/GeneratorGeneratedImage.cpp:
1550 (WebCore::GeneratorGeneratedImage::drawPattern): Passed the space values to the image buffer.
1551 * platform/graphics/Image.cpp:
1552 (WebCore::Image::drawTiled): Added the space values when computing the location of the tile.
1553 * platform/graphics/Image.h: Added the space property.
1554 (WebCore::Image::spaceSize):
1555 (WebCore::Image::setSpaceSize):
1556 * platform/graphics/ImageBuffer.h: Added the space property.
1557 (WebCore::ImageBuffer::spaceSize):
1558 (WebCore::ImageBuffer::setSpaceSize):
1559 * platform/graphics/cg/ImageBufferCG.cpp: Passed the space values when copying an image.
1560 (WebCore::ImageBuffer::copyImage):
1561 * platform/graphics/cg/ImageCG.cpp: Added the space values when creating a platform pattern.
1562 (WebCore::Image::drawPattern):
1563 * rendering/RenderBoxModelObject.cpp:
1564 (WebCore::RenderBoxModelObject::paintFillLayerExtended): Computed the space values on x and y axis.
1565 (WebCore::getSpace):
1566 (WebCore::RenderBoxModelObject::calculateBackgroundImageGeometry): Pass the space values to the Image class.
1567 * rendering/RenderBoxModelObject.h: Added the space property.
1568 (WebCore::RenderBoxModelObject::BackgroundImageGeometry::spaceSize):
1569 (WebCore::RenderBoxModelObject::BackgroundImageGeometry::setSpaceSize):
1570 * svg/graphics/SVGImage.cpp: Passed the space property to the created image.
1571 (WebCore::SVGImage::drawPatternForContainer):
1572 * svg/graphics/SVGImageForContainer.cpp: Passed the space property to the image property.
1573 (WebCore::SVGImageForContainer::drawPattern):
1575 2013-08-30 Antti Koivisto <antti@apple.com>
1577 Remove AttachContext
1578 https://bugs.webkit.org/show_bug.cgi?id=120518
1580 Reviewed by Andreas Kling.
1582 This type is not useful anymore. Just pass the precomputed style to attachRenderTree and reattach-or-not flag to detachRenderTree.
1584 * style/StyleResolveTree.cpp:
1585 (WebCore::Style::createRendererIfNeeded):
1586 (WebCore::Style::attachChildren):
1587 (WebCore::Style::attachShadowRoot):
1588 (WebCore::Style::attachRenderTree):
1589 (WebCore::Style::detachChildren):
1590 (WebCore::Style::detachShadowRoot):
1591 (WebCore::Style::detachRenderTree):
1592 (WebCore::Style::reattachRenderTree):
1593 (WebCore::Style::resolveLocal):
1594 * style/StyleResolveTree.h:
1596 2013-08-30 Commit Queue <commit-queue@webkit.org>
1598 Unreviewed, rolling out r154826.
1599 http://trac.webkit.org/changeset/154826
1600 https://bugs.webkit.org/show_bug.cgi?id=120517
1602 Still breaks icloud.com (Requested by mwenge_ on #webkit).
1604 * rendering/RenderBox.cpp:
1605 (WebCore::RenderBox::computePositionedLogicalWidth):
1606 (WebCore::RenderBox::computePositionedLogicalHeight):
1607 * rendering/RenderBox.h:
1608 (WebCore::RenderBox::intrinsicSize):
1609 * rendering/RenderButton.h:
1610 * rendering/RenderFileUploadControl.cpp:
1611 (WebCore::RenderFileUploadControl::computePreferredLogicalWidths):
1612 * rendering/RenderListBox.cpp:
1613 (WebCore::RenderListBox::RenderListBox):
1614 (WebCore::RenderListBox::computePreferredLogicalWidths):
1615 (WebCore::RenderListBox::computeLogicalHeight):
1616 * rendering/RenderListBox.h:
1617 * rendering/RenderMenuList.h:
1618 * rendering/RenderSlider.cpp:
1619 (WebCore::RenderSlider::computePreferredLogicalWidths):
1620 * rendering/RenderTextControl.cpp:
1621 (WebCore::RenderTextControl::RenderTextControl):
1622 (WebCore::RenderTextControl::computeLogicalHeight):
1623 * rendering/RenderTextControl.h:
1625 2013-08-29 Ryosuke Niwa <rniwa@webkit.org>
1627 Unduplicate the code to convert between VisiblePosition and index
1628 https://bugs.webkit.org/show_bug.cgi?id=120506
1630 Reviewed by Darin Adler.
1632 Encapsulate the conversion between VisiblePosition and index into indexForVisiblePosition
1633 and visiblePositionForIndexUsingCharacterIterator. It's unfortunate that these two functions
1634 are different from the two other existing functions of similar names but we've at least
1635 confined problems into a single cpp file now.
1637 * accessibility/AccessibilityRenderObject.cpp:
1638 (WebCore::AccessibilityRenderObject::visiblePositionForIndex):
1639 (WebCore::AccessibilityRenderObject::indexForVisiblePosition):
1640 * editing/Editor.cpp:
1641 (WebCore::findFirstMarkable):
1642 * editing/htmlediting.cpp:
1643 (WebCore::indexForVisiblePosition):
1644 (WebCore::visiblePositionForIndexUsingCharacterIterator):
1645 * editing/htmlediting.h:
1646 * html/HTMLTextFormControlElement.cpp:
1647 (WebCore::hasVisibleTextArea):
1648 (WebCore::HTMLTextFormControlElement::setSelectionRange):
1649 (WebCore::HTMLTextFormControlElement::indexForVisiblePosition):
1650 (WebCore::HTMLTextFormControlElement::visiblePositionForIndex):
1651 * html/HTMLTextFormControlElement.h:
1652 * rendering/RenderTextControl.cpp:
1653 * rendering/RenderTextControl.h:
1655 2013-08-29 Ryosuke Niwa <rniwa@webkit.org>
1657 Avoid Node references from AXObjectCache from leaking
1658 https://bugs.webkit.org/show_bug.cgi?id=120501
1660 Reviewed by Darin Adler.
1662 Merge https://chromium.googlesource.com/chromium/blink/+/454f31497613b6d0fbcfb0df757254b64a177c06
1663 without any tests since we don't have the same infrastructure to detect leaks in WebKit.
1665 A real world example of this would be selecting an <option> item inside frame by keyboard. The node will not be deref()-ed until the topDocument() is detached.
1667 The issue was that AccessibilityMenuListOption is created in childrenChanged()
1668 hook called when its RenderObject is being destroyed. This patch modifies AccessibilityMenuListPopup so it won't create AccessibilityMenuListOption if its
1669 element is already detached.
1671 * accessibility/AccessibilityMenuListPopup.cpp:
1672 (WebCore::AccessibilityMenuListPopup::menuListOptionAccessibilityObject):
1673 * rendering/RenderMenuList.cpp:
1674 (WebCore::RenderMenuList::didUpdateActiveOption):
1676 2013-08-26 Simon Fraser <simon.fraser@apple.com>
1678 Implement object-fit CSS property
1679 https://bugs.webkit.org/show_bug.cgi?id=52040
1681 Reviewed by Antti Koivisto, Sam Weinig.
1683 Merge object-fit patch from Blink r156535, which started as a patch
1686 Since then, the spec has gone to CR. This patch is an
1687 implementation of object-fit as described in
1688 http://www.w3.org/TR/2012/CR-css3-images-20120417/#object-fit
1690 Object-fit is used to maintain the aspect ratio of replaced content
1691 within its content box. All object-fit values but the initial one
1692 ('fill') will always ensure that the aspect ratio is retained, in
1693 different ways (fit inside the content box, cover the content box, or
1694 use intrinsic size). Painting is always clipped against the content
1695 box, regardless of the 'overflow' property.
1697 Tests: fast/css/object-fit/object-fit-canvas.html
1698 fast/css/object-fit/object-fit-embed.html
1699 fast/css/object-fit/object-fit-grow-landscape.html
1700 fast/css/object-fit/object-fit-grow-portrait.html
1701 fast/css/object-fit/object-fit-img-svg.html
1702 fast/css/object-fit/object-fit-img-svg2.html
1703 fast/css/object-fit/object-fit-img.html
1704 fast/css/object-fit/object-fit-input-image.html
1705 fast/css/object-fit/object-fit-object.html
1706 fast/css/object-fit/object-fit-shrink.html
1707 fast/css/object-fit/object-fit-video-poster.html
1708 fast/css/parsing-object-fit.html
1709 http/tests/css/object-fit-delayed-img-svg.html
1710 media/video-object-fit-change.html
1711 media/video-object-fit.html
1713 * css/CSSComputedStyleDeclaration.cpp:
1714 (WebCore::ComputedStyleExtractor::propertyValue):
1715 * css/CSSParser.cpp:
1716 (WebCore::isValidKeywordPropertyAndValue):
1717 (WebCore::isKeywordPropertyID):
1718 (WebCore::CSSParser::parseValue):
1719 * css/CSSPrimitiveValueMappings.h:
1720 (WebCore::CSSPrimitiveValue::CSSPrimitiveValue):
1721 (WebCore::CSSPrimitiveValue::operator EObjectFit):
1722 * css/CSSProperty.cpp:
1723 (WebCore::CSSProperty::isInheritedProperty):
1724 * css/CSSPropertyNames.in:
1725 * css/CSSValueKeywords.in:
1726 * css/DeprecatedStyleBuilder.cpp:
1727 (WebCore::DeprecatedStyleBuilder::DeprecatedStyleBuilder):
1728 * css/StyleResolver.cpp:
1729 (WebCore::StyleResolver::applyProperty):
1731 (video): Set object-fit to 'contain'. This is how VIDEO elements
1733 * loader/cache/CachedImage.cpp:
1734 (WebCore::CachedImage::imageSizeForRenderer):
1735 * loader/cache/CachedImage.h:
1736 * platform/graphics/LayoutSize.h:
1737 (WebCore::fitLayoutSizeToAspectRatio): New function to grow or shrink
1738 in one dimension to fit to the aspect ratio.
1739 * rendering/RenderHTMLCanvas.cpp:
1740 (WebCore::RenderHTMLCanvas::paintReplaced): Apply object-fit and
1742 * rendering/RenderImage.cpp:
1743 (WebCore::RenderImage::updateInnerContentRect):
1744 (WebCore::RenderImage::imageDimensionsChanged): Update intrinsic
1745 size properly, and recalculate the inner content rectangle (the
1746 exact area occupied by the replaced content) again if appropriate.
1747 (WebCore::RenderImage::paintReplaced): Apply object-fit and clip
1749 (WebCore::RenderImage::foregroundIsKnownToBeOpaqueInRect):
1750 object-fit may leave parts of the content box empty, in which case
1751 it won't be fully obscured.
1752 (WebCore::RenderImage::layout):
1753 * rendering/RenderImage.h:
1754 * rendering/RenderImageResource.cpp:
1755 (WebCore::RenderImageResource::intrinsicSize): Need this to
1756 differentiate between intrinsic and extrinsic size for SVG images.
1757 * rendering/RenderImageResource.h:
1758 * rendering/RenderImageResourceStyleImage.h:
1759 * rendering/RenderReplaced.cpp:
1760 (WebCore::RenderReplaced::replacedContentRect): Return the
1761 rectangle occupied by the replaced content. This will be identical
1762 to the content box if object-fit is 'fill', but will typically be
1763 something else for other values.
