1 2015-06-24 Commit Queue <commit-queue@webkit.org>
3 Unreviewed, rolling out r185906.
4 https://bugs.webkit.org/show_bug.cgi?id=146276
6 MSVC doesn't yet provide a const-qualified std::array<T,
7 N>::size(), failing to compile the static_assert (Requested by
12 "Improve the source code generated by make_names.pl"
13 https://bugs.webkit.org/show_bug.cgi?id=146208
14 http://trac.webkit.org/changeset/185906
16 2015-06-24 Zan Dobersek <zdobersek@igalia.com>
18 Improve the source code generated by make_names.pl
19 https://bugs.webkit.org/show_bug.cgi?id=146208
21 Reviewed by Darin Adler.
23 Clean up and optimize the output that's generated by the make_names.pl script
24 when generating large sets of DOM names for attributes and tags.
26 The GenerateStrings() function in the StaticString.pm module is split into
27 GenerateStringData() and GenerateASCIILiteral() so that the two new functions
28 can be used independently, with the original function still being called when
29 generating font names.
31 Tags and attributes have the corresponding static QualifiedName globals defined
32 as before. After that, two static const std::array<> objects are defined for
33 both types -- the first is an ordered array of addresses of the QualifiedName
34 objects (corresponding to the C array that was defined in getHTMLTags(),
35 getSVGAttrs() etc.), and the second is an ordered array of StringImpl::StaticASCIILiteral
36 objects that replaces separately defined StringImpl::StaticASCIILiteral objects
37 and the additional tables that contained pairs of QualifiedName object addresses
38 and the corresponding StaticASCIILiteral object references in the init() function.
39 This is all generated by the printStaticData() function in make_names.pl.
41 The printQualifiedNameCreation() function generates a static_assert() that ensures
42 that the corresponding std::array<QualifiedName*> and std::array<StaticASCIILiteral>
43 objects have the same amount of items, and then sets up a loop that walks through
44 the two arrays and properly constructs the QualifiedName objects from the static
47 On the GTK port, this shaves off ~54kB from the final stripped shared library
48 on a 64-bit build, and ~21kB on a 32-bit build.
50 * bindings/scripts/StaticString.pm:
52 (GenerateASCIILiteral):
58 (printQualifiedNameCreation):
60 (printDefinitions): Deleted.
62 2015-06-24 Youenn Fablet <youenn.fablet@crf.canon.fr>
64 Refactor UserMediaRequest to share more codes between MediaDevices.getUserMedia and legacy webkitGetUserMedia
65 https://bugs.webkit.org/show_bug.cgi?id=146237
67 Reviewed by Darin Adler.
69 Covered by existing tests, no change in behavior.
71 * Modules/mediastream/MediaConstraintsImpl.cpp:
72 (WebCore::MediaConstraintsImpl::create): Simplified error handling by removing exception code parameter.
73 * Modules/mediastream/MediaConstraintsImpl.h: Ditto.
74 * Modules/mediastream/MediaDevices.cpp:
75 (WebCore::MediaDevices::getUserMedia): Removed code now in UserMediaRequest::start.
76 * Modules/mediastream/NavigatorUserMedia.cpp:
77 (WebCore::NavigatorUserMedia::webkitGetUserMedia): Ditto.
78 * Modules/mediastream/UserMediaRequest.cpp:
79 (WebCore::parseOptions): Simplified error handling by removing exception code parameter.
80 (WebCore::UserMediaRequest::start): Renamed create in start and added common code.
81 * Modules/mediastream/UserMediaRequest.h:
83 2015-06-23 Andreas Kling <akling@apple.com>
85 Should reduce tile coverage for the first paint after a tab switch.
86 <https://webkit.org/b/146252>
87 <rdar://problem/19821583>
89 Reviewed by Darin Adler.
91 Reduce the number of tiles we need to paint after switching tabs,
92 to shorten the time it takes before we can flush pixels to screen.
94 We accomplish this by piggybacking on the "speculative tiling enabled"
95 mode of FrameView, which was previously only used to throttle painting
96 and layer flushes during page load.
98 When a FrameView becomes visible, which is what happens when you
99 switch to its tab, we revert the speculative tiling optimization to
100 its initial state, and reset the "scrolled by user" flag.
102 In practice this means that after switching tabs, we only generate
103 enough tiles to fill the viewport. Then, after 500ms has passed
104 or the user scrolls the page, we go back to the usual speculative
107 * page/FrameView.cpp:
108 (WebCore::FrameView::show):
110 2015-06-23 Anders Carlsson <andersca@apple.com>
112 Remove windowResizerRect code, nobody is using it anymore
113 https://bugs.webkit.org/show_bug.cgi?id=146265
115 Reviewed by Beth Dakin.
117 * loader/EmptyClients.h:
119 (WebCore::Chrome::windowResizerRect): Deleted.
121 * page/ChromeClient.h:
122 * page/FrameView.cpp:
123 (WebCore::FrameView::windowResizerRect): Deleted.
125 * platform/ScrollView.cpp:
126 (WebCore::ScrollView::ScrollView): Deleted.
127 (WebCore::ScrollView::windowResizerRectChanged): Deleted.
128 (WebCore::ScrollView::containsScrollbarsAvoidingResizer): Deleted.
129 (WebCore::ScrollView::adjustScrollbarsAvoidingResizerCount): Deleted.
130 (WebCore::ScrollView::setParent): Deleted.
131 * platform/ScrollView.h:
132 (WebCore::ScrollView::windowResizerRect): Deleted.
133 * platform/Scrollbar.cpp:
134 (WebCore::Scrollbar::Scrollbar): Deleted.
135 (WebCore::Scrollbar::setFrameRect): Deleted.
136 (WebCore::Scrollbar::setParent): Deleted.
137 * platform/Scrollbar.h:
139 2015-06-23 Dean Jackson <dino@apple.com>
141 Media controls are missing the white backdrop in UIWebViews
142 https://bugs.webkit.org/show_bug.cgi?id=146251
143 <rdar://problem/20181345>
145 Reviewed by Simon Fraser.
147 Implement two new CALayer subclasses that explicitly set
148 something that resembles the system appearance for
149 media-controls-dark-bar-background and media-controls-light-bar-background.
150 This way, WebKit1 clients get a visible result.
152 Creating the actual system recipes is tracked by:
153 https://bugs.webkit.org/show_bug.cgi?id=146250
155 * WebCore.xcodeproj/project.pbxproj: Add new files.
156 * platform/graphics/ca/cocoa/PlatformCALayerCocoa.mm:
157 (PlatformCALayerCocoa::PlatformCALayerCocoa): Use the new CALayer subclasses for
158 the appropriate layer types.
159 * platform/graphics/ca/cocoa/WebSystemBackdropLayer.h: Added.
160 * platform/graphics/ca/cocoa/WebSystemBackdropLayer.mm: Added.
161 (-[WebLightSystemBackdropLayer init]):
162 (-[WebLightSystemBackdropLayer setBackgroundColor:]): Only set to a light grey.
163 (-[WebDarkSystemBackdropLayer init]):
164 (-[WebDarkSystemBackdropLayer setBackgroundColor:]): Only set to a dark grey.
166 2015-06-18 Matt Rajca <mrajca@apple.com>
168 Support releasing media sessions
169 https://bugs.webkit.org/show_bug.cgi?id=146132
171 Reviewed by Darin Adler.
173 * Modules/mediasession/MediaSession.cpp: Implemented as described in the Media Session spec.
174 (WebCore::MediaSession::releaseSession):
175 (WebCore::MediaSession::releaseInternal):
176 * Modules/mediasession/MediaSession.h:
178 2015-06-23 Chris Fleizach <cfleizach@apple.com>
180 AX: iOS: VoiceOver and ARIA: has popup property not announced
181 https://bugs.webkit.org/show_bug.cgi?id=146188
183 Reviewed by Darin Adler.
185 Expose the existing "has popup" property to the iOS Accessibility API.
187 * accessibility/ios/WebAccessibilityObjectWrapperIOS.mm:
188 (-[WebAccessibilityObjectWrapper accessibilityHasPopup]):
190 2015-06-23 Javier Fernandez <jfernandez@igalia.com>
192 [CSS Grid Layout] Performance optimization: avoid computing overflow alignment if not needed
193 https://bugs.webkit.org/show_bug.cgi?id=146231
195 Reviewed by Sergio Villar Senin.
197 We don't need to apply any overflow handling if alignment value don't have a potential
198 risk of data loss, as it's the case of 'start' value.
200 This patch avoid computing the overflow in all the cases, since it adds an unneeded
201 overhead which affects performance.
203 New code improves performance around 3%-8%, depending on the grid tests.
205 No new tests, no new funcitonality.
207 * rendering/RenderGrid.cpp:
208 (WebCore::RenderGrid::rowPositionForChild):
209 (WebCore::RenderGrid::columnPositionForChild):
211 2015-06-23 Youenn Fablet <youenn.fablet@crf.canon.fr>
213 MediaDevices.getUserMedia should migrate from callbacks to DOMPromise
214 https://bugs.webkit.org/show_bug.cgi?id=146200
216 Reviewed by Darin Adler.
218 Introduced DOMPromiseWithCallback to resolve/reject promises while allowing promise clients to use a typed callback approach.
219 Migrated MediaDevices.getUserMedia from callbacks to DOMPromiseWithCallback.
220 Removed MediaDevices.getUserMedia custom binding.
222 Covered by existing tests.
224 * CMakeLists.txt: Removing custom binding.
225 * Modules/mediastream/MediaDevices.cpp:
226 (WebCore::MediaDevices::getUserMedia): Moving from callback to promise.
227 * Modules/mediastream/MediaDevices.h: Ditto.
228 * Modules/mediastream/MediaDevices.idl: Removing custom binding.
229 * Modules/mediastream/NavigatorUserMedia.cpp:
230 (WebCore::NavigatorUserMedia::webkitGetUserMedia): Converting from promise callback to API callbacks.
231 * Modules/mediastream/UserMediaRequest.cpp:
232 (WebCore::UserMediaRequest::create): Moving from callback to promise.
233 (WebCore::UserMediaRequest::UserMediaRequest): Ditto.
234 (WebCore::UserMediaRequest::didCreateStream): Ditto.
235 (WebCore::UserMediaRequest::failedToCreateStreamWithConstraintsError): Ditto.
236 (WebCore::UserMediaRequest::failedToCreateStreamWithPermissionError): Ditto.
237 * Modules/mediastream/UserMediaRequest.h: Ditto.
238 * bindings/js/JSDOMPromise.h: Introducing DOMPromiseWithCallback and removing crypto specific header.
239 (WebCore::DOMPromiseWithCallback::DOMPromiseWithCallback):
240 (WebCore::Error>::resolve):
241 (WebCore::Error>::reject):
242 * bindings/js/JSMediaDevicesCustom.cpp: Removed.
243 * bindings/js/JSSubtleCryptoCustom.cpp: Updating headers.
245 2015-06-23 Xabier Rodriguez Calvar <calvaris@igalia.com> and Youenn Fablet <youenn.fablet@crf.canon.fr>
247 [Streams API] Implement ReadableStream js source "'cancel" callback
248 https://bugs.webkit.org/show_bug.cgi?id=146204
250 Reviewed by Darin Adler.
252 Calling "cancel" JS function when web app is cancelling a JS readable stream.
253 Handling of promise returned value in case of async cancel.
255 Covered by rebased tests.
257 * bindings/js/ReadableJSStream.cpp:
258 (WebCore::ReadableJSStream::invoke): Refactoring to pass cancel reason or controller to the JS function.
259 (WebCore::ReadableJSStream::doStart): Ditto.
260 (WebCore::startReadableStreamAsync): Renaming readableStream as protectedStream.
261 (WebCore::createPullResultFulfilledFunction): Ditto.
262 (WebCore::ReadableJSStream::doPull): Refactoring to pass cancel reason or controller to the JS function.
263 (WebCore::createCancelResultFulfilledFunction): Cancel promise fullfil callback.
264 (WebCore::createCancelResultRejectedFunction): Cancel promise reject callback.
265 (WebCore::ReadableJSStream::doCancel): Calling cancel JS callback and handling promise returned value.
266 * bindings/js/ReadableJSStream.h: Refactoring to pass cancel reason or controller to the JS function.
268 2015-06-22 Ryuan Choi <ryuan.choi@navercorp.com>
270 [EFL] Hyphenation is not supported
271 https://bugs.webkit.org/show_bug.cgi?id=89830
273 Reviewed by Gyuyoung Kim.
275 Share libHyphen backend of GTK port.
277 Rebased fast/text/hyphenate-*.html
281 * platform/efl/FileSystemEfl.cpp:
282 (WebCore::listDirectory): Deleted because of lack of functionality.
283 eina_file_ls returns full directory path so fnmatch fails to check dict file.
284 This patch reuse Posix implementation instead of EFL port specific function.
285 * platform/posix/FileSystemPOSIX.cpp: Ditto.
286 * platform/text/gtk/HyphenationLibHyphen.cpp: Moved to platform/text/hyphen
287 * platform/text/hyphen/HyphenationLibHyphen.cpp:
288 Renamed from Source/WebCore/platform/text/gtk/HyphenationLibHyphen.cpp.
289 (WebCore::scanTestDictionariesDirectoryIfNecessary):
290 Added PLATFORM guard and EFL implementation for the test directory
292 2015-06-22 Chris Dumez <cdumez@apple.com>
294 [WK1] WebAllowDenyPolicyListener.denyOnlyThisRequest() should not start a new permission request
295 https://bugs.webkit.org/show_bug.cgi?id=146228
296 <rdar://problem/15179262>
298 Reviewed by Daniel Bates.
300 Add Geolocation::resetIsAllowed() API that merely resets
301 m_allowGeolocation to Unknown, so that we will request the permission
302 again the next time a position is requested.
304 * Modules/geolocation/Geolocation.h:
305 (WebCore::Geolocation::resetIsAllowed):
307 2015-06-22 Simon Fraser <simon.fraser@apple.com>
309 ASSERT(!m_zOrderListsDirty) when mousing over web view with incremental rendering suppressed
310 https://bugs.webkit.org/show_bug.cgi?id=146225
312 Reviewed by Zalan Bujtas.
314 Update RenderLayer's z-order lists when hit testing. There's no guarantee that they've
315 been updated; this happens to work most of the time because painting updates them,
316 but if incremental rendering is suppressed, we may not have painted yet.
318 Easy to hit on webkit.org in MiniBrowser, but I wasn't able to make a reduced testcase.
320 * rendering/RenderLayer.cpp:
321 (WebCore::RenderLayer::hitTest):
322 (WebCore::RenderLayer::updateLayerListsIfNeeded): Flip the order of the tests, since checking
323 dirty bits is cheaper than calling isStackingContext().
325 2015-06-22 Dean Jackson <dino@apple.com>
327 Rename PlatformCA*Mac to PlatformCA*Cocoa
328 https://bugs.webkit.org/show_bug.cgi?id=146224
329 <rdar://problem/21497182>
331 Reviewed by Simon Fraser.
333 Rename PlatformCALayerMac and related files in platform/graphics/ca/mac
334 to the more accurate Cocoa suffix.
336 * WebCore.xcodeproj/project.pbxproj:
337 * page/mac/ServicesOverlayController.mm:
338 * platform/graphics/ca/GraphicsLayerCA.cpp:
339 * platform/graphics/ca/PlatformCAAnimation.h:
340 * platform/graphics/ca/PlatformCALayer.h:
341 * platform/graphics/ca/cocoa/LayerFlushSchedulerMac.cpp: Renamed from Source/WebCore/platform/graphics/ca/mac/LayerFlushSchedulerMac.cpp.
342 * platform/graphics/ca/cocoa/PlatformCAAnimationCocoa.h: Renamed from Source/WebCore/platform/graphics/ca/mac/PlatformCAAnimationMac.h.
343 * platform/graphics/ca/cocoa/PlatformCAAnimationCocoa.mm: Renamed from Source/WebCore/platform/graphics/ca/mac/PlatformCAAnimationMac.mm.
344 * platform/graphics/ca/cocoa/PlatformCAFiltersCocoa.mm: Renamed from Source/WebCore/platform/graphics/ca/mac/PlatformCAFiltersMac.mm.
345 * platform/graphics/ca/cocoa/PlatformCALayerCocoa.h: Renamed from Source/WebCore/platform/graphics/ca/mac/PlatformCALayerMac.h.
346 * platform/graphics/ca/cocoa/PlatformCALayerCocoa.mm: Renamed from Source/WebCore/platform/graphics/ca/mac/PlatformCALayerMac.mm.
347 * platform/graphics/ca/cocoa/WebTiledBackingLayer.h: Renamed from Source/WebCore/platform/graphics/ca/mac/WebTiledBackingLayer.h.
348 * platform/graphics/ca/cocoa/WebTiledBackingLayer.mm: Renamed from Source/WebCore/platform/graphics/ca/mac/WebTiledBackingLayer.mm.
349 * platform/graphics/ca/mac/LayerFlushSchedulerMac.cpp:
350 * platform/graphics/ca/mac/PlatformCAAnimationMac.mm:
351 * platform/graphics/ca/mac/PlatformCAFiltersMac.mm:
352 * platform/graphics/ca/mac/PlatformCALayerMac.mm:
353 * platform/graphics/ca/mac/WebTiledBackingLayer.mm:
355 2015-06-22 Alex Christensen <achristensen@webkit.org>
357 [cssjit] Disable compiling scrollbar pseudoclass selectors
358 https://bugs.webkit.org/show_bug.cgi?id=146220
360 Reviewed by Benjamin Poulain.
362 * cssjit/SelectorCompiler.cpp:
363 (WebCore::SelectorCompiler::addScrollbarPseudoClassType):
364 Don't compile selectors with scrollbar pseudoclasses.
366 2015-06-22 Simon Fraser <simon.fraser@apple.com>
368 -webkit-clip-path clips incorrectly if the element bounds go beyond the top edge of the page
369 https://bugs.webkit.org/show_bug.cgi?id=146218
370 rdar://problem/21127840
372 Reviewed by Zalan Bujtas.
374 The clip path is computed using the RenderLayer's bounding box, so needs to be offset
375 by the offsetFromRenderer when set on the mask layer.
377 Test: compositing/masks/compositing-clip-path-origin.html
379 * rendering/RenderLayerBacking.cpp:
380 (WebCore::RenderLayerBacking::updateMaskingLayerGeometry):
382 2015-06-22 Michael Catanzaro <mcatanzaro@igalia.com>
384 Web sockets should be treated as active mixed content
385 https://bugs.webkit.org/show_bug.cgi?id=140624
387 Reviewed by Sam Weinig.
389 Tests: http/tests/security/mixedContent/websocket/insecure-websocket-in-iframe.html
390 http/tests/security/mixedContent/websocket/insecure-websocket-in-main-frame.html
392 * Modules/websockets/WebSocket.cpp:
393 (WebCore::WebSocket::connect): Block ws:// WebSocket connections from https:// pages, and
394 emit the onerror event after doing so.
395 * platform/SchemeRegistry.cpp:
396 (WebCore::secureSchemes): Add wss:// to the list of secure schemes.
398 2015-06-22 Dean Jackson <dino@apple.com>
400 Element with blur backdrop-filter shows edge duplication and dark edges
401 https://bugs.webkit.org/show_bug.cgi?id=146215
402 <rdar://problem/20367695>
404 Reviewed by Tim Horton.
406 The input images to backdrop filters should duplicate their edge pixels
407 outwards, rather than using transparent pixels. This is a flag we
408 set on the Gaussian blur, but means we have to remember if the
409 FilterOperations list came from a regular filter or a backdrop filter.
411 Test: css3/filters/backdrop/blur-input-bounds.html
413 * css/CSSPropertyNames.in: New custom convertor for backdrop-filter.
414 * css/StyleBuilderConverter.h:
415 (WebCore::StyleBuilderConverter::convertBackdropFilterOperations): New convertor
416 that sets the backdrop flag, but is otherwise the same as a normal filter
418 * page/animation/CSSPropertyAnimation.cpp:
419 (WebCore::blendFilterOperations): Inherit the backdrop flag if either of our
421 * platform/graphics/ca/mac/PlatformCAFiltersMac.mm: Set the inputNormalizeEdges
422 key on the CAFilter if necessary.
423 * platform/graphics/filters/FilterOperations.cpp: Add a new flag indicating if
424 these operations are intended for backdrops.
425 (WebCore::FilterOperations::operator=):
426 (WebCore::FilterOperations::operator==):
427 * platform/graphics/filters/FilterOperations.h:
428 (WebCore::FilterOperations::isUsedForBackdropFilters):
429 (WebCore::FilterOperations::setUsedForBackdropFilters):
431 2015-06-22 Tim Horton <timothy_horton@apple.com>
433 [TextIndicator] Text shifts one pixel to the left when I force click to bring up Lookup in Mail messages
434 https://bugs.webkit.org/show_bug.cgi?id=146214
435 <rdar://problem/20782970>
437 Reviewed by Dean Jackson.
439 * page/mac/TextIndicatorWindow.mm:
440 (-[WebTextIndicatorView initWithFrame:textIndicator:margin:offset:]):
441 (WebCore::TextIndicatorWindow::setTextIndicator):
442 (-[WebTextIndicatorView initWithFrame:textIndicator:margin:]): Deleted.
443 When the WebView is at a nonintegral position, we can end up needing a TextIndicator
444 with a nonintegral position. We need to round the window position, so we need to apply
445 the fractional part to the indicator layers inside, not to the window.
