1 2018-01-11 Basuke Suzuki <Basuke.Suzuki@sony.com>
3 [Curl] Extract multipart handling from ResourceHandle to CurlRequest.
4 https://bugs.webkit.org/show_bug.cgi?id=181506
6 Reviewed by Alex Christensen.
8 Rename old MultipartHandle class to CurlMultipartHandle and modernize it. Also move the responsibility
9 of handling multi part from ResourceHandle to CurlRequest. This is required for upcoming NetworkLoadTask.
11 No new tests because no new behavior.
13 * platform/Curl.cmake:
14 * platform/network/curl/CurlMultipartHandle.cpp: Renamed from Source/WebCore/platform/network/curl/MultipartHandle.cpp.
15 (WebCore::CurlMultipartHandle::createIfNeeded):
16 (WebCore::CurlMultipartHandle::extractBoundary):
17 (WebCore::CurlMultipartHandle::extractBoundaryFromContentType):
18 (WebCore::CurlMultipartHandle::CurlMultipartHandle):
19 (WebCore::CurlMultipartHandle::didReceiveData):
20 (WebCore::CurlMultipartHandle::didComplete):
21 (WebCore::CurlMultipartHandle::processContent):
22 (WebCore::CurlMultipartHandle::checkForBoundary):
23 (WebCore::CurlMultipartHandle::matchedLength):
24 (WebCore::CurlMultipartHandle::parseHeadersIfPossible):
25 * platform/network/curl/CurlMultipartHandle.h: Renamed from Source/WebCore/platform/network/curl/MultipartHandle.h.
26 (WebCore::CurlMultipartHandle::~CurlMultipartHandle):
27 * platform/network/curl/CurlMultipartHandleClient.h: Added.
28 (WebCore::CurlMultipartHandleClient::~CurlMultipartHandleClient):
29 * platform/network/curl/CurlRequest.cpp:
30 (WebCore::CurlRequest::CurlRequest):
31 (WebCore::CurlRequest::didReceiveHeader):
32 (WebCore::CurlRequest::didReceiveData):
33 (WebCore::CurlRequest::didReceiveHeaderFromMultipart):
34 (WebCore::CurlRequest::didReceiveDataFromMultipart):
35 (WebCore::CurlRequest::didCompleteTransfer):
36 (WebCore::CurlRequest::finalizeTransfer):
37 (WebCore::CurlRequest::invokeDidReceiveResponseForFile):
38 (WebCore::CurlRequest::invokeDidReceiveResponse):
39 (WebCore::CurlRequest::completeDidReceiveResponse):
40 * platform/network/curl/CurlRequest.h:
41 (WebCore::CurlRequest::create):
42 * platform/network/curl/ResourceHandleCurlDelegate.cpp:
43 (WebCore::ResourceHandleCurlDelegate::createCurlRequest):
44 (WebCore::ResourceHandleCurlDelegate::curlDidReceiveResponse):
45 (WebCore::ResourceHandleCurlDelegate::curlDidReceiveBuffer):
46 (WebCore::ResourceHandleCurlDelegate::curlDidComplete):
47 * platform/network/curl/ResourceHandleCurlDelegate.h:
49 2018-01-11 Zalan Bujtas <zalan@apple.com>
51 RenderTreeUpdater::current() returns null_ptr when mutation is done through Document::resolveStyle.
52 https://bugs.webkit.org/show_bug.cgi?id=181513
53 <rdar://problem/36367085>
55 Reviewed by Antti Koivisto.
57 This patch ensures that we use a valid RenderTreeBuilder even when
58 Document::resolveStyle (incorrectly) triggers tree mutation.
59 It can be reverted soon after the incorrect mutations are taken care of.
61 Test: fast/forms/button-set-text-crash.html
63 * rendering/RenderButton.cpp:
64 (WebCore::RenderButton::setText):
65 * rendering/RenderMenuList.cpp:
66 (RenderMenuList::setText):
68 2018-01-11 Antoine Quint <graouts@apple.com>
70 Only listen to webkitplaybacktargetavailabilitychanged when media controls are visible to the user
71 https://bugs.webkit.org/show_bug.cgi?id=181547
72 <rdar://problem/35947650>
74 Reviewed by Eric Carlson.
76 Because listening to "webkitplaybacktargetavailabilitychanged" events incurs some higher power usage on iOS,
77 we only listen to such events when controls are visible to the user. In other words, the MediaControls need to
78 have both "visible" set to "true" and "faded" set to "false". To support this, we add a delegate method on
79 MediaControls such that it can tell the MediaController that the "visible" property changed. With this message,
80 MediaController can inform its MediaControllerSupport objects that user visibility of the controls changed, which
81 lets AirplaySupport disable itself when controls are no longer visible.
83 Test: media/modern-media-controls/airplay-support/airplay-support-disable-event-listeners-with-hidden-controls.html
85 * Modules/modern-media-controls/controls/media-controls.js:
86 (MediaControls.prototype.set visible):
87 * Modules/modern-media-controls/media/airplay-support.js:
88 (AirplaySupport.prototype.controlsUserVisibilityDidChange):
89 * Modules/modern-media-controls/media/media-controller-support.js:
90 (MediaControllerSupport.prototype.controlsUserVisibilityDidChange):
91 * Modules/modern-media-controls/media/media-controller.js:
92 (MediaController.prototype.mediaControlsVisibilityDidChange):
93 (MediaController.prototype.mediaControlsFadedStateDidChange):
94 (MediaController.prototype._controlsUserVisibilityDidChange):
96 2018-01-11 Antoine Quint <graouts@apple.com>
98 [iOS] There should be no controls markup generated in fullscreen
99 https://bugs.webkit.org/show_bug.cgi?id=181540
100 <rdar://problem/35060379>
102 Reviewed by Eric Carlson.
104 We completely forgo the display of any content when fullscreen on iOS by setting the
105 "visible" flag to "false" on the MediaControls, which will prevent any DOM content from
108 * Modules/modern-media-controls/media/media-controller.js:
109 (MediaController.prototype.handleEvent):
110 (MediaController.prototype._updateiOSFullscreenProperties):
112 (MediaController.prototype._updateSupportingObjectsEnabledState): Deleted.
114 2018-01-11 Alex Christensen <achristensen@webkit.org>
116 Revert changes accidentally committed with r226789.
117 https://bugs.webkit.org/show_bug.cgi?id=181423
119 I had some local changes I did not mean to commit.
121 * platform/network/mac/WebCoreResourceHandleAsOperationQueueDelegate.mm:
122 (scheduledWithCustomRunLoopMode):
123 (-[WebCoreResourceHandleAsOperationQueueDelegate callFunctionOnMainThread:]):
124 (WebCore::if): Deleted.
125 (WebCore::>::fromCallable): Deleted.
127 2018-01-10 Simon Fraser <simon.fraser@apple.com>
129 On macOS, getBoundingClientRect gives incorrect values when pinch-zoomed
130 https://bugs.webkit.org/show_bug.cgi?id=181511
131 rdar://problem/33741427
133 Reviewed by Zalan Bujtas.
135 When reverting "client coordinates are relative to layout viewport" in r219829
136 I broke documentToClientOffset() on macOS by failing to take pinch zoom scale into
137 account (frameScaleFactor() is always 1 on iOs, so this bug doesn't manifest there).