1764 * rendering/RenderReplaced.h:
1765 * rendering/RenderVideo.cpp:
1766 (WebCore::RenderVideo::videoBox): Not much left to do here, with
1767 the new RenderReplaced::replacedContentRect() method in place.
1768 (WebCore::RenderVideo::paintReplaced): Apply object-fit and clip
1770 * rendering/style/RenderStyle.cpp:
1771 (WebCore::RenderStyle::changeRequiresRepaint):
1772 * rendering/style/RenderStyle.h:
1773 * rendering/style/RenderStyleConstants.h:
1774 * rendering/style/StyleRareNonInheritedData.cpp:
1775 (WebCore::StyleRareNonInheritedData::StyleRareNonInheritedData):
1776 (WebCore::StyleRareNonInheritedData::operator==):
1777 * rendering/style/StyleRareNonInheritedData.h:
1779 2013-08-29 Tim Horton <timothy_horton@apple.com>
1781 SVG clipping, masking, and gradients-on-text do not respect the device scale factor
1782 https://bugs.webkit.org/show_bug.cgi?id=120377
1783 <rdar://problem/14777944>
1785 Reviewed by Simon Fraser.
1787 Tests: svg/custom/masking-clipping-hidpi.svg
1789 calculateTransformationToOutermostCoordinateSystem should include the page's
1790 device scale factor in its transform.
1792 * rendering/svg/SVGRenderingContext.cpp:
1793 (WebCore::SVGRenderingContext::calculateTransformationToOutermostCoordinateSystem):
1795 2013-08-29 Chris Fleizach <cfleizach@apple.com>
1797 AX: Crash when VoiceOver reference a TextMarker from an iframe that has been removed
1798 https://bugs.webkit.org/show_bug.cgi?id=120318
1800 Reviewed by Darin Adler.
1802 These crashes occur because an AXTextMarker is retrieved that reference a Node in an iframe.
1803 The iframe goes away, but when it tries to clean up and remove that Node from the InUse cache,
1804 the document() tree has already been detached, so it never clears the actual InUse cache.
1806 The fix here is to pre-emptively clean up any nodes in the document going away when the frame is about to disconnect.
1808 I'm removing the clearAXObjectCache() at the disconnectOwnerElement because it
1809 1) Cleared the AXObjectCache for the iframe document (which is always empty -- only the top level doc maintains the cache), because
1810 the document() tree had already been detached... so it essentially did nothing.
1811 2) And if it did work, we wouldn't want this behavior -- that is to say, when an iframe goes away, we don't want to clear the entire cache for all
1812 the other documents (there's even an existing layout test to verify this behavior).
1814 Test: platform/mac/accessibility/stale-textmarker-crash.html
1816 * accessibility/AXObjectCache.cpp:
1817 (WebCore::AXObjectCache::clearTextMarkerNodesInUse):
1818 * accessibility/AXObjectCache.h:
1820 (WebCore::Frame::disconnectOwnerElement):
1822 2013-08-29 Brent Fulgham <bfulgham@apple.com>
1824 [Windows] More unreviewed gardening of project file.
1825 Move css files inline with related source code.
1827 * WebCore.vcxproj/WebCore.vcxproj:
1828 * WebCore.vcxproj/WebCore.vcxproj.filters:
1830 2013-08-29 Sam Weinig <sam@webkit.org>
1832 Add ENABLE guards for Promises
1833 https://bugs.webkit.org/show_bug.cgi?id=120488
1835 Reviewed by Andreas Kling.
1837 * Configurations/FeatureDefines.xcconfig:
1839 2013-08-29 Andreas Kling <akling@apple.com>
1841 Inspector: Use a bit of RefPtr::release() in generated code.
1842 <https://webkit.org/b/120485>
1844 Reviewed by Darin Adler.
1846 Just something I spotted while reading InspectorBackendDispatcher.cpp.
1847 Avoid 3 cases of reference count churnery.
1849 * inspector/CodeGeneratorInspectorStrings.py:
1851 2013-08-29 Christophe Dumez <ch.dumez@sisa.samsung.com>
1853 According to DOM4, all DocType nodes should have a document
1854 https://bugs.webkit.org/show_bug.cgi?id=99244
1856 Reviewed by Darin Adler.
1858 Doctypes now always have a node document and can be moved across document boundaries as per
1859 the latest DOM4 specification:
1860 http://dom.spec.whatwg.org/#dom-domimplementation-createdocumenttype
1861 http://dom.spec.whatwg.org/#dom-node-ownerdocument
1863 This means that DOMImplementation.createDocumentType() now sets the ownerDocument of the
1864 new DocumentType Node to the associated document of the current "context" object. In
1865 DOM4, all nodes have a document at all times. DocumentType nodes can now be moved across
1866 document boundaries so that the node can be added to a Document after being created.
1868 This means we will no longer need to special case DocumentType nodes in the code and
1869 Node::document() can no longer return NULL, which means that we'll be able to remove
1870 NULL checks in call sites.
1872 Firefox stable and since recently Blink already follow DOM4 here while IE10 does not (yet).
1874 Test: fast/dom/createDocumentType-ownerDocument.html
1876 * dom/ContainerNode.cpp:
1877 (WebCore::checkAcceptChild):
1878 * dom/DOMImplementation.cpp:
1879 (WebCore::DOMImplementation::createDocumentType):
1880 (WebCore::DOMImplementation::createDocument):
1882 (WebCore::Node::document):
1884 2013-08-29 Darin Adler <darin@apple.com>
1886 Pasteboard::writeSelection violates layering (first step, fixes it for Mac platform only)
1887 https://bugs.webkit.org/show_bug.cgi?id=120483
1889 Reviewed by Anders Carlsson.
1891 * editing/Editor.cpp:
1892 (WebCore::Editor::cut): Added some comments. Use the new writeSelectionToPasteboard
1893 function on Mac instead of Pasteboard::writeSelection.
1894 (WebCore::Editor::copy): Ditto.
1896 * editing/Editor.h: Removed an old unused Mac-only writeSelectionToPasteboard function
1897 that was a cover that called through to Pasteboard::writeSelectionForTypes. Added a new
1898 Mac-only writeSelectionToPasteboard function that is destined to become cross-platform soon.
1900 * editing/mac/EditorMac.mm:
1901 (WebCore::Editor::writeSelectionToPasteboard): Added. Uses a new pattern where the Editor
1902 puts all the data into a structure called PasteboardWebContent then calls the Pasteboard to
1903 do the work. The platform-specific aspect of PasteboardWebContent is what formats are needed
1906 * page/DragController.cpp:
1907 (WebCore::DragController::startDrag): Added some comments. Use the new writeSelectionToPasteboard
1908 function on Mac instead of Pasteboard::writeSelection.
1910 * platform/Pasteboard.h: Remove some unneeded forward declarations. Added comments for all functions
1911 that don't belong in this class because they are layering violations; this becomes the to do list for
1912 the project we are beginning here. Added the new PasteboardWebContent structure, empty on all platforms
1913 except for Mac for now. Removed writeSelectionForTypes, a Mac-only function that is no longer used.
1914 Added setTypes and writeAfterSettingTypes, the two halves of the future function named writeWebContent.
1915 Put the writeSelection function inside a "not Mac" if statement. Later to be deleted entirely.
1917 * platform/mac/PasteboardMac.mm: Removed now-unneeded selectionPasteboardTypes,
1918 Pasteboard::writeSelectionForTypes, and writeSelection functions.
1919 (WebCore::Pasteboard::setTypes): Added. First half of writing web content to the pasteboard.
1920 (WebCore::Pasteboard::writeAfterSettingTypes): Added. Second half of writing web content to the pasteboard.
1922 2013-08-29 Antti Koivisto <antti@apple.com>
1924 Remove code behind ENABLE(DIALOG_ELEMENT)
1925 https://bugs.webkit.org/show_bug.cgi?id=120467
1927 Reviewed by Darin Adler.
1929 It is incomplete and no one is building it.
1932 * Configurations/FeatureDefines.xcconfig:
1933 * DerivedSources.make:
1934 * DerivedSources.pri:
1935 * GNUmakefile.list.am:
1938 * WebCore.xcodeproj/project.pbxproj:
1939 * bindings/generic/RuntimeEnabledFeatures.cpp:
1940 * bindings/generic/RuntimeEnabledFeatures.h:
1941 * css/StyleResolver.cpp:
1942 (WebCore::StyleResolver::adjustRenderStyle):
1944 * dom/ContextFeatures.cpp:
1945 * dom/ContextFeatures.h:
1949 (WebCore::Element::removedFrom):
1950 (WebCore::Element::setContainsFullScreenElementOnAncestorsCrossingFrameBoundaries):
1952 (WebCore::Element::isDisabledFormControl):
1953 * dom/ElementRareData.h:
1954 (WebCore::ElementRareData::ElementRareData):
1955 * html/HTMLDialogElement.cpp: Removed.
1956 * html/HTMLDialogElement.h: Removed.
1957 * html/HTMLDialogElement.idl: Removed.
1958 * html/HTMLElementsAllInOne.cpp:
1959 * html/HTMLTagNames.in:
1960 * rendering/RenderDialog.cpp: Removed.
1961 * rendering/RenderDialog.h: Removed.
1962 * rendering/RenderLayer.cpp:
1963 (WebCore::RenderLayer::rebuildZOrderLists):
1964 (WebCore::RenderLayer::collectLayers):
1965 * rendering/RenderLayer.h:
1966 * rendering/RenderObject.h:
1967 (WebCore::RenderObject::isCounter):
1968 * rendering/RenderingAllInOne.cpp:
1969 * testing/InternalSettings.cpp:
1970 (WebCore::InternalSettings::Backup::Backup):
1971 (WebCore::InternalSettings::Backup::restoreTo):
1972 * testing/InternalSettings.h:
1974 2013-08-29 Simon Fraser <simon.fraser@apple.com>
1976 Fix slider thumb event handling to use local, not absolute coordinates
1977 https://bugs.webkit.org/show_bug.cgi?id=120480
1979 Reviewed by Darin Adler.
1981 SliderThumbElement::setPositionFromPoint() did all of its coordinate
1982 math by mapping renderer rects into absolute coordinates, which was
1983 unnecessary and expensive.
1985 Fix by doing all the math in the coordinate space of the input's
1986 renderer. This simplified the code. Also, currentPosition
1987 was computed but unused, so was removed.
1989 No behavior change. Tested by fast/forms/range/slider-transformed.html
1991 * html/shadow/SliderThumbElement.cpp:
1992 (WebCore::SliderThumbElement::setPositionFromPoint):
1994 2013-08-29 Zan Dobersek <zdobersek@igalia.com>
1996 [Automake] libWebCoreDOM.la could use a better name
1997 https://bugs.webkit.org/show_bug.cgi?id=120232
1999 Reviewed by Martin Robinson.
2001 * bindings/gobject/GNUmakefile.am: Rename the libWebCoreDOM library to libGObjectDOMBindings.