447 2015-06-22 Myles C. Maxfield <mmaxfield@apple.com>
449 [iOS] Arabic text styled with Georgia is rendered as boxes
450 https://bugs.webkit.org/show_bug.cgi?id=145681
451 <rdar://problem/21169844>
453 Reviewed by Darin Adler.
455 Georgia doesn't support Arabic, so we ask CoreText what font does support Arabic. It returns
456 TimesNewRomanPSMT. However, WebKit explicitly disallows this font on iOS. Therefore, instead
457 of using TimesNewRomanPSMT, we will simply just use GeezaPro.
459 Test: fast/text/arabic-times-new-roman.html
461 * platform/graphics/ios/FontCacheIOS.mm:
462 (WebCore::FontCache::systemFallbackForCharacters):
463 * platform/graphics/Font.h: Let FontCacheIOS call fontFamilyShouldNotBeUsedForArabic()
464 * platform/graphics/cocoa/FontCocoa.mm:
465 (WebCore::fontFamilyShouldNotBeUsedForArabic):
467 2015-06-22 Alex Christensen <achristensen@webkit.org>
469 Unreviewed non-mac debug build fix after r185840.
471 * loader/ResourceLoader.cpp:
472 (WebCore::ResourceLoader::willSendRequest):
473 Added enable flag around assertion.
475 2015-06-20 Alex Christensen <achristensen@webkit.org>
477 [Content Extensions] Add SPI to reload without content blocking.
478 https://bugs.webkit.org/show_bug.cgi?id=146128
479 rdar://problem/20351903
481 Reviewed by Sam Weinig.
483 * html/HTMLMediaElement.cpp:
484 (WebCore::HTMLMediaElement::loadResource):
485 * loader/ResourceLoader.cpp:
486 (WebCore::ResourceLoader::willSendRequest):
487 * loader/cache/CachedResourceLoader.cpp:
488 (WebCore::CachedResourceLoader::requestResource):
489 * page/DOMWindow.cpp:
490 (WebCore::DOMWindow::open):
492 (WebCore::Page::userContentController):
493 (WebCore::Page::userContentExtensionsEnabled):
494 (WebCore::Page::setUserContentExtensionsEnabled):
495 (WebCore::Page::group):
496 * page/UserContentController.cpp:
497 (WebCore::UserContentController::removeAllUserContentExtensions):
498 (WebCore::UserContentController::processContentExtensionRulesForLoad):
499 (WebCore::UserContentController::actionsForResourceLoad):
500 * page/UserContentController.h:
502 2015-06-22 Zalan Bujtas <zalan@apple.com>
504 REGRESSION(r169105) Dangling renderer pointer in SelectionSubtreeRoot::SelectionSubtreeData.
505 https://bugs.webkit.org/show_bug.cgi?id=146116
506 rdar://problem/20959369
508 Reviewed by Brent Fulgham.
510 This patch ensures that we don't adjust the selection unless the visual selection still matches this subtree root.
512 When multiple selection roots are present we need to ensure that a RenderObject
513 only shows up in one of them.
514 RenderView::splitSelectionBetweenSubtrees(), as the name implies, splits the
515 selection and sets the selection range (start/end) on each selection root.
516 However, SelectionSubtreeRoot::adjustForVisibleSelection() later recomputes the range
517 based on visible selection and that could end up collecting renderers as selection start/end
518 from another selection subtree.
519 RenderObject's holds the last selection state (RenderObject::setSelectionState).
520 If we set a renderer first as "on selection border" and later "inside" using multiple selection roots,
521 we can't clean up selections properly when this object gets destroyed.
522 One of the roots ends up with a dangling RenderObject pointer.
524 Test: fast/regions/crash-when-renderer-is-in-multiple-selection-subtrees.html
526 * rendering/SelectionSubtreeRoot.cpp:
527 (WebCore::SelectionSubtreeRoot::adjustForVisibleSelection):
529 2015-06-22 Jeremy Jones <jeremyj@apple.com>
531 Do not exit fullscreen when starting PiP since this is done automatically.
532 https://bugs.webkit.org/show_bug.cgi?id=144871
534 Reviewed by Darin Adler.
536 Since we don't explicitly exit fullscreen, update state in shouldExitFullscreenWithReason()
538 * platform/ios/WebVideoFullscreenInterfaceAVKit.h: Declare shouldExitFullscreenWithReason().
539 * platform/ios/WebVideoFullscreenInterfaceAVKit.mm:
540 (-[WebAVPlayerController playerViewController:shouldExitFullScreenWithReason:]): Forward to WebVideoFullscreenInterfaceAVKit.
541 (WebVideoFullscreenInterfaceAVKit::shouldExitFullscreenWithReason): Added.
542 (WebVideoFullscreenInterfaceAVKit::willStartPictureInPicture): Remove enter fullscreen code.
543 * platform/spi/cocoa/AVKitSPI.h: Add missing enums.
545 2015-06-22 Daniel Bates <dabates@apple.com>
547 AX: UI Automation cannot find AutoFill or search cancel buttons
548 https://bugs.webkit.org/show_bug.cgi?id=145241
549 <rdar://problem/21051411>
551 Reviewed by Chris Fleizach.
553 Add support for hit testing the search cancel button and AutoFill button so that
554 they can be accessed by UI Automation.
556 Currently the accessibility hit test machinery ignores nodes in a shadow tree.
557 So, it neither finds the <input type="search"> cancel button nor the AutoFill button
558 when it performs a hit test. Therefore these buttons cannot be accessed using
561 Tests: accessibility/hit-test-input-auto-fill-button.html
562 accessibility/hit-test-input-search-cancel-button.html
563 accessibility/input-search-cancel-button.html
565 * accessibility/AccessibilityRenderObject.cpp:
566 (WebCore::AccessibilityRenderObject::accessibilityTextFieldDecorationHitTest): Added; returns
567 the AccessibilityObject for the search cancel button or AutoFill text field decoration as applicable.
568 (WebCore::AccessibilityRenderObject::accessibilityHitTest): Check whether the hit node
569 is a text field decoration.
571 2015-06-22 Xabier Rodriguez Calvar <calvaris@igalia.com> and Youenn Fablet <youenn.fablet@crf.canon.fr>
573 [Streams API] Implement ReadableStream cancel (abstract part)
574 https://bugs.webkit.org/show_bug.cgi?id=146111
576 Reviewed by Darin Adler.
578 This patch implements ReadableStream and ReadableStreamReader cancel.
579 The reader delegates cancellation to its stream.
581 This patch also ensures that controller.close() will not throw in case cancellation is on-going.
583 A follow-up patch will implement the calling of 'cancel' JS callback for JS sources.
585 Covered by rebased tests.
587 * Modules/streams/ReadableStream.cpp:
588 (WebCore::ReadableStream::cancel): Checks whether locked or not before cancelling.
589 (WebCore::ReadableStream::cancelNoCheck): Cancel without lock check.
590 (WebCore::ReadableStream::notifyCancelSucceeded): Async cancel callback.
591 (WebCore::ReadableStream::notifyCancelFailed): Ditto.
592 * Modules/streams/ReadableStream.h:
593 * Modules/streams/ReadableStream.idl: Cleaned up IDL.
594 * Modules/streams/ReadableStreamReader.cpp:
595 (WebCore::ReadableStreamReader::cancel):
596 * Modules/streams/ReadableStreamReader.h:
597 * Modules/streams/ReadableStreamReader.idl: Cleaned up IDL
598 * bindings/js/JSReadableStreamControllerCustom.cpp:
599 (WebCore::JSReadableStreamController::close):
600 * bindings/js/JSReadableStreamCustom.cpp:
601 (WebCore::JSReadableStream::cancel):
602 * bindings/js/JSReadableStreamReaderCustom.cpp:
603 (WebCore::JSReadableStreamReader::cancel):
604 * bindings/js/ReadableJSStream.cpp:
605 (WebCore::ReadableJSStream::doCancel):
606 * bindings/js/ReadableJSStream.h:
608 2015-06-22 Adam Bergkvist <adam.bergkvist@ericsson.com>
610 WebRTC: Navigator.webkitGetUserMedia() requires three arguments
611 https://bugs.webkit.org/show_bug.cgi?id=146022
613 Reviewed by Eric Carlson.
615 Updated custom binding to make the third error callback argument
616 mandatory. Updated and unskipped three tests (for the GTK+ port).
618 * bindings/js/JSNavigatorCustom.cpp:
619 (WebCore::JSNavigator::webkitGetUserMedia):
621 2015-06-22 Carlos Garcia Campos <cgarcia@igalia.com>
623 Unreviewed. Fix GTK+ build after r185818.
625 Actually rollout r185320.
627 * platform/network/soup/DNSSoup.cpp:
628 (WebCore::DNSResolveQueue::platformProxyIsEnabledInSystemPreferences):
629 (WebCore::DNSResolveQueue::platformResolve):
630 (WebCore::gotProxySettingsCallback): Deleted.
631 (WebCore::DNSResolveQueue::platformMaybeResolveHost): Deleted.
633 2015-06-16 Gavin Barraclough <barraclough@apple.com>
635 Page load performance regression due to bugs.webkit.org/show_bug.cgi?id=145542
636 https://bugs.webkit.org/show_bug.cgi?id=146198
640 * platform/network/DNSResolveQueue.cpp:
641 (WebCore::DNSResolveQueue::DNSResolveQueue):
642 (WebCore::DNSResolveQueue::isUsingProxy):
643 (WebCore::DNSResolveQueue::add):
644 (WebCore::DNSResolveQueue::timerFired):
645 * platform/network/DNSResolveQueue.h:
646 * platform/network/cf/DNSCFNet.cpp:
647 (WebCore::DNSResolveQueue::platformProxyIsEnabledInSystemPreferences):
648 (WebCore::clientCallback):
649 (WebCore::DNSResolveQueue::platformResolve):
650 (WebCore::proxyIsEnabledInSystemPreferences): Deleted.
651 (WebCore::isUsingProxy): Deleted.
652 (WebCore::DNSResolveQueue::platformMaybeResolveHost): Deleted.
653 * platform/network/soup/DNSSoup.cpp:
655 2015-06-21 Alexey Proskuryakov <ap@apple.com>
657 REGRESSION (r172975): navigator.language unable to tell region for Traditional Chinese users
658 https://bugs.webkit.org/show_bug.cgi?id=146121
659 rdar://problem/21395180
661 Reviewed by Darin Adler.
663 Revert to previous behavior, which is wrong in many ways, but not as wrong as the new one.
665 * platform/mac/Language.mm:
666 (WebCore::httpStyleLanguageCode):
667 (WebCore::platformUserPreferredLanguages):
668 * platform/spi/cf/CFBundleSPI.h:
670 2015-06-19 Andy Estes <aestes@apple.com>
672 Give Node::didNotifySubtreeInsertions() a better name
673 https://bugs.webkit.org/show_bug.cgi?id=146170
675 Reviewed by Darin Adler.
677 didNotifySubtreeInsertions() is not a good name. It sounds like we are notifying the subtree insertions, which doesn't make sense.
679 This function is really about notifying the Node that the subtree it's a part of has finished being inserted into the DOM
680 (i.e. all nodes have received their call to insertedInto()). Change the name to finishedInsertingSubtree() to better reflect this.
682 * dom/ContainerNode.cpp:
683 (WebCore::ContainerNode::notifyChildInserted):
684 * dom/ContainerNodeAlgorithms.h:
685 (WebCore::ChildNodeInsertionNotifier::notifyNodeInsertedIntoDocument):
686 (WebCore::ChildNodeInsertionNotifier::notifyNodeInsertedIntoTree):
688 (WebCore::Element::addShadowRoot):
690 (WebCore::Node::finishedInsertingSubtree): Renamed from didNotifySubtreeInsertions.
691 (WebCore::Node::didNotifySubtreeInsertions): Renamed to finishedInsertingSubtree.
692 * dom/ScriptElement.cpp:
693 (WebCore::ScriptElement::finishedInsertingSubtree): Renamed from didNotifySubtreeInsertions.
694 (WebCore::ScriptElement::didNotifySubtreeInsertions): Renamed to finishedInsertingSubtree.
695 * dom/ScriptElement.h:
696 * html/HTMLFrameElementBase.cpp:
697 (WebCore::HTMLFrameElementBase::insertedInto):
698 (WebCore::HTMLFrameElementBase::finishedInsertingSubtree): Renamed from didNotifySubtreeInsertions.
699 (WebCore::HTMLFrameElementBase::didNotifySubtreeInsertions): Renamed to finishedInsertingSubtree.
700 * html/HTMLFrameElementBase.h:
701 * html/HTMLScriptElement.cpp:
702 (WebCore::HTMLScriptElement::insertedInto):
703 (WebCore::HTMLScriptElement::finishedInsertingSubtree): Renamed from didNotifySubtreeInsertions.
704 (WebCore::HTMLScriptElement::didNotifySubtreeInsertions): Renamed to finishedInsertingSubtree.
705 * html/HTMLScriptElement.h:
706 * svg/SVGFEImageElement.cpp:
707 (WebCore::SVGFEImageElement::insertedInto):
708 (WebCore::SVGFEImageElement::finishedInsertingSubtree): Renamed from didNotifySubtreeInsertions.
709 (WebCore::SVGFEImageElement::didNotifySubtreeInsertions): Renamed to finishedInsertingSubtree.
710 * svg/SVGFEImageElement.h:
711 * svg/SVGMPathElement.cpp:
712 (WebCore::SVGMPathElement::insertedInto):
713 (WebCore::SVGMPathElement::finishedInsertingSubtree): Renamed from didNotifySubtreeInsertions.
714 (WebCore::SVGMPathElement::didNotifySubtreeInsertions): Renamed to finishedInsertingSubtree.
715 * svg/SVGMPathElement.h:
716 * svg/SVGScriptElement.cpp:
717 (WebCore::SVGScriptElement::insertedInto):
718 (WebCore::SVGScriptElement::finishedInsertingSubtree): Renamed from didNotifySubtreeInsertions.
719 (WebCore::SVGScriptElement::didNotifySubtreeInsertions): Renamed to finishedInsertingSubtree.
720 * svg/SVGScriptElement.h:
721 * svg/SVGTRefElement.cpp:
722 (WebCore::SVGTRefElement::insertedInto):
723 (WebCore::SVGTRefElement::finishedInsertingSubtree): Renamed from didNotifySubtreeInsertions.
724 (WebCore::SVGTRefElement::didNotifySubtreeInsertions): Renamed to finishedInsertingSubtree.
725 * svg/SVGTRefElement.h:
726 * svg/SVGTextPathElement.cpp:
727 (WebCore::SVGTextPathElement::insertedInto):
728 (WebCore::SVGTextPathElement::finishedInsertingSubtree): Renamed from didNotifySubtreeInsertions.
729 (WebCore::SVGTextPathElement::didNotifySubtreeInsertions): Renamed to finishedInsertingSubtree.
730 * svg/SVGTextPathElement.h:
731 * svg/animation/SVGSMILElement.cpp:
732 (WebCore::SVGSMILElement::insertedInto):
733 (WebCore::SVGSMILElement::finishedInsertingSubtree): Renamed from didNotifySubtreeInsertions.
734 (WebCore::SVGSMILElement::didNotifySubtreeInsertions): Renamed to finishedInsertingSubtree.
735 * svg/animation/SVGSMILElement.h:
737 2015-06-21 Philip Chimento <philip.chimento@gmail.com>
739 libwebkit2gtk fails to link without opengl
740 https://bugs.webkit.org/show_bug.cgi?id=138332
742 Reviewed by Carlos Garcia Campos.
744 * CMakeLists.txt: The third-party ANGLE directories need to be
745 included even if ENABLE(GRAPHICS_CONTEXT_3D) is false. They must
746 be included after the OpenGL headers as the comment says.
747 * platform/graphics/texmap/BitmapTexturePool.cpp: Remove
748 reference to no longer existent header file.
750 2015-06-20 Tim Horton <timothy_horton@apple.com>
752 Deselection of text causes a noticeable jump on force touch machines
753 https://bugs.webkit.org/show_bug.cgi?id=146173
754 <rdar://problem/20992842>
756 Reviewed by Sam Weinig.
758 When we have a TextIndicator of type Crossfade, we end up putting
759 a layer with the blue highlight + text painted into it on top of the
760 content, and cross-fading that layer to the yellow-highlighted text.
762 This is necessary for BounceAndCrossfade TextIndicators, because the
763 blue highlight has to bounce, but is not necessary for Crossfade-only
764 ones; we can just fade in the yellow highlight on top of the
765 existing blue page highlight, and all is well.
767 So, get rid of the Crossfade TextIndicator type and use FadeIn, separately
768 keeping track of whether or not we can add a margin (we still can't
769 add a margin to TextIndicators that indicate the page's current selection,
770 because the blue highlight cannot have the margin applied to it, and we
771 want the bounds to match exactly).
773 * page/TextIndicator.cpp:
774 (WebCore::TextIndicator::createWithRange):
775 If the range is the same as the selection, turn off the margin.
776 We were previously doing this based on the presentation transition, but now
777 there's no difference in presentation transition in this case.
779 (WebCore::TextIndicator::createWithSelectionInFrame):
780 (WebCore::TextIndicator::wantsBounce):
781 (WebCore::TextIndicator::wantsContentCrossfade):
782 (WebCore::TextIndicator::wantsFadeIn):
783 (WebCore::TextIndicator::wantsManualAnimation):
784 * page/TextIndicator.h:
785 Get rid of TextIndicatorPresentationTransition::Crossfade.
787 (WebCore::TextIndicator::setWantsMargin):
788 (WebCore::TextIndicator::wantsMargin):
789 Keep track of whether we want a margin.
791 * page/mac/TextIndicatorWindow.mm:
792 (-[WebTextIndicatorView initWithFrame:textIndicator:margin:]):
793 Determine if we should use a margin based on wantsMargin instead of the
794 presentation transition.
796 2015-06-20 Ryuan Choi <ryuan.choi@navercorp.com>
798 [EFL] Do not consider test directories when DEVELOPER_MODE is OFF
799 https://bugs.webkit.org/show_bug.cgi?id=146171
801 Reviewed by Gyuyoung Kim.
803 * platform/efl/EflInspectorUtilities.cpp:
804 (WebCore::inspectorResourcePath): Only used WEB_INSPECTOR_DIR which CMake decides properly.
806 2015-06-19 Michael Catanzaro <mcatanzaro@igalia.com>
808 Fix absolute value warning in LocalizedStringsGtk.cpp
809 https://bugs.webkit.org/show_bug.cgi?id=145919
811 Reviewed by Martin Robinson.
813 Use abs(static_cast<int>(time)) rather than static_cast<int>(abs(time)) to avoid clang's
814 warnings about passing a float to abs() instead of std::abs(). Also, because casting an int
817 * platform/gtk/LocalizedStringsGtk.cpp:
818 (WebCore::localizedMediaTimeDescription):
820 2015-06-19 Devin Rousso <drousso@apple.com>
822 Web Inspector: Highlight currently edited CSS selector
823 https://bugs.webkit.org/show_bug.cgi?id=145658
825 Reviewed by Joseph Pecoraro.
827 Test: inspector/dom/highlight-multiple-shapes.html
829 * inspector/InspectorController.cpp:
830 (WebCore::InspectorController::buildObjectForHighlightedNode):
831 * inspector/InspectorController.h:
832 * inspector/InspectorDOMAgent.cpp:
833 (WebCore::InspectorDOMAgent::highlightSelector): Gets a list of all nodes matching a given selector string and highlights each of them.
834 * inspector/InspectorDOMAgent.h:
835 * inspector/InspectorOverlay.cpp:
836 (WebCore::InspectorOverlay::hideHighlight):
837 (WebCore::InspectorOverlay::highlightNodeList): Loops through a given NodeList to create highlightObjects for each of them.
838 (WebCore::InspectorOverlay::shouldShowOverlay):
839 (WebCore::buildObjectForElementData): Don't show flow fragments when highlighting multiple nodes.
840 (WebCore::InspectorOverlay::buildHighlightObjectForNode):
841 (WebCore::InspectorOverlay::buildObjectForHighlightedNode): Now returns an array containing the highlightObject for each highligthed node.
842 (WebCore::InspectorOverlay::drawNodeHighlight): Now sends an array to the InspectorOverlayPage.js to provide support for highlighting multiple nodes.
843 * inspector/InspectorOverlay.h:
844 * inspector/InspectorOverlayPage.js: Now expects an array as its parameter and loops through it to highlight each node given.
845 If the parameter array contains more than one element, do not draw the textbox containing info on that node.
848 2015-06-19 Joseph Pecoraro <pecoraro@apple.com>
850 Crash under WebCore::PageConsoleClient::addMessage attempting to log insecure content message in ImageDocument
851 https://bugs.webkit.org/show_bug.cgi?id=146096
853 Reviewed by Timothy Hatcher.
855 Was able to reproduce this using a user stylesheet with an http css font
856 on a pdf (ImageDocument) main document loaded over https. Was unable to
857 create a reliable test for this scenario.