139 Covered by existing tests.
141 * page/FrameView.cpp:
142 (WebCore::FrameView::documentToClientOffset const):
144 2018-01-11 Youenn Fablet <youenn@apple.com>
146 Replace WebRTCLegacyAPIDisabled by WebRTCLegacyAPIEnabled and switch off WebRTC legacy flag by default
147 https://bugs.webkit.org/show_bug.cgi?id=181480
149 Reviewed by Eric Carlson.
151 No change of behavior.
153 * page/RuntimeEnabledFeatures.h: Set default value to false.
155 2018-01-11 Wenson Hsieh <wenson_hsieh@apple.com>
157 [Attachment Support] Support moving attachment elements in editable areas using drag and drop
158 https://bugs.webkit.org/show_bug.cgi?id=181337
159 <rdar://problem/36324813>
161 Reviewed by Tim Horton.
163 Makes slight adjustments to attachment-specific drag and drop logic to ensure that moving attachments via drag
164 and drop behaves correctly. See per-change comments for more detail.
166 Tests: WKAttachmentTests.DragInPlaceVideoAttachmentElement
167 WKAttachmentTests.MoveAttachmentElementAsIconByDragging
168 WKAttachmentTests.MoveInPlaceAttachmentElementByDragging
170 * editing/cocoa/EditorCocoa.mm:
171 (WebCore::Editor::getPasteboardTypesAndDataForAttachment):
173 Stop vending the private web archive pasteboard type for attachments, for now. This works around issues where an
174 attachment element that is dragged and dropped within the same page may lose its blob backing data if we try to
175 remove and insert it as a fragment from the archive. Providing a web archive would allow us to avoid destroying
176 and recreating an attachment element when dragging within the same page, but this is a nice-to-have optimization
177 we can re-enable after investigation in a subsequent patch.
179 * html/HTMLAttachmentElement.cpp:
180 (WebCore::HTMLAttachmentElement::populateShadowRootIfNecessary):
182 Add `draggable=false` to the image element of an in-place attachment element.
184 * page/DragController.cpp:
185 (WebCore::enclosingAttachmentElement):
186 (WebCore::DragController::draggableElement const):
188 Tweak single-selected-attachment handling to account for in-place attachments. Since the hit-tested node is
189 inside the shadow subtree of the attachment element, the condition needs to check for the startElement as well
190 as the startElement's shadow host.
192 (WebCore::DragController::startDrag):
194 Make two tweaks here. First, don't require a RenderAttachment to drag an attachment element (this is required
195 for dragging in-place attachments). This was added in r217083 to address <rdar://problem/32282831>, but is no
196 longer correct, since attachments may now be displayed in-place.
198 Secondly, only restore the previous selection if the attachment is in a richly contenteditable area. This was
199 added to prevent the selection highlight from appearing in when dragging non-editable attachment elements in the
200 Mail viewer. However, to allow drag moves to occur, we need the selection to persist after drag start.
202 2018-01-04 Filip Pizlo <fpizlo@apple.com>
204 CodeBlocks should be in IsoSubspaces
205 https://bugs.webkit.org/show_bug.cgi?id=180884
207 Reviewed by Saam Barati.
209 No new tests because no new behavior.
211 Adopting new parallel constraint API, so that more of the logic of doing parallel
212 constraint solving is shared between the DOM's output constraints and JSC's output
215 * bindings/js/DOMGCOutputConstraint.cpp:
216 (WebCore::DOMGCOutputConstraint::executeImpl):
217 (WebCore::DOMGCOutputConstraint::doParallelWorkImpl): Deleted.
218 (WebCore::DOMGCOutputConstraint::finishParallelWorkImpl): Deleted.
219 * bindings/js/DOMGCOutputConstraint.h:
221 2018-01-11 Wenson Hsieh <wenson_hsieh@apple.com>
223 [Attachment Support] Support dragging attachment elements out as files on iOS
224 https://bugs.webkit.org/show_bug.cgi?id=181199
225 <rdar://problem/36299316>
227 Reviewed by Tim Horton, Andy Estes and Joseph Pecoraro.
229 Adds support for dragging "files" (i.e. creating item providers with preferred attachment presentation styles)
230 from attachment elements on iOS for Mail. See below for more detail.
232 Tests: WKAttachmentTestsIOS.DragAttachmentInsertedAsData
233 WKAttachmentTestsIOS.DragAttachmentInsertedAsFile
235 * page/DragController.cpp:
236 (WebCore::DragController::platformContentTypeForBlobType const):
237 (WebCore::DragController::dragAttachmentElement):
238 * page/DragController.h:
239 * page/mac/DragControllerMac.mm:
240 (WebCore::DragController::platformContentTypeForBlobType const):
242 Add a private method to convert the type of a promised blob to a platform type. For Cocoa platforms, this
243 converts the blob type (either a UTI or a MIME type) to a UTI for the platform to consume.
245 * platform/ios/WebItemProviderPasteboard.h:
246 * platform/ios/WebItemProviderPasteboard.mm:
248 Refactor WebItemProviderRegistrationInfo. WebItemProviderRegistrationInfo currently encapsulates a single item
249 provider registration call, and contains either a type identifier and data buffer, or an NSItemProviderWriting-
250 conformant object. To register an item provider using a WebItemProviderRegistrationInfo, the item provider
251 pasteboard currently checks to see whether the info contains an object or a type and data.
253 This patch removes WebItemProviderRegistrationInfo and replaces it with WebItemProviderDataRegistrar. Objects
254 that implement this protocol know how to take an NSItemProvider and register data to it. So far, there are
255 three implementations below.
257 (-[WebItemProviderDataRegistrar initWithData:type:]):
258 (-[WebItemProviderDataRegistrar typeIdentifier]):
259 (-[WebItemProviderDataRegistrar data]):
260 (-[WebItemProviderDataRegistrar typeIdentifierForClient]):
261 (-[WebItemProviderDataRegistrar dataForClient]):
262 (-[WebItemProviderDataRegistrar registerItemProvider:]):
263 (-[WebItemProviderDataRegistrar description]):
265 A data registrar takes a UTI and data buffer, and registers the UTI to the data. This replaces a
266 WebItemProviderRegistrationInfo with both a type and data, but no representing object.
268 (-[WebItemProviderWritableObjectRegistrar initWithObject:]):
269 (-[WebItemProviderWritableObjectRegistrar representingObjectForClient]):
270 (-[WebItemProviderWritableObjectRegistrar registerItemProvider:]):
271 (-[WebItemProviderWritableObjectRegistrar description]):
273 The writable object registrar writes an NSItemProviderWriting-conformant object to an item provider. This
274 replaces a WebItemProviderRegistrationInfo with only a representing object.
276 (-[WebItemProviderPromisedFileRegistrar initWithType:callback:]):
277 (-[WebItemProviderPromisedFileRegistrar registerItemProvider:]):
278 (-[WebItemProviderPromisedFileRegistrar description]):
279 (-[WebItemProviderRegistrationInfoList addData:forType:]):
280 (-[WebItemProviderRegistrationInfoList addRepresentingObject:]):
281 (-[WebItemProviderRegistrationInfoList addPromisedType:fileCallback:]):
283 Helper methods to add new registrars to a registration info list.
285 (-[WebItemProviderRegistrationInfoList itemAtIndex:]):
286 (-[WebItemProviderRegistrationInfoList enumerateItems:]):
287 (-[WebItemProviderRegistrationInfoList itemProvider]):
288 (-[WebItemProviderRegistrationInfoList description]):
289 (-[WebItemProviderRegistrationInfo initWithRepresentingObject:typeIdentifier:data:]): Deleted.