2003 2013-08-29 Joseph Pecoraro <pecoraro@apple.com>
2005 Web Inspector: Download Web Archive of Inspected Page
2006 https://bugs.webkit.org/show_bug.cgi?id=119774
2008 Reviewed by Timothy Hatcher.
2010 Add PageAgent.archive which will return a Base-64 encoded web archive
2011 when successful. In order to then allow saving non-string files, extend
2012 InspectorFrontendHost.save to allow for Base-64 encoded data.
2014 * inspector/Inspector.json:
2015 * inspector/InspectorPageAgent.cpp:
2016 (WebCore::InspectorPageAgent::archive):
2017 * inspector/InspectorPageAgent.h:
2018 Introduce and implement PageAgent.archive. Create a Web Archive of the
2021 * inspector/InspectorFrontendClient.h:
2022 * inspector/InspectorFrontendClientLocal.h:
2023 (WebCore::InspectorFrontendClientLocal::save):
2024 * inspector/InspectorFrontendHost.cpp:
2025 (WebCore::InspectorFrontendHost::save):
2026 * inspector/InspectorFrontendHost.h:
2027 * inspector/InspectorFrontendHost.idl:
2028 Extend InspectorFrontendHost.save to include a base64Encoded param.
2030 * inspector/front-end/FileManager.js:
2031 * inspector/front-end/InspectorFrontendHostStub.js:
2032 (.WebInspector.InspectorFrontendHostStub.prototype.save):
2033 Misc. updates to the old inspector for function changes.
2035 2013-08-29 Robert Hogan <robert@webkit.org>
2037 Positioned Replaced Elements That Aren't RenderReplaced get Incorrect Width
2038 https://bugs.webkit.org/show_bug.cgi?id=93735
2040 Reviewed by David Hyatt.
2042 Replaced elements that aren't RenderReplaced aren't |isReplaced| and don't have an
2043 intrinsic height or width. This causes them to go down the wrong height and width computation
2044 path in RenderBox when they are absolute positioned.
2046 The notion of |isReplaced| is entwined with the notion of being |isInline| so it isn't really
2047 possible to make them isReplaced without re-wiring a lot of code. So instead use an ad-hoc definition
2048 of isReplacedElement in RenderBox to bring all replaced elements into the height and width calculation.
2049 To make sure we get the right height and width in there, give non-RenderReplaced replaced renderers
2050 the helpers for returning their approximation of intrinsic height and width.
2052 The initial attempt at landing this patch had to be rolled out because it used LayoutUnit() for default
2053 intrinsic height of some replaced elements and this made the layout of the elements unstable in some sites.
2054 The fix for this issue is captured in intrinsic-button-and-input-height.html.
2056 Tests: fast/replaced/intrinsic-button-and-input-height.html
2057 fast/replaced/width-and-height-of-positioned-replaced-elements.html
2059 * rendering/RenderBox.cpp:
2060 (WebCore::isReplacedElement):
2061 (WebCore::RenderBox::computePositionedLogicalWidth):
2062 (WebCore::RenderBox::computePositionedLogicalHeight):
2063 * rendering/RenderBox.h:
2064 (WebCore::RenderBox::intrinsicSize):
2065 * rendering/RenderButton.h:
2066 * rendering/RenderFileUploadControl.cpp:
2067 (WebCore::RenderFileUploadControl::computePreferredLogicalWidths):
2068 * rendering/RenderListBox.cpp:
2069 (WebCore::RenderListBox::RenderListBox):
2070 (WebCore::RenderListBox::computePreferredLogicalWidths):
2071 (WebCore::RenderListBox::computeLogicalHeight):
2072 * rendering/RenderListBox.h:
2073 * rendering/RenderMenuList.h:
2074 * rendering/RenderReplaced.h:
2075 * rendering/RenderSlider.cpp:
2076 (WebCore::RenderSlider::computePreferredLogicalWidths):
2077 * rendering/RenderTextControl.cpp:
2078 (WebCore::RenderTextControl::RenderTextControl):
2079 (WebCore::RenderTextControl::computeLogicalHeight):
2080 * rendering/RenderTextControl.h:
2082 2013-08-29 Brent Fulgham <bfulgham@apple.com>
2084 [Windows] Unreviewed gardening. Add missing *.css files from project.
2086 * WebCore.vcxproj/WebCore.vcxproj:
2087 * WebCore.vcxproj/WebCore.vcxproj.filters:
2089 2013-08-29 Pratik Solanki <pratik.solanki@gmail.com>
2091 SharedBuffer m_segments and m_dataArray must be exclusive
2092 https://bugs.webkit.org/show_bug.cgi?id=77715
2094 Reviewed by Benjamin Poulain.
2096 When USE(NETWORK_CFDATA_ARRAY_CALLBACK) is enabled, we use m_dataArray to hold the incoming
2097 data. We do not use m_segments. Since they are exclusive in practice, do not define or use
2098 m_segments when NETWORK_CFDATA_ARRAY_CALLBACK is enabled.
2100 No new tests because no functional changes.
2102 * platform/SharedBuffer.cpp:
2103 (WebCore::SharedBuffer::append):
2104 (WebCore::SharedBuffer::clear):
2105 (WebCore::SharedBuffer::copy):
2106 (WebCore::SharedBuffer::buffer):
2107 (WebCore::SharedBuffer::getSomeData):
2108 * platform/SharedBuffer.h:
2110 2013-08-29 Daniel Bates <dabates@apple.com>
2112 [iOS] Upstream changes to WebCore/style
2113 https://bugs.webkit.org/show_bug.cgi?id=120173
2115 Reviewed by Darin Adler.
2117 * style/StyleResolveTree.cpp:
2118 (WebCore::Style::elementImplicitVisibility): Added; specific to iOS.
2119 (WebCore::Style::CheckForVisibilityChangeOnRecalcStyle::CheckForVisibilityChangeOnRecalcStyle): Added; specific to iOS.
2120 (WebCore::Style::CheckForVisibilityChangeOnRecalcStyle::~CheckForVisibilityChangeOnRecalcStyle): Added; specific to iOS.
2121 (WebCore::Style::resolveTree): Modified to instantiate CheckForVisibilityChangeOnRecalcStyle when building on iOS.
2123 2013-08-29 Arnaud Renevier <a.renevier@samsung.com>
2125 [cairo] canvas drawing on itself doesn't work with accelerated canvas
2126 https://bugs.webkit.org/show_bug.cgi?id=118808
2128 Reviewed by Martin Robinson.
2130 When copying an accelerated image, we try to get its dimensions with
2131 cairo_image_surface_get_width/cairo_image_surface_get_height. As
2132 surface is not an image, this returns width and height of 0.
2134 Many other places use cairo_image_surface_get although the surface may
2137 This patch fixes those issues by implementing a cairoSurfaceSize
2138 helper that returns the surface size whatever type it is.
2140 It use cairo_surface_create_similar instead of
2141 cairo_image_surface_create in copyCairoImageSurface. It also calls
2142 cairo_paint in encodeImage when a drawing over a black background is
2145 It copies the surface to an image surface if needed in extractImage.
2147 No new tests. Covered by existing tests.
2149 * platform/graphics/cairo/BitmapImageCairo.cpp:
2150 (WebCore::BitmapImage::BitmapImage):
2151 (WebCore::BitmapImage::draw):
2152 (WebCore::BitmapImage::checkForSolidColor):
2153 * platform/graphics/cairo/CairoUtilities.cpp:
2154 (WebCore::copyCairoImageSurface):
2155 (WebCore::cairoSurfaceSize):
2156 * platform/graphics/cairo/CairoUtilities.h:
2157 * platform/graphics/cairo/GraphicsContext3DCairo.cpp:
2158 (WebCore::GraphicsContext3D::ImageExtractor::extractImage):
2159 * platform/graphics/gtk/GdkCairoUtilities.cpp:
2160 (cairoSurfaceToGdkPixbuf):
2161 * platform/graphics/gtk/GdkCairoUtilities.h:
2162 * platform/graphics/gtk/ImageBufferGtk.cpp:
2163 (WebCore::encodeImage):
2164 * platform/graphics/gtk/ImageGtk.cpp:
2165 (WebCore::BitmapImage::getGdkPixbuf):
2166 * platform/gtk/DragIcon.cpp:
2167 (WebCore::DragIcon::setImage):
2169 2013-08-29 Tamas Czene <tczene@inf.u-szeged.hu>
2171 Buildfix after r154806 for !ENABLE(CSS_REGIONS) platforms.
2172 https://bugs.webkit.org/show_bug.cgi?id=120461
2174 Reviewed by Csaba Osztrogonác.
2176 moveToFlowThreadIfNeeded() is used only inside of ENABLE(CSS_REGIONS)
2177 block (and only once), so it should be guarded too.
2179 * style/StyleResolveTree.cpp:
2181 2013-08-29 David Kilzer <ddkilzer@apple.com>
2183 BUILD FIX (r154778): AXSearchFieldCancelButtonText() is not defined on iOS
2185 See: AX: Cancel button in search field not accessible.
2186 <https://webkit.org/b/120322>
2188 Fixes the following build error:
2190 In file included from Source/WebCore/accessibility/AccessibilityAllInOne.cpp:44:
2191 Source/WebCore/accessibility/AccessibilitySearchFieldButtons.cpp:46:12: error: use of undeclared identifier 'AXSearchFieldCancelButtonText'
2192 return AXSearchFieldCancelButtonText();
2196 * accessibility/AccessibilitySearchFieldButtons.cpp:
2197 (WebCore::AccessibilitySearchFieldCancelButton::accessibilityDescription):
2198 Return String() instead of calling AXSearchFieldCancelButtonText() on iOS.
2200 2013-08-29 Antti Koivisto <antti@apple.com>
2202 Remove NodeRenderingContext
2203 https://bugs.webkit.org/show_bug.cgi?id=120466
2205 Reviewed by Andreas Kling.
2207 Switch the few remaining clients of this class to call the underlying code directly.
2210 * GNUmakefile.list.am:
2213 * WebCore.vcxproj/WebCore.vcxproj:
2214 * WebCore.xcodeproj/project.pbxproj:
2215 * css/StyleResolver.cpp:
2216 (WebCore::StyleResolver::State::initForStyleResolve):
2217 * dom/DOMAllInOne.cpp:
2221 Remove Node::parentNodeForRenderingAndStyle(). Make clients invoke NodeRenderingTraversal::parent() directly.
2223 * dom/NodeRenderingContext.cpp: Removed.
2224 * dom/NodeRenderingContext.h: Removed.
2228 * dom/TreeScope.cpp:
2231 Remove virtual TreeScope::resetStyleInheritance. Make clients cast to ShadowRoot when needed.
2233 * html/HTMLSummaryElement.cpp:
2234 (WebCore::HTMLSummaryElement::detailsElement):
2235 * style/StyleResolveTree.cpp:
2236 (WebCore::Style::createTextRendererIfNeeded):
2238 For consistency switch resetStyleInheritance check to use parentNode() instead of renderingParentNode. This matches the NodeRenderingTraversal implementation.