859 * page/PageConsoleClient.cpp:
860 (WebCore::getParserLocationForConsoleMessage):
861 The scriptableDocumentParser could be null, such as in an ImageDocument.
863 2015-06-19 Dean Jackson <dino@apple.com>
865 Extremely large canvas crashes on pre-El Capitan machines
866 https://bugs.webkit.org/show_bug.cgi?id=146169
867 <rdar://problem/21410046>
869 Reviewed by Tim Horton.
871 On machines before El Capitan, make the maximum canvas size
874 Covered by existing tests.
876 * html/HTMLCanvasElement.cpp:
878 2015-06-19 Tim Horton <timothy_horton@apple.com>
880 Selection services cannot be invoked when force click is enabled
881 https://bugs.webkit.org/show_bug.cgi?id=146166
882 <rdar://problem/21468362>
884 Reviewed by Darin Adler.
886 * page/mac/ServicesOverlayController.h:
887 Turn Highlight::Type into something we can use for dirty flags.
889 * page/mac/ServicesOverlayController.mm:
890 (WebCore::ServicesOverlayController::Highlight::createForSelection):
891 (WebCore::ServicesOverlayController::Highlight::createForTelephoneNumber):
892 (WebCore::ServicesOverlayController::ServicesOverlayController):
893 (WebCore::ServicesOverlayController::selectionRectsDidChange):
894 (WebCore::ServicesOverlayController::selectedTelephoneNumberRangesChanged):
895 (WebCore::ServicesOverlayController::invalidateHighlightsOfType):
896 (WebCore::ServicesOverlayController::buildPotentialHighlightsIfNeeded):
897 (WebCore::ServicesOverlayController::remainingTimeUntilHighlightShouldBeShown):
898 (WebCore::ServicesOverlayController::buildPhoneNumberHighlights):
899 (WebCore::ServicesOverlayController::buildSelectionHighlight):
900 (WebCore::ServicesOverlayController::findTelephoneNumberHighlightContainingSelectionHighlight):
901 (WebCore::ServicesOverlayController::determineActiveHighlight):
902 (WebCore::ServicesOverlayController::didScrollFrame):
903 (WebCore::ServicesOverlayController::handleClick):
904 Coalesce highlight rebuilding so that things (like TextIndicator creation)
905 that change the selection and then reset it immediately don't cause us
906 to lose the active highlight.
908 This also means that if the selection changes multiple times in a runloop
909 (easily possible from script), we won't waste a lot of time rebuilding highlights.
911 (WebCore::ServicesOverlayController::didRebuildPotentialHighlights):
912 Merged into buildPotentialHighlightsIfNeeded.
914 2015-06-19 Matt Baker <mattbaker@apple.com>
916 Web Inspector: TimelineAgent needs to handle nested runloops
917 https://bugs.webkit.org/show_bug.cgi?id=145090
919 Reviewed by Joseph Pecoraro.
921 Previously nested run loops caused InspectorTimelineAgent to prematurely pop the current run loop record. This
922 patch adds a counter to track the run loop nesting level, and rendering frame records are only pushed/popped
923 when the nesting level is zero. Run loop entry/exit notifications received while the debugger is paused do not
924 affect the nesting level.
926 * inspector/InspectorTimelineAgent.cpp:
927 (WebCore::InspectorTimelineAgent::internalStart):
928 (WebCore::InspectorTimelineAgent::internalStop):
929 (WebCore::InspectorTimelineAgent::InspectorTimelineAgent):
930 * inspector/InspectorTimelineAgent.h:
932 2015-06-19 Brent Fulgham <bfulgham@apple.com>
934 Follow-up fix to r185766.
935 https://bugs.webkit.org/show_bug.cgi?id=22132
937 Reviewed by Zalan Bujtas.
939 Suggested by Darin Adler in the bug.
941 * platform/graphics/filters/FETile.cpp:
942 (WebCore::FETile::platformApplySoftware): Use WTF::move when passing
943 the new tileImageCopy RefPtr.
945 2015-06-19 Michael Catanzaro <mcatanzaro@igalia.com>
947 [SOUP] Fix return-type-c-linkage warning after r185553
948 https://bugs.webkit.org/show_bug.cgi?id=146014
950 Reviewed by Martin Robinson.
952 * platform/network/soup/WebKitSoupRequestGeneric.cpp:
953 (webkitSoupRequestGenericGetRequest): Return a pointer rather than a reference.
954 * platform/network/soup/WebKitSoupRequestGeneric.h: webkitSoupRequestGenericGetRequest now
955 returns a pointer rather than a reference.
957 2015-06-19 Dean Jackson <dino@apple.com>
959 Played <audio> looks invisible against the gray background
960 https://bugs.webkit.org/show_bug.cgi?id=146164
961 <rdar://problem/21014284>
963 Reviewed by Brent Fulgham.
965 The plus-darker blend mode was not allowing any white to
966 show through in the rendering. We don't need this for
967 audio controls, where we draw on an opaque grey background.
969 * Modules/mediacontrols/mediaControlsiOS.css:
970 (audio::-webkit-media-controls-panel): Darken the color of the controls a
971 little to make white stand out more.
972 (audio::-webkit-media-controls-timeline): Remove the plus-darker blending.
973 (video::-webkit-media-controls-timeline): Apply blending only to video.
975 2015-06-19 Andy Estes <aestes@apple.com>
977 Various assertion failures occur when executing script in the midst of DOM insertion
978 https://bugs.webkit.org/show_bug.cgi?id=132482
980 Reviewed by Darin Adler.
982 Prior to this change, when an element containing a <script> child was inserted into a document, the script was
983 executed in ScriptElement::insertedInto(). That script can access nodes that follow it in the newly-inserted
984 hierarchy but are not yet fully inserted, leading to at least the following problems:
986 - The script could remove a node that is not yet marked as in the document.
987 - The script could remove a named <map> that has yet to be added to TreeScope::m_imageMapsByName.
988 - The script could remove a form control that has yet to be added to FormController::m_formElementsWithState.
990 These scenarios all result in assertion failures. This change ensures that each node in the newly-inserted
991 hierarchy is fully inserted before executing any scripts.
993 Tests: fast/dom/element-removed-while-inserting-parent-crash.html
994 fast/dom/named-map-removed-while-inserting-parent-crash.html
995 fast/forms/form-control-removed-while-inserting-parent-crash.html
996 svg/dom/element-removed-while-inserting-parent-crash.html
998 * dom/ScriptElement.cpp:
999 (WebCore::ScriptElement::shouldNotifySubtreeInsertions): Renamed from insertedInto().
1000 Returned true in the case where insertedInto() would've called prepareScript().
1001 (WebCore::ScriptElement::didNotifySubtreeInsertions): Called prepareScript().
1002 (WebCore::ScriptElement::insertedInto): Renamed to shouldNotifySubtreeInsertions().
1003 * dom/ScriptElement.h:
1004 * html/HTMLScriptElement.cpp:
1005 (WebCore::HTMLScriptElement::insertedInto): If shouldNotifySubtreeInsertions() is true, returned InsertionShouldCallDidNotifySubtreeInsertions.
1006 Otherwise, returned InsertionDone.
1007 (WebCore::HTMLScriptElement::didNotifySubtreeInsertions): Called ScriptElement::didNotifySubtreeInsertions().
1008 * html/HTMLScriptElement.h:
1009 * svg/SVGScriptElement.cpp:
1010 (WebCore::SVGScriptElement::insertedInto): Did the same as HTMLScriptElement::insertedInto().
1011 (WebCore::SVGScriptElement::didNotifySubtreeInsertions): Called ScriptElement::didNotifySubtreeInsertions().
1012 * svg/SVGScriptElement.h:
1014 2015-06-19 Brent Fulgham <bfulgham@apple.com>
1016 All calls of ImageBuffer::create should null check the return value
1017 https://bugs.webkit.org/show_bug.cgi?id=22132
1019 Reviewed by Zalan Bujtas.
1021 ImageBuffer::create returns nullptr for a number of reasons, and should be
1022 expected to do so. We missed this check in a few places, resulting in
1023 crashes on some systems. Likewise, ImageBuffer::copyImage may return nullptr
1024 in normal use and should be checked.
1026 * platform/graphics/BitmapImage.cpp:
1027 (WebCore::BitmapImage::drawPattern): Add nullptr check for create and copyImage. Remove
1028 extra call to 'setImageObserver'.
1029 * platform/graphics/cairo/ImageBufferCairo.cpp:
1030 (WebCore::ImageBuffer::drawPattern): Add nullptr check for copyImage.
1031 * platform/graphics/cg/ImageBufferCG.cpp:
1032 (WebCore::ImageBuffer::drawPattern): Add nullptr checks for copyImage.
1033 * platform/graphics/filters/FETile.cpp:
1034 (WebCore::FETile::platformApplySoftware): Add nullptr check for copyImage.
1035 * platform/graphics/filters/FilterEffect.cpp:
1036 (WebCore::FilterEffect::asImageBuffer): Add nullptr check for create.
1037 (WebCore::FilterEffect::openCLImageToImageBuffer): Ditto.
1038 * platform/graphics/texmap/BitmapTexture.cpp:
1039 (WebCore::BitmapTexture::updateContents): Add nullptr checks for create and copyImage.
1040 * svg/graphics/SVGImage.cpp:
1041 (WebCore::SVGImage::drawPatternForContainer): Add nullptr check for copyImage.
1043 2015-06-19 Jeremy Jones <jeremyj@apple.com>
1045 Get CAContext directly for CALayer instead of walking the layer tree.
1046 https://bugs.webkit.org/show_bug.cgi?id=146138
1047 <rdar://problem/21455974>
1049 Reviewed by Darin Adler.
1051 This will get the context directly from the CALayer instead of getting all CAContexts, walking the layer tree
1052 to the root and comparing that against each CAContext's root layer.
1054 * platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.mm:
1055 (WebCore::MediaPlayerPrivateAVFoundationObjC::setVideoFullscreenLayer):
1057 2015-06-18 Brent Fulgham <bfulgham@apple.com>
1059 [iOS] scrollIntoViewIfNeeded is not working with scroll-snap points
1060 https://bugs.webkit.org/show_bug.cgi?id=145318
1061 <rdar://problem/21081501>
1063 Reviewed by Simon Fraser.
1065 Use the ScrollController in iOS to track the scroll snap point state.
1066 We do not need the animation implementation or timers since the actual
1067 animation is handled by UIKit.
1069 This change lets us communicate the current offset into the scroll snap
1070 offset vector between the WebProcess and RemoteScrollingTree so that
1071 both sides stay in sync regardless of whether user gestures or style
1072 updates have caused us to shift to a different snap point.
1074 * page/scrolling/AsyncScrollingCoordinator.cpp:
1075 (WebCore::AsyncScrollingCoordinator::frameViewLayoutUpdated): Set the
1076 current horizontal and vertical scroll snap offset indices.
1077 (WebCore::AsyncScrollingCoordinator::updateOverflowScrollingNode): Ditto.
1078 * page/scrolling/AsyncScrollingCoordinator.h: Mark the setActiveScrollSnapIndices
1079 for export so that it can be reached by the UIProcess.
1080 * page/scrolling/ScrollingCoordinator.h: Keep track of horizontal and
1081 vertical scroll snap offset indices.
1082 * page/scrolling/ScrollingStateScrollingNode.cpp:
1083 (WebCore::ScrollingStateScrollingNode::setCurrentHorizontalSnapPointIndex): Added.
1084 (WebCore::ScrollingStateScrollingNode::setCurrentVerticalSnapPointIndex): Added.
1085 * page/scrolling/ScrollingStateScrollingNode.h:
1086 (WebCore::ScrollingStateScrollingNode::currentHorizontalSnapPointIndex): Added.
1087 (WebCore::ScrollingStateScrollingNode::currentVerticalSnapPointIndex): Added.
1088 * page/scrolling/ScrollingTree.h:
1089 * page/scrolling/ScrollingTreeScrollingNode.cpp:
1090 (WebCore::ScrollingTreeScrollingNode::updateBeforeChildren): Update the scroll snap
1091 point offset indices if either has changed.
1092 * page/scrolling/ScrollingTreeScrollingNode.h:
1093 (WebCore::ScrollingTreeScrollingNode::currentHorizontalSnapPointIndex): Added.
1094 (WebCore::ScrollingTreeScrollingNode::currentVerticalSnapPointIndex): Added.
1095 (WebCore::ScrollingTreeScrollingNode::setCurrentHorizontalSnapPointIndex): Added.
1096 (WebCore::ScrollingTreeScrollingNode::setCurrentVerticalSnapPointIndex): Added.
1097 * page/scrolling/ThreadedScrollingTree.cpp:
1098 (WebCore::ThreadedScrollingTree::currentSnapPointIndicesDidChange): New method
1099 to handle notifications about scroll snap index changes from the UIProcess.
1100 * page/scrolling/ThreadedScrollingTree.h:
1101 * page/scrolling/ios/ScrollingTreeIOS.cpp:
1102 (WebCore::ScrollingTreeIOS::currentSnapPointIndicesDidChange): New method
1103 to handle notifications about scroll snap index changes from the UIProcess.
1104 * page/scrolling/ios/ScrollingTreeIOS.h:
1105 * page/scrolling/mac/ScrollingTreeFrameScrollingNodeMac.mm:
1106 (WebCore::ScrollingTreeFrameScrollingNodeMac::updateBeforeChildren): Update scroll
1107 snap point current offset indices if they have changed.
1108 (WebCore::ScrollingTreeFrameScrollingNodeMac::scrollOffsetOnAxis): Remove unneeded
1109 PLATFORM(MAC) macro.
1110 * platform/ScrollAnimator.cpp:
1111 (WebCore::ScrollAnimator::ScrollAnimator): We have a ScrollController if we are
1112 supporting scroll snap points or rubber banding.
1113 (WebCore::ScrollAnimator::processWheelEventForScrollSnap): This method is not needed
1115 (WebCore::ScrollAnimator::updateActiveScrollSnapIndexForOffset): Enable this on iOS.
1116 (WebCore::ScrollAnimator::updateScrollSnapState): Renamed from 'updateScrollAnimatorsAndTimers'
1118 (WebCore::ScrollAnimator::updateScrollAnimatorsAndTimers): Deleted.
1119 * platform/ScrollAnimator.h: Enable some scroll snap methods on iOS.
1120 * platform/ScrollableArea.cpp:
1121 (WebCore::ScrollableArea::handleWheelEvent): Enable scroll snap index bookkeeping on iOS, too.
1122 (WebCore::ScrollableArea::updateScrollSnapState): Revise to call 'updateScrollSnapState' instead
1123 of 'updateScrollAnimatorsAndTimers'.
1124 * platform/cocoa/ScrollController.h: Enable some methods on iOS. Reorder methods to
1125 reduce the number of macros needed to do so.
1126 * platform/cocoa/ScrollController.mm:
1127 (systemUptime): Only build for Mac.
1128 (WebCore::ScrollController::ScrollController): Disable rubber band-specific members on iOS.
1129 (WebCore::ScrollController::handleWheelEvent): Only build this on Mac.
1130 (WebCore::ScrollController::isRubberBandInProgress): Always return 'false' on iOS.
1131 (WebCore::ScrollController::startSnapRubberbandTimer): Only build this on Mac.
1132 (WebCore::ScrollController::shouldRubberBandInHorizontalDirection): Ditto.
1133 (WebCore::ScrollController::scrollSnapPointState): Enable on iOS.
1134 (WebCore::ScrollController::hasActiveScrollSnapTimerForAxis): Only build on Mac.
1135 (WebCore::ScrollController::updateScrollSnapState): renamed from 'updateScrollAnimatorsAndTimers'
1136 (WebCore::ScrollController::startScrollSnapTimer): Only build on Mac.
1137 (WebCore::ScrollController::initializeGlideParameters): Ditto.
1138 (WebCore::ScrollController::activeScrollSnapIndexForAxis): Enable on iOS.
1139 (WebCore::ScrollController::setActiveScrollSnapIndicesForOffset): Ditto.
1140 (WebCore::ScrollController::beginScrollSnapAnimation): Only build on Mac.
1141 (WebCore::ScrollController::computeGlideDelta): Ditto.
1142 (WebCore::ScrollController::updateScrollAnimatorsAndTimers): Deleted.
1143 * rendering/RenderLayerCompositor.cpp:
1144 (WebCore::RenderLayerCompositor::updateScrollCoordinatedLayer): Capture any changes in scroll
1145 snap offset indices.
1147 2015-06-19 Jeremy Jones <jeremyj@apple.com>
1149 Fullscreen view should not update bounds of video when in PiP.
1150 https://bugs.webkit.org/show_bug.cgi?id=146134
1152 Reviewed by Darin Adler.
1154 Don't update bounds on video layer when it is not a child.
1156 * platform/ios/WebVideoFullscreenInterfaceAVKit.mm:
1157 (-[WebAVVideoLayer setBounds:]):
1159 2015-06-19 Zalan Bujtas <zalan@apple.com>
1161 RenderRubyText requires RenderRubyRun parent.
1162 https://bugs.webkit.org/show_bug.cgi?id=146148
1163 rdar://problem/21423319
1165 Reviewed by Simon Fraser.
1167 RenderRubyText expects its parent to be RenderRubyRun and since a
1168 a non-block <rt> requires anonymous wrapper, we should check whether
1169 the display type is actually block.
1171 Test: fast/ruby/crash-when-ruby-rt-is-non-block.html
1173 * html/RubyTextElement.cpp:
1174 (WebCore::RubyTextElement::createElementRenderer):
1176 2015-06-19 Jeremy Jones <jeremyj@apple.com>
1178 cancelPreviousPerformRequestsWithTarget for -resolveBounds in wrong class.
1179 https://bugs.webkit.org/show_bug.cgi?id=146140
1181 Reviewed by Eric Carlson.
1183 * platform/ios/WebVideoFullscreenInterfaceAVKit.mm:
1184 (-[WebCALayerHostWrapper dealloc]): Added.
1185 (-[WebAVVideoLayer dealloc]): Deleted.
1187 2015-06-19 Per Arne Vollan <peavo@outlook.com>
1189 [WinCairo] Null pointer crash in BitmapTexture::updateContents.
1190 https://bugs.webkit.org/show_bug.cgi?id=146147
1192 Reviewed by Brent Fulgham.
1194 Added null pointer check.
1196 * platform/graphics/texmap/BitmapTexture.cpp:
1197 (WebCore::BitmapTexture::updateContents):
1199 2015-06-19 Anders Carlsson <andersca@apple.com>
1201 Spintracer treats the web process as hung when it's showing JavaScript dialogs in the UI process
1202 https://bugs.webkit.org/show_bug.cgi?id=146124
1203 rdar://problem/21449395
1205 Reviewed by Geoffrey Garen.
1207 * platform/spi/cg/CoreGraphicsSPI.h:
1209 2015-06-19 Csaba Osztrogonác <ossy@webkit.org>
1211 Remove unnecessary svn:executable flags
1212 https://bugs.webkit.org/show_bug.cgi?id=146107
1214 Reviewed by Alexey Proskuryakov.
1216 * html/canvas/CanvasRenderingContext2D.cpp: Removed property svn:executable.
1217 * mathml/MathMLMencloseElement.cpp: Removed property svn:executable.
1218 * mathml/MathMLMencloseElement.h: Removed property svn:executable.
1219 * platform/efl/RenderThemeEfl.cpp: Removed property svn:executable.
1220 * rendering/mathml/RenderMathMLMenclose.cpp: Removed property svn:executable.
1221 * rendering/mathml/RenderMathMLMenclose.h: Removed property svn:executable.
1223 2015-06-19 Youenn Fablet <youenn.fablet@crf.canon.fr>
1225 Bindings generator should generate code to catch exception and reject promises for Promise-based APIs
1226 https://bugs.webkit.org/show_bug.cgi?id=146060
1228 Reviewed by Darin Adler.
1230 The binding generator splits the function that binds JS to the DOM class implementation in two for functions returning promise.
1231 The first function, called from JS, is responsible of casting this to the expected JSXXX class.
1232 If casting fails, an exception is raised. Otherwise, it calls the second function.
1233 After calling the second function, it checks whether an exception is raised, in which case it returns a rejected promise.
1234 The second function is responsible of argument conversion and calling the DOM class function.
1236 Covered by expectations and AudioContext promise still working.
1237 A test case is added for a promise returning function taking a typed argument as input (if argument value cannot be typed, the promise is rejected).
1238 A second test case is a promise-returning function that can raise an exception. In that case the DOMException is used as rejection value.
1240 As can be seen from generated code, this generalized code adds a mandatory check (is there an exception?) at the end of the function.
1241 This check is done even in cases we know there will be no exception.
1242 This may be covered by another patch if this optimization is thought useful enough.
1244 * bindings/js/JSDOMPromise.cpp:
1245 (WebCore::rejectPromiseWithExceptionIfAny): Utility method for the binding code.
1246 (WebCore::callPromiseFunction): Ditto.
1247 * bindings/js/JSDOMPromise.h:
1248 * bindings/scripts/CodeGeneratorJS.pm:
1249 (GenerateImplementation):
1250 (GenerateFunctionCastedThis): Extracted from GenerateImplementationFunctionCall to reuse it in case of promise-returning functions.
1251 (GenerateImplementationFunctionCall):
1252 (GenerateCallbackImplementation): Deleted.
1253 * bindings/scripts/test/JS/JSTestObj.cpp:
1254 (WebCore::jsTestObjPrototypeFunctionTestPromiseFunction):
1255 (WebCore::jsTestObjPrototypeFunctionTestPromiseFunctionPromise):
1256 (WebCore::jsTestObjPrototypeFunctionTestPromiseFunctionWithFloatArgument):
1257 (WebCore::jsTestObjPrototypeFunctionTestPromiseFunctionWithFloatArgumentPromise):
1258 (WebCore::jsTestObjPrototypeFunctionTestPromiseFunctionWithException):
1259 (WebCore::jsTestObjPrototypeFunctionTestPromiseFunctionWithExceptionPromise):
1260 * bindings/scripts/test/TestObj.idl:
1262 2015-06-18 Jeremy Jones <jeremyj@apple.com>
1264 Disable UIWindow for fullscreen video for selected clients.
1265 https://bugs.webkit.org/show_bug.cgi?id=145852
1267 Reviewed by Simon Fraser.
1269 Disable UIWindow for fullscreen video doesn't work everywhere (rdar://problem/21315993), so just disable it when creating a UIWindow won't work.