290 (-[WebItemProviderRegistrationInfo representingObject]): Deleted.
291 (-[WebItemProviderRegistrationInfo typeIdentifier]): Deleted.
293 2018-01-11 Michael Saboff <msaboff@apple.com>
295 Add a DOM gadget for Spectre testing
296 https://bugs.webkit.org/show_bug.cgi?id=181351
298 Reviewed by Ryosuke Niwa.
300 This change is used to test Spectre mitigations.
302 Added a new DOM class to test for Spectre issues in the DOM layer.
303 This additional functionality is disabled by default and must be enabled
304 through the JSC option "enableSpectreGadgets".
307 * DerivedSources.make:
309 * WebCore.xcodeproj/project.pbxproj:
310 * bindings/js/WebCoreBuiltinNames.h:
311 * dom/SpectreGadget.cpp: Added.
312 (WebCore::SpectreGadget::SpectreGadget):
313 (WebCore::SpectreGadget::create):
314 (WebCore::SpectreGadget::setReadLength):
315 (WebCore::SpectreGadget::charCodeAt):
316 (WebCore::SpectreGadget::clflushReadLength):
317 * dom/SpectreGadget.h: Added.
318 * dom/SpectreGadget.idl: Added.
319 * page/RuntimeEnabledFeatures.cpp:
320 (WebCore::RuntimeEnabledFeatures::spectreGadgetsEnabled const):
321 * page/RuntimeEnabledFeatures.h:
323 2018-01-11 Philippe Normand <pnormand@igalia.com>
325 [GTK] media/muted-video-is-playing-audio.html is timing out
326 https://bugs.webkit.org/show_bug.cgi?id=163781
328 Reviewed by Carlos Garcia Campos.
330 * platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp:
331 (WebCore::MediaPlayerPrivateGStreamer::notifyPlayerOfVideo): Sprinkle some debugging.
332 (WebCore::MediaPlayerPrivateGStreamer::notifyPlayerOfAudio): Ditto.
333 * platform/graphics/gstreamer/MediaPlayerPrivateGStreamerBase.cpp:
334 (WebCore::MediaPlayerPrivateGStreamerBase::setMuted): Ditto.
336 2018-01-10 Andy Estes <aestes@apple.com>
338 [Payment Request] Rename ApplePayMerchantValidationEvent to MerchantValidationEvent
339 https://bugs.webkit.org/show_bug.cgi?id=181437
340 <rdar://problem/36376481>
342 Reviewed by Tim Horton.
344 Renamed ApplePayMerchantValidationEvent to MerchantValidationEvent and changed complete() to accept a Promise.
346 Test: http/tests/paymentrequest/payment-request-merchant-validation.https.html
348 * DerivedSources.make:
349 * Modules/applepay/paymentrequest/ApplePayPaymentHandler.cpp:
350 (WebCore::ApplePayPaymentHandler::merchantValidationCompleted):
351 (WebCore::ApplePayPaymentHandler::validateMerchant):
352 * Modules/applepay/paymentrequest/ApplePayPaymentHandler.h:
353 * Modules/paymentrequest/MerchantValidationEvent.cpp: Renamed from Source/WebCore/Modules/applepay/paymentrequest/ApplePayMerchantValidationEvent.cpp.
354 (WebCore::MerchantValidationEvent::create):
355 (WebCore::MerchantValidationEvent::MerchantValidationEvent):
356 (WebCore::MerchantValidationEvent::eventInterface const):
357 (WebCore::MerchantValidationEvent::complete):
358 * Modules/paymentrequest/MerchantValidationEvent.h: Renamed from Source/WebCore/Modules/applepay/paymentrequest/ApplePayMerchantValidationEvent.h.
359 * Modules/paymentrequest/MerchantValidationEvent.idl: Renamed from Source/WebCore/Modules/applepay/paymentrequest/ApplePayMerchantValidationEvent.idl.
360 * Modules/paymentrequest/PaymentHandler.h:
361 * Modules/paymentrequest/PaymentRequest.cpp:
362 (WebCore::PaymentRequest::completeMerchantValidation):
363 * Modules/paymentrequest/PaymentRequest.h:
364 * Modules/paymentrequest/PaymentRequest.idl:
365 * WebCore.xcodeproj/project.pbxproj:
369 2018-01-10 Basuke Suzuki <Basuke.Suzuki@sony.com>
371 [Curl] Cross-protocol, cross-site scripting (XPXSS) using HTML forms
372 Nhttps://bugs.webkit.org/show_bug.cgi?id=153088
374 Reviewed by Alex Christensen.
376 No new tests because it's covered by existing tests.
378 * platform/network/curl/CurlContext.cpp:
379 (WebCore::CurlHandle::getHttpVersion):
380 * platform/network/curl/CurlContext.h:
381 * platform/network/curl/CurlRequest.cpp:
382 (WebCore::CurlRequest::didReceiveHeader):
383 * platform/network/curl/CurlResponse.h:
384 (WebCore::CurlResponse::isolatedCopy const):
385 * platform/network/curl/ResourceResponseCurl.cpp:
386 (WebCore::ResourceResponse::ResourceResponse):
387 (WebCore::ResourceResponse::setStatusLine):
389 2018-01-10 Wenson Hsieh <wenson_hsieh@apple.com>
391 REGRESSION(r222507): Composition highlight doesn't render when using IME
392 https://bugs.webkit.org/show_bug.cgi?id=181485
393 <rdar://problem/35896516>
395 Reviewed by Ryosuke Niwa.
397 Fixes the order of arguments passed to paintTextSubrangeBackground from paintCompositionBackground.
399 Test: editing/marked-text-appearance.html
401 * rendering/InlineTextBox.cpp:
402 (WebCore::InlineTextBox::paintCompositionBackground):
404 2018-01-10 Youenn Fablet <youenn@apple.com>
406 Use no-cache fetch mode when loading main documents with location.reload()
407 https://bugs.webkit.org/show_bug.cgi?id=181285
409 Reviewed by Alex Christensen.
411 Covered by rebased tests.
413 Start to translate cache policy used for navigation as FetchOptions::Cache.
414 This allows ensuring service workers receive the right cache mode when intercepting navigation loads.
415 To not change current navigation behavior, ReturnCacheDataElseLoad and ReturnCacheDataDontLoad still trigger default fetch cache mode.
417 For Reload and ReloadExpiredOnly frame load types, using no-cache mode is more efficient than reload mode,
418 as a conditional request will be sent if possible. This applies to location.reload which is consistent with other browsers.