2239 This difference is probably not testable in current trunk.
2241 (WebCore::Style::resolveTree):
2243 2013-08-29 Antti Koivisto <antti@apple.com>
2245 Move element renderer creation out of NodeRenderingContext
2246 https://bugs.webkit.org/show_bug.cgi?id=120461
2248 Reviewed by Andreas Kling.
2250 Move NodeRenderingContext::createRendererIfNeeded() and the related utility functions to StyleResolveTree.
2252 Tighten typing and constness. Refactor sligthly to be more understandable.
2255 (WebCore::Element::shouldMoveToFlowThread):
2257 * dom/NodeRenderingContext.cpp:
2258 (WebCore::NodeRenderingContext::NodeRenderingContext):
2259 (WebCore::NodeRenderingContext::nextRenderer):
2260 (WebCore::NodeRenderingContext::previousRenderer):
2261 (WebCore::NodeRenderingContext::parentRenderer):
2262 * dom/NodeRenderingContext.h:
2263 * dom/PseudoElement.h:
2264 * style/StyleResolveTree.cpp:
2265 (WebCore::Style::nextSiblingRenderer):
2266 (WebCore::Style::shouldCreateRenderer):
2267 (WebCore::Style::elementInsideRegionNeedsRenderer):
2268 (WebCore::Style::moveToFlowThreadIfNeeded):
2269 (WebCore::Style::createRendererIfNeeded):
2270 (WebCore::Style::attachRenderTree):
2271 * svg/SVGElement.cpp:
2272 (WebCore::SVGElement::shouldMoveToFlowThread):
2275 2013-08-28 Chris Fleizach <cfleizach@apple.com>
2277 AX: WebProcess at com.apple.WebCore: WebCore::AXObjectCache::rootObject + 27
2278 https://bugs.webkit.org/show_bug.cgi?id=120434
2280 Reviewed by Darin Adler.
2282 Crash logs indicate that there's a null pointer access in rootObject. That seems like it could only
2283 happen in Document was null.
2285 Unfortunately, there are no reproducible steps and no other information to construct a test case.
2287 * accessibility/AXObjectCache.cpp:
2288 (WebCore::AXObjectCache::rootObject):
2290 2013-08-28 Ryosuke Niwa <rniwa@webkit.org>
2292 The code to look for an ancestor form element is duplicated in three different places
2293 https://bugs.webkit.org/show_bug.cgi?id=120391
2295 Reviewed by Darin Adler.
2297 Unduplicated the code by putting a single implementation in HTMLFormElement.cpp.
2300 * html/FormAssociatedElement.cpp:
2301 (WebCore::FormAssociatedElement::findAssociatedForm):
2302 (WebCore::FormAssociatedElement::formAttributeChanged):
2303 * html/HTMLElement.cpp:
2304 (WebCore::HTMLElement::virtualForm):
2305 * html/HTMLElement.h:
2306 * html/HTMLFormControlElement.cpp:
2307 (WebCore::HTMLFormControlElement::HTMLFormControlElement):
2308 * html/HTMLFormElement.cpp:
2309 (WebCore::HTMLFormElement::findClosestFormAncestor):
2310 * html/HTMLFormElement.h:
2311 * html/HTMLImageElement.cpp:
2312 (WebCore::HTMLImageElement::insertedInto):
2313 * html/HTMLObjectElement.cpp:
2314 (WebCore::HTMLObjectElement::HTMLObjectElement):
2315 * html/parser/HTMLTreeBuilder.cpp:
2316 (WebCore::HTMLTreeBuilder::HTMLTreeBuilder):
2318 2013-08-28 Ryosuke Niwa <rniwa@webkit.org>
2320 Stop throwing DOM exceptions in internal 'XMLHttpRequest' response getters
2321 https://bugs.webkit.org/show_bug.cgi?id=120446
2323 Reviewed by Alexey Proskuryakov.
2325 Merge https://chromium.googlesource.com/chromium/blink/+/c8188c21452501b68950a9fcc1f5cbc7b4de4df5
2327 Unlike 'responseText' and 'responseXML', 'responseBlob' and
2328 'responseArrayBuffer' are not exposed to JavaScript (they don't
2329 appear in the IDL or in the specification). As they are only called from
2330 custom bindings in response to a JavaScript call to the 'response' getter,
2331 we can safely replace the exception-throwing code in the implementation
2332 with an ASSERT that the request type is correct.
2334 * bindings/js/JSXMLHttpRequestCustom.cpp:
2335 (WebCore::JSXMLHttpRequest::response):
2336 * xml/XMLHttpRequest.cpp:
2337 (WebCore::XMLHttpRequest::responseBlob):
2338 (WebCore::XMLHttpRequest::responseArrayBuffer):
2339 * xml/XMLHttpRequest.h:
2341 2013-08-28 Chris Curtis <chris_curtis@apple.com>
2343 https://bugs.webkit.org/show_bug.cgi?id=119548
2344 Refactoring Exception throws.
2346 Reviewed by Geoffrey Garen.
2348 Gets column information from the error object for reporting exceptions.
2349 * bindings/js/JSDOMBinding.cpp:
2350 (WebCore::reportException):
2351 * bindings/js/ScriptCallStackFactory.cpp:
2352 (WebCore::createScriptCallStackFromException):
2354 Moved setting an exception into the vm, These functions changed to use the new functionality.
2356 * bindings/js/JSAudioBufferSourceNodeCustom.cpp:
2357 (WebCore::JSAudioBufferSourceNode::setBuffer):
2358 * bindings/js/JSBiquadFilterNodeCustom.cpp:
2359 (WebCore::JSBiquadFilterNode::setType):
2360 * bindings/js/JSCryptoCustom.cpp:
2361 (WebCore::JSCrypto::getRandomValues):
2362 * bindings/js/JSDOMBinding.cpp:
2363 (WebCore::setDOMException):
2364 * bindings/js/JSInjectedScriptHostCustom.cpp:
2365 (WebCore::JSInjectedScriptHost::setFunctionVariableValue):
2366 * bindings/js/JSJavaScriptCallFrameCustom.cpp:
2367 (WebCore::JSJavaScriptCallFrame::evaluate):
2368 (WebCore::JSJavaScriptCallFrame::setVariableValue):
2369 * bindings/js/JSNodeFilterCondition.cpp:
2370 (WebCore::JSNodeFilterCondition::acceptNode):
2371 * bindings/js/JSOscillatorNodeCustom.cpp:
2372 (WebCore::JSOscillatorNode::setType):
2373 * bindings/js/JSPannerNodeCustom.cpp:
2374 (WebCore::JSPannerNode::setPanningModel):
2375 (WebCore::JSPannerNode::setDistanceModel):
2376 * bindings/js/JSSVGLengthCustom.cpp:
2377 (WebCore::JSSVGLength::convertToSpecifiedUnits):
2378 * bindings/js/JSWebGLRenderingContextCustom.cpp:
2379 (WebCore::getObjectParameter):
2380 (WebCore::JSWebGLRenderingContext::getAttachedShaders):
2381 (WebCore::JSWebGLRenderingContext::getExtension):
2382 (WebCore::JSWebGLRenderingContext::getFramebufferAttachmentParameter):
2383 (WebCore::JSWebGLRenderingContext::getParameter):
2384 (WebCore::JSWebGLRenderingContext::getProgramParameter):
2385 (WebCore::JSWebGLRenderingContext::getShaderParameter):
2386 (WebCore::JSWebGLRenderingContext::getUniform):
2387 (WebCore::dataFunctionf):
2388 (WebCore::dataFunctioni):
2389 (WebCore::dataFunctionMatrix):
2390 * bindings/js/JSXMLHttpRequestCustom.cpp:
2391 (WebCore::JSXMLHttpRequest::open):
2392 * bindings/js/SerializedScriptValue.cpp:
2393 (WebCore::CloneBase::throwStackOverflow):
2394 (WebCore::CloneDeserializer::throwValidationError):
2395 (WebCore::SerializedScriptValue::maybeThrowExceptionIfSerializationFailed):
2396 * bindings/js/WorkerScriptController.cpp:
2397 (WebCore::WorkerScriptController::evaluate):
2398 (WebCore::WorkerScriptController::setException):
2399 * bridge/c/c_instance.cpp:
2400 (JSC::Bindings::CInstance::moveGlobalExceptionToExecState):
2401 (JSC::Bindings::CInstance::invokeMethod):
2402 (JSC::Bindings::CInstance::invokeDefaultMethod):
2403 (JSC::Bindings::CInstance::invokeConstruct):
2404 (JSC::Bindings::CInstance::toJSPrimitive):
2405 * bridge/objc/objc_instance.mm:
2406 (ObjcInstance::invokeMethod):
2407 * bridge/objc/objc_runtime.mm:
2408 (JSC::Bindings::ObjcArray::setValueAt):
2409 (JSC::Bindings::ObjcArray::valueAt):
2410 * bridge/objc/objc_utility.mm:
2411 (JSC::Bindings::throwError):
2412 * bridge/qt/qt_instance.cpp:
2413 (JSC::Bindings::QtField::valueFromInstance):
2414 (JSC::Bindings::QtField::setValueToInstance):
2415 * bridge/runtime_array.cpp:
2416 (JSC::RuntimeArray::put):
2417 (JSC::RuntimeArray::putByIndex):
2418 * bridge/runtime_object.cpp:
2419 (JSC::Bindings::RuntimeObject::throwInvalidAccessError):
2421 2013-08-28 Alexey Proskuryakov <ap@apple.com>
2423 Remove an unused data member from Page.
2425 Rubber-stamped by Brady Eidson.
2429 Removed m_cookieEnabled. This was completely dead code, long obsoleted by PageSettings.
2431 2013-08-28 Gustavo Noronha Silva <gns@gnome.org>
2433 Unreviewed build fix - copy/paste failure, copied too much.
2435 * bindings/gobject/WebKitDOMCustom.h:
2437 2013-08-28 Dean Jackson <dino@apple.com>
2439 [WebGL] CoreGraphics can provide greyscale image data
2440 https://webkit.org/b/120443
2442 Reviewed by Simon Fraser.
2444 CoreGraphics can decode greyscale or greyscale+alpha images
2445 while preserving the format. Our WebGL texture unpacker
2446 was seeing this and assuming it meant the data did not come
2447 from an <img> element. Since that method already special cased
2448 CoreGraphics, the fix was to simply return true for these
2451 I also renamed srcFormatComeFromDOMElementOrImageData
2452 to srcFormatComesFromDOMElementOrImageData.
2454 Test: fast/canvas/webgl/tex-image-with-greyscale-image.html
2456 * platform/graphics/GraphicsContext3D.cpp: Call new name.
2457 * platform/graphics/GraphicsContext3D.h:
2458 (WebCore::GraphicsContext3D::srcFormatComesFromDOMElementOrImageData):
2459 Add support for R8, AR8, A8, and RA8 data formats.
2461 2013-08-28 Gustavo Noronha Silva <gns@gnome.org>
2463 [GTK] HTMLElement lost setID and getID - need to add compatibility symbols
2464 https://bugs.webkit.org/show_bug.cgi?id=120440
2466 Reviewed by Martin Robinson.
2468 No tests, just adding compatibility symbols.
2470 setID and getID were removed, and the parent class (Element) ones should be used instead.
2471 We need to keep our ABI compatible, though, so add compatibility symbols.
2473 * bindings/gobject/WebKitDOMCustom.cpp:
2474 (webkit_dom_html_element_get_id):
2475 (webkit_dom_html_element_set_id):
2476 * bindings/gobject/WebKitDOMCustom.h:
2478 2013-08-28 Simon Fraser <simon.fraser@apple.com>
2480 Fix compositing layers in columns
2481 https://bugs.webkit.org/show_bug.cgi?id=120436
2483 Reviewed by Dave Hyatt.
2485 Remove the old hack in RenderLayer::updateLayerPosition() for placing
2486 layers in columns, which changed the layer position for composited
2487 layers; this broke hit-testing.