1270 Fix some interface hiding and layout problems that showed up in the non UIWindow code path.
1272 * platform/RuntimeApplicationChecksIOS.h:
1273 * platform/RuntimeApplicationChecksIOS.mm: Remove iAD bundle identifier.
1274 * platform/ios/WebVideoFullscreenInterfaceAVKit.mm:
1275 (WebVideoFullscreenInterfaceAVKit::setupFullscreen): Opt out of UIWindow when hosted in another process. And fix view parenting for non-window case.
1276 (WebVideoFullscreenInterfaceAVKit::exitFullscreen): Fix for AVKit exit fullscreen complaining about -needsLayout.
1278 (WebVideoFullscreenInterfaceAVKit::requestHideAndExitFullscreen):
1279 (WebVideoFullscreenInterfaceAVKit::preparedToReturnToInline):
1280 (WebVideoFullscreenInterfaceAVKit::willStartOptimizedFullscreen):
1281 (WebVideoFullscreenInterfaceAVKit::didStartOptimizedFullscreen):
1282 (WebVideoFullscreenInterfaceAVKit::willStopOptimizedFullscreen):
1283 (WebVideoFullscreenInterfaceAVKit::didStopOptimizedFullscreen):
1284 These hide and show the view controller where we hide and show the window.
1286 2015-06-18 Dean Jackson <dino@apple.com>
1288 Provide a way for web developers to draw a Theme-specific Wireless Playback icon
1289 https://bugs.webkit.org/show_bug.cgi?id=146123
1290 <rdar://problem/21119287>
1292 Reviewed by Simon Fraser.
1294 Implement a -webkit-named-image() CSS <image> generator that allows a site to
1295 request artwork by name and get the platform variant. At the moment
1296 we only support "wireless-playback" which returns a generic image everywhere
1297 but Cocoa platforms, where we render the AirPlay icon.
1299 In order to do this I added a ThemeCocoa to share any Theme code between
1302 Test: fast/css/named-icons.html
1304 * WebCore.xcodeproj/project.pbxproj: Add new files CSSNamedImageValue, NamedImageGeneratedImage and ThemeCocoa.
1306 * css/CSSImageGeneratorValue.cpp: Handle the new NamedImageClass in the switch statements for downcasting.
1307 (WebCore::CSSImageGeneratorValue::image):
1308 (WebCore::CSSImageGeneratorValue::isFixedSize):
1309 (WebCore::CSSImageGeneratorValue::isPending):
1310 (WebCore::CSSImageGeneratorValue::knownToBeOpaque):
1312 * css/CSSNamedImageValue.cpp: New class. Just holds a name String.
1313 (WebCore::CSSNamedImageValue::customCSSText):
1314 (WebCore::CSSNamedImageValue::image):
1315 (WebCore::CSSNamedImageValue::equals):
1316 * css/CSSNamedImageValue.h:
1317 (WebCore::CSSNamedImageValue::create):
1318 (WebCore::CSSNamedImageValue::isFixedSize):
1319 (WebCore::CSSNamedImageValue::isPending):
1320 (WebCore::CSSNamedImageValue::CSSNamedImageValue):
1322 * css/CSSParser.cpp:
1323 (WebCore::CSSParser::isGeneratedImageValue): Allow "-webkit-named-image(".
1324 (WebCore::CSSParser::parseGeneratedImage): Call parseNamedImage if we hit named-icon.
1325 (WebCore::CSSParser::parseNamedImage): Parse the function looking for a CSS ident.
1328 * css/CSSValue.cpp: Handle NamedImageClass in the various switch statements.
1329 (WebCore::CSSValue::equals):
1330 (WebCore::CSSValue::cssText):
1331 (WebCore::CSSValue::destroy):
1333 (WebCore::CSSValue::isNamedImageValue): Helper to detect the correct CSSValue subclass.
1335 * platform/Theme.cpp:
1336 (WebCore::Theme::drawNamedImage): Draw a generic wireless playback icon.
1337 * platform/Theme.h: Add drawNamedImage.
1339 * platform/cocoa/ThemeCocoa.cpp: New shared base class for ThemeMac and ThemeIOS.
1340 (WebCore::fitContextToBox):
1341 (WebCore::ThemeCocoa::drawNamedImage): Draw an AirPlay icon for wireless playback.
1342 * platform/cocoa/ThemeCocoa.h:
1344 * platform/graphics/CrossfadeGeneratedImage.h: Drive-by removal of unnecessary forward class definition.
1346 * platform/graphics/ImageBuffer.h: Add NamedImageGeneratedImage as a friend class.
1348 * platform/graphics/NamedImageGeneratedImage.cpp: New class. Calls into the Theme to render the artwork.
1349 (WebCore::NamedImageGeneratedImage::NamedImageGeneratedImage):
1350 (WebCore::NamedImageGeneratedImage::draw):
1351 (WebCore::NamedImageGeneratedImage::drawPattern):
1352 * platform/graphics/NamedImageGeneratedImage.h:
1354 * platform/ios/ThemeIOS.h: Inherit from ThemeCocoa.
1355 * platform/mac/ThemeMac.h: Ditto.
1357 2015-06-18 KyungTae Kim <ktf.kim@samsung.com> and Myles C. Maxfield <mmaxfield@apple.com>
1359 [CSS3] Add support for the word-break:keep-all CSS property
1360 https://bugs.webkit.org/show_bug.cgi?id=123782
1362 Reviewed by Darin Adler.
1364 Add support for word-break:keep-all CSS property by CSS3 spec:
1365 http://www.w3.org/TR/2013/WD-css-text-3-20131010/#word-break-property
1367 Test: fast/text/word-break-keep-all.html
1369 * css/CSSParser.cpp:
1370 (WebCore::isValidKeywordPropertyAndValue):
1371 * css/CSSPrimitiveValueMappings.h:
1372 (WebCore::CSSPrimitiveValue::CSSPrimitiveValue):
1373 (WebCore::CSSPrimitiveValue::operator EWordBreak):
1374 * css/CSSValueKeywords.in:
1375 * rendering/RenderText.cpp:
1376 (WebCore::RenderText::computePreferredLogicalWidths):
1377 * rendering/break_lines.h:
1378 (WebCore::nextBreakablePositionKeepingAllWords):
1379 (WebCore::nextBreakablePositionKeepingAllWordsIgnoringNBSP):
1380 (WebCore::isBreakable):
1381 * rendering/line/BreakingContext.h:
1382 (WebCore::BreakingContext::handleText):
1383 (WebCore::BreakingContext::optimalLineBreakLocationForTrailingWord):
1384 * rendering/style/RenderStyleConstants.h:
1386 2015-06-18 Jon Lee <jonlee@apple.com>
1388 Update AVKit usage of pip
1389 https://bugs.webkit.org/show_bug.cgi?id=146095
1390 <rdar://problem/21386853>
1392 Reviewed by Eric Carlson.
1394 - Rename enum VideoFullscreenModeOptimized to VideoFullscreenModePictureInPicture
1395 - Rename MediaElementSession::allowsAlternateFullscreen to allowsPictureInPicture
1396 - Rename Settings::allowsAlternateFullscreen to allowsPictureInPictureMediaPlayback
1397 - Update AVKit calls and AVKitSPI.h
1398 - Rename WebVideoFullscreenInterfaceAVKit delegate functions and member variables
1400 * html/HTMLMediaElement.cpp:
1401 * html/HTMLVideoElement.cpp:
1402 * html/MediaElementSession.cpp:
1403 (WebCore::MediaElementSession::allowsPictureInPicture): Renamed.
1404 (WebCore::MediaElementSession::allowsAlternateFullscreen): Deleted.
1405 * html/MediaElementSession.h:
1406 * page/Settings.cpp:
1408 * platform/graphics/MediaPlayerEnums.h:
1409 * platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.mm:
1410 * platform/ios/WebVideoFullscreenControllerAVKit.mm:
1411 * platform/ios/WebVideoFullscreenInterfaceAVKit.h: Remove unused setIsOptimized.
1412 * platform/ios/WebVideoFullscreenInterfaceAVKit.mm:
1413 * platform/spi/cocoa/AVKitSPI.h: Remove unused typedef.
1414 * platform/spi/mac/AVFoundationSPI.h:
1416 2015-06-18 Jeremy Jones <jeremyj@apple.com>
1418 Fix crash when entering fullscreen during exit fullscreen animation.
1419 https://bugs.webkit.org/show_bug.cgi?id=146117
1421 Reviewed by Simon Fraser.
1423 Because enterFullscreen can be called during exitFullscreen animation, the exit fullscreen teardown
1424 should not imply a fullscreen state change on video element.
1426 * platform/ios/WebVideoFullscreenControllerAVKit.mm:
1427 (WebVideoFullscreenControllerContext::exitFullscreen): May be called from mainThread
1428 (-[WebVideoFullscreenController exitFullscreen]): May be called from mainThread
1429 * platform/ios/WebVideoFullscreenModelVideoElement.mm:
1430 (WebVideoFullscreenModelVideoElement::setVideoElement):
1431 Changing associated video element does not imply fullscreen mode change.
1433 2015-06-18 Brian J. Burg <burg@cs.washington.edu>
1435 Web Inspector: improve generated types for objects passed to backend commands
1436 https://bugs.webkit.org/show_bug.cgi?id=146091
1438 Reviewed by Joseph Pecoraro.
1440 Update type signatures for backend command implementations. In a few cases, clean
1441 up relevant helper function signatures and copy data out of parameter objects where
1442 the code previously held onto a reference.
1444 No new tests, no behavior changed.
1446 * inspector/InspectorCSSAgent.cpp:
1447 (WebCore::computePseudoClassMask):
1448 (WebCore::InspectorCSSAgent::setStyleText):
1449 (WebCore::InspectorCSSAgent::setRuleSelector):
1450 (WebCore::InspectorCSSAgent::forcePseudoState):
1451 * inspector/InspectorCSSAgent.h:
1452 * inspector/InspectorDOMAgent.cpp:
1453 (WebCore::parseColor):
1454 (WebCore::parseConfigColor):
1455 (WebCore::parseQuad):
1456 (WebCore::InspectorDOMAgent::performSearch):
1457 (WebCore::InspectorDOMAgent::setSearchingForNode):
1458 (WebCore::InspectorDOMAgent::highlightConfigFromInspectorObject):
1459 (WebCore::InspectorDOMAgent::setInspectModeEnabled):
1460 (WebCore::InspectorDOMAgent::highlightRect):
1461 (WebCore::InspectorDOMAgent::highlightQuad):
1462 (WebCore::InspectorDOMAgent::innerHighlightQuad):
1463 (WebCore::InspectorDOMAgent::highlightNode):
1464 (WebCore::InspectorDOMAgent::highlightFrame):
1465 * inspector/InspectorDOMAgent.h:
1466 * inspector/InspectorDOMStorageAgent.cpp:
1467 (WebCore::InspectorDOMStorageAgent::getDOMStorageItems):
1468 (WebCore::InspectorDOMStorageAgent::setDOMStorageItem):
1469 (WebCore::InspectorDOMStorageAgent::removeDOMStorageItem):
1470 (WebCore::InspectorDOMStorageAgent::findStorageArea):
1471 * inspector/InspectorDOMStorageAgent.h:
1472 * inspector/InspectorIndexedDBAgent.cpp:
1473 (WebCore::InspectorIndexedDBAgent::requestData):
1474 * inspector/InspectorIndexedDBAgent.h:
1475 * inspector/InspectorReplayAgent.cpp:
1476 (WebCore::InspectorReplayAgent::replayToPosition):
1477 * inspector/InspectorReplayAgent.h:
1478 * inspector/InspectorResourceAgent.cpp:
1479 (WebCore::InspectorResourceAgent::willSendRequest):
1480 (WebCore::InspectorResourceAgent::setExtraHTTPHeaders):
1481 Clean up extraHTTPHeaders to copy header key/values out of the InspectorObject, rather
1482 than retaining the protocol object indefinitely. This matches the ownership scheme used
1485 * inspector/InspectorResourceAgent.h:
1486 * inspector/InspectorStyleSheet.h:
1487 (WebCore::InspectorCSSId::InspectorCSSId):
1488 * inspector/InspectorWorkerAgent.cpp:
1489 (WebCore::InspectorWorkerAgent::sendMessageToWorker):
1490 * inspector/InspectorWorkerAgent.h:
1492 2015-06-18 Anders Carlsson <andersca@apple.com>
1494 Remove shouldInterruptJavaScript
1495 https://bugs.webkit.org/show_bug.cgi?id=146118
1497 Reviewed by Antti Koivisto.
1499 The WebKit SPI methods for deciding whether JavaScript execution should be interrupted hasn't been used
1500 for many releases. Furthermore, they don't make sense in the multi-process architecture since it's still possible
1501 to interrupt execution (by closing the browser tab or window) from the UI process.
1503 * bindings/js/JSDOMWindowBase.cpp:
1504 (WebCore::JSDOMWindowBase::shouldInterruptScript):
1505 * loader/EmptyClients.h:
1507 (WebCore::Chrome::shouldInterruptJavaScript): Deleted.
1509 * page/ChromeClient.h:
1511 2015-06-18 Benjamin Poulain <bpoulain@apple.com>
1513 [CSS JIT][ARMv7] The pseudo element early exit trashes r6
1514 https://bugs.webkit.org/show_bug.cgi?id=146078
1516 Reviewed by Alex Christensen.
1518 The pseudo element early failure runs before we generate the prologue.
1519 The reason is that we can often exit immediately on function entry, before
1520 we even touch any memory.
1522 On ARMv7, we don't have many spare registers so the MacroAssembler
1523 uses r6 as a scratch register and the client code is expected to save
1526 In the early failure case, we were not pushing r6 before using the MacroAssembler
1527 and its value could be trashed.
1529 This patch push the macro assembler registers separately from the prologue.
1531 For restoring the registers, a new function generateFunctionEnding() encapsulate
1532 the pop() and ret().
1534 * cssjit/SelectorCompiler.cpp:
1535 (WebCore::SelectorCompiler::SelectorCodeGenerator::pushMacroAssemblerRegisters):
1536 (WebCore::SelectorCompiler::SelectorCodeGenerator::popMacroAssemblerRegisters):
1537 (WebCore::SelectorCompiler::SelectorCodeGenerator::generatePrologue):
1538 (WebCore::SelectorCompiler::SelectorCodeGenerator::generateEpilogue):
1539 (WebCore::SelectorCompiler::SelectorCodeGenerator::generateSelectorChecker):
1541 * cssjit/StackAllocator.h:
1542 (WebCore::StackAllocator::operator=):
1543 We have a new case for the stack allocator: some stack changes are conditional
1544 at compile time instead of runtime. This is easy to deal with by overriding
1545 the stack if a path is not taken at compile time.
1547 2015-06-17 Conrad Shultz <conrad_shultz@apple.com>
1549 REGRESSION: js/dom/navigator-plugins-crash.html asserts a lot
1550 https://bugs.webkit.org/show_bug.cgi?id=144399
1552 Reviewed by Darin Adler.
1554 Earlier work made the array of web-visible plug-ins dynamic, but allowed DOMPlugin (and, indirectly by extension,
1555 DOMMimeType) to continue keeping a reference to a plug-in in terms of an index into that array. This superficially
1556 appeared correct since DOMPlugin immutably holds onto a PluginData instance, which in turn immutably holds onto a
1557 Page instance. PluginStrategy::getWebVisiblePluginInfo() is passed this Page, which is used to determine the contents
1558 of the plugin array. The expectation was that keeping an index would still be safe since the Page is not changing,
1559 but this is not strictly correct since relevant attributes of the Page and/or the available plugins may still change.
1561 It's not entirely clear why the test failures are intermittent and occur only on certain configurations, but address
1562 them by eliminating the incorrect storage of indexes in favor of keeping copies of the relevant plugin info itself.
1564 * plugins/DOMMimeType.cpp:
1565 (WebCore::DOMMimeType::DOMMimeType):
1566 Instead of storing the MIME type index, retrieve and store the MIME class info and plugin info.
1567 (WebCore::DOMMimeType::type):
1568 Directly access the m_mimeClassInfo member.
1569 (WebCore::DOMMimeType::suffixes):
1571 (WebCore::DOMMimeType::description):
1573 (WebCore::DOMMimeType::enabledPlugin):
1574 Directly access the m_pluginInfo member.
1575 (WebCore::DOMMimeType::mimeClassInfo): Deleted.
1577 * plugins/DOMMimeType.h:
1578 Update member variables.
1580 * plugins/DOMPlugin.cpp:
1581 (WebCore::DOMPlugin::DOMPlugin):
1582 Instead of storing the plugin index, store the plugin info directly.
1583 (WebCore::DOMPlugin::name):
1584 Directly access m_pluginInfo.
1585 (WebCore::DOMPlugin::filename):
1587 (WebCore::DOMPlugin::description):
1589 (WebCore::DOMPlugin::length):
1591 (WebCore::DOMPlugin::item):
1592 Access m_pluginInfo directly; find the matching plug-in based on matching PluginInfo (for which an overloaded
1593 comparator is supplied below).
1594 (WebCore::DOMPlugin::pluginInfo): Deleted.
1596 * plugins/DOMPlugin.h:
1597 Update member variables.
1598 (WebCore::DOMPlugin::create):
1599 Accept a PluginInfo instead of a plugin index.
1601 * plugins/DOMPluginArray.cpp:
1602 (WebCore::DOMPluginArray::item):
1603 (WebCore::DOMPluginArray::namedItem):
1605 * plugins/PluginData.h:
1606 (WebCore::operator==):
1607 Added; compare PluginInfo structs on the basis of member equality.
1609 2015-06-17 Alex Christensen <achristensen@webkit.org>
1611 [Content Extensions] Log blocked loads to the WebInspector console
1612 https://bugs.webkit.org/show_bug.cgi?id=146089
1614 Reviewed by Joseph Pecoraro.
1616 * contentextensions/ContentExtensionsBackend.cpp:
1617 (WebCore::ContentExtensions::ContentExtensionsBackend::processContentExtensionRulesForLoad):
1618 (WebCore::ContentExtensions::ContentExtensionsBackend::displayNoneCSSRule):
1619 Log which URLs are blocked and the URL of the page they are blocked from.
1621 2015-06-18 Joseph Pecoraro <pecoraro@apple.com>
1623 Crash under WebCore::DOMWindow::dispatchMessageEventWithOriginCheck attempting to log console message
1624 https://bugs.webkit.org/show_bug.cgi?id=146093
1626 Reviewed by Timothy Hatcher.
1628 * page/DOMWindow.cpp:
1629 (WebCore::DOMWindow::dispatchMessageEventWithOriginCheck):
1630 The console could be null so null check its use.
1632 2015-06-18 Csaba Osztrogonác <ossy@webkit.org>
1634 Suppress null-conversion warnings in ANGLE
1635 https://bugs.webkit.org/show_bug.cgi?id=145125
1637 Reviewed by Alex Christensen.
1641 2015-06-18 Youenn Fablet <youenn.fablet@crf.canon.fr> and Xabier Rodriguez Calvar <calvaris@igalia.com>
1643 [Streams API] Implement ReadableStreamReader.releaseLock
1644 https://bugs.webkit.org/show_bug.cgi?id=145299
1646 Reviewed by Darin Adler.
1648 Covered by rebased tests.
1650 * Modules/streams/ReadableStream.cpp:
1651 (WebCore::ReadableStream::close): Moving some close code to newly added releaseReader.
1652 (WebCore::ReadableStream::releaseReader): Implements reader release and callbacks finalization.
1653 (WebCore::ReadableStream::changeStateToErrored): Calls releaseReader.
1654 * Modules/streams/ReadableStream.h:
1655 (WebCore::ReadableStream::hasReadPendingRequests): Added to enable reader.releaseLock throwing if read requests are pending.
1656 * Modules/streams/ReadableStreamReader.cpp:
1657 (WebCore::ReadableStreamReader::releaseLock): Implementation of releaseLock
1658 * Modules/streams/ReadableStreamReader.h:
1659 * Modules/streams/ReadableStreamReader.idl:
1660 (WebCore::releaseLock): Deleted.
1662 2015-06-18 Youenn Fablet <youenn.fablet@crf.canon.fr>
1664 GObject and ObjC bindings generator should not generate code for promise-based APIs
1665 https://bugs.webkit.org/show_bug.cgi?id=146059
1667 Reviewed by Darin Adler.
1669 Covered by rebased expectations.
1671 * bindings/scripts/CodeGeneratorGObject.pm:
1672 (SkipFunction): Disabling GObject DOM binding for functions returning promises.
1673 * bindings/scripts/CodeGeneratorObjC.pm:
1674 (SkipFunction): Disabling ObjC DOM binding for functions returning promises.
1675 * bindings/scripts/test/GObject/WebKitDOMTestObj.cpp: Rebasing expectation.
1676 (webkit_dom_test_obj_get_read_only_long_attr): Deleted.
1677 (webkit_dom_test_obj_get_read_only_string_attr): Deleted.
1678 * bindings/scripts/test/GObject/WebKitDOMTestObj.h: Rebasing expectation.
1679 * bindings/scripts/test/ObjC/DOMTestObj.h: Ditto.
1680 * bindings/scripts/test/ObjC/DOMTestObj.mm: Ditto.
1683 2015-06-17 Ryuan Choi <ryuan.choi@navercorp.com>
1685 [EFL] test_ewk2_context_url_scheme_register has been crashed since r185553
1686 https://bugs.webkit.org/show_bug.cgi?id=146075
1688 Reviewed by Carlos Garcia Campos.
1690 Since r185553, CustomProtocolManager sends StartLoading message to UIProcess
1691 with request of SoupGenericRequest instead of request itself.