419 Keep reload mode for ReloadFromOrigin.
421 * loader/DocumentLoader.cpp:
422 (WebCore::toFetchOptionsCache):
423 (WebCore::DocumentLoader::loadMainResource):
424 * loader/FrameLoader.cpp:
425 (WebCore::FrameLoader::loadFrameRequest):
426 (WebCore::FrameLoader::loadURL):
427 (WebCore::FrameLoader::load):
428 (WebCore::FrameLoader::reload):
429 (WebCore::FrameLoader::defaultRequestCachingPolicy):
430 (WebCore::FrameLoader::loadDifferentDocumentItem):
431 * loader/NavigationScheduler.cpp:
433 2018-01-10 Commit Queue <commit-queue@webkit.org>
435 Unreviewed, rolling out r226667 and r226673.
436 https://bugs.webkit.org/show_bug.cgi?id=181488
438 This caused a flaky crash. (Requested by mlewis13 on #webkit).
442 "CodeBlocks should be in IsoSubspaces"
443 https://bugs.webkit.org/show_bug.cgi?id=180884
444 https://trac.webkit.org/changeset/226667
446 "REGRESSION (r226667): CodeBlocks should be in IsoSubspaces"
447 https://bugs.webkit.org/show_bug.cgi?id=180884
448 https://trac.webkit.org/changeset/226673
450 2018-01-10 Antti Koivisto <antti@apple.com>
452 REGRESSION(r225650): The scores of MotionMark tests Multiply and Leaves dropped by 8%
453 https://bugs.webkit.org/show_bug.cgi?id=181460
454 <rdar://problem/36379776>
456 Reviewed by Ryosuke Niwa.
458 * css/parser/CSSParser.cpp:
459 (WebCore::CSSParserContext::CSSParserContext):
461 Don't do the expensive security origin test if the sheet base URL and document URL are identical.
462 This is true for inline style and inline stylesheets.
464 2018-01-10 Antti Koivisto <antti@apple.com>
466 Try to fix windows build.
468 * css/RuleFeature.cpp:
469 (WebCore::RuleFeatureSet::computeNextMatchElement):
471 2018-01-10 Zalan Bujtas <zalan@apple.com>
473 [RenderTreeBuilder] Move RenderRuby's moveChildren logic to RenderTreeBuilder
474 https://bugs.webkit.org/show_bug.cgi?id=181470
475 <rdar://problem/36397683>
477 Reviewed by Antti Koivisto.
479 This is about moving code, no cleanup and/or normalization (unfortunately it also means
480 some temporary changes).
482 No change in functionality.
484 * rendering/RenderBlockFlow.h:
485 * rendering/RenderRubyBase.cpp:
486 (WebCore::RenderRubyBase::moveChildren): Deleted.
487 (WebCore::RenderRubyBase::mergeChildrenWithBase): Deleted.
488 (WebCore::RenderRubyBase::moveInlineChildren): Deleted.
489 (WebCore::RenderRubyBase::moveBlockChildren): Deleted.
490 * rendering/RenderRubyBase.h:
491 * rendering/RenderRubyRun.cpp:
492 (WebCore::RenderRubyRun::takeChild):
493 * rendering/updating/RenderTreeBuilder.cpp:
494 (WebCore::RenderTreeBuilder::moveRubyChildren):
495 * rendering/updating/RenderTreeBuilder.h:
496 * rendering/updating/RenderTreeBuilderRuby.cpp:
497 (WebCore::RenderTreeBuilder::Ruby::moveInlineChildren):
498 (WebCore::RenderTreeBuilder::Ruby::moveBlockChildren):
499 (WebCore::RenderTreeBuilder::Ruby::moveChildren):
500 (WebCore::RenderTreeBuilder::Ruby::moveChildrenInternal):
501 (WebCore::RenderTreeBuilder::Ruby::insertChild):
502 * rendering/updating/RenderTreeBuilderRuby.h:
504 2018-01-10 Philippe Normand <pnormand@igalia.com>
506 [GStreamer] fix critical GObject warning
508 Rubber-stamped by Carlos Garcia Campos.
510 * platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp:
511 (WebCore::MediaPlayerPrivateGStreamer::handleMessage): No need to
512 resort to complicated things to get the element name...
514 2018-01-10 Philippe Normand <pnormand@igalia.com>
516 [GStreamer] REGRESSION(r226629): broke media/video-interruption-with-resume-allowing-play.html
517 https://bugs.webkit.org/show_bug.cgi?id=181471
518 <rdar://problem/36402323>
520 Reviewed by Carlos Garcia Campos.
522 This patch mainly reduces the amount of playback state changes
523 emitted by the GStreamer player to its client. Emitting those
524 notifications too often has bad side effects.
526 * platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp:
527 (WebCore::MediaPlayerPrivateGStreamer::paused const): Add debug messages.
528 (WebCore::MediaPlayerPrivateGStreamer::handleMessage): This debug message appears too much. Demote.
529 (WebCore::MediaPlayerPrivateGStreamer::maxTimeLoaded const): Ditto.
530 (WebCore::MediaPlayerPrivateGStreamer::didLoadingProgress const): Ditto.
531 (WebCore::MediaPlayerPrivateGStreamer::updateStates): Try to emit
532 playback state change notification only when going to PLAYING.
533 (WebCore::MediaPlayerPrivateGStreamer::loadingFailed): Add warning message.
535 2018-01-10 Youenn Fablet <youenn@apple.com>
537 Add Service Worker CSP persistency
538 https://bugs.webkit.org/show_bug.cgi?id=181434
540 Reviewed by Alex Christensen.
542 Covered by manual testing.
543 Future work on service worker test infrastructure should allow automating such tests.
545 Add support for service worker CSP data persistency.
546 Add a version parameter to increment each time the schema is changing.
547 This allows the same store to be used by multiple WebKits.
549 * workers/service/server/RegistrationDatabase.cpp:
550 (WebCore::v1RecordsTableSchema):
551 (WebCore::RegistrationDatabase::openSQLiteDatabase):
552 (WebCore::RegistrationDatabase::doPushChanges):
553 (WebCore::RegistrationDatabase::importRecords):
555 2018-01-10 Antti Koivisto <antti@apple.com>
557 Invalidate current element style on class change accurately
558 https://bugs.webkit.org/show_bug.cgi?id=181210
560 Reviewed by Zalan Bujtas.
562 * css/DocumentRuleSets.cpp:
563 (WebCore::DocumentRuleSets::collectFeatures const):
564 (WebCore::DocumentRuleSets::subjectClassRules const):
566 New rule set containing class rules affecting the subject element.
568 (WebCore::DocumentRuleSets::ancestorClassRules const):
569 * css/DocumentRuleSets.h:
570 * css/RuleFeature.cpp:
571 (WebCore::RuleFeatureSet::recursivelyCollectFeaturesFromSelector):
573 Classify selector components into various buckets based on the elements they match relative to
574 the subject element. There are more categories than this patch strictly needs, for future use.
576 (WebCore::RuleFeatureSet::collectFeatures):
577 (WebCore::RuleFeatureSet::add):
578 (WebCore::RuleFeatureSet::clear):
579 (WebCore::RuleFeatureSet::shrinkToFit):
581 * css/StyleResolver.h:
582 (WebCore::StyleResolver::hasSelectorForClass const): Deleted.
583 * style/ClassChangeInvalidation.cpp:
584 (WebCore::Style::elementNeedsInvalidation):
585 (WebCore::Style::ClassChangeInvalidation::computeInvalidation):
587 Don't invalidate current element unconditionally on class change. Instead find the subject rulesets
588 that might affect it use them to perform invalidation.