2489 Fix a better way by moving compositing layers to the correct
2490 positions that take column offsets into account, by fixing
2491 RenderLayer::convertToLayerCoords() to optionally apply column
2492 adjustment, and using this in the code which positions compositing layers.
2494 Tests: compositing/columns/ancestor-clipped-in-paginated.html
2495 compositing/columns/clipped-in-paginated.html
2496 compositing/columns/composited-columns-vertical-rl.html
2497 compositing/columns/composited-columns.html
2498 compositing/columns/composited-in-paginated-rl.html
2499 compositing/columns/composited-in-paginated-writing-mode-rl.html
2500 compositing/columns/composited-lr-paginated-repaint.html
2501 compositing/columns/composited-rl-paginated-repaint.html
2502 compositing/columns/hittest-composited-in-paginated.html
2503 compositing/columns/rotated-in-paginated.html
2504 compositing/columns/untransformed-composited-in-paginated.html
2506 * rendering/RenderLayer.cpp:
2507 (WebCore::RenderLayer::updateLayerPosition):
2508 (WebCore::RenderLayer::convertToPixelSnappedLayerCoords):
2509 (WebCore::accumulateOffsetTowardsAncestor):
2510 (WebCore::RenderLayer::convertToLayerCoords):
2511 * rendering/RenderLayer.h:
2512 * rendering/RenderLayerBacking.cpp:
2513 (WebCore::RenderLayerBacking::updateCompositedBounds):
2514 (WebCore::RenderLayerBacking::updateGraphicsLayerGeometry):
2516 2013-08-28 Brent Fulgham <bfulgham@webkit.org>
2518 [WinCairo] Unreviewed build fix.
2520 * WebCore.vcxproj/WebCore.vcxproj: Don't exclude the full screen
2521 window from the build.
2522 * WebCore.vcxproj/WebCore.vcxproj.filters: Ditto.
2524 2013-08-28 Benjamin Poulain <benjamin@webkit.org>
2526 Fix the build after r154780
2528 * dom/SpaceSplitString.cpp:
2529 (WebCore::SpaceSplitStringData::create):
2531 2013-08-27 Benjamin Poulain <bpoulain@apple.com>
2533 Simplify and clean SpaceSplitString
2534 https://bugs.webkit.org/show_bug.cgi?id=120385
2536 Reviewed by Ryosuke Niwa.
2538 Clean up of SpaceSplitString following the cleaning of the DOMTokenList hierarchy.
2539 This brings the following:
2540 - Fix the coding style of the header.
2541 - Remove the concepts of empty and null. The list can now be empty or have something.
2542 There is no null state.
2543 - Put the tokens directly following SpaceSplitStringData instead of using a Vector in between.
2546 * dom/ElementData.h:
2547 (WebCore::ElementData::hasClass):
2548 * dom/SpaceSplitString.cpp:
2549 (WebCore::tokenizeSpaceSplitString):
2550 (WebCore::SpaceSplitStringData::containsAll):
2551 (WebCore::SpaceSplitString::set):
2552 (WebCore::SpaceSplitString::spaceSplitStringContainsValue):
2553 (WebCore::TokenCounterProcessor::TokenCounterProcessor):
2554 (WebCore::TokenCounterProcessor::processToken):
2555 (WebCore::TokenCounterProcessor::tokenCount):
2556 (WebCore::TokenInitializerProcessor::TokenInitializerProcessor):
2557 (WebCore::TokenInitializerProcessor::processToken):
2558 (WebCore::TokenInitializerProcessor::nextMemoryBucket):
2559 (WebCore::SpaceSplitStringData::create):
2560 (WebCore::SpaceSplitStringData::destroy):
2561 * dom/SpaceSplitString.h:
2562 (WebCore::SpaceSplitStringData::contains):
2563 (WebCore::SpaceSplitStringData::size):
2564 (WebCore::SpaceSplitStringData::operator[]):
2565 (WebCore::SpaceSplitStringData::ref):
2566 (WebCore::SpaceSplitStringData::deref):
2567 (WebCore::SpaceSplitStringData::SpaceSplitStringData):
2568 (WebCore::SpaceSplitStringData::~SpaceSplitStringData):
2569 (WebCore::SpaceSplitStringData::tokenArrayStart):
2570 (WebCore::SpaceSplitString::SpaceSplitString):
2571 (WebCore::SpaceSplitString::operator!=):
2572 (WebCore::SpaceSplitString::clear):
2573 (WebCore::SpaceSplitString::contains):
2574 (WebCore::SpaceSplitString::containsAll):
2575 (WebCore::SpaceSplitString::size):
2576 (WebCore::SpaceSplitString::isEmpty):
2577 (WebCore::SpaceSplitString::operator[]):
2578 (WebCore::SpaceSplitString::spaceSplitStringContainsValue):
2579 * html/ClassList.cpp:
2580 (WebCore::ClassList::classNames):
2581 * page/EventHandler.cpp:
2582 (WebCore::findDropZone):
2585 2013-08-28 Benjamin Poulain <benjamin@webkit.org>
2587 Simplify and clean SpaceSplitString
2588 https://bugs.webkit.org/show_bug.cgi?id=120385
2590 Reviewed by Ryosuke Niwa.
2592 Clean up of SpaceSplitString following the cleaning of the DOMTokenList hierarchy.
2593 - Remove the concepts of empty and null. The list can now be empty or have something.
2594 There is no null state.
2595 - Put the tokens directly following SpaceSplitStringData instead of using a Vector in between.
2598 * dom/ElementData.h:
2599 (WebCore::ElementData::hasClass):
2600 * dom/SpaceSplitString.cpp:
2601 (WebCore::tokenizeSpaceSplitString):
2602 (WebCore::SpaceSplitStringData::containsAll):
2603 (WebCore::SpaceSplitString::set):
2604 (WebCore::SpaceSplitString::spaceSplitStringContainsValue):
2605 (WebCore::TokenCounter::TokenCounter):
2606 (WebCore::TokenCounter::processToken):
2607 (WebCore::TokenCounter::tokenCount):
2608 (WebCore::TokenAtomicStringInitializer::TokenAtomicStringInitializer):
2609 (WebCore::TokenAtomicStringInitializer::processToken):
2610 (WebCore::TokenAtomicStringInitializer::nextMemoryBucket):
2611 (WebCore::SpaceSplitStringData::create):
2612 (WebCore::SpaceSplitStringData::destroy):
2613 * dom/SpaceSplitString.h:
2614 (WebCore::SpaceSplitStringData::contains):
2615 (WebCore::SpaceSplitStringData::size):
2616 (WebCore::SpaceSplitStringData::operator[]):
2617 (WebCore::SpaceSplitStringData::ref):
2618 (WebCore::SpaceSplitStringData::deref):
2619 (WebCore::SpaceSplitStringData::SpaceSplitStringData):
2620 (WebCore::SpaceSplitStringData::~SpaceSplitStringData):
2621 (WebCore::SpaceSplitStringData::tokenArrayStart):
2622 (WebCore::SpaceSplitString::containsAll):
2623 (WebCore::SpaceSplitString::isEmpty):
2624 (WebCore::SpaceSplitString::operator[]):
2625 * html/ClassList.cpp:
2626 (WebCore::ClassList::classNames):
2627 * page/EventHandler.cpp:
2628 (WebCore::findDropZone):
2631 2013-08-28 Rob Buis <rwlbuis@webkit.org>
2633 Namespace prefix is blindly followed when serializing
2634 https://bugs.webkit.org/show_bug.cgi?id=19121
2635 Serializer doesn't handling inconsistent prefixes properly
2636 https://bugs.webkit.org/show_bug.cgi?id=117764
2637 Attribute namespaces are serialized as if they were element ones
2638 https://bugs.webkit.org/show_bug.cgi?id=22958
2640 Reviewed by Ryosuke Niwa.
2642 Add code to make sure unique prefixes and namespace declarations are generated.
2643 Unique prefix generation happens when:
2644 - the same prefix is used to map to different namespaces or
2645 - no prefix is given but the attribute is in a namespace.
2647 This is done in order to not violate constraints listed in http://www.w3.org/TR/xml-names11/. In general
2648 the pseudo code listed in http://www.w3.org/TR/DOM-Level-3-Core/namespaces-algorithms.html#normalizeDocumentAlgo
2649 is used, doing the following for attributes:
2650 if the attribute has a namespace then
2651 if the attribute has no prefix OR prefix is not declared OR conflicts with existing prefix mapping to different NS then
2652 try to find the matching in-scope declaration by looking up the prefix in the namespace -> prefix mapping, if found use that prefix
2653 else if the attribute prefix is not null AND not mapped in-scope, declare the prefix
2654 else generate a unique prefix for the namespace
2656 To keep track of in-scope namespaces a prefix to namespace mapping is used.
2658 Tests: fast/dom/XMLSerializer-attribute-namespace-prefix-conflicts.html
2659 fast/dom/XMLSerializer-same-prefix-different-namespaces-conflict.html
2660 fast/dom/XMLSerializer-setAttributeNS-namespace-no-prefix.html
2661 svg/custom/xlink-prefix-generation-in-attributes.html
2663 * editing/MarkupAccumulator.cpp:
2664 (WebCore::MarkupAccumulator::MarkupAccumulator):
2665 (WebCore::MarkupAccumulator::shouldAddNamespaceAttribute):
2666 (WebCore::MarkupAccumulator::appendNamespace):
2667 (WebCore::MarkupAccumulator::generateUniquePrefix):
2668 (WebCore::MarkupAccumulator::appendAttribute):
2669 * editing/MarkupAccumulator.h:
2671 2013-08-28 Sam White <samuel_white@apple.com>
2673 AX: Cancel button in search field not accessible.
2674 <https://webkit.org/b/120322>
2676 Expose the cancel button that shows in an input element of
2677 type search to accessibility.
2679 Reviewed by Chris Fleizach.
2681 Test: platform/mac/accessibility/search-field-cancel-button.html
2684 * English.lproj/Localizable.strings:
2685 * GNUmakefile.list.am:
2687 * WebCore.vcxproj/WebCore.vcxproj:
2688 * WebCore.xcodeproj/project.pbxproj:
2689 * accessibility/AXObjectCache.cpp:
2690 (WebCore::createFromRenderer):
2691 * accessibility/AccessibilityAllInOne.cpp:
2692 * accessibility/AccessibilitySearchFieldButtons.cpp: Added.
2693 (WebCore::AccessibilitySearchFieldCancelButton::create):
2694 (WebCore::AccessibilitySearchFieldCancelButton::AccessibilitySearchFieldCancelButton):
2695 (WebCore::AccessibilitySearchFieldCancelButton::accessibilityDescription):
2696 (WebCore::AccessibilitySearchFieldCancelButton::accessibilityText):
2697 (WebCore::AccessibilitySearchFieldCancelButton::press):
2698 (WebCore::AccessibilitySearchFieldCancelButton::computeAccessibilityIsIgnored):
2699 * accessibility/AccessibilitySearchFieldButtons.h: Added.