1692 But, request of SoupGenericRequest is nullptr in EFL port because EFL port
1693 does not use m_initiatingPageID.
1695 This patch updates request of SoupGenericRequest although m_initiatingPageID is null.
1697 * platform/network/soup/ResourceRequestSoup.cpp:
1698 (WebCore::ResourceRequest::updateSoupRequest):
1700 2015-06-17 Daniel Bates <dabates@apple.com>
1702 Client may receive began editing callback for already focused text field
1703 https://bugs.webkit.org/show_bug.cgi?id=146074
1704 <rdar://problem/21293562>
1706 Reviewed by Darin Adler.
1708 Fixes an issue where the client would be notified that began editing in a text field
1709 for each programmatic DOM focus event dispatched at the text field regardless of
1710 whether the field was focused. The client should only be notified that began editing
1711 exactly once when a text field becomes focused (either programmatically or by user interaction).
1713 * html/TextFieldInputType.cpp:
1714 (WebCore::TextFieldInputType::forwardEvent): Move logic to dispatch editing began callback from here...
1715 (WebCore::TextFieldInputType::handleFocusEvent): to here. This function is called when the
1716 text field becomes newly focused.
1717 * html/TextFieldInputType.h:
1719 2015-06-17 Alex Christensen <achristensen@webkit.org>
1721 [Content Extensions] Fail to parse invalid arrays
1722 https://bugs.webkit.org/show_bug.cgi?id=146079
1723 rdar://problem/21422649
1725 Reviewed by Benjamin Poulain.
1727 Covered by new and corrected API tests.
1729 * contentextensions/ContentExtensionParser.cpp:
1730 (WebCore::ContentExtensions::loadTrigger):
1731 Fail to parse invalid arrays for if-domain, unless-domain, resource-type, and load-type arrays.
1733 2015-06-16 Jon Honeycutt <jhoneycutt@apple.com>
1735 Position::findParent() should take a reference
1736 https://bugs.webkit.org/show_bug.cgi?id=146038
1738 Reviewed by Darin Adler.
1741 (WebCore::Position::containerNode):
1742 (WebCore::Position::parentAnchoredEquivalent):
1743 Pass a reference; there is already a null check.
1744 (WebCore::Position::previous):
1745 Add a missing null check. Code below this expects that node is non-null.
1746 (WebCore::Position::next):
1748 (WebCore::Position::atStartOfTree):
1749 (WebCore::Position::atEndOfTree):
1751 (WebCore::Position::findParent):
1752 Changed to take a reference.
1757 2015-06-17 Brent Fulgham <bfulgham@apple.com>
1759 Overflow regions with scroll snap points are not reliably rubber banding
1760 https://bugs.webkit.org/show_bug.cgi?id=142522
1761 <rdar://problem/20100726>
1763 Reviewed by Darin Adler.
1765 When computing the target scroll destination, update the nearest snap point index
1766 and other bookkeeping, but keep the original gesture target if it would have taken
1767 us beyond either limit of the scroll container.
1769 * page/scrolling/mac/ScrollingTreeFrameScrollingNodeMac.h:
1770 * page/scrolling/mac/ScrollingTreeFrameScrollingNodeMac.mm:
1771 (WebCore::ScrollingTreeFrameScrollingNodeMac::scrollExtents): Add new method
1772 to support client API.
1773 * platform/ScrollAnimator.cpp:
1774 (WebCore::ScrollAnimator::scrollExtents): Add new method to support client API.
1775 * platform/ScrollAnimator.h:
1776 * platform/cocoa/ScrollController.h:
1777 (WebCore::ScrollControllerClient::scrollExtents): Added new pure virtual method to API.
1778 * platform/cocoa/ScrollController.mm:
1779 (WebCore::ScrollController::beginScrollSnapAnimation): Hold onto original user gesture
1780 target, and use that instead of our nearest snap point if the gesture takes us past
1781 either extreme of the scroll container.
1783 2015-06-17 Tim Horton <timothy_horton@apple.com>
1785 Swipe gesture can get stuck, preventing scrolling and other gestures
1786 https://bugs.webkit.org/show_bug.cgi?id=146088
1787 <rdar://problem/16056087>
1789 Reviewed by Darin Adler.
1791 * WebCore.xcodeproj/project.pbxproj:
1792 * platform/spi/mac/NSEventSPI.h: Added.
1795 2015-06-16 Matt Rajca <mrajca@apple.com>
1797 MediaSession: handle MediaEventTrackNext and MediaEventTrackPrevious events
1798 https://bugs.webkit.org/show_bug.cgi?id=146028
1800 Reviewed by Darin Adler.
1802 * Modules/mediasession/MediaRemoteControls.idl: Added nexttrack/previoustrack event handlers.
1803 * Modules/mediasession/MediaSession.cpp: Dispatch the nexttrack/previoustrack events.
1804 (WebCore::MediaSession::skipToNextTrack):
1805 (WebCore::MediaSession::skipToPreviousTrack):
1806 * Modules/mediasession/MediaSession.h:
1807 * Modules/mediasession/MediaSessionManager.cpp: Skip to the next/previous track as described in the media session spec.
1808 (WebCore::MediaSessionManager::skipToNextTrack):
1809 (WebCore::MediaSessionManager::skipToPreviousTrack):
1810 * Modules/mediasession/MediaSessionManager.h:
1811 * dom/EventNames.h: Added the nexttrack/previoustrack event names.
1812 * page/Page.cpp: Tell MediaSessionManager to handle the new track-skipping events.
1813 (WebCore::Page::handleMediaEvent):
1815 2015-06-17 Chris Fleizach <cfleizach@apple.com>
1817 AX: VoiceOver in iOS not announcing generic WAI-ARIA region, even if labelled properly
1818 https://bugs.webkit.org/show_bug.cgi?id=146066
1820 Reviewed by Darin Adler.
1822 Allow the region role to identify as a landmark type.
1824 Updated test: platform/ios-simulator/accessibility/landmark-types.html
1826 * accessibility/ios/WebAccessibilityObjectWrapperIOS.mm:
1827 (-[WebAccessibilityObjectWrapper _accessibilityIsLandmarkRole:]):
1829 2015-06-17 Simon Fraser <simon.fraser@apple.com>
1831 REGRESSION (r173283-r173296): Amazon.com front page has no caret in the search field
1832 https://bugs.webkit.org/show_bug.cgi?id=146073
1833 rdar://problem/21022203
1835 Reviewed by Tim Horton.
1837 Text controls (text inputs and textareas) need backing store even when empty, because
1838 they need to be able to paint a caret.
1840 Test: compositing/backing/form-controls-backing.html
1842 * rendering/RenderLayerBacking.cpp:
1843 (WebCore::RenderLayerBacking::isSimpleContainerCompositingLayer):
1845 2015-06-17 Zalan Bujtas <zalan@apple.com>
1847 Selection cache produces invalid result when ancestor has float element.
1848 https://bugs.webkit.org/show_bug.cgi?id=146042
1849 rdar://problem/20604592
1851 Reviewed by Ryosuke Niwa.
1853 Selection cache already takes floats into account, however it's not enough to check current
1854 block against floats. Any of the ancestor's float starting from the selection root block
1855 can impact the selection offsets.
1857 Test: fast/block/selection-cache-is-incorrect-when-non-direct-parent-has-float.html
1859 * rendering/LogicalSelectionOffsetCaches.h:
1860 (WebCore::LogicalSelectionOffsetCaches::ContainingBlockInfo::ContainingBlockInfo):
1861 (WebCore::LogicalSelectionOffsetCaches::ContainingBlockInfo::setBlock):
1862 (WebCore::LogicalSelectionOffsetCaches::ContainingBlockInfo::block):
1863 (WebCore::LogicalSelectionOffsetCaches::ContainingBlockInfo::cache):
1864 (WebCore::LogicalSelectionOffsetCaches::ContainingBlockInfo::hasFloatsOrFlowThreads):
1865 (WebCore::LogicalSelectionOffsetCaches::LogicalSelectionOffsetCaches):
1867 2015-06-17 Joanmarie Diggs <jdiggs@igalia.com>
1869 AX: [ATK] Expose element tag name as an object attribute
1870 https://bugs.webkit.org/show_bug.cgi?id=146062
1872 Reviewed by Mario Sanchez Prada.
1874 Expose the element tag name as an object attribute with name "tag" and
1875 value being the lowercase tag name, both being what Gecko does for ATK.
1877 No new tests. We already have sufficient coverage for AtkObject attributes.
1878 These tests have been updated to reflect the addition of the new attribute.
1880 * accessibility/atk/WebKitAccessibleWrapperAtk.cpp:
1881 (webkitAccessibleGetAttributes):
1883 2015-06-17 Antti Koivisto <antti@apple.com>
1885 iOS WebKit1: [LegacyTileLayer drawInContext:] should ensure it has web lock
1886 https://bugs.webkit.org/show_bug.cgi?id=146072
1887 rdar://problem/21149759
1889 Reviewed by Simon Fraser
1891 There are some scenarios where we end up drawing without web lock due to client or system issues.
1892 This can cause crashes.
1894 * platform/ios/LegacyTileLayer.mm:
1895 (-[LegacyTileLayer setNeedsDisplayInRect:]):
1896 (-[LegacyTileLayer drawInContext:]):
1898 Ensure we have the web lock when called in main thread (even though we should have it already).
1900 2015-06-17 Brent Fulgham <bfulgham@apple.com>
1902 CSS scroll snap: defining snap points on axis that does not scroll does not work properly
1903 https://bugs.webkit.org/show_bug.cgi?id=146043
1904 <rdar://problem/20125511>
1906 Reviewed by Simon Fraser.
1908 Tested by css3/scroll-snap/scroll-snap-mismatch.html
1910 We always seed the set of scroll snap points with the start and end of the scroll container. This is not
1911 the right behavior if there are no scroll points defined, because we end up creating a snap for the start
1912 and end of the container, and any scroll gesture just takes us across the entire element.
1914 Instead, when we do not find any scroll snap points, we should clear the snap point state for the container.
1916 * page/scrolling/AxisScrollSnapOffsets.cpp:
1917 (WebCore::updateFromStyle): If we did not find any snap points (i.e., the snapOffsets container
1918 only holds '0', return an empty Vector.
1919 (WebCore::updateSnapOffsetsForScrollableArea): If the set of snap points produced by 'updateFromStyle' is empty,
1920 clear the horizontal (or vertical) snap offsets for the scroll area.
1923 2015-06-17 Chris Fleizach <cfleizach@apple.com>
1925 AX: input role="spinbutton" gets skipped in voiceover
1926 https://bugs.webkit.org/show_bug.cgi?id=145514
1928 Reviewed by Mario Sanchez Prada.
1930 SpinButton role was added, but left out of iOS.
1931 To prevent this from happening again, explicitly list every role in the switch statement that
1932 determines accessible visibility.
1934 Test: platform/ios-simulator/accessibility/spinbutton.html
1936 * accessibility/ios/WebAccessibilityObjectWrapperIOS.mm:
1937 (-[WebAccessibilityObjectWrapper determineIsAccessibilityElement]):
1938 (-[WebAccessibilityObjectWrapper isAccessibilityElement]):
1940 2015-06-17 Xabier Rodriguez Calvar <calvaris@igalia.com> and Youenn Fablet <youenn.fablet@crf.canon.fr>
1942 [Streams API] ReadableJSStream should handle promises returned by JS source pull callback
1943 https://bugs.webkit.org/show_bug.cgi?id=145965
1945 Reviewed by Darin Adler.
1947 Implemented asynchronous pulling.
1948 In particular, ensuring that doPull is not called as long as previous call to doPull is finished.
1949 Storing whether to pull automatically when the current pull is finished.
1951 Covered by rebased tests.
1953 * Modules/streams/ReadableStream.cpp:
1954 (WebCore::ReadableStream::pull): stores whether to pull again.
1955 (WebCore::ReadableStream::finishPulling): called when pulling finishes.
1956 * Modules/streams/ReadableStream.h:
1957 * bindings/js/ReadableJSStream.cpp:
1958 (WebCore::createPullResultFulfilledFunction): The promise resolve callback.
1959 (WebCore::ReadableJSStream::doPull): Handling of promise.
1960 * bindings/js/ReadableJSStream.h:
1962 2015-06-16 Carlos Garcia Campos <cgarcia@igalia.com>
1964 WebProcess crashes after too many redirect error when there's an active NPAPI plugin
1965 https://bugs.webkit.org/show_bug.cgi?id=146019
1967 Reviewed by Darin Adler.
1969 This happens with the GTK+ port after a navigation action ends up
1970 in an infinite redirection and the ResourceHandle fails with too
1971 many redirections error. I should actually happen after any error
1972 is reported by the ResourceHnalder before the load is
1973 committed. But tt only happens if there's an active NPAPI
1974 plugin. The problem is that FrameLoader::receivedMainResourceError()
1975 is called recursively because DocumentLoader::stopLoading() ends up
1976 calling mainReceivedError() that calls FrameLoader::receivedMainResourceError()
1977 again. DocumentLoader::stopLoading() checks if the document is
1978 still loading, which can happen if the main resource is loading,
1979 if there's any subresource loading or if there's a plugin
1980 loading. So, in case of being loading, those cases are handled
1981 individually to cancel the main resource, or set an error in the
1982 document loader and cancel subresources and plugins, except for
1983 this case of plugins, that mainReceivedError is called instead of
1984 setting cancelled error on the document loader.
1986 * loader/DocumentLoader.cpp:
1987 (WebCore::DocumentLoader::stopLoading): If the document is still
1988 loading because there are active plugins, set the cancelled error
1989 on the document instead of calling mainReceivedError again.
1991 2015-06-16 Youenn Fablet <youenn.fablet@crf.canon.fr> and Xabier Rodriguez Calvar <calvaris@igalia.com>
1993 [Streams API] Implement ReadableStream locked property
1994 https://bugs.webkit.org/show_bug.cgi?id=146023
1996 Reviewed by Darin Adler.
1998 Covered by rebased tests.
2000 * Modules/streams/ReadableStream.h:
2001 (WebCore::ReadableStream::locked): Renamed isLocked by locked.
2002 * Modules/streams/ReadableStream.idl: Adding locked.
2003 * bindings/js/JSReadableStreamCustom.cpp:
2004 (WebCore::JSReadableStream::getReader): Using isLocked.
2005 * bindings/js/JSReadableStreamReaderCustom.cpp:
2006 (WebCore::constructJSReadableStreamReader): Using isLocked.
2008 2015-06-16 Myles C. Maxfield <mmaxfield@apple.com>
2010 REGRESSION(r184899): [Cocoa] font-variant: small-caps is not honored with web fonts
2011 https://bugs.webkit.org/show_bug.cgi?id=145873
2012 <rdar://problem/21314282>
2014 Reviewed by Dean Jackson.
2016 When font-variant: small-caps is applied, we create a smaller version of the original font
2017 and draw capital characters in that smaller font. CGFontRefs do not have an intrinsic size,
2018 and web fonts historically only had a CGFontRef, which means that there was no need to
2019 convert the CGFontRef to be smaller (as opposed to regular fonts, which had a CTFontRef and
2020 therefore needed the conversion). Instead, we just changed m_size, which represents
2021 the size that the text should be drawn in.
2023 However, r184899 gave CTFontRefs to web fonts. This means that now the FontPlatformData's
2024 m_size variable disagreed with the CTFontRef member. The solution here is to unify the web
2025 font and regular font codepaths, and treat them the same throughout.
2027 Note that this patch removes the last use of the m_isCustomFont variable. As soon as we
2028 entirely migrate to CORETEXT_WEB_FONTS, we should delete this variable.
2030 Test: fast/text/small-caps-web-font.html
2032 * platform/graphics/cocoa/FontCocoa.mm:
2033 (WebCore::Font::platformCreateScaledFont): Treat web fonts the same as regular fonts.
2034 * platform/text/TextFlags.h: Add a comment regarding teh deletion of m_isCustomFont.
2036 2015-06-16 Alex Christensen <achristensen@webkit.org>
2038 [Content Extensions] Implement branch compaction for DFA bytecode.
2039 https://bugs.webkit.org/show_bug.cgi?id=145619
2041 Reviewed by Benjamin Poulain.
2043 This patch adds another pass to the DFABytecodeCompiler which finds where the bytecode from each node
2044 would be if it were compiled with no branch compaction, then uses that as a worst-case value to determine
2045 how many bytes are needed to store the relative jump distance. Then when linking, it will fill in the
2046 value as it already did, but with a variable size jump. The jumps are also now signed distances relative to
2047 where the jump is stored.
2049 This patch is covered by existing tests, which have many jumps that are near the -128/127 byte boundary,
2050 and the switch from 16-bit jumps to 32-bit jumps near the -65536/65535 byte boundary is analogous.
2052 * contentextensions/ContentExtensionCompiler.cpp:
2053 (WebCore::ContentExtensions::compileRuleList):
2054 * contentextensions/DFABytecode.h:
2055 (WebCore::ContentExtensions::smallestPossibleJumpSize):
2056 (WebCore::ContentExtensions::instructionSizeWithArguments):
2057 * contentextensions/DFABytecodeCompiler.cpp:
2058 (WebCore::ContentExtensions::append):
2059 (WebCore::ContentExtensions::appendZeroes):
2060 (WebCore::ContentExtensions::setBits):
2061 (WebCore::ContentExtensions::appendActionBytecodeSize):
2062 (WebCore::ContentExtensions::DFABytecodeCompiler::emitAppendAction):
2063 (WebCore::ContentExtensions::DFABytecodeCompiler::longestPossibleJump):
2064 (WebCore::ContentExtensions::DFABytecodeCompiler::emitJump):
2065 (WebCore::ContentExtensions::DFABytecodeCompiler::emitCheckValue):
2066 (WebCore::ContentExtensions::DFABytecodeCompiler::emitCheckValueRange):
2067 (WebCore::ContentExtensions::DFABytecodeCompiler::emitTerminate):
2068 (WebCore::ContentExtensions::DFABytecodeCompiler::compileNode):
2069 (WebCore::ContentExtensions::DFABytecodeCompiler::compiledNodeMaxBytecodeSize):
2070 (WebCore::ContentExtensions::DFABytecodeCompiler::ranges):
2071 (WebCore::ContentExtensions::DFABytecodeCompiler::checkForRangeMaxBytecodeSize):
2072 (WebCore::ContentExtensions::DFABytecodeCompiler::compileCheckForRange):
2073 (WebCore::ContentExtensions::DFABytecodeCompiler::nodeTransitionsMaxBytecodeSize):
2074 (WebCore::ContentExtensions::DFABytecodeCompiler::compileNodeTransitions):
2075 (WebCore::ContentExtensions::DFABytecodeCompiler::compile):
2076 (WebCore::ContentExtensions::set32Bits): Deleted.
2077 * contentextensions/DFABytecodeCompiler.h:
2078 * contentextensions/DFABytecodeInterpreter.cpp:
2079 (WebCore::ContentExtensions::getBits):
2080 (WebCore::ContentExtensions::getInstruction):
2081 (WebCore::ContentExtensions::jumpSizeInBytes):
2082 (WebCore::ContentExtensions::getJumpSize):
2083 (WebCore::ContentExtensions::getJumpDistance):
2084 (WebCore::ContentExtensions::DFABytecodeInterpreter::interpretAppendAction):
2085 (WebCore::ContentExtensions::DFABytecodeInterpreter::interpretTestFlagsAndAppendAction):
2086 (WebCore::ContentExtensions::DFABytecodeInterpreter::actionsForDefaultStylesheetFromDFARoot):
2087 (WebCore::ContentExtensions::DFABytecodeInterpreter::interpret):
2088 * loader/ResourceLoadInfo.h:
2090 2015-06-16 Carlos Alberto Lopez Perez <clopez@igalia.com>
2092 [GTK] [Wayland] Should be possible to build with support for both X11 and Wayland.
2093 https://bugs.webkit.org/show_bug.cgi?id=145701
2095 Reviewed by Darin Adler.
2097 No new tests, no behavior changes.
2099 When building both targets, we have to include the wayland-egl
2100 headers in order to build the Wayland target. This causes that
2101 EGLNativePixmapType and EGLNativeWindowType get defined as
2102 different types than when building only the X11 target.
2104 By type casting them to the ones that are expected, we are able
2105 to build both targets at the same time.
2107 I have done tests (building each target alone as also both targets
2108 at the same time), and everything seems to works as expected.
2110 Once built for both targets, if you try to launch the MiniBrowser
2111 from inside a Wayland compositor (Weston on top of X for example),
2112 it will trigger the X11 target if the DISPLAY environment variable
2113 is set and the environment variable GDK_BACKEND is not set to wayland,
2114 otherwise it will trigger the Wayland target.
2116 * platform/graphics/GLContext.cpp:
2117 (WebCore::GLContext::createContextForWindow): Add type casts. We have
2118 to consider here two different type casts depending on the type of
2119 GLNativeWindowType to avoid a build failure on 32-bits platforms.
2120 The static_cast one was already beeing done as an implicit cast
2121 (from uint64_t to XID), the reinterpret_cast is the new one that
2122 we need to do only when building on both platforms.
2123 * platform/graphics/egl/GLContextEGL.cpp: Add missing include when
2124 building both targets that is required for defining DefaultRootWindow().
2125 (WebCore::GLContextEGL::createPixmapContext): Add type cast.