590 (WebCore::Style::ClassChangeInvalidation::invalidateStyleWithRuleSets):
591 (WebCore::Style::ClassChangeInvalidation::invalidateStyle): Deleted.
592 (WebCore::Style::ClassChangeInvalidation::invalidateDescendantStyle): Deleted.
593 * style/ClassChangeInvalidation.h:
594 (WebCore::Style::ClassChangeInvalidation::ClassChangeInvalidation):
595 (WebCore::Style::ClassChangeInvalidation::~ClassChangeInvalidation):
596 * style/StyleSharingResolver.cpp:
597 (WebCore::Style::SharingResolver::classNamesAffectedByRules const):
599 2018-01-09 Antoine Quint <graouts@apple.com>
601 [Web Animations] Expose the id property on Animation
602 https://bugs.webkit.org/show_bug.cgi?id=181450
603 <rdar://problem/36383600>
605 Reviewed by Dean Jackson.
607 Expose the "id" property on Animation and handle the "id" property on the optional KeyframeAnimationOptions object
608 passed to Element.animate(). All of the WPT tests related to this property are now passing.
610 * animation/WebAnimation.h:
611 * animation/WebAnimation.idl:
613 (WebCore::Element::animate):
615 2018-01-09 Chris Dumez <cdumez@apple.com>
617 Make service workers behave correctly with regards to Page Cache
618 https://bugs.webkit.org/show_bug.cgi?id=181446
619 <rdar://problem/36164291>
621 Reviewed by Youenn Fablet.
623 Make service workers behave correctly with regards to Page Cache:
624 1. If a document has an active service worker, do not let it go into PageCache
625 2. When a document goes into page cache, unregister it from the list of service worker clients
626 3. When a document is restored from page cache, add it nack to the list of service worker clients
628 Tests: http/tests/workers/service/client-added-to-clients-when-restored-from-page-cache.html
629 http/tests/workers/service/client-removed-from-clients-while-in-page-cache.html
630 http/tests/workers/service/no-page-cache-when-controlled.html
631 http/tests/workers/service/other_resources/test.html
634 (WebCore::Document::suspend):
635 (WebCore::Document::resume):
636 * history/PageCache.cpp:
637 (WebCore::canCacheFrame):
638 * page/DiagnosticLoggingKeys.cpp:
639 (WebCore::DiagnosticLoggingKeys::serviceWorkerKey):
640 * page/DiagnosticLoggingKeys.h:
642 2018-01-09 Chris Dumez <cdumez@apple.com>
644 We should not return undefined for most properties of a detached Window
645 https://bugs.webkit.org/show_bug.cgi?id=181416
646 <rdar://problem/36162489>
648 Reviewed by Ryosuke Niwa.
650 We should not return undefined for most properties on a detached Window. WebKit previously only exposed "closed"
651 and "close" properties on detached / frameless windows. However, this does not match the HTML specification [1]
652 or the behavior of Firefox and Chrome.
654 Note that Chrome does not seem to fully follow the HTML specification either, it seems to treat detached windows
655 the same way as cross-origin ones. As a result, it only exposed properties that are visible cross-origin when
656 a window is detached / frameless.
658 [1] https://html.spec.whatwg.org/#windowproxy-get
660 No new tests, updated existingt tests.
662 * bindings/js/JSDOMWindowCustom.cpp:
663 (WebCore::jsDOMWindowGetOwnPropertySlotRestrictedAccess):
664 (WebCore::JSDOMWindow::getOwnPropertySlot):
665 (WebCore::JSDOMWindow::getOwnPropertySlotByIndex):
667 2018-01-09 Darin Adler <darin@apple.com>
669 Further refinement to list item and counter code after "list-item" counter fix
670 https://bugs.webkit.org/show_bug.cgi?id=181426
672 Reviewed by Zalan Bujtas.
674 * css/StyleBuilderCustom.h:
675 (WebCore::StyleBuilderCustom::applyInheritCounter): Use auto.
676 (WebCore::StyleBuilderCustom::applyValueCounter): Use auto, removed unneeded
677 null checks for things that can never be null, moved the saturated addition
678 here and got rid of the addIncrementValue function from CounterDirectives.
679 Use the saturatedAddition function from SaturatedArithmetic.h instead of the
680 much less efficient one that did the same thing, CounterDirectives::addClamped.
682 * rendering/RenderCounter.cpp:
683 (WebCore::listItemCounterDirectives): Use aggregate syntax for the return
685 (WebCore::planCounter): Changed to use a struct return value instead of two
686 out arguments. Use the saturatedAddition function from SaturatedArithmetic.h
687 instead of the much less efficient one that did the same thing,
688 CounterDirectives::addClamped.
689 (WebCore::findPlaceForCounter): Changed to use a struct return value instead
690 of two out arguments.
691 (WebCore::makeCounterNode): Updated for the above changes. Changed code to
692 use add instead of both get and set. Updated to keep the counter maps inside
693 the values of the "map of maps" instead of using a unique_ptr and allocating
694 each one on the heap.
695 (WebCore::destroyCounterNodeWithoutMapRemoval): Changed argument to a reference
696 instead of a pointer. Updated for changes to the map. Use RefPtr more
698 (WebCore::RenderCounter::destroyCounterNodes): Use iterators less.
699 (WebCore::RenderCounter::destroyCounterNode): Ditto.
700 (WebCore::RenderCounter::rendererRemovedFromTree): Add a check of
701 hasCounterNodeMap here before calling destroyCounterNodes, so that function
702 can assume the flag is true (both other callers already check it).
703 (WebCore::updateCounters): Use auto and update for changes above.
704 (WebCore::RenderCounter::rendererStyleChanged): Use modern for loops instead
706 (showCounterRendererTree): Use auto and udpate for changes above.
708 * rendering/RenderListItem.cpp:
709 (WebCore::enclosingList): Stop referring to elements as "nodes". Changed
710 the local variable names for clarity.
711 (WebCore::nextListItemHelper): Renamed from nextListItem since it's not
712 intended to be called directly and we want to use a function pointer to
713 nextListItem. Fixed the algorithm to correctly handle ad hoc "lists" that
714 are not actually HTML list elements, using the definition in the enclosingList
715 function as the previousListItem function already did.
716 (WebCore::nextListItem): Updated for name changes.
717 (WebCore::firstListItem): Renamed from nextListItem for clarity.
718 (WebCore::previousListItem): Rewrote loop so it doesn't have to do things
719 so strangely when we find another list.
720 (WebCore::RenderListItem::updateItemValuesForOrderedList): Use auto and
721 update local variable names.
722 (WebCore::RenderListItem::itemCountForOrderedList): Ditto.
723 (WebCore::RenderListItem::updateValueNow const): Rewrote to use an iterative
724 algorithm instead of a recursive one. Fixes the FIXME here.
725 (WebCore::RenderListItem::updateValue): Use m_valueWasSetExplicitly
726 instead of m_explicitValue.
727 (WebCore::RenderListItem::explicitValueChanged): Use auto and simplified
729 (WebCore::RenderListItem::setExplicitValue): Set m_valueWasSetExplicitly
730 instead of m_explicitValue.