2701 (WebCore::Element::isSearchFieldCancelButtonElement):
2702 * html/shadow/TextControlInnerElements.h:
2703 * platform/LocalizedStrings.cpp:
2704 (WebCore::AXSearchFieldCancelButtonText):
2705 * platform/LocalizedStrings.h:
2706 * platform/efl/LocalizedStringsEfl.cpp:
2707 (WebCore::AXSearchFieldCancelButtonText):
2708 * platform/gtk/LocalizedStringsGtk.cpp:
2709 (WebCore::AXSearchFieldCancelButtonText):
2710 * platform/qt/LocalizedStringsQt.cpp:
2711 (WebCore::AXSearchFieldCancelButtonText):
2714 2013-08-28 Eric Carlson <eric.carlson@apple.com>
2716 MediaPlayerPrivateAVFoundationObjC is painting video frames under the video layer
2717 https://bugs.webkit.org/show_bug.cgi?id=120170
2719 Reviewed by Simon Fraser.
2721 No new tests, it is only possible to test in the debugger.
2723 * platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.mm:
2724 (WebCore::MediaPlayerPrivateAVFoundationObjC::hasAvailableVideoFrame): Drive by optimization.
2725 (WebCore::MediaPlayerPrivateAVFoundationObjC::paintCurrentFrameInContext): Move logic from paint here.
2726 (WebCore::MediaPlayerPrivateAVFoundationObjC::paint): Do nothing if we already have a video layer,
2727 otherwise call paint().
2729 2013-08-28 Antti Koivisto <antti@apple.com>
2731 Add child and descendant const iterators
2732 https://bugs.webkit.org/show_bug.cgi?id=120430
2734 Reviewed by Andreas Kling
2736 This patch adds const-correct DOM tree traversal iterators. It also uses them in a few places.
2738 Some const_casts have been applied where constness breaks.
2740 * dom/ChildIterator.h:
2741 (WebCore::::ChildConstIterator):
2742 (WebCore::::operator):
2744 (WebCore::::ChildConstIteratorAdapter):
2747 (WebCore::elementChildren):
2748 (WebCore::childrenOfType):
2749 * dom/DescendantIterator.h:
2750 (WebCore::::DescendantConstIterator):
2751 (WebCore::::operator):
2753 (WebCore::::DescendantConstIteratorAdapter):
2756 (WebCore::elementDescendants):
2757 (WebCore::descendantsOfType):
2759 (WebCore::Node::numberOfScopedHTMLStyleChildren):
2760 * html/HTMLFieldSetElement.cpp:
2761 (WebCore::HTMLFieldSetElement::legend):
2762 * html/HTMLFieldSetElement.h:
2763 * html/HTMLMediaElement.cpp:
2764 (WebCore::HTMLMediaElement::finishParsingChildren):
2765 * html/HTMLObjectElement.cpp:
2766 (WebCore::HTMLObjectElement::containsJavaApplet):
2767 * svg/SVGElement.cpp:
2768 (WebCore::SVGElement::title):
2769 * svg/SVGSVGElement.cpp:
2770 (WebCore::SVGSVGElement::collectIntersectionOrEnclosureList):
2771 (WebCore::SVGSVGElement::checkIntersection):
2772 (WebCore::SVGSVGElement::checkEnclosure):
2773 (WebCore::SVGSVGElement::getElementById):
2774 * svg/SVGSVGElement.h:
2776 2013-08-28 Lukasz Gajowy <l.gajowy@samsung.com>
2778 AX:Null pointer may be dereferenced.
2779 https://bugs.webkit.org/show_bug.cgi?id=120300
2781 Reviewed by Chris Fleizach.
2783 Added a check if newObj is not null and an assert in order to avoid dereferecing null pointer.
2785 * accessibility/AXObjectCache.cpp:
2786 (WebCore::AXObjectCache::getOrCreate):
2788 2013-08-28 Pratik Solanki <psolanki@apple.com>
2790 Document::elementSheet() should return a reference
2791 https://bugs.webkit.org/show_bug.cgi?id=120433
2793 Reviewed by Andreas Kling.
2795 Since elementSheet() always retruns a valid pointer, we can simply return a reference
2796 instead. Also rename m_elemSheet to m_elementSheet.
2798 * css/CSSParser.cpp:
2799 (WebCore::CSSParser::parseInlineStyleDeclaration):
2800 * css/PropertySetCSSStyleDeclaration.cpp:
2801 (WebCore::InlineCSSStyleDeclaration::parentStyleSheet):
2803 (WebCore::Document::~Document):
2804 (WebCore::Document::recalcStyle):
2805 (WebCore::Document::updateBaseURL):
2806 (WebCore::Document::elementSheet):
2808 * dom/StyledElement.cpp:
2809 (WebCore::StyledElement::setInlineStyleFromString):
2810 (WebCore::StyledElement::setInlineStyleProperty):
2811 (WebCore::StyledElement::addSubresourceAttributeURLs):
2812 (WebCore::StyledElement::addPropertyToPresentationAttributeStyle):
2813 * inspector/InspectorStyleSheet.cpp:
2814 (WebCore::InspectorStyleSheetForInlineStyle::getStyleAttributeRanges):
2816 2013-08-28 Ryosuke Niwa <rniwa@webkit.org>
2818 REGRESSION(r154586): Past names map should only be used when named item is empty
2819 https://bugs.webkit.org/show_bug.cgi?id=120432
2821 Reviewed by Anders Carlsson.
2823 Don't add the element from the past names map if we've found elements of the given name.
2825 Test: fast/forms/past-names-map-should-be-used-only-when-named-item-is-empty.html
2827 * html/HTMLFormElement.cpp:
2828 (WebCore::HTMLFormElement::getNamedElements):
2830 2013-08-27 Ryosuke Niwa <rniwa@webkit.org>
2832 Don't keep unassociated elements in the past names map
2833 https://bugs.webkit.org/show_bug.cgi?id=120328
2835 Reviewed by Darin Adler.
2837 Remove elements from the past names map of a form element when they are disassociated with the form to match
2838 the behaviors of Firefox 24 and Internet Explorer 10. The specification feedback has been submitted to WHATWG
2839 in http://lists.whatwg.org/pipermail/whatwg-whatwg.org/2013-August/040586.html
2841 Also fix a memory leak via the past names map when the elements in the map becomes an ancestor of the form
2842 element by storing a raw pointer in the map. This is safe because the form associated elements are kept alive
2843 by another mechanism.
2845 Because ~FormAssociatedElement removes entries from the past names map, we could no longer store HTMLElement*
2846 in HTMLFormElement::m_pastNamesMap as that requires casting FormAssociatedElement* to HTMLElement*, which is
2847 not possible in ~FormAssociatedElement. We instead store pointers to FormNamedItem, new base class of
2848 FormAssociatedElement and HTMLImageElement.
2850 Test: fast/forms/past-names-map-should-not-contained-disassociated-elements.html
2854 * WebCore.vcxproj/WebCore.vcxproj:
2855 * WebCore.vcxproj/WebCore.vcxproj.filters:
2856 * WebCore.xcodeproj/project.pbxproj:
2857 * html/FormAssociatedElement.cpp:
2858 * html/FormAssociatedElement.h:
2859 (WebCore::toHTMLElement):
2861 * html/FormNamedItem.h: Added.
2862 (WebCore::FormNamedItem::~FormNamedItem):
2864 * html/HTMLElement.h:
2865 (WebCore::HTMLElement::asFormNamedItem): Added. This allows the conversion from a HTMLFormControlElement,
2866 HTMLObjectElement, HTMLImageElement to FormNamedItem in getNamedElements to update the past names map.
2868 * html/HTMLFormControlElement.h:
2869 * html/HTMLFormElement.cpp:
2870 (WebCore::HTMLFormElement::removeFormElement):
2871 (WebCore::HTMLFormElement::removeImgElement):
2872 (WebCore::HTMLFormElement::assertItemCanBeInPastNamesMap): Asserts that FormNamedItem added to or obtained
2873 from the past names map is either a form associated element or an image element; the condition guarantees
2874 that the item will be removed from the map before its element gets destructed.
2876 (WebCore::HTMLFormElement::elementFromPastNamesMap):
2877 (WebCore::HTMLFormElement::addToPastNamesMap):
2878 (WebCore::HTMLFormElement::removeFromPastNamesMap): Finds and removes the obsolete item from the map in O(n).
2879 Note that removeFromVector, which is called on m_associatedElements or m_imageElements before this function is called,
2882 (WebCore::HTMLFormElement::getNamedElements):
2884 * html/HTMLFormElement.h:
2885 * html/HTMLImageElement.h:
2886 * html/HTMLObjectElement.h:
2888 2013-08-28 Brendan Long <b.long@cablelabs.com>
2890 Duplicate in-band tracks when switching <source> elements
2891 https://bugs.webkit.org/show_bug.cgi?id=120369
2893 Reviewed by Eric Carlson.
2895 Test: media/track/track-in-band-duplicate-tracks-when-source-changes.html
2897 * html/HTMLMediaElement.cpp:
2898 (WebCore::HTMLMediaElement::createMediaPlayer):
2899 Delete existing in-band tracks before creating a new media player.
2901 2013-08-28 Bem Jones-Bey <bjonesbe@adobe.com>
2903 Code cleanup: rename FloatIntervalSearchAdapter and remove unnecessary inlines
2904 https://bugs.webkit.org/show_bug.cgi?id=120378
2906 Reviewed by Darin Adler.
2908 Rename FloatIntervalSearchAdapter to ComputeFloatOffsetAdapter. The
2909 naming of this adapter has caused much confusion in reading the code,
2910 as it wasn't apparent that calls to it were actually doing anything
2911 other than searching the interval tree. The new name is a much better
2912 description of what it actually does.
2914 Also, rename m_lowValue and m_highValue member variables to make it
2915 easier to read the code that uses them.
2917 Removed the inlines based on a change by eseidel in Blink.
2919 No new tests, no behavior change.
2921 * rendering/RenderBlock.cpp:
2922 (WebCore::::updateOffsetIfNeeded): Update for renames.
2923 (WebCore::::collectIfNeeded): Ditto.
2924 (WebCore::::getHeightRemaining): Ditto.
2925 (WebCore::RenderBlock::logicalLeftFloatOffsetForLine): Ditto.
2926 (WebCore::RenderBlock::logicalRightFloatOffsetForLine): Ditto.
2927 * rendering/RenderBlock.h:
2928 (WebCore::RenderBlock::FloatingObject::x): Remove unnecessary inline.
2929 (WebCore::RenderBlock::FloatingObject::maxX): Ditto.
2930 (WebCore::RenderBlock::FloatingObject::y): Ditto.
2931 (WebCore::RenderBlock::FloatingObject::maxY): Ditto.
2932 (WebCore::RenderBlock::FloatingObject::width): Ditto.
2933 (WebCore::RenderBlock::FloatingObject::height): Ditto.
2934 (WebCore::RenderBlock::ComputeFloatOffsetAdapter::ComputeFloatOffsetAdapter): Rename.
2935 (WebCore::RenderBlock::ComputeFloatOffsetAdapter::lowValue): Rename m_lowValue.
2936 (WebCore::RenderBlock::ComputeFloatOffsetAdapter::highValue): Rename m_highValue.