2127 2015-06-15 Jon Honeycutt <jhoneycutt@apple.com>
2129 [iOS] Crash long pressing on <input type=file>
2130 https://bugs.webkit.org/show_bug.cgi?id=146009
2131 <rdar://problem/21234453>
2133 Reviewed by Ryosuke Niwa.
2136 (WebCore::Position::atStartOfTree):
2137 (WebCore::Position::atEndOfTree):
2138 Null check the container node before passing it to findParent().
2140 2015-06-15 Chris Fleizach <cfleizach@apple.com>
2142 AX: iOS accessibility tests are not running because we need WKTR support
2143 https://bugs.webkit.org/show_bug.cgi?id=145991
2145 Reviewed by Daniel Bates.
2147 Make some minor modifications to support notification handling in WKTR.
2149 * accessibility/ios/AXObjectCacheIOS.mm:
2150 (WebCore::AXObjectCache::postPlatformNotification):
2151 (WebCore::AXObjectCache::postTextStateChangePlatformNotification):
2152 * accessibility/ios/WebAccessibilityObjectWrapperIOS.mm:
2153 (-[WebAccessibilityObjectWrapper accessibilityClickPoint]):
2154 (-[WebAccessibilityObjectWrapper description]):
2155 (-[WebAccessibilityObjectWrapper accessibilitySetPostedNotificationCallback:withContext:]): Deleted.
2156 (-[WebAccessibilityObjectWrapper accessibilityPostedNotification:]): Deleted.
2158 2015-06-16 Mark Lam <mark.lam@apple.com>
2160 Use NakedPtr<Exception>& to return exception results.
2161 https://bugs.webkit.org/show_bug.cgi?id=145870
2163 Reviewed by Anders Carlsson and Filip Pizlo.
2165 No new WebCore tests because this functionality is already covered by existing tests.
2166 API tests added for WTF::NakedPtr.
2168 * bindings/js/JSCallbackData.cpp:
2169 (WebCore::JSCallbackData::invokeCallback):
2170 * bindings/js/JSCustomXPathNSResolver.cpp:
2171 (WebCore::JSCustomXPathNSResolver::lookupNamespaceURI):
2172 * bindings/js/JSErrorHandler.cpp:
2173 (WebCore::JSErrorHandler::handleEvent):
2174 * bindings/js/JSEventListener.cpp:
2175 (WebCore::JSEventListener::handleEvent):
2176 * bindings/js/JSMainThreadExecState.cpp:
2177 (WebCore::JSMainThreadExecState::didLeaveScriptContext):
2178 (WebCore::functionCallHandlerFromAnyThread):
2179 (WebCore::evaluateHandlerFromAnyThread):
2180 * bindings/js/JSMainThreadExecState.h:
2181 (WebCore::JSMainThreadExecState::currentState):
2182 (WebCore::JSMainThreadExecState::call):
2183 (WebCore::JSMainThreadExecState::evaluate):
2184 * bindings/js/JSMutationCallback.cpp:
2185 (WebCore::JSMutationCallback::call):
2186 * bindings/js/ScheduledAction.cpp:
2187 (WebCore::ScheduledAction::executeFunctionInContext):
2188 * bindings/js/ScriptController.cpp:
2189 (WebCore::ScriptController::evaluateInWorld):
2190 * bindings/js/WorkerScriptController.cpp:
2191 (WebCore::WorkerScriptController::evaluate):
2192 (WebCore::WorkerScriptController::setException):
2193 * bindings/js/WorkerScriptController.h:
2194 (WebCore::WorkerScriptController::workerGlobalScopeWrapper):
2195 * bindings/objc/WebScriptObject.mm:
2196 (-[WebScriptObject callWebScriptMethod:withArguments:]):
2197 * workers/WorkerGlobalScope.cpp:
2198 (WebCore::WorkerGlobalScope::importScripts):
2200 2015-06-16 Brent Fulgham <bfulgham@apple.com>
2202 CSS Scroll Snap - support snapping to nested elements
2203 https://bugs.webkit.org/show_bug.cgi?id=145843
2204 <rdar://problem/21339581>
2206 Reviewed by Darin Adler.
2208 Tested by css3/scroll-snap/nested-elements.html
2210 The Scroll Snap Point implementation was not properly handling nested elements.
2211 This could be resolved by recursively calling 'appendChildSnapOffsets', but this
2212 seemed like an inefficient approach, especially considering how often this method
2213 is called during various scaling and other operations.
2215 Instead, do the following:
2216 (1) Add a new HashSet to RenderView that holds a collection of RenderElements that
2217 have scroll-snap-coordinates.
2218 (2) During RenderElement::styleWillChange, register all elements that have the
2219 scroll-snap-coordinates style with the RenderView.
2220 (3) When performing 'appendChildSnapOffsets', refer to the HashSet of elements, select the
2221 subset of these entries relevant to the current scrolling container, and build up the
2222 set of scroll-snap-coordinates needed for the current scrolling container.
2224 * page/scrolling/AxisScrollSnapOffsets.cpp:
2225 (WebCore::appendChildSnapOffsets): Check the scroll-snap-coordinate RenderElement HashSet
2226 for the RenderView to find all elements that are children of the current scrolling container.
2227 Add the scroll-snap-coordinates for these RenderElements to the current set of snap points.
2228 * rendering/RenderElement.cpp:
2229 (WebCore::findEnclosingScrollableContainer): New helper function.
2230 (WebCore::RenderElement::styleWillChange): If the current element has scroll-snap-coordinate
2231 defined, remember it for later so we can use it with the relevant scrolling container
2232 after layout completes.
2233 (WebCore::RenderElement::willBeRemovedFromTree): Unregister the current element from the
2235 (WebCore::RenderElement::findEnclosingScrollableContainer): Added. Locate the relevant
2236 scrolling container for the current object.
2237 * rendering/RenderElement.h:
2238 * rendering/RenderView.cpp:
2239 (WebCore::Document::registerRenderElementWithScrollSnapCoordinates): Added.
2240 (WebCore::Document::unregisterRenderElementWithScrollSnapCoordinates): Added.
2241 * rendering/RenderView.h:
2243 2015-06-16 Brady Eidson <beidson@apple.com>
2245 [IndexedDB] array index keys are concatenated across cursor lifetime
2246 <rdar://problem/19684902> and https://bugs.webkit.org/show_bug.cgi?id=138504
2248 Reviewed by Brady Eidson, patch by Mark Dixon <mark@lowla.io>
2251 storage/indexeddb/keypath-arrays.html
2253 IDBKeyData and IDBKeyPath need to clear any existing array values before calling
2254 decodeObjects to update the value of an existing object.
2256 * Modules/indexeddb/IDBKeyData.cpp:
2257 (WebCore::IDBKeyData::decode):
2258 * Modules/indexeddb/IDBKeyPath.cpp:
2259 (WebCore::IDBKeyPath::decode):
2261 2015-06-16 Said Abou-Hallawa <sabouhallawa@apple.com>
2263 Canvas dimensions should be limited to 4096x4096 pixels on iOS devices.
2264 https://bugs.webkit.org/show_bug.cgi?id=145998
2266 Reviewed by Darin Adler.
2268 The value of MaxCanvasArea should depend on the platform. If the platform
2269 is iOS, the limit should be 64M. Otherwise the limit should be 1G.
2271 Test: fast/canvas/pattern-too-large-to-create-2.html
2273 * html/HTMLCanvasElement.cpp: Change MaxCanvasArea value based on the platform.
2275 * rendering/svg/RenderSVGShape.h:
2276 (WebCore::RenderSVGShape::graphicsElement): Remove un-implemented constructor.
2278 2015-06-16 Chris Dumez <cdumez@apple.com>
2280 REGRESSION(r185012): chat frame in Gmail now says "Something's not right"
2281 https://bugs.webkit.org/show_bug.cgi?id=146025
2282 <rdar://problem/21391412>
2284 Reviewed by Darin Adler.
2286 Only throttle timers in non-visible iframes once they've reached the
2287 max nesting level to avoid throttling critical one-shot timers. This is
2288 consistent with the default DOMTimer throttling behavior that is
2289 defined in the specification.
2291 Power-wise, we are mostly interested in DOMTimers that fire frequently
2292 and cause high CPU usage over an extended period of time anyway.
2295 (WebCore::Document::setTimerThrottlingEnabled):
2296 (WebCore::Document::timerAlignmentInterval):
2298 * dom/ScriptExecutionContext.cpp:
2299 (WebCore::ScriptExecutionContext::timerAlignmentInterval):
2300 * dom/ScriptExecutionContext.h:
2301 * page/DOMTimer.cpp:
2302 (WebCore::DOMTimer::alignedFireTime):
2304 2015-06-16 sylvain-galineau <galineau@adobe.com>
2306 Incorrect order of arguments in initial-letter property
2307 https://bugs.webkit.org/show_bug.cgi?id=139667
2309 Reviewed by Sam Weinig.
2311 The CSS specification swapped the order of the initial-letters numeric values.
2312 The drop cap's height now comes first, followed by its optional vertical position.
2313 See http://www.w3.org/TR/css-inline/#sizing-drop-initials.
2315 No new tests. Existing tests updated.
2317 * css/CSSParser.cpp:
2318 (WebCore::CSSParser::parseValue): swap arguments to reflect new spec order.
2320 2015-06-16 Alex Christensen <achristensen@webkit.org>
2322 Remove some unused values.
2323 https://bugs.webkit.org/show_bug.cgi?id=145997
2325 Reviewed by Gyuyoung Kim.
2327 This patch should have no change in behavior.
2329 * accessibility/AccessibilityObject.cpp:
2330 (WebCore::computeBestScrollOffset):
2331 (WebCore::AccessibilityObject::scrollToMakeVisibleWithSubFocus):
2332 (WebCore::AccessibilityObject::scrollToGlobalPoint):
2333 * accessibility/mac/WebAccessibilityObjectWrapperMac.mm:
2334 (-[WebAccessibilityObjectWrapper accessibilityAttributeValue:forParameter:]):
2335 * html/canvas/WebGL2RenderingContext.cpp:
2336 (WebCore::WebGL2RenderingContext::validateTexFuncData):
2337 * html/canvas/WebGLRenderingContext.cpp:
2338 (WebCore::WebGLRenderingContext::validateTexFuncData):
2339 * platform/graphics/StringTruncator.cpp:
2340 (WebCore::leftTruncateToBuffer):
2341 * platform/graphics/mac/MediaPlayerPrivateQTKit.mm:
2342 (WebCore::MediaPlayerPrivateQTKit::paintCurrentFrameInContext):
2343 * rendering/InlineTextBox.cpp:
2344 (WebCore::InlineTextBox::localSelectionRect):
2345 * rendering/RenderElement.cpp:
2346 (WebCore::RenderElement::anchorRect):
2347 * rendering/SimpleLineLayoutTextFragmentIterator.cpp:
2348 (WebCore::SimpleLineLayout::TextFragmentIterator::skipToNextPosition):
2349 * rendering/svg/SVGTextQuery.cpp:
2350 (WebCore::SVGTextQuery::modifyStartEndPositionsRespectingLigatures):
2351 Remove unused values.
2353 2015-06-16 Youenn Fablet <youenn.fablet@crf.canon.fr> and Xabier Rodriguez Calvar <calvaris@igalia.com>
2355 [Streams API] Calling controller.error() should trigger storing an undefined error
2356 https://bugs.webkit.org/show_bug.cgi?id=145976
2358 Reviewed by Darin Adler.
2360 Covered by rebased test.
2362 * bindings/js/JSReadableStreamControllerCustom.cpp:
2363 (WebCore::JSReadableStreamController::error): Storing undefined if no error value passed.
2364 * bindings/js/ReadableJSStream.cpp: Removed storeError(ExecState*).
2365 (WebCore::ReadableJSStream::ReadableJSStream):
2367 2015-06-16 Chris Dumez <cdumez@apple.com>
2369 Purge StyledElement's presentation attribute cache on memory pressure
2370 https://bugs.webkit.org/show_bug.cgi?id=145999
2371 <rdar://problem/21359252>
2373 Reviewed by Andreas Kling.
2375 Purge StyledElement's presentation attribute cache on memory pressure.
2377 * dom/StyledElement.cpp:
2378 (WebCore::presentationAttributeCache):
2379 (WebCore::presentationAttributeCacheCleaner):
2380 (WebCore::StyledElement::clearPresentationAttributeCache):
2381 * dom/StyledElement.h:
2382 * platform/MemoryPressureHandler.cpp:
2383 (WebCore::MemoryPressureHandler::releaseNoncriticalMemory):
2385 2015-06-15 Brent Fulgham <bfulgham@apple.com>
2387 REGRESSION(r175251, Mavericks Only): Playback may stall
2388 https://bugs.webkit.org/show_bug.cgi?id=145989
2389 <rdar://problem/21271919>
2391 Unreviewed post-review correction.
2393 Dave Kilzer pointed out that the macro around the waitForVideoOutputMediaDataWillChange
2394 call was incorrect. This patch corrects this error.
2396 * platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.mm:
2397 (WebCore::MediaPlayerPrivateAVFoundationObjC::createVideoOutput): Correct the
2400 2015-06-15 Chris Fleizach <cfleizach@apple.com>
2402 AX: Changing state of radio buttons causes VoiceOver to go busy for a short time.
2403 https://bugs.webkit.org/show_bug.cgi?id=145933
2405 Reviewed by Dean Jackson.
2407 When radio buttons animate the new focus selection state, the thread activity looks a lot like short burst of
2408 activity to draw, then wait on CoreAnimation to apply those changes.
2410 Since those periods of activity during animation are so short, VoiceOver is not able to query for all the
2411 attributes it needs, and gets stuck in the queue behind rendering.
2413 The fix here is to turn off button state animations while VoiceOver is running.
2415 * platform/mac/ThemeMac.mm:
2416 (WebCore::updateStates):
2418 2015-06-15 Zalan Bujtas <zalan@apple.com>
2420 RootInlineBox::m_lineBreakObj becomes invalid when a child renderer is removed and the line does not get marked dirty.
2421 https://bugs.webkit.org/show_bug.cgi?id=145988
2422 rdar://problem/20959137
2424 Reviewed by David Hyatt.
2426 This patch ensures that we find the right first inline box so that we can dirty the
2427 the appropriate line boxes.
2428 With marking the right line boxes dirty, now we can update RootInlineBox::m_lineBreakObj at the next layout.
2430 Test: fast/inline/crash-when-child-renderer-is-removed-and-line-stays-clean.html
2432 * rendering/RenderInline.cpp:
2433 (WebCore::RenderInline::culledInlineFirstLineBox):
2434 (WebCore::RenderInline::culledInlineLastLineBox):
2435 * rendering/RootInlineBox.cpp:
2436 (WebCore::RootInlineBox::setLineBreakInfo): Deleted. Remove misleading assert and comment.
2438 2015-06-15 Matt Rajca <mrajca@apple.com>
2440 Media Session: Improve the safety of playback toggling
2441 https://bugs.webkit.org/show_bug.cgi?id=145986
2443 Reviewed by Darin Adler.
2445 * Modules/mediasession/MediaSession.cpp:
2446 (WebCore::MediaSession::togglePlayback): Improved the safety of the loop so that we don't re-visit elements that
2447 may have been deleted underneath us.
2448 * Modules/mediasession/MediaSession.h: Added a pointer to the set of iterated active participating elements so
2449 we can remove any elements that are deleted from the underlying "real" set.
2451 2015-06-15 Brent Fulgham <bfulgham@apple.com>
2453 REGRESSION(r175251, Mavericks Only): Playback may stall
2454 https://bugs.webkit.org/show_bug.cgi?id=145989
2455 <rdar://problem/21271919>
2457 Reviewed by Dean Jackson.
2459 Revert r175251 for Mavericks build targets.
2461 * platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.mm:
2462 (WebCore::MediaPlayerPrivateAVFoundationObjC::createVideoLayer):
2463 (WebCore::MediaPlayerPrivateAVFoundationObjC::createVideoOutput):
2464 (WebCore::MediaPlayerPrivateAVFoundationObjC::paintWithVideoOutput):
2466 2015-06-15 Darin Adler <darin@apple.com>
2468 REGRESSION (r182215): Reproducible crash at drawsvg.org due to reentrant layout
2469 https://bugs.webkit.org/show_bug.cgi?id=145964
2471 Reviewed by Simon Fraser.
2473 Test: svg/as-object/mutate-on-load.html
2475 * page/FrameView.cpp:
2476 (WebCore::FrameView::forceLayoutParentViewIfNeeded): Don't do a synchronous layout here,
2477 because it can lead indirectly to unwanted layout reentrancy. Instead schedule a layout.
2479 2015-06-15 Matt Rajca <mrajca@apple.com>
2481 Media Session: Active participating elements can change while being iterated
2482 https://bugs.webkit.org/show_bug.cgi?id=145978
2484 Reviewed by Alex Christensen.
2486 * Modules/mediasession/MediaSession.cpp:
2487 (WebCore::MediaSession::togglePlayback): Iterate through a copy of m_activeParticipatingElements since its contents
2488 can be modified in the loop.
2490 2015-06-15 Chris Fleizach <cfleizach@apple.com>
2492 AX: no accessibility support for details element
2493 https://bugs.webkit.org/show_bug.cgi?id=131111
2495 Reviewed by Darin Adler.
2497 Add accessibility support for Mac for details element by:
2498 1) Returning new subroles for <details> and <summary>
2499 2) Exposing isExpanded property for <details> element.
2501 Test: platform/mac/accessibility/details-summary.html
2503 * accessibility/AccessibilityObject.cpp:
2504 (WebCore::AccessibilityObject::supportsARIAPressed):
2505 (WebCore::AccessibilityObject::supportsExpanded):
2506 (WebCore::AccessibilityObject::isExpanded):
2507 (WebCore::AccessibilityObject::supportsARIAExpanded): Deleted.
2508 * accessibility/AccessibilityObject.h:
2509 (WebCore::AccessibilityObject::canvasHasFallbackContent):
2510 * accessibility/AccessibilityRenderObject.cpp:
2511 (WebCore::AccessibilityRenderObject::computeAccessibilityIsIgnored):
2512 (WebCore::AccessibilityRenderObject::determineAccessibilityRole):
2513 * accessibility/ios/WebAccessibilityObjectWrapperIOS.mm:
2514 (-[WebAccessibilityObjectWrapper accessibilitySupportsARIAExpanded]):
2515 (-[WebAccessibilityObjectWrapper accessibilityIsExpanded]):
2516 * accessibility/mac/WebAccessibilityObjectWrapperMac.mm:
2517 (-[WebAccessibilityObjectWrapper additionalAccessibilityAttributeNames]):
2518 (createAccessibilityRoleMap):
2519 (-[WebAccessibilityObjectWrapper subrole]):
2520 * html/HTMLDetailsElement.h:
2521 * inspector/InspectorDOMAgent.cpp:
2522 (WebCore::InspectorDOMAgent::buildObjectForAccessibilityProperties):
2524 2015-06-15 Alex Christensen <achristensen@webkit.org>
2526 [Content Extensions] Limit number of rules.
2527 https://bugs.webkit.org/show_bug.cgi?id=145663
2529 Reviewed by Benjamin Poulain.
2531 Added an API test to make sure that parsing fails when there are too many rules.
2533 * contentextensions/ContentExtensionError.cpp:
2534 (WebCore::ContentExtensions::contentExtensionErrorCategory):
2535 * contentextensions/ContentExtensionError.h:
2536 * contentextensions/ContentExtensionParser.cpp:
2537 (WebCore::ContentExtensions::loadEncodedRules):
2538 Fail to parse a content extension with more than 50000 rules.
2540 2015-06-12 Alexey Proskuryakov <ap@apple.com>
2542 -[WKWebView evaluateJavaScript] provides a misleading error when the return cannot be serialized
2543 https://bugs.webkit.org/show_bug.cgi?id=145900
2545 Reviewed by Sam Weinig.
2547 * English.lproj/Localizable.strings:
2549 2015-06-15 Carlos Garcia Campos <cgarcia@igalia.com>
2551 [SOUP] Custom URI schemes don't work for requests containing a fragment identifier
2552 https://bugs.webkit.org/show_bug.cgi?id=145969
2554 Reviewed by Sergio Villar Senin.
2556 For URIs like foo:bar#baz, what the custom protocol manager
2557 receives in the UI process is foo:bar, so the user can't handle fragments.
2559 * platform/network/soup/ResourceRequestSoup.cpp:
2560 (WebCore::ResourceRequest::updateSoupRequest): If the SoupRequest
2561 is a WebKitSoupRequestGeneric, call
2562 webkitSoupRequestGenericSetRequest with the ResourceRequest.
2563 * platform/network/soup/WebKitSoupRequestGeneric.cpp:
2564 (webkitSoupRequestGenericSetRequest):
2565 (webkitSoupRequestGenericGetRequest):
2566 * platform/network/soup/WebKitSoupRequestGeneric.h:
2568 2015-06-15 Carlos Garcia Campos <cgarcia@igalia.com>
2570 [SOUP] Move WebKitSoupRequestGeneric to platform layer
2571 https://bugs.webkit.org/show_bug.cgi?id=145968
2573 Reviewed by Sergio Villar Senin.
2575 * PlatformEfl.cmake:
2576 * PlatformGTK.cmake:
2577 * platform/network/soup/WebKitSoupRequestGeneric.cpp: Renamed from Source/WebKit2/WebProcess/soup/WebKitSoupRequestGeneric.cpp.
2578 (webkitSoupRequestGenericFinalize):
2579 (webkit_soup_request_generic_init):
2580 (webkitSoupRequestGenericSendAsync):
2581 (webkitSoupRequestGenericSendFinish):
2582 (webkitSoupRequestGenericGetContentLength):
2583 (webkitSoupRequestGenericGetContentType):
2584 (webkit_soup_request_generic_class_init):
2585 (webkitSoupRequestGenericSetContentLength):
2586 (webkitSoupRequestGenericSetContentType):
2587 * platform/network/soup/WebKitSoupRequestGeneric.h: Renamed from Source/WebKit2/WebProcess/soup/WebKitSoupRequestGeneric.h.