731 (WebCore::previousOrNextItem): Deleted.
732 (WebCore::RenderListItem::updateListMarkerNumbers): Streamlined the loop
733 a bit and used a fucntion pointer to handle the two different directions.
734 (WebCore::RenderListItem::isInReversedOrderedList const): Simplified by
735 getting rid of an unneeded use of pointers and local variables.
737 * rendering/RenderListItem.h: Use a boolean, m_valueWasSetExplicitly,
738 instead of a separate optional m_explicitValue.
740 * rendering/style/CounterDirectives.h: Since all the code in this file was
741 rewritten, removed old copyrights. Deleted the addIncrementValue function,
742 since it is clear enough in the one call site in the style builder.
743 Deleted the addClamped function because it was just a much slower
744 version of the saturatedAddition function. Made == and != into constexpr
745 functions since they are simple enough to be.
747 * rendering/style/RenderStyle.cpp:
748 (WebCore::RenderStyle::getCounterDirectives const): Deleted. Caller can
749 handle this just fine without a helper function.
750 * rendering/style/RenderStyle.h: Ditto.
752 2018-01-09 Myles C. Maxfield <mmaxfield@apple.com>
754 font-display:fallback can cause a visual flash (which is supposed to be impossible)
755 https://bugs.webkit.org/show_bug.cgi?id=181374
757 Reviewed by Simon Fraser.
759 A FontCascade represents an entire font-family fallback list, but sometimes we need to pull out a single
760 representative font from the list to calculate things like line height. Previously, if the first item in
761 the font-family list was in the middle of being downloaded, this representative font was hardcoded to be
762 Times. However, when actually laying out and drawing the glyphs, we have logic to skip the interstitial
763 Times if there are any installed fonts present in the font-family list (so you wouldn't ever actually
764 see Times). This means that line height (among other things) was being calculated as if Times was used,
765 but in reality, some other font from the font-family list was being used.
767 Alone, this isn't a huge problem, but font-display:fallback makes a font transition between "timed out"
768 and "failed," and when the font hits the failed state, the representative font skips over the cancelled
769 item and hits the next item in the fallback list. This means that line heights will change, which causes
770 a visual flash, even when font-display:fallback is specified.
772 The solution is simply to educate the logic which identifies this representative font so that it
773 understands what to do for currently-loading fonts.
775 Tests: fast/text/font-display/swap-flash.html
777 * platform/graphics/FontCascadeFonts.h:
778 (WebCore::FontCascadeFonts::primaryFont):
779 * rendering/line/BreakingContext.h:
780 (WebCore::textWidth):
782 2018-01-04 Filip Pizlo <fpizlo@apple.com>
784 CodeBlocks should be in IsoSubspaces
785 https://bugs.webkit.org/show_bug.cgi?id=180884
787 Reviewed by Saam Barati.
789 No new tests because no new behavior.
791 Adopting new parallel constraint API, so that more of the logic of doing parallel
792 constraint solving is shared between the DOM's output constraints and JSC's output
795 * bindings/js/DOMGCOutputConstraint.cpp:
796 (WebCore::DOMGCOutputConstraint::executeImpl):
797 (WebCore::DOMGCOutputConstraint::doParallelWorkImpl): Deleted.
798 (WebCore::DOMGCOutputConstraint::finishParallelWorkImpl): Deleted.
799 * bindings/js/DOMGCOutputConstraint.h:
801 2018-01-08 Simon Fraser <simon.fraser@apple.com>
803 Clean up Marquee-related enums
804 https://bugs.webkit.org/show_bug.cgi?id=181347
806 Reviewed by Anders Carlsson.
808 Modernize EMarqueeBehavior and EMarqueeDirection enums. Stop using the weird negative
809 values in the MarqueeDirection and do manual reverse direction mapping.
811 Make some member functions of RenderMarquee private.
813 Stop using bitfields in RenderMarquee because the memory saving is not worth it, and doing so
814 allows us to use modern initialization.
818 * css/CSSPrimitiveValueMappings.h:
819 (WebCore::CSSPrimitiveValue::CSSPrimitiveValue):
820 (WebCore::CSSPrimitiveValue::operator MarqueeBehavior const):
821 (WebCore::CSSPrimitiveValue::operator MarqueeDirection const):
822 (WebCore::CSSPrimitiveValue::operator EMarqueeBehavior const): Deleted.
823 (WebCore::CSSPrimitiveValue::operator EMarqueeDirection const): Deleted.
824 * css/StyleResolver.cpp:
825 (WebCore::StyleResolver::adjustRenderStyle):
826 * rendering/RenderBox.cpp:
827 (WebCore::RenderBox::sizesLogicalWidthToFitContent const):
828 * rendering/RenderLayer.cpp:
829 (WebCore::RenderLayer::calculateClipRects const):
830 * rendering/RenderMarquee.cpp:
831 (WebCore::RenderMarquee::RenderMarquee):
832 (WebCore::reverseDirection):
833 (WebCore::RenderMarquee::direction const):
834 (WebCore::RenderMarquee::isHorizontal const):
835 (WebCore::RenderMarquee::computePosition):
836 (WebCore::RenderMarquee::start):
837 (WebCore::RenderMarquee::updateMarqueePosition):
838 (WebCore::RenderMarquee::updateMarqueeStyle):
839 (WebCore::RenderMarquee::timerFired):
840 * rendering/RenderMarquee.h:
841 * rendering/style/RenderStyle.h:
842 (WebCore::RenderStyle::marqueeBehavior const):
843 (WebCore::RenderStyle::marqueeDirection const):
844 (WebCore::RenderStyle::setMarqueeDirection):
845 (WebCore::RenderStyle::setMarqueeBehavior):
846 (WebCore::RenderStyle::initialMarqueeBehavior):
847 (WebCore::RenderStyle::initialMarqueeDirection):
848 * rendering/style/RenderStyleConstants.h:
849 * rendering/style/StyleMarqueeData.cpp:
850 (WebCore::StyleMarqueeData::StyleMarqueeData):
851 * rendering/style/StyleMarqueeData.h:
853 2018-01-09 Jer Noble <jer.noble@apple.com>
855 Many CVDisplayLink threads created and destroyed while watching a YouTube video
856 https://bugs.webkit.org/show_bug.cgi?id=181396
858 Reviewed by Simon Fraser.
860 When watching some YouTube videos (or any video with default controls), event handlers for
861 the "timeupdate" event which use rAF will cause the underlying platform objects to be
862 destroyed in between "timeupdate" events being fired, since they occur every 250ms, and rAF
863 objects are destroyed every 166ms (or 10/60hz). Update this constant to destroy the
864 underlying objects every 333ms (or 20/60hz) so that this common pattern doesn't lead to
865 excessive rAF platform object turnover.
867 * platform/Logging.h:
868 * platform/graphics/DisplayRefreshMonitor.h:
869 (WebCore::DisplayRefreshMonitor::shouldBeTerminated const):
870 * platform/graphics/DisplayRefreshMonitor.cpp:
871 (WebCore::DisplayRefreshMonitor::displayDidRefresh):
872 * platform/graphics/DisplayRefreshMonitorManager.cpp:
873 (WebCore::DisplayRefreshMonitorManager::createMonitorForClient):
874 (WebCore::DisplayRefreshMonitorManager::displayDidRefresh):
876 2018-01-09 Zalan Bujtas <zalan@apple.com>
878 [RenderTreeBuilder] Move RenderElement addChild mutation logic to RenderTreeBuilder
879 https://bugs.webkit.org/show_bug.cgi?id=181451
880 <rdar://problem/36385562>
882 Reviewed by Antti Koivisto.
884 This is about moving code, no cleanup and/or normalization (unfortunately it also means
885 some temporary changes).