2938 2013-08-28 Tamas Czene <tczene@inf.u-szeged.hu>
2940 Resolve unused parameter warning in ScriptedAnimationController.cpp.
2941 https://bugs.webkit.org/show_bug.cgi?id=120408
2943 Reviewed by Darin Adler.
2945 * dom/ScriptedAnimationController.cpp:
2946 (WebCore::ScriptedAnimationController::setThrottled):
2948 2013-08-28 Sergio Villar Senin <svillar@igalia.com>
2950 [CSS Grid Layout] Handle 'span' positions during layout
2951 https://bugs.webkit.org/show_bug.cgi?id=119756
2953 Reviewed by Andreas Kling.
2955 From Blink r149133 by <jchaffraix@chromium.org>
2957 Properly handle the 'span' keyword during layout. We only had
2958 parsing support so far but with this change we are able to
2959 recognize these positions and act accordingly.
2961 * rendering/RenderGrid.cpp:
2962 (WebCore::RenderGrid::resolveGridPositionsFromStyle):
2963 (WebCore::RenderGrid::resolveGridPositionAgainstOppositePosition):
2964 * rendering/RenderGrid.h:
2965 * rendering/style/GridPosition.h:
2966 (WebCore::GridPosition::shouldBeResolvedAgainstOppositePosition):
2968 2013-08-28 Antti Koivisto <antti@apple.com>
2970 Factor descendant iterator assertions into a class.
2971 https://bugs.webkit.org/show_bug.cgi?id=120422
2973 Reviewed by Darin Adler.
2975 Share the assertions between ChildIterator and DescendantIterator. We can use it for future const iterators too.
2977 * GNUmakefile.list.am:
2979 * WebCore.vcxproj/WebCore.vcxproj:
2980 * WebCore.xcodeproj/project.pbxproj:
2981 * dom/ChildIterator.h:
2982 (WebCore::::ChildIterator):
2983 (WebCore::::operator):
2985 * dom/DescendantIterator.h:
2986 (WebCore::::DescendantIterator):
2987 (WebCore::::operator):
2989 * dom/DescendantIteratorAssertions.h: Added.
2990 (WebCore::DescendantIteratorAssertions::DescendantIteratorAssertions):
2991 (WebCore::DescendantIteratorAssertions::domTreeHasMutated):
2992 (WebCore::DescendantIteratorAssertions::dropEventDispatchAssertion):
2994 2013-08-28 Darin Adler <darin@apple.com>
2996 Eliminate Pasteboard::generalPasteboard
2997 https://bugs.webkit.org/show_bug.cgi?id=120392
2999 Reviewed by Anders Carlsson.
3001 * WebCore.exp.in: Removed the generalPasteboard function.
3002 It didn't need to be exported, because no one was using it.
3004 * editing/Editor.cpp:
3005 (WebCore::Editor::paste): Added an overload that takes a Pasteboard.
3006 (WebCore::Editor::copyURL): Ditto.
3007 * editing/Editor.h: Added overloads.
3009 * editing/EditorCommand.cpp:
3010 (WebCore::executePasteGlobalSelection): Put this function inside the same
3011 platform #if that the global selection code in the Pasteboard class was in.
3012 Changed to use Pasteboard::createForGlobalSelection instead of using the
3013 Pasteboard::setSelectionMode approach.
3014 (WebCore::createCommandMap): Put PasteGlobalSelection inside the platform #if.
3016 * inspector/InjectedScriptHost.cpp:
3017 (WebCore::InjectedScriptHost::copyText): Use Pasteboard::createForCopyAndPaste()
3018 instead of Pasteboard::generalPasteboard().
3019 * inspector/InspectorFrontendHost.cpp:
3020 (WebCore::InspectorFrontendHost::copyText): Ditto.
3022 * platform/Pasteboard.h: Removed generalPasteboard. Replaced isSelectionMode
3023 and setSelectionMode with createForGlobalSelection.
3025 * platform/blackberry/PasteboardBlackBerry.cpp: Deleted generalPasteboard.
3026 * platform/efl/PasteboardEfl.cpp: Deleted generalPasteboard.
3028 * platform/gtk/PasteboardGtk.cpp: Deleted selectionClipboard, primaryClipboard,
3029 generalPasteboard, isSelectionMode, and setSelectionMode.
3030 (WebCore::Pasteboard::createForGlobalSelection): Added.
3031 * platform/gtk/PasteboardHelper.cpp: Deleted m_usePrimarySelectionClipboard,
3032 getCurrentClipboard, and getClipboard.
3033 * platform/gtk/PasteboardHelper.h: Deleted the above, plus
3034 setUsePrimarySelectionClipboard and usePrimarySelectionClipboard.
3036 * platform/ios/PasteboardIOS.mm: Deleted generalPasteboard.
3037 * platform/mac/PasteboardMac.mm: Deleted generalPasteboard.
3039 * platform/qt/PasteboardQt.cpp: Deleted generalPasteboard, isSelectionMode,
3040 and setSelectionMode.
3041 (WebCore::Pasteboard::createForGlobalSelection): Added.
3043 * platform/win/PasteboardWin.cpp: Deleted generalPasteboard.
3045 2013-08-28 Antti Koivisto <antti@apple.com>
3047 Share attach loops between Elements and ShadowRoots
3048 https://bugs.webkit.org/show_bug.cgi?id=120414
3050 Reviewed Andreas Kling.
3052 * style/StyleResolveTree.cpp:
3053 (WebCore::Style::attachChildren):
3054 (WebCore::Style::attachShadowRoot):
3055 (WebCore::Style::detachChildren):
3056 (WebCore::Style::detachShadowRoot):
3058 2013-08-28 Anders Carlsson <andersca@apple.com>
3060 Clean up XPathExpressionNode and XPath::Function
3061 https://bugs.webkit.org/show_bug.cgi?id=120411
3063 Reviewed by Antti Koivisto.
3065 Rename the subexpression and function argument getters to be more descriptive,
3066 remove the non-const overloads (they were never used) and change the getters to return
3067 references since they can never be null.
3069 * xml/XPathExpressionNode.cpp:
3070 * xml/XPathExpressionNode.h:
3071 (WebCore::XPath::ParseNode::~ParseNode):
3072 (WebCore::XPath::Expression::addSubExpression):
3073 (WebCore::XPath::Expression::isContextNodeSensitive):
3074 (WebCore::XPath::Expression::setIsContextNodeSensitive):
3075 (WebCore::XPath::Expression::isContextPositionSensitive):
3076 (WebCore::XPath::Expression::setIsContextPositionSensitive):
3077 (WebCore::XPath::Expression::isContextSizeSensitive):
3078 (WebCore::XPath::Expression::setIsContextSizeSensitive):
3079 (WebCore::XPath::Expression::subExpressionCount):
3080 (WebCore::XPath::Expression::subExpression):
3081 * xml/XPathFunctions.cpp:
3082 (WebCore::XPath::Function::setArguments):
3083 (WebCore::XPath::FunId::evaluate):
3084 (WebCore::XPath::FunLocalName::evaluate):
3085 (WebCore::XPath::FunNamespaceURI::evaluate):
3086 (WebCore::XPath::FunName::evaluate):
3087 (WebCore::XPath::FunCount::evaluate):
3088 (WebCore::XPath::FunString::evaluate):
3089 (WebCore::XPath::FunConcat::evaluate):
3090 (WebCore::XPath::FunStartsWith::evaluate):
3091 (WebCore::XPath::FunContains::evaluate):
3092 (WebCore::XPath::FunSubstringBefore::evaluate):
3093 (WebCore::XPath::FunSubstringAfter::evaluate):
3094 (WebCore::XPath::FunSubstring::evaluate):
3095 (WebCore::XPath::FunStringLength::evaluate):
3096 (WebCore::XPath::FunNormalizeSpace::evaluate):
3097 (WebCore::XPath::FunTranslate::evaluate):
3098 (WebCore::XPath::FunBoolean::evaluate):
3099 (WebCore::XPath::FunNot::evaluate):
3100 (WebCore::XPath::FunLang::evaluate):
3101 (WebCore::XPath::FunNumber::evaluate):
3102 (WebCore::XPath::FunSum::evaluate):
3103 (WebCore::XPath::FunFloor::evaluate):
3104 (WebCore::XPath::FunCeiling::evaluate):
3105 (WebCore::XPath::FunRound::evaluate):
3106 * xml/XPathFunctions.h:
3107 (WebCore::XPath::Function::setName):
3108 (WebCore::XPath::Function::argumentCount):
3109 (WebCore::XPath::Function::argument):
3110 (WebCore::XPath::Function::name):
3111 * xml/XPathPredicate.cpp:
3112 (WebCore::XPath::Negative::evaluate):
3113 (WebCore::XPath::NumericOp::evaluate):
3114 (WebCore::XPath::EqTestOp::evaluate):
3115 (WebCore::XPath::LogicalOp::evaluate):
3116 (WebCore::XPath::Union::evaluate):
3118 2013-08-28 Andreas Kling <akling@apple.com>
3120 Page::pluginData() should return a reference.
3121 <https://webkit.org/b/120386>
3123 Reviewed by Darin Adler.
3125 The PluginData is lazily constructed by pluginData(); it never returns null.
3126 A small number of null checks were harmed in the making of this patch.
3128 2013-08-28 Andrei Parvu <parvu@adobe.com>
3130 <https://webkit.org/b/120002> [CSS Masking] Add -webkit-mask-source-type shorthand property
3132 Added the -webkit-mask-source-type property to the -webkit-mask shorthand property.
3134 Reviewed by Dirk Schulze.
3136 Test cases added in LayoutTests/fast/masking/parsing-mask.html
3138 * css/CSSParser.cpp: Added the CSSPropertyWebkitMaskSourceType property to the array of shorthand properties.
3139 (WebCore::CSSParser::parseValue):
3140 * css/StylePropertyShorthand.cpp: Added the CSSPropertyWebkitMaskSourceType to the list of shorthands.
3141 (WebCore::webkitMaskShorthand):
3142 (WebCore::matchingShorthandsForLonghand):
3144 2013-08-28 Zan Dobersek <zdobersek@igalia.com>
3146 REGRESSION(r154708): It broke all plugin tests on GTK and Qt WK1
3147 https://bugs.webkit.org/show_bug.cgi?id=120398
3149 Reviewed by Anders Carlsson.
3151 * bridge/c/c_class.cpp:
3152 (JSC::Bindings::CClass::methodNamed): Keep the pointer of the new CMethod object
3153 to return it after it's adopted by the new HashMap entry.
3154 (JSC::Bindings::CClass::fieldNamed): The pointer to the newly created CField object
3155 should be returned in this branch, matching the behavior before r154708.
3157 2013-08-28 Antti Koivisto <antti@apple.com>
3159 Fix Qt no-libxml2 build.
3163 * xml/parser/XMLDocumentParserQt.cpp:
3164 (WebCore::XMLDocumentParser::parseCdata):
3166 2013-08-28 Antti Koivisto <antti@apple.com>
3168 Don't use NodeRenderingContext when attaching text renderers
3169 https://bugs.webkit.org/show_bug.cgi?id=120402
3171 Reviewed by Andreas Kling.
3173 This patch moves various functions for creating text renderers from NodeRenderingContext and Text to StyleResolveTree.