2588 * platform/network/soup/WebKitSoupRequestGenericClient.h: Renamed from Source/WebKit2/WebProcess/soup/WebKitSoupRequestGenericClient.h.
2590 2015-06-13 Chris Dumez <cdumez@apple.com>
2592 [WK2] API::Navigation objects are leaked on history navigation to HistoryItems in PageCache
2593 https://bugs.webkit.org/show_bug.cgi?id=145948
2595 Reviewed by Darin Adler.
2597 API::Navigation objects were leaked on history navigation to
2598 HistoryItems in PageCache. In such case, we would create 2 Navigation
2599 objects instead of 1 and the first one would be leaked. The reason
2600 we create the second one is because we fail to pass along the
2601 navigationID from the UIProcess to the WebProcess and then back to the
2602 UIProcess. On the IPC back to the UIProcess, the navigationID ends up
2603 being 0 so the UIProcess creates a new Navigation object, thinking that
2604 the load was triggered by the WebContent process.
2606 We now pass along the navigationID, even if the HistoryItem is in the
2607 PageCache and we end up reusing the cached DocumentLoader, instead of
2608 creating a new one. A new updateCachedDocumentLoader() delegate is
2609 added to the FrameLoaderClient, similarly to the pre-existing
2610 createDocumentLoader() but for the case where the DocumentLoader gets
2613 * loader/EmptyClients.h:
2614 * loader/FrameLoader.cpp:
2615 (WebCore::FrameLoader::loadDifferentDocumentItem):
2616 * loader/FrameLoaderClient.h:
2618 2015-06-13 Xabier Rodriguez Calvar <calvaris@igalia.com> and Youenn Fablet <youenn.fablet@crf.canon.fr>
2620 [Streams API] ReadableJSStream should handle promises returned by JS source start callback
2621 https://bugs.webkit.org/show_bug.cgi?id=145792
2623 Reviewed by Darin Adler.
2625 Covered by rebased tests.
2627 When calling start callback, the returned value is checked.
2628 If it is not a promise, we do as if it is a resolved promise.
2629 If it is a promise, we call its then() method with two resolve/reject JS functions.
2631 * Modules/streams/ReadableStream.cpp:
2632 * bindings/js/ReadableJSStream.cpp:
2633 (WebCore::ReadableJSStream::invoke): Returns a JSPromise* if any is returned by the JS source callback.
2634 (WebCore::thenPromise): Utility method to call the promise.
2635 (WebCore::createStartResultFulfilledFunction): The promise resolve callback.
2636 (WebCore::ReadableJSStream::doStart): Calls thenPromise if a JSPromise* is returned by invoke.
2637 (WebCore::ReadableJSStream::ReadableJSStream):
2638 * bindings/js/ReadableJSStream.h:
2640 2015-06-13 Andres Gonzalez <agonzalez334@nc.rr.com>
2642 AX: WebKit exposes all Ruby Text as Unknown (Japanese EPUB accessibility blocker)
2643 https://bugs.webkit.org/show_bug.cgi?id=141303
2645 Reviewed by Chris Fleizach.
2647 Test: accessibility/ruby-hierarchy-roles.html
2649 * accessibility/AccessibilityObject.h:
2650 * accessibility/AccessibilityRenderObject.cpp:
2651 (WebCore::AccessibilityRenderObject::computeAccessibilityIsIgnored):
2652 (WebCore::AccessibilityRenderObject::determineAccessibilityRole):
2653 * accessibility/mac/WebAccessibilityObjectWrapperMac.mm:
2654 (createAccessibilityRoleMap):
2655 (-[WebAccessibilityObjectWrapper subrole]):
2657 2015-06-12 Chris Fleizach <cfleizach@apple.com>
2659 AX: iOS: after updating control, VoiceOver speaks aria-expanded states reversed (says "collapsed" when "expanded")
2660 https://bugs.webkit.org/show_bug.cgi?id=145943
2662 Reviewed by Darin Adler.
2664 iOS Accessibility platform needs to be notified of when aria expanded changes.
2666 * accessibility/ios/AXObjectCacheIOS.mm:
2667 (WebCore::AXObjectCache::postPlatformNotification):
2668 * accessibility/ios/WebAccessibilityObjectWrapperIOS.h:
2669 * accessibility/ios/WebAccessibilityObjectWrapperIOS.mm:
2670 (-[WebAccessibilityObjectWrapper postValueChangedNotification]):
2671 (-[WebAccessibilityObjectWrapper postExpandedChangedNotification]):
2672 (-[WebAccessibilityObjectWrapper postScrollStatusChangeNotification]):
2674 2015-06-12 Chris Fleizach <cfleizach@apple.com>
2676 AX:ARIA Toggle buttons aren't properly conveyed on iOS using VoiceOver
2677 https://bugs.webkit.org/show_bug.cgi?id=145949
2679 Reviewed by Darin Adler.
2681 Expose pressed state information to the iOS platform AX API.
2683 Test: Updated inspector-protocol/dom/getAccessibilityPropertiesForNode.html
2684 There was a FIXME for this issue in that test: https://bugs.webkit.org/show_bug.cgi?id=129830
2686 * accessibility/AccessibilityNodeObject.cpp:
2687 (WebCore::AccessibilityNodeObject::isPressed):
2688 * accessibility/AccessibilityObject.cpp:
2689 (WebCore::AccessibilityObject::classList):
2690 (WebCore::AccessibilityObject::supportsARIAPressed):
2691 (WebCore::AccessibilityObject::supportsARIAExpanded):
2692 * accessibility/AccessibilityObject.h:
2693 * accessibility/ios/WebAccessibilityObjectWrapperIOS.mm:
2694 (-[WebAccessibilityObjectWrapper accessibilityARIALiveRegionIsAtomic]):
2695 (-[WebAccessibilityObjectWrapper accessibilitySupportsARIAPressed]):
2696 (-[WebAccessibilityObjectWrapper accessibilityIsPressed]):
2697 (-[WebAccessibilityObjectWrapper accessibilitySupportsARIAExpanded]):
2699 2015-06-12 Gyuyoung Kim <gyuyoung.kim@webkit.org>
2701 Purge PassRefPtr in JavaScriptCore - 2
2702 https://bugs.webkit.org/show_bug.cgi?id=145834
2704 Reviewed by Darin Adler.
2706 Fix call sites depends on changing of JSC.
2708 * html/canvas/WebGL2RenderingContext.cpp:
2709 (WebCore::WebGL2RenderingContext::getParameter):
2710 * html/canvas/WebGLRenderingContext.cpp:
2711 (WebCore::WebGLRenderingContext::getParameter):
2712 * html/canvas/WebGLRenderingContextBase.cpp:
2713 (WebCore::WebGLRenderingContextBase::getUniform):
2714 (WebCore::WebGLRenderingContextBase::getVertexAttrib):
2715 (WebCore::WebGLRenderingContextBase::getWebGLFloatArrayParameter):
2716 (WebCore::WebGLRenderingContextBase::getWebGLIntArrayParameter):
2718 2015-06-12 Zalan Bujtas <zalan@apple.com>
2720 Be more defensive at renderer type checking when initializing flow segments.
2721 https://bugs.webkit.org/show_bug.cgi?id=145942
2723 Reviewed by Simon Fraser.
2725 FlowContents::initializeSegments should ignore unsupported renderers so that when we miss
2726 a simple line layout path invalidation, we don't downcast the unsupported renderer to RenderText.
2728 I have not reproduced this issue (but related to rdar://problem/21312452)
2730 * rendering/SimpleLineLayoutFlowContents.cpp:
2731 (WebCore::SimpleLineLayout::initializeSegments):
2733 2015-06-12 Anders Carlsson <andersca@apple.com>
2735 deleteEmptyDirectory should delete .DS_Store files on OS X
2736 https://bugs.webkit.org/show_bug.cgi?id=145944
2738 Reviewed by Dan Bernstein.
2740 deleteEmptyDirectory is often used when clearing website data, so we should
2741 take extra care to delete empty directories so the user won't think that clearing
2742 website data didn't do anything because it would leave directories with .DS_Store
2745 * platform/mac/FileSystemMac.mm:
2746 (WebCore::deleteEmptyDirectory):
2747 * platform/posix/FileSystemPOSIX.cpp:
2749 2015-06-12 Manuel Rego Casasnovas <rego@igalia.com>
2751 [CSS Grid Layout] Fix grid-template-areas parsing to avoid spaces
2752 https://bugs.webkit.org/show_bug.cgi?id=145860
2754 Reviewed by Sergio Villar Senin.
2756 The spec doesn't require to have spaces between unnamed and named areas
2757 in grid-template-areas syntax. But spaces are currently required in our
2760 This was confirmed in the CSS WG mailing list:
2761 https://lists.w3.org/Archives/Public/www-style/2015May/0239.html
2763 This patch updates grid-template-areas parsing to allow the possibility
2764 of removing spaces between unnamed and named areas.
2766 Added new cases to fast/css-grid-layout/grid-template-areas-get-set.html.
2768 * css/CSSParser.cpp:
2769 (WebCore::parseGridTemplateAreasColumnNames): New helper method to
2770 determine the column names split by white spaces or dots.
2771 (WebCore::CSSParser::parseGridTemplateAreasRow): Use the new helper
2772 method to get the column names.
2773 (WebCore::containsOnlyDots): Deleted. Not needed anymore as
2774 parseGridTemplateAreasColumnNames() is using a single dot for unnamed
2775 grid areas (despite of being defined with 1 or more dots).
2777 2015-06-12 Eric Carlson <eric.carlson@apple.com>
2779 [Mac] AirPlay menu button doesn't always show on page load
2780 https://bugs.webkit.org/show_bug.cgi?id=145936
2782 Reviewed by Brent Fulgham.
2784 * Modules/mediasession/WebMediaSessionManager.cpp:
2785 (WebCore::mediaProducerStateString): New, return a string representing MediaStateFlags.
2786 (WebCore::WebMediaSessionManager::clientStateDidChange): Log the states as strings.
2787 (WebCore::WebMediaSessionManager::toString): New, return a string representing ConfigurationTasks.
2788 (WebCore::WebMediaSessionManager::scheduleDelayedTask): Add logging.
2789 (WebCore::WebMediaSessionManager::taskTimerFired): Add logging.
2790 * Modules/mediasession/WebMediaSessionManager.h:
2792 * platform/graphics/avfoundation/objc/MediaPlayerPrivateMediaSourceAVFObjC.h: Override wirelessVideoPlaybackDisabled
2793 so it is possible to activate route monitoring for a movie loaded with this engine.
2794 * platform/graphics/mac/MediaPlayerPrivateQTKit.h: Ditto.
2796 2015-06-12 Zan Dobersek <zdobersek@igalia.com>
2798 Add the remaining missing override specifiers under Source/WebCore/
2799 https://bugs.webkit.org/show_bug.cgi?id=145907
2801 Reviewed by Darin Adler.
2803 Fix the remaining compiler warnings about missing override specifiers
2804 for overriding method declarations in classes under Source/WebCore/.
2806 Where the addition had to be performed on all virtual methods of the
2807 class, the virtual specifier was removed so now only the override
2808 specifier is in use.
2810 * Modules/indexeddb/IDBOpenDBRequest.h:
2811 * Modules/indexeddb/IDBRequest.h:
2812 * Modules/webdatabase/DatabaseServer.h:
2813 * bindings/js/ReadableJSStream.h:
2814 * editing/InsertTextCommand.h:
2815 * mathml/MathMLInlineContainerElement.h:
2816 * platform/audio/EqualPowerPanner.h:
2817 (WebCore::EqualPowerPanner::reset):
2818 * platform/graphics/MediaPlayer.cpp:
2819 (WebCore::NullMediaPlayerPrivate::load):
2820 (WebCore::NullMediaPlayerPrivate::cancelLoad):
2821 (WebCore::NullMediaPlayerPrivate::prepareToPlay):
2822 (WebCore::NullMediaPlayerPrivate::play):
2823 (WebCore::NullMediaPlayerPrivate::pause):
2824 (WebCore::NullMediaPlayerPrivate::platformMedia):
2825 (WebCore::NullMediaPlayerPrivate::platformLayer):
2826 (WebCore::NullMediaPlayerPrivate::naturalSize):
2827 (WebCore::NullMediaPlayerPrivate::hasVideo):
2828 (WebCore::NullMediaPlayerPrivate::hasAudio):
2829 (WebCore::NullMediaPlayerPrivate::setVisible):
2830 (WebCore::NullMediaPlayerPrivate::durationDouble):
2831 (WebCore::NullMediaPlayerPrivate::currentTimeDouble):
2832 (WebCore::NullMediaPlayerPrivate::seekDouble):
2833 (WebCore::NullMediaPlayerPrivate::seeking):
2834 (WebCore::NullMediaPlayerPrivate::setRateDouble):
2835 (WebCore::NullMediaPlayerPrivate::setPreservesPitch):
2836 (WebCore::NullMediaPlayerPrivate::paused):
2837 (WebCore::NullMediaPlayerPrivate::setVolumeDouble):
2838 (WebCore::NullMediaPlayerPrivate::supportsMuting):
2839 (WebCore::NullMediaPlayerPrivate::setMuted):
2840 (WebCore::NullMediaPlayerPrivate::hasClosedCaptions):
2841 (WebCore::NullMediaPlayerPrivate::setClosedCaptionsVisible):
2842 (WebCore::NullMediaPlayerPrivate::networkState):
2843 (WebCore::NullMediaPlayerPrivate::readyState):
2844 (WebCore::NullMediaPlayerPrivate::maxTimeSeekableDouble):
2845 (WebCore::NullMediaPlayerPrivate::minTimeSeekable):
2846 (WebCore::NullMediaPlayerPrivate::buffered):
2847 (WebCore::NullMediaPlayerPrivate::totalBytes):
2848 (WebCore::NullMediaPlayerPrivate::didLoadingProgress):
2849 (WebCore::NullMediaPlayerPrivate::setSize):
2850 (WebCore::NullMediaPlayerPrivate::canLoadPoster):
2851 (WebCore::NullMediaPlayerPrivate::setPoster):
2852 (WebCore::NullMediaPlayerPrivate::hasSingleSecurityOrigin):
2853 * platform/graphics/filters/DistantLightSource.h:
2854 * platform/graphics/filters/FEComposite.h:
2855 * platform/graphics/filters/FEDisplacementMap.h:
2856 (WebCore::FEDisplacementMap::determineAbsolutePaintRect):
2857 * platform/graphics/filters/FEFlood.h:
2858 (WebCore::FEFlood::determineAbsolutePaintRect):
2859 * platform/graphics/filters/PointLightSource.h:
2860 * platform/graphics/filters/SpotLightSource.h:
2861 * platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.h:
2862 (WebCore::MediaPlayerPrivateGStreamer::hasVideo):
2863 (WebCore::MediaPlayerPrivateGStreamer::hasAudio):
2864 (WebCore::MediaPlayerPrivateGStreamer::audioSourceProvider):
2865 (WebCore::MediaPlayerPrivateGStreamer::engineDescription):
2866 (WebCore::MediaPlayerPrivateGStreamer::isLiveStream):
2867 (WebCore::MediaPlayerPrivateGStreamer::totalVideoFrames):
2868 (WebCore::MediaPlayerPrivateGStreamer::droppedVideoFrames):
2869 (WebCore::MediaPlayerPrivateGStreamer::corruptedVideoFrames):
2870 (WebCore::MediaPlayerPrivateGStreamer::totalFrameDelay):
2871 * platform/graphics/texmap/BitmapTextureGL.h:
2872 (WebCore::BitmapTextureGL::isBackedByOpenGL):
2873 * platform/graphics/transforms/Matrix3DTransformOperation.h:
2874 * platform/graphics/transforms/MatrixTransformOperation.h:
2875 * platform/graphics/transforms/PerspectiveTransformOperation.h:
2876 * platform/graphics/transforms/RotateTransformOperation.h:
2877 * platform/graphics/transforms/ScaleTransformOperation.h:
2878 * platform/graphics/transforms/SkewTransformOperation.h:
2879 * platform/image-decoders/png/PNGImageDecoder.h:
2880 (WebCore::PNGImageDecoder::filenameExtension):
2881 * platform/mediastream/openwebrtc/RealtimeMediaSourceCenterOwr.h:
2882 * platform/mock/MockRealtimeMediaSourceCenter.h:
2884 2015-06-12 Commit Queue <commit-queue@webkit.org>
2886 Unreviewed, rolling out r185512.
2887 https://bugs.webkit.org/show_bug.cgi?id=145932
2889 Many asserts on layout tests and on API tests (Requested by
2894 "Use modern for-loops in WebCore/rendering - 1"
2895 https://bugs.webkit.org/show_bug.cgi?id=145831
2896 http://trac.webkit.org/changeset/185512
2898 2015-06-12 Simon Fraser <simon.fraser@apple.com>
2900 OpenGLESSPI.h doesn't need to include UIKit.h
2901 https://bugs.webkit.org/show_bug.cgi?id=145931
2903 Reviewed by Darin Adler.
2905 No need to suck in all of UIKit from OpenGLESSPI.h.
2907 * platform/spi/ios/OpenGLESSPI.h:
2909 2015-06-12 Matt Rajca <mrajca@apple.com>
2911 Add barebones implementation of media session invocation algorithm.
2912 https://bugs.webkit.org/show_bug.cgi?id=145847
2914 Reviewed by Darin Adler.
2916 * Modules/mediasession/MediaSession.cpp:
2917 (WebCore::MediaSession::invoke): Move the media session to an active state.
2918 * Modules/mediasession/MediaSession.h:
2919 * html/HTMLMediaElement.cpp:
2920 (WebCore::HTMLMediaElement::playInternal): Call the media session invocation algorithm as described in the
2923 2015-06-12 Hunseop Jeong <hs85.jeong@samsung.com>
2925 Use modern for-loops in WebCore/rendering - 1
2926 https://bugs.webkit.org/show_bug.cgi?id=145831
2928 Reviewed by Darin Adler.
2930 No new tests because there is no behavior change.
2932 * rendering/AutoTableLayout.cpp:
2933 (WebCore::AutoTableLayout::computeIntrinsicLogicalWidths):
2934 (WebCore::AutoTableLayout::calcEffectiveLogicalWidth):
2935 * rendering/FilterEffectRenderer.cpp:
2936 (WebCore::FilterEffectRenderer::build):
2937 (WebCore::FilterEffectRenderer::clearIntermediateResults):
2938 * rendering/FilterEffectRenderer.h:
2939 * rendering/FloatingObjects.cpp:
2940 (WebCore::FloatingObjects::clearLineBoxTreePointers):
2941 (WebCore::FloatingObjects::moveAllToFloatInfoMap):
2942 (WebCore::FloatingObjects::computePlacedFloatsTree):
2943 * rendering/FlowThreadController.cpp:
2944 (WebCore::FlowThreadController::collectFixedPositionedLayers):
2945 * rendering/HitTestResult.cpp:
2946 (WebCore::HitTestResult::append):
2947 * rendering/ImageQualityController.cpp:
2948 (WebCore::ImageQualityController::highQualityRepaintTimerFired):
2949 * rendering/InlineTextBox.cpp:
2950 (WebCore::InlineTextBox::paint):
2951 * rendering/RenderBlock.cpp:
2952 (WebCore::removeBlockFromDescendantAndContainerMaps):
2953 (WebCore::RenderBlock::addOverflowFromPositionedObjects):
2954 (WebCore::RenderBlock::dirtyForLayoutFromPercentageHeightDescendants):
2955 (WebCore::RenderBlock::simplifiedNormalFlowLayout):
2956 (WebCore::RenderBlock::layoutPositionedObjects):
2957 (WebCore::RenderBlock::markPositionedObjectsForLayout):
2958 (WebCore::RenderBlock::paintContinuationOutlines):
2959 (WebCore::clipOutPositionedObjects):
2960 (WebCore::RenderBlock::removeFromTrackedRendererMaps):
2961 (WebCore::RenderBlock::removePositionedObjects):
2962 (WebCore::RenderBlock::checkPositionedObjectsNeedLayout):
2963 * rendering/RenderBlockFlow.cpp:
2964 (WebCore::RenderBlockFlow::rebuildFloatingObjectSetFromIntrudingFloats):
2965 (WebCore::RenderBlockFlow::styleDidChange):
2966 (WebCore::RenderBlockFlow::moveFloatsTo):
2967 (WebCore::RenderBlockFlow::addOverflowFromFloats):
2968 (WebCore::RenderBlockFlow::repaintOverhangingFloats):
2969 (WebCore::RenderBlockFlow::paintFloats):
2970 (WebCore::RenderBlockFlow::clipOutFloatingObjects):
2971 (WebCore::RenderBlockFlow::lowestFloatLogicalBottom):
2972 (WebCore::RenderBlockFlow::lowestInitialLetterLogicalBottom):
2973 (WebCore::RenderBlockFlow::addOverhangingFloats):
2974 (WebCore::RenderBlockFlow::addIntrudingFloats):
2975 (WebCore::RenderBlockFlow::markSiblingsWithFloatsForLayout):
2976 (WebCore::RenderBlockFlow::adjustForBorderFit):
2977 * rendering/RenderBlockLineLayout.cpp:
2978 (WebCore::setLogicalWidthForTextRun):
2979 (WebCore::RenderBlockFlow::layoutRunsAndFloatsInRange):
2980 (WebCore::RenderBlockFlow::linkToEndLineIfNeeded):
2981 (WebCore::RenderBlockFlow::repaintDirtyFloats):
2982 (WebCore::RenderBlockFlow::layoutLineBoxes):
2983 (WebCore::RenderBlockFlow::checkFloatsInCleanLine):
2984 (WebCore::RenderBlockFlow::determineStartPosition):
2985 (WebCore::RenderBlockFlow::checkPaginationAndFloatsAtEndLine):
2986 * rendering/RenderCounter.cpp:
2987 (WebCore::RenderCounter::destroyCounterNodes):
2988 (WebCore::updateCounters):
2989 (WebCore::RenderCounter::rendererStyleChanged):
2990 * rendering/RenderFlexibleBox.cpp:
2991 (WebCore::RenderFlexibleBox::autoMarginOffsetInMainAxis):
2992 (WebCore::RenderFlexibleBox::freezeViolations):
2993 (WebCore::RenderFlexibleBox::resolveFlexibleLengths):
2994 (WebCore::RenderFlexibleBox::numberOfInFlowPositionedChildren):
2995 (WebCore::RenderFlexibleBox::layoutColumnReverse):
2996 (WebCore::RenderFlexibleBox::alignFlexLines):
2997 (WebCore::RenderFlexibleBox::alignChildren):
2998 (WebCore::RenderFlexibleBox::flipForWrapReverse):
2999 * rendering/RenderFlowThread.cpp:
3000 (WebCore::RenderFlowThread::updateAllLayerToRegionMappings):
3001 * rendering/RenderInline.cpp:
3002 (WebCore::RenderInline::addAnnotatedRegions):
3003 * rendering/RenderLayer.cpp:
3004 (WebCore::RenderLayer::updateDescendantsAreContiguousInStackingOrder):
3005 (WebCore::RenderLayer::updateDescendantDependentFlags):
3006 * rendering/RenderLayerBacking.cpp:
3007 (WebCore::descendantLayerPaintsIntoAncestor):
3008 (WebCore::RenderLayerBacking::startAnimation):
3009 * rendering/RenderLineBoxList.cpp:
3010 (WebCore::RenderLineBoxList::paint):
3011 * rendering/RenderListBox.cpp:
3012 (WebCore::RenderListBox::updateFromElement):
3013 * rendering/RenderMenuList.cpp:
3014 (WebCore::selectedOptionCount):
3015 (RenderMenuList::updateOptionsWidth):
3016 * rendering/RenderMultiColumnSet.cpp:
3017 (WebCore::RenderMultiColumnSet::distributeImplicitBreaks):