887 No change in functionality.
889 * rendering/RenderElement.cpp:
890 (WebCore::RenderElement::addChild):
891 (WebCore::RenderElement::childRequiresTable const): Deleted.
892 * rendering/RenderElement.h:
893 * rendering/updating/RenderTreeBuilder.cpp:
894 (WebCore::RenderTreeBuilder::insertChildToRenderElement):
895 (WebCore::RenderTreeBuilder::childRequiresTable):
896 * rendering/updating/RenderTreeBuilder.h:
897 * rendering/updating/RenderTreeBuilderInline.cpp:
898 (WebCore::RenderTreeBuilder::Inline::insertChildToContinuation):
899 (WebCore::RenderTreeBuilder::Inline::insertChildIgnoringContinuation):
900 (WebCore::RenderTreeBuilder::Inline::newChildIsInline):
901 (WebCore::newChildIsInline): Deleted.
902 * rendering/updating/RenderTreeBuilderInline.h:
904 2018-01-09 Matt Lewis <jlewis3@apple.com>
906 Unreviewed, rolling out r226531.
908 This caused test failures on macOS WK2.
912 "Use no-cache fetch mode when loading main documents with
914 https://bugs.webkit.org/show_bug.cgi?id=181285
915 https://trac.webkit.org/changeset/226531
917 2018-01-09 Michael Saboff <msaboff@apple.com>
919 Unreviewed, rolling out r226600 and r226603
920 https://bugs.webkit.org/show_bug.cgi?id=181351
922 Add a DOM gadget for Spectre testing
925 (WebCore::Comment::Comment):
926 (WebCore::Comment::setReadLength): Deleted.
927 (WebCore::Comment::charCodeAt): Deleted.
928 (WebCore::Comment::clflushReadLength): Deleted.
931 * page/RuntimeEnabledFeatures.cpp:
932 (WebCore::RuntimeEnabledFeatures::spectreGadgetsEnabled const): Deleted.
933 * page/RuntimeEnabledFeatures.h:
935 2018-01-09 Don Olmstead <don.olmstead@sony.com>
937 Add additional WEBCORE_EXPORTs
938 https://bugs.webkit.org/show_bug.cgi?id=181414
940 Reviewed by Alex Christensen.
942 No new tests. No change in behavior.
945 * dom/ViewportArguments.h:
948 * page/PageOverlayController.h:
949 * platform/ContextMenuItem.h:
950 * platform/Pasteboard.h:
951 * platform/SharedBuffer.h:
952 * platform/UserAgent.h:
953 * platform/graphics/GLContext.h:
954 * platform/graphics/GraphicsContext.h:
955 * platform/graphics/GraphicsLayerTransform.h:
956 * platform/graphics/PlatformDisplay.h:
957 * platform/graphics/cairo/BackingStoreBackendCairoImpl.h:
958 * platform/graphics/cairo/PlatformContextCairo.h:
959 * platform/graphics/cairo/RefPtrCairo.h:
960 * platform/graphics/nicosia/NicosiaBuffer.h:
961 * platform/graphics/nicosia/NicosiaPaintingEngine.h:
962 * platform/graphics/texmap/TextureMapper.h:
963 * platform/graphics/texmap/TextureMapperAnimation.h:
964 * platform/graphics/texmap/TextureMapperBackingStore.h:
965 * platform/graphics/texmap/TextureMapperFPSCounter.h:
966 * platform/graphics/texmap/TextureMapperLayer.h:
967 (WebCore::TextureMapperLayer::TextureMapperLayer): Deleted.
968 (WebCore::TextureMapperLayer::setID): Deleted.
969 (WebCore::TextureMapperLayer::id): Deleted.
970 (WebCore::TextureMapperLayer:: const): Deleted.
971 (WebCore::TextureMapperLayer::setScrollClient): Deleted.
972 (WebCore::TextureMapperLayer::setIsScrollable): Deleted.
973 (WebCore::TextureMapperLayer::isScrollable const): Deleted.
974 (WebCore::TextureMapperLayer::textureMapper const): Deleted.
975 (WebCore::TextureMapperLayer::setTextureMapper): Deleted.
976 (WebCore::TextureMapperLayer::drawsContent const): Deleted.
977 (WebCore::TextureMapperLayer::contentsAreVisible const): Deleted.
978 (WebCore::TextureMapperLayer::size const): Deleted.
979 (WebCore::TextureMapperLayer::opacity const): Deleted.
980 (WebCore::TextureMapperLayer::transform const): Deleted.
981 (WebCore::TextureMapperLayer::hasFilters const): Deleted.
982 (WebCore::TextureMapperLayer::isShowingRepaintCounter const): Deleted.
983 (WebCore::TextureMapperLayer::fixedToViewport const): Deleted.
984 (WebCore::TextureMapperLayer::rootLayer const): Deleted.
985 (WebCore::TextureMapperLayer::texture): Deleted.
986 (WebCore::TextureMapperLayer::adjustedPosition const): Deleted.
987 (WebCore::TextureMapperLayer::layerRect const): Deleted.
988 (WebCore::TextureMapperLayer::State::State): Deleted.
989 * platform/graphics/texmap/TextureMapperPlatformLayerProxy.h:
990 * platform/graphics/texmap/TextureMapperTile.h:
991 * platform/graphics/texmap/coordinated/CoordinatedGraphicsLayer.h:
992 (WebCore::CoordinatedGraphicsLayer::isScrollable const): Deleted.
993 (WebCore::CoordinatedGraphicsLayer::id const): Deleted.
994 (WebCore::CoordinatedGraphicsLayer::coverRect const): Deleted.
995 (WebCore::CoordinatedGraphicsLayer::fixedToViewport const): Deleted.
996 * platform/graphics/texmap/coordinated/CoordinatedImageBacking.h:
997 (WebCore::CoordinatedImageBacking::id const): Deleted.
998 * platform/graphics/texmap/coordinated/TiledBackingStore.h:
999 * platform/graphics/texmap/coordinated/UpdateAtlas.h:
1000 * platform/network/AuthenticationChallengeBase.h:
1001 * platform/network/CredentialBase.h:
1002 * platform/network/ProtectionSpaceBase.h:
1003 * platform/network/curl/SocketStreamHandleImpl.h:
1004 * replay/UserInputBridge.h:
1006 2018-01-09 Basuke Suzuki <Basuke.Suzuki@sony.com>
1008 [Curl] Move FormData related tasks into new CurlFormDataStream class.
1009 https://bugs.webkit.org/show_bug.cgi?id=181106
1011 Reviewed by Alex Christensen.
1013 No new tests because no behavior change.
1015 * platform/Curl.cmake:
1016 * platform/network/curl/CurlFormDataStream.cpp: Added.