3174 It also tightens the logic and combines some functions.
3176 * dom/CharacterData.cpp:
3177 (WebCore::CharacterData::parserAppendData):
3178 (WebCore::CharacterData::setDataAndUpdate):
3179 * dom/ContainerNode.cpp:
3180 (WebCore::attachChild):
3181 (WebCore::detachChild):
3182 * dom/NodeRenderingContext.cpp:
3183 * dom/NodeRenderingContext.h:
3185 (WebCore::Text::~Text):
3187 * html/HTMLViewSourceDocument.cpp:
3188 (WebCore::HTMLViewSourceDocument::addText):
3189 * html/parser/HTMLConstructionSite.cpp:
3190 (WebCore::executeTask):
3191 * html/shadow/InsertionPoint.cpp:
3192 (WebCore::InsertionPoint::willAttachRenderers):
3193 (WebCore::InsertionPoint::willDetachRenderers):
3194 * style/StyleResolveTree.cpp:
3195 (WebCore::Style::isRendererReparented):
3196 (WebCore::Style::previousSiblingRenderer):
3197 (WebCore::Style::nextSiblingRenderer):
3199 From NodeRenderingContext::next/previousRenderer
3201 (WebCore::Style::createTextRenderersForSiblingsAfterAttachIfNeeded):
3203 From Text::createTextRenderersForSiblingsAfterAttachIfNeeded()
3205 (WebCore::Style::textRendererIsNeeded):
3207 From Text::textRendererIsNeeded
3209 (WebCore::Style::createTextRendererIfNeeded):
3211 Combines code from Text::createTextRendererIfNeeded, NodeRenderingContext::createRendererForTextIfNeeded,
3212 NodeRenderingContext constructor and text node relevant code NodeRenderingContext::shouldCreateRenderer.
3214 (WebCore::Style::attachTextRenderer):
3215 (WebCore::Style::detachTextRenderer):
3217 New functions of attaching text renderers. From Text::attach/detachText()
3219 (WebCore::Style::updateTextRendererAfterContentChange):
3221 From Text::updateTextRenderer.
3223 (WebCore::Style::attachShadowRoot):
3224 (WebCore::Style::attachChildren):
3225 (WebCore::Style::attachRenderTree):
3226 (WebCore::Style::detachShadowRoot):
3227 (WebCore::Style::detachChildren):
3228 (WebCore::Style::updateTextStyle):
3229 * style/StyleResolveTree.h:
3230 * xml/parser/XMLDocumentParser.cpp:
3231 (WebCore::XMLDocumentParser::exitText):
3232 * xml/parser/XMLDocumentParserLibxml2.cpp:
3233 (WebCore::XMLDocumentParser::cdataBlock):
3235 2013-08-28 Antti Koivisto <antti@apple.com>
3237 Make descendant iterators always require ContainerNode root
3238 https://bugs.webkit.org/show_bug.cgi?id=120393
3240 Reviewed by Andreas Kling.
3242 Remove Node* root versions of the iterators.
3243 Fix the few call sites that required them to have tighter typing.
3245 * accessibility/AccessibilityNodeObject.cpp:
3246 (WebCore::AccessibilityNodeObject::canvasHasFallbackContent):
3247 (WebCore::siblingWithAriaRole):
3248 * dom/ChildIterator.h:
3249 (WebCore::::ChildIteratorAdapter):
3252 (WebCore::elementChildren):
3253 (WebCore::childrenOfType):
3254 * dom/DescendantIterator.h:
3255 (WebCore::::DescendantIterator):
3256 (WebCore::::DescendantIteratorAdapter):
3259 (WebCore::elementDescendants):
3260 (WebCore::descendantsOfType):
3261 * editing/ApplyStyleCommand.cpp:
3262 (WebCore::dummySpanAncestorForNode):
3263 (WebCore::ApplyStyleCommand::cleanupUnstyledAppleStyleSpans):
3264 (WebCore::ApplyStyleCommand::applyInlineStyle):
3265 * editing/ApplyStyleCommand.h:
3267 2013-08-28 Sergio Villar Senin <svillar@igalia.com>
3269 WorkerGlobalScopeWebDatabase requires ENABLE(WORKERS)
3270 https://bugs.webkit.org/show_bug.cgi?id=120395
3272 Reviewed by Christophe Dumez.
3274 WorkerGlobalScopeDatabase uses the WorkerGlobalScope object which is
3275 defined only when WORKERS are enabled. We should guard that code.
3277 * Modules/webdatabase/WorkerGlobalScopeWebDatabase.cpp:
3278 * Modules/webdatabase/WorkerGlobalScopeWebDatabase.h:
3279 * Modules/webdatabase/WorkerGlobalScopeWebDatabase.idl:
3281 2013-08-28 Sergio Villar Senin <svillar@igalia.com>
3283 [CSS Grid Layout] Fix grid position resolution
3284 https://bugs.webkit.org/show_bug.cgi?id=119801
3286 Reviewed by Andreas Kling.
3288 From Blink r148833, r148878, r150403 by <jchaffraix@chromium.org>
3290 Both grid-{column|row}-end and negative positions were not
3291 properly handled in our grid position resolution code. We were
3292 using the same code to resolve all the grid positions without
3293 considering the edges of the grid.
3295 Also refactored the grid size estimation in
3296 resolveGridPositionsFromStyle() so we can use it for the grid size
3297 estimation. The code no longer requires the grid to be filled at
3298 that moment as the specs changed to use the "explicit grid" which
3299 is independent of grid items (only depends on style).
3301 Test: fast/css-grid-layout/grid-item-negative-position-resolution.html
3303 * rendering/RenderGrid.cpp:
3304 (WebCore::RenderGrid::maximumIndexInDirection):
3305 (WebCore::RenderGrid::resolveGridPositionsFromStyle):
3306 (WebCore::adjustGridPositionForSide):
3307 (WebCore::RenderGrid::resolveGridPositionFromStyle):
3308 * rendering/RenderGrid.h:
3310 2013-08-28 Sergio Villar Senin <svillar@igalia.com>
3312 [CSS Grid Layout] infinity should be defined as a negative value
3313 https://bugs.webkit.org/show_bug.cgi?id=107053
3315 Reviewed by Andreas Kling.
3317 From Blink r154805 by <jchaffraix@chromium.org>
3319 Reject negative values for track-breadth at parse time as
3320 mentioned in the latest versions of the spec.
3322 Added some extra checks to the existing tests.
3324 * css/CSSParser.cpp:
3325 (WebCore::CSSParser::parseGridBreadth):
3326 * rendering/RenderGrid.cpp:
3327 (WebCore::RenderGrid::computeUsedBreadthOfMaxLength):
3329 2013-08-28 Sergio Villar Senin <svillar@igalia.com>
3331 [Soup] WebTiming information not shown in the inspector
3332 https://bugs.webkit.org/show_bug.cgi?id=118395
3334 Reviewed by Martin Robinson.
3336 WebTiming information was not correctly provided to WebCore
3337 because the gotHeadersCallback was incorrectly resetting the
3338 original ResourceResponse (which had the ResourceLoadTiming
3339 object) instead of simply updating their contents using the
3342 No new test required as this feature is already covered by the
3343 existing webtiming tests. In any case this change includes a fix
3344 for the http/tests/misc/webtiming-ssl.php test which was not
3345 failing even if it should because it was not correct.
3347 * platform/network/soup/ResourceHandleSoup.cpp:
3348 (WebCore::gotHeadersCallback):
3349 (WebCore::restartedCallback): restartedCallback should be
3350 available only if WEB_TIMING is defined.
3351 (WebCore::createSoupMessageForHandleAndRequest): Ditto.
3353 2013-08-26 Antonio Gomes <a1.gomes@sisa.samsung.com>
3355 Scrolling allowed when overflow:hidden (seen on Acid2)
3356 https://bugs.webkit.org/show_bug.cgi?id=22769
3358 Reviewed by Darin Adler.
3360 Autoscroll, as well as other user-driven scroll actions,
3361 has to respect the scrollability styled into the web page.
3362 More specifically, if a html or body tags are styled with
3363 overflow:hidden, autoscroll should not scroll the containing document.
3365 In order to fix this, patch hardens RenderBox::canAutoscroll as
3366 following: previously, ::canAutoscroll was relying solemnly in
3367 ::canBeScrolledAndHasScrollableArea to determine the scrollability
3368 of #document node, which was unconditionally returned as 'true'.
3369 Patch extends ::canAutoscroll to handle the #document case for
3370 main and inner frames, and now it asks through ::isScrollable if
3371 the corresponding document's FrameView is actually user-scrollable.
3373 Note, that the patch change ::canAutoscroll to cover the non-mainFrame
3374 #document case, so the comment specific to Apple's Mail app can be omited
3377 Test: fast/events/autoscroll-in-overflow-hidden-html.html
3379 * rendering/RenderBox.cpp:
3380 (WebCore::RenderBox::canAutoscroll):
3382 2013-08-27 Santosh Mahto <santosh.ma@samsung.com>
3384 Adding "explicit" keyword in forms related classes constructor
3385 https://bugs.webkit.org/show_bug.cgi?id=120366
3387 Reviewed by Darin Adler.
3389 Adding "explicit" keyword in constructors.
3391 * html/BaseButtonInputType.h:
3392 (WebCore::BaseButtonInputType::BaseButtonInputType):
3393 * html/BaseCheckableInputType.h:
3394 (WebCore::BaseCheckableInputType::BaseCheckableInputType):
3395 * html/BaseChooserOnlyDateAndTimeInputType.h:
3396 (WebCore::BaseChooserOnlyDateAndTimeInputType::BaseChooserOnlyDateAndTimeInputType):
3397 * html/BaseClickableWithKeyInputType.h:
3398 (WebCore::BaseClickableWithKeyInputType::BaseClickableWithKeyInputType):
3399 * html/BaseTextInputType.h:
3400 (WebCore::BaseTextInputType::BaseTextInputType):
3401 * html/ColorInputType.h:
3402 (WebCore::ColorInputType::ColorInputType):
3403 * html/DateInputType.h:
3404 * html/DateTimeInputType.h:
3405 (WebCore::DateTimeInputType::DateTimeInputType):
3406 * html/DateTimeLocalInputType.h:
3407 (WebCore::DateTimeLocalInputType::DateTimeLocalInputType):
3408 * html/EmailInputType.h:
3409 (WebCore::EmailInputType::EmailInputType):
3410 * html/FileInputType.h:
3411 * html/HiddenInputType.h:
3412 (WebCore::HiddenInputType::HiddenInputType):
3414 * html/ImageInputType.h:
3416 (WebCore::InputType::InputType):
3417 * html/MediaController.h:
3418 * html/MonthInputType.h:
3419 (WebCore::MonthInputType::MonthInputType):
3420 * html/RadioInputType.h:
3421 (WebCore::RadioInputType::RadioInputType):
3422 * html/RangeInputType.h:
3423 * html/ResetInputType.h:
3424 (WebCore::ResetInputType::ResetInputType):
3425 * html/SearchInputType.h:
3426 * html/SubmitInputType.h:
3427 (WebCore::SubmitInputType::SubmitInputType):
3428 * html/TelephoneInputType.h:
3429 (WebCore::TelephoneInputType::TelephoneInputType):
3430 * html/TextFieldInputType.h:
3431 * html/TextInputType.h:
3432 (WebCore::TextInputType::TextInputType):
3433 * html/TimeInputType.h:
3434 * html/URLInputType.h:
3435 (WebCore::URLInputType::URLInputType):
3436 * html/ValidationMessage.h:
3437 * html/WeekInputType.h:
3438 (WebCore::WeekInputType::WeekInputType):
git://git.webkit.org / WebKit-https.git / blob