3019 2015-06-12 Csaba Osztrogonác <ossy@webkit.org>
3021 Fix unused private field warning in ResourceHandleSoup.cpp
3022 https://bugs.webkit.org/show_bug.cgi?id=145910
3024 Reviewed by Darin Adler.
3026 * platform/network/soup/ResourceHandleSoup.cpp:
3028 2015-06-12 Joseph Pecoraro <pecoraro@apple.com>
3030 Web Inspector: CRASH trying to inspect text that was removed/replaced
3031 https://bugs.webkit.org/show_bug.cgi?id=145898
3033 Reviewed by Darin Adler.
3035 * inspector/InspectorDOMAgent.cpp:
3036 (WebCore::InspectorDOMAgent::inspect):
3037 (WebCore::InspectorDOMAgent::focusNode):
3038 Ensure we only cause focusNode with a node to focus. If the original
3039 node that was provided is no longer in the document, then we won't
3040 actually find a node to inspect.
3042 2015-06-12 KwangHyuk Kim <hyuki.kim@samsung.com>
3044 [EFL] canvas/philip/tests/toDataURL.jpeg.foo tests have been failed since r185417.
3045 https://bugs.webkit.org/show_bug.cgi?id=145878
3047 Reviewed by Gyuyoung Kim.
3049 Fix input buffer alignment issue.
3051 No new tests, canvas/philip/tests/toDataURL.jpeg.foo can be used.
3053 * platform/graphics/efl/ImageBufferEfl.cpp:
3054 (WebCore::encodeImagePNG):
3055 (WebCore::encodeImageJPEG):
3056 (WebCore::ImageBuffer::toDataURL):
3058 2015-06-12 Zan Dobersek <zdobersek@igalia.com>
3060 Add missing override specifiers under Source/WebCore/svg/
3061 https://bugs.webkit.org/show_bug.cgi?id=145841
3063 Reviewed by Darin Adler.
3065 Add missing override specifiers to virtual method overrides for classes
3066 under Source/WebCore/svg/, suppressing a bunch of warnings when compiling
3069 Add the BEGIN_DECLARE_ANIMATED_PROPERTIES_BASE() macro that doesn't
3070 override the localAttributeToPropertyMap() method declaration, and
3071 have the BEGIN_DECLARE_ANIMATED_PROPERTIES() macro add the override
3072 specifier. The new macro is used in SVGElement.
3074 Change the DECLARE_ANIMATED_PROPERTY() to accept the optional override
3075 specifier as the fifth parameter. Current DECLARE_ANIMATED_*() macros
3076 are modified to just pass an empty argument, not adding any specifier.
3077 DECLARE_ANIMATED_BOOLEAN_OVERRIDE() and DECLARE_ANIMATED_STRING_OVERRIDE()
3078 macros are added to override the methods for the externalResourcesRequired
3079 and href animated properties, where required.
3081 * svg/SVGAElement.h:
3082 * svg/SVGAltGlyphElement.h:
3083 * svg/SVGAnimatedAngle.h:
3084 * svg/SVGAnimatedBoolean.h:
3085 * svg/SVGAnimatedEnumeration.h:
3086 * svg/SVGAnimatedInteger.h:
3087 * svg/SVGAnimatedLength.h:
3088 * svg/SVGAnimatedNumber.h:
3089 * svg/SVGAnimatedPreserveAspectRatio.h:
3090 * svg/SVGAnimatedRect.h:
3091 * svg/SVGAnimatedString.h:
3092 * svg/SVGAnimationElement.h:
3093 * svg/SVGCircleElement.h:
3094 * svg/SVGClipPathElement.h:
3095 * svg/SVGCursorElement.h:
3096 * svg/SVGDefsElement.h:
3098 * svg/SVGEllipseElement.h:
3099 * svg/SVGFEImageElement.h:
3100 * svg/SVGFilterElement.h:
3101 * svg/SVGFontElement.h:
3102 * svg/SVGForeignObjectElement.h:
3103 * svg/SVGGElement.h:
3104 * svg/SVGGlyphRefElement.h:
3105 * svg/SVGGradientElement.h:
3106 * svg/SVGImageElement.h:
3107 * svg/SVGLineElement.h:
3108 * svg/SVGMPathElement.h:
3109 * svg/SVGMarkerElement.h:
3110 * svg/SVGMaskElement.h:
3111 * svg/SVGPathElement.h:
3112 * svg/SVGPatternElement.h:
3113 * svg/SVGPolyElement.h:
3114 * svg/SVGRectElement.h:
3115 * svg/SVGSVGElement.h:
3116 * svg/SVGScriptElement.h:
3117 * svg/SVGSwitchElement.h:
3118 * svg/SVGSymbolElement.h:
3119 * svg/SVGTRefElement.h:
3120 * svg/SVGTextContentElement.h:
3121 * svg/SVGTextPathElement.h:
3122 * svg/SVGURIReference.h: Add the hrefBaseValue() pure virtual method
3123 so it can be overridden by any class that also overrides the
3124 corresponding setHrefBaseValue() method.
3125 * svg/SVGUseElement.h:
3126 * svg/SVGViewElement.h:
3127 * svg/properties/SVGAnimatedPropertyMacros.h:
3128 * svg/properties/SVGMatrixTearOff.h:
3129 (WebCore::SVGMatrixTearOff::commitChange):
3131 2015-06-12 Zan Dobersek <zdobersek@igalia.com>
3133 [GLib] Move files under Source/WTF/wtf/gobject to Source/WTF/wtf/glib
3134 https://bugs.webkit.org/show_bug.cgi?id=145799
3136 Reviewed by Carlos Garcia Campos.
3138 Update header inclusions for headers that have been moved
3139 to Source/WTF/wtf/glib/.
3141 * accessibility/atk/AXObjectCacheAtk.cpp:
3142 * accessibility/atk/WebKitAccessibleInterfaceText.cpp:
3143 * bindings/gobject/DOMObjectCache.cpp:
3144 * bindings/gobject/GObjectEventListener.h:
3145 * bindings/gobject/GObjectNodeFilterCondition.h:
3146 * bindings/gobject/GObjectXPathNSResolver.h:
3147 * bindings/gobject/WebKitDOMEventTarget.cpp:
3148 * platform/Pasteboard.h:
3149 * platform/audio/gstreamer/AudioDestinationGStreamer.cpp:
3150 * platform/audio/gstreamer/AudioFileReaderGStreamer.cpp:
3151 * platform/audio/gstreamer/AudioSourceProviderGStreamer.cpp:
3152 * platform/audio/gstreamer/WebKitWebAudioSourceGStreamer.cpp:
3153 * platform/audio/gtk/AudioBusGtk.cpp:
3154 * platform/geoclue/GeolocationProviderGeoclue.h:
3155 * platform/geoclue/GeolocationProviderGeoclue2.cpp:
3156 * platform/glib/BatteryProviderUPower.cpp:
3157 * platform/glib/BatteryProviderUPower.h:
3158 * platform/graphics/gstreamer/GRefPtrGStreamer.h:
3159 * platform/graphics/gstreamer/GStreamerUtilities.cpp:
3160 * platform/graphics/gstreamer/GUniquePtrGStreamer.h:
3161 * platform/graphics/gstreamer/InbandTextTrackPrivateGStreamer.h:
3162 * platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp:
3163 * platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.h:
3164 * platform/graphics/gstreamer/MediaPlayerPrivateGStreamerBase.cpp:
3165 * platform/graphics/gstreamer/MediaPlayerPrivateGStreamerBase.h:
3166 * platform/graphics/gstreamer/MediaSourceGStreamer.cpp:
3167 * platform/graphics/gstreamer/TrackPrivateBaseGStreamer.cpp:
3168 * platform/graphics/gstreamer/TrackPrivateBaseGStreamer.h:
3169 * platform/graphics/gstreamer/VideoSinkGStreamer.cpp:
3170 * platform/graphics/gstreamer/WebKitMediaSourceGStreamer.cpp:
3171 * platform/graphics/gstreamer/WebKitWebSourceGStreamer.cpp:
3172 * platform/graphics/gtk/ImageBufferGtk.cpp:
3173 * platform/graphics/gtk/ImageGtk.cpp:
3174 * platform/gtk/ContextMenuGtk.cpp:
3175 * platform/gtk/ContextMenuItemGtk.cpp:
3176 * platform/gtk/DataObjectGtk.cpp:
3177 * platform/gtk/DataObjectGtk.h:
3178 * platform/gtk/FileSystemGtk.cpp:
3179 * platform/gtk/GRefPtrGtk.h:
3180 * platform/gtk/GUniquePtrGtk.h:
3181 * platform/gtk/GamepadsGtk.cpp:
3182 * platform/gtk/GtkUtilities.cpp:
3183 * platform/gtk/LanguageGtk.cpp:
3184 * platform/gtk/LocalizedStringsGtk.cpp:
3185 * platform/gtk/PasteboardHelper.cpp:
3186 * platform/gtk/ScrollbarThemeGtk.cpp:
3187 * platform/gtk/SharedBufferGtk.cpp:
3188 * platform/gtk/SharedTimerGtk.cpp:
3189 * platform/mediastream/openwebrtc/RealtimeMediaSourceCenterOwr.cpp:
3190 * platform/network/ResourceHandleInternal.h:
3191 * platform/network/gtk/CredentialBackingStore.cpp:
3192 * platform/network/soup/CertificateInfo.h:
3193 * platform/network/soup/CookieJarSoup.cpp:
3194 * platform/network/soup/DNSSoup.cpp:
3195 * platform/network/soup/GRefPtrSoup.h:
3196 * platform/network/soup/GUniquePtrSoup.h:
3197 * platform/network/soup/ResourceError.h:
3198 * platform/network/soup/ResourceErrorSoup.cpp:
3199 * platform/network/soup/ResourceHandleSoup.cpp:
3200 * platform/network/soup/ResourceResponse.h:
3201 * platform/network/soup/SocketStreamHandle.h:
3202 * platform/network/soup/SocketStreamHandleSoup.cpp:
3203 * platform/network/soup/SoupNetworkSession.h:
3204 * platform/text/gtk/HyphenationLibHyphen.cpp:
3205 * rendering/RenderThemeGtk.cpp:
3207 2015-06-12 Yoav Weiss <yoav@yoav.ws>
3209 Fix the build when the PICTURE_SIZES flag is off
3210 https://bugs.webkit.org/show_bug.cgi?id=145926
3212 Reviewed by Csaba Osztrogonác.
3214 No new tests since there's no functionality change.
3216 * html/parser/HTMLPreloadScanner.cpp: Remove the guard around the definition of m_sizesAttribute.
3218 2015-06-12 Commit Queue <commit-queue@webkit.org>
3220 Unreviewed, rolling out r185492.
3221 https://bugs.webkit.org/show_bug.cgi?id=145927
3223 Causes crashes on debug (Requested by rego on #webkit).
3227 "[CSS Grid Layout] Fix grid-template-areas parsing to avoid
3229 https://bugs.webkit.org/show_bug.cgi?id=145860
3230 http://trac.webkit.org/changeset/185492
3232 2015-06-12 Youenn Fablet <youenn.fablet@crf.canon.fr>
3234 Bindings generator should generate code for Promise-based APIs
3235 https://bugs.webkit.org/show_bug.cgi?id=145833
3237 Reviewed by Darin Adler.
3239 Covered by existing tests.
3241 * Modules/webaudio/AudioContext.idl: Removing custom binding for resume, suspend and close.
3242 * bindings/js/JSAudioContextCustom.cpp: Ditto.
3243 * bindings/scripts/CodeGeneratorJS.pm:
3244 (GenerateHeader): Refactoring to use IsReturningPromise.
3245 (GenerateImplementation): Disabling include for return type if it is a promise.
3246 (GenerateParametersCheck): Adding DeferredWrapper() as argument to the DOM method if JS method returns a promise.
3247 (GenerateImplementationFunctionCall): Added support for promise-returning API.
3248 (IsReturningPromise): Checking whether function is returning a promise.
3249 * bindings/scripts/test/GObject/WebKitDOMTestObj.cpp:
3250 (webkit_dom_test_obj_test_promise_function):
3251 * bindings/scripts/test/GObject/WebKitDOMTestObj.h:
3252 * bindings/scripts/test/JS/JSTestObj.cpp:
3253 (WebCore::jsTestObjPrototypeFunctionTestPromiseFunction):
3254 * bindings/scripts/test/ObjC/DOMTestObj.h:
3255 * bindings/scripts/test/ObjC/DOMTestObj.mm:
3256 (-[DOMTestObj testPromiseFunction]):
3257 * bindings/scripts/test/TestObj.idl: Adding promise returning function.
3260 2015-06-12 Manuel Rego Casasnovas <rego@igalia.com>
3262 [CSS Grid Layout] Fix grid-template-areas parsing to avoid spaces
3263 https://bugs.webkit.org/show_bug.cgi?id=145860
3265 Reviewed by Sergio Villar Senin.
3267 The spec doesn't require to have spaces between unnamed and named areas
3268 in grid-template-areas syntax. But spaces are currently required in our
3271 This was confirmed in the CSS WG mailing list:
3272 https://lists.w3.org/Archives/Public/www-style/2015May/0239.html
3274 This patch updates grid-template-areas parsing to allow the possibility
3275 of removing spaces between unnamed and named areas.
3277 Added new cases to fast/css-grid-layout/grid-template-areas-get-set.html.
3279 * css/CSSParser.cpp:
3280 (WebCore::parseGridTemplateAreasColumnNames): New helper method to
3281 determine the column names split by white spaces or dots.
3282 (WebCore::CSSParser::parseGridTemplateAreasRow): Use the new helper
3283 method to get the column names.
3284 (WebCore::containsOnlyDots): Deleted. Not needed anymore as
3285 parseGridTemplateAreasColumnNames() is using a single dot for unnamed
3286 grid areas (despite of being defined with 1 or more dots).
3288 2015-06-11 Mark Lam <mark.lam@apple.com>
3290 WebCore::reportException() needs to be able to accept a raw thrown value in addition to Exception objects.
3291 https://bugs.webkit.org/show_bug.cgi?id=145872
3293 Reviewed by Michael Saboff.
3295 API test added: WebKit1.ReportExceptionTest.
3297 * bindings/js/JSDOMBinding.cpp:
3298 (WebCore::reportException):
3299 - Added a version of reportException() that takes a JSValue instead of an Exception
3300 object. This version will ensure that we have an Exception object before calling
3301 into the original reportException() as follows:
3303 1. If the JSValue is already an Exception object, we're good to go.
3305 2. Else, if VM::lastException() is available, use that as the exception.
3306 This is how the old code use to behave (in terms of where it gets the exception
3307 stack trace). The assumption is that reportException() will be called in
3308 a timely manner before the exception stack trace has been purged.
3310 3. Else, create an Exception object with no stack trace. This is the fall back
3311 in case the client did not call reportException() in a timely manner.
3313 - Also clear the last exception after we've reported it. This is how the old code
3314 before r185259 behave (in terms of the lifecycle of the last exception stack
3315 trace). We're restoring that behavior here.
3317 * bindings/js/JSDOMBinding.h:
3319 2015-06-11 Zalan Bujtas <zalan@apple.com>
3321 Do not crash when the descendant frame tree is destroyed during layout.
3322 https://bugs.webkit.org/show_bug.cgi?id=144540
3323 rdar://problem/20793184
3325 Reviewed by Andreas Kling.
3327 Widget::setFrameRect(), through WebHTMLView layout, could trigger a style recalc, which in turn
3328 could initiate an onBeforeLoad callback.
3329 If javascript happens to destroy the current iframe in the onBeforeLoad callback, we lose the descendant
3330 render tree, including the child FrameView (the iframe element's view). However the RenderIFrame
3331 object stays protected until after the layout is done. (see protectRenderWidgetUntilLayoutIsDone())
3333 Climbing back on the callstack, we need to make sure that
3334 1. the root widget of the descendant render tree (FrameView) stays valid as long as it is needed.
3335 2. RenderFrameBase::layoutWithFlattening() can handle the case when the associated widget (child FrameView) is set to nullptr.
3336 (see RenderWidget::willBeDestroyed() -> setWidget(nullptr))
3338 (and later, when layout is finished this (RenderIFrame) object gets destroyed too.)
3340 Covered by fast/frames/flattening/crash-remove-iframe-during-object-beforeload.html.
3342 * page/FrameView.cpp:
3343 (WebCore::FrameView::setFrameRect):
3344 (WebCore::FrameView::updateEmbeddedObject):
3345 (WebCore::FrameView::updateWidgetPositions):
3346 * platform/ScrollView.cpp:
3347 (WebCore::ScrollView::setFrameRect):
3348 * platform/mac/WidgetMac.mm:
3349 (WebCore::Widget::setFrameRect):
3350 * rendering/RenderFrameBase.cpp:
3351 (WebCore::RenderFrameBase::layoutWithFlattening):
3352 (WebCore::RenderFrameBase::childRenderView):
3353 (WebCore::RenderFrameBase::peformLayoutWithFlattening):
3354 * rendering/RenderFrameBase.h:
3355 * rendering/RenderWidget.cpp:
3356 (WebCore::RenderWidget::updateWidgetPosition):
3357 * rendering/RenderWidget.h:
3359 2015-06-11 Commit Queue <commit-queue@webkit.org>
3361 Unreviewed, rolling out r185470.
3362 https://bugs.webkit.org/show_bug.cgi?id=145902
3364 Caused ASan violations (Requested by ap on #webkit).
3368 "Add an appearance keyword for wireless playback / airplay
3370 https://bugs.webkit.org/show_bug.cgi?id=145892
3371 http://trac.webkit.org/changeset/185470
3373 2015-06-11 Brady Eidson <beidson@apple.com>
3375 IndexedDB onupgradeneeded event has incorrect value for oldVersion.
3376 <rdar://problem/18309792> and https://bugs.webkit.org/show_bug.cgi?id=136888
3378 Reviewed by Sam Weinig.
3380 Test: storage/indexeddb/version-change-event-basic.html
3382 "NoIntVersion" is an internal bookkeeping concept that we never should've been passing to Javascript.
3384 This cleans up things by:
3385 - Adjusting an old version of "NoIntVersion" to "0" before making the version change callback.
3386 - Removing the VersionNullness parameter from almost everywhere.
3387 - Removing the nullability of the newVersion parameter from the IDL.
3389 * Modules/indexeddb/IDBDatabase.cpp:
3390 (WebCore::IDBDatabase::onVersionChange):
3391 * Modules/indexeddb/IDBDatabase.h:
3393 * Modules/indexeddb/IDBDatabaseBackend.cpp:
3394 (WebCore::IDBDatabaseBackend::runIntVersionChangeTransaction):
3395 (WebCore::IDBDatabaseBackend::deleteDatabase):
3397 * Modules/indexeddb/IDBDatabaseCallbacks.h:
3398 * Modules/indexeddb/IDBDatabaseCallbacksImpl.cpp:
3399 (WebCore::IDBDatabaseCallbacksImpl::onVersionChange):
3400 * Modules/indexeddb/IDBDatabaseCallbacksImpl.h:
3402 * Modules/indexeddb/IDBOpenDBRequest.cpp:
3403 (WebCore::IDBOpenDBRequest::onBlocked):
3404 (WebCore::IDBOpenDBRequest::onUpgradeNeeded):
git://git.webkit.org / WebKit-https.git / blob