1017 (WebCore::CurlFormDataStream::CurlFormDataStream):
1018 (WebCore::CurlFormDataStream::~CurlFormDataStream):
1019 (WebCore::CurlFormDataStream::clean):
1020 (WebCore::CurlFormDataStream::shouldUseChunkTransfer):
1021 (WebCore::CurlFormDataStream::totalSize):
1022 (WebCore::CurlFormDataStream::computeContentLength):
1023 (WebCore::CurlFormDataStream::read):
1024 (WebCore::CurlFormDataStream::readFromFile):
1025 (WebCore::CurlFormDataStream::readFromData):
1026 * platform/network/curl/CurlFormDataStream.h: Renamed from Source/WebCore/platform/network/curl/FormDataStreamCurl.h.
1027 (WebCore::CurlFormDataStream::elementSize):
1028 (WebCore::CurlFormDataStream::totalReadSize):
1029 * platform/network/curl/CurlRequest.cpp:
1030 (WebCore::CurlRequest::CurlRequest):
1031 (WebCore::CurlRequest::willSendData):
1032 (WebCore::CurlRequest::finalizeTransfer):
1033 (WebCore::CurlRequest::setupPUT):
1034 (WebCore::CurlRequest::setupPOST):
1035 (WebCore::CurlRequest::setupSendData):
1036 (WebCore::CurlRequest::resolveBlobReferences): Deleted.
1037 (WebCore::CurlRequest::setupFormData): Deleted.
1038 * platform/network/curl/CurlRequest.h:
1039 * platform/network/curl/FormDataStreamCurl.cpp: Removed.
1041 2018-01-09 Zalan Bujtas <zalan@apple.com>
1043 [RenderTreeBuilder] Move MathML addChild logic to RenderTreeBuilder
1044 https://bugs.webkit.org/show_bug.cgi?id=181443
1045 <rdar://problem/36380228>
1047 Reviewed by Antti Koivisto.
1049 This is about moving code, no cleanup and/or normalization (unfortunately it also means
1050 some temporary changes).
1052 No change in functionality.
1055 * WebCore.xcodeproj/project.pbxproj:
1056 * rendering/mathml/RenderMathMLFenced.cpp:
1057 (WebCore::RenderMathMLFenced::addChild):
1058 (WebCore::RenderMathMLFenced::createMathMLOperator): Deleted.
1059 (WebCore::RenderMathMLFenced::makeFences): Deleted.
1060 * rendering/mathml/RenderMathMLFenced.h:
1061 * rendering/updating/RenderTreeBuilder.cpp:
1062 (WebCore::RenderTreeBuilder::RenderTreeBuilder):
1063 (WebCore::RenderTreeBuilder::insertChildToRenderMathMLFenced):
1064 * rendering/updating/RenderTreeBuilder.h:
1065 (WebCore::RenderTreeBuilder::mathMLBuilder):
1067 2018-01-09 Antti Koivisto <antti@apple.com>
1069 Blank page except for inner iframes because pending stylesheets cause style.isNotFinal() to be true
1070 https://bugs.webkit.org/show_bug.cgi?id=180940
1071 <rdar://problem/36116507>
1073 Reviewed by Darin Adler.
1075 Test: http/tests/local/loading-stylesheet-import-remove.html
1077 If a <link> referencing a stylesheet containing an @import that was still loading was removed
1078 from the document, the loading state was never cleared. For head stylesheets this blocked
1079 rendering permanently.
1081 Test reduction by Justin Ridgewell.
1083 * html/HTMLLinkElement.cpp:
1084 (WebCore::HTMLLinkElement::removedFromAncestor):
1086 Test if the stylesheet it loading before clearing the pointer.
1088 2018-01-09 Dan Bernstein <mitz@apple.com>
1090 Removed some empty directories that were left behind
1092 * Modules/indieui: Removed.
1093 * Modules/vibration: Removed.
1094 * platform/graphics/ca/mac: Removed.
1095 * platform/mediastream/gstreamer: Removed.
1097 2018-01-09 Antoine Quint <graouts@apple.com>
1099 Refactor timing function solving code
1100 https://bugs.webkit.org/show_bug.cgi?id=181428
1102 Reviewed by Dean Jackson.
1104 We have duplicated code to solve "cubic" and "steps" timing functions in AnimationBase and TextureMapperAnimation,
1105 and we will soon need similar code to deal with timing functions in Web Animations. We move this code into TimingFunction
1106 with a single transformTime() function that can be called directly on the timing function rather than having callers
1107 figure out what type timing functions are to solve them.
1109 No test as there shouldn't be any behavior change.
1111 * page/animation/AnimationBase.cpp:
1112 (WebCore::AnimationBase::progress const):
1113 (WebCore::solveEpsilon): Deleted.
1114 (WebCore::solveCubicBezierFunction): Deleted.
1115 (WebCore::solveStepsFunction): Deleted.
1116 (WebCore::solveSpringFunction): Deleted.
1117 * platform/animation/TimingFunction.cpp:
1118 (WebCore::TimingFunction::transformTime const):
1119 * platform/animation/TimingFunction.h:
1120 * platform/graphics/texmap/TextureMapperAnimation.cpp:
1121 (WebCore::TextureMapperAnimation::apply):
1122 (WebCore::solveEpsilon): Deleted.
1123 (WebCore::solveCubicBezierFunction): Deleted.
1124 (WebCore::solveStepsFunction): Deleted.
1125 (WebCore::applyTimingFunction): Deleted.
1127 2018-01-09 Michael Catanzaro <mcatanzaro@igalia.com>
1129 Unreviewed, silence -Wunused-parameter warning
1131 * testing/js/WebCoreTestSupport.cpp:
1132 (WebCoreTestSupport::setupNewlyCreatedServiceWorker):
1134 2018-01-09 Said Abou-Hallawa <sabouhallawa@apple.com>
1136 Check Image::m_image is not null in ImageLoader::decode()
1137 https://bugs.webkit.org/show_bug.cgi?id=180386
1138 <rdar://problem/34634483>
1140 Reviewed by Tim Horton.
1142 Ensure ImageLoader::m_image is not null before referencing it.
1144 * loader/ImageLoader.cpp:
1145 (WebCore::ImageLoader::decode):
1147 2018-01-09 Yusuke Suzuki <utatane.tea@gmail.com>
1149 [FreeType] Use FastMalloc for FreeType
1150 https://bugs.webkit.org/show_bug.cgi?id=181387
1152 Reviewed by Michael Catanzaro.
1154 Add custom memory allocator for FreeType to use FastMalloc.
1156 * platform/graphics/freetype/FontCustomPlatformDataFreeType.cpp:
1157 (WebCore::initializeFreeTypeLibrary):
1158 (WebCore::createFontCustomPlatformData):
1160 2018-01-09 Zalan Bujtas <zalan@apple.com>
1162 [RenderTreeBuilder] Move RenderTable* addChild mutation logic to RenderTreeBuilder
1163 https://bugs.webkit.org/show_bug.cgi?id=181424
1164 <rdar://problem/36368628>
1166 Reviewed by Antti Koivisto.
1168 This is about moving code, no cleanup and/or normalization (unfortunately it also means
1169 some temporary changes).
1171 Covered by existing tests.
1173 * rendering/RenderTable.cpp:
1174 (WebCore::RenderTable::addChild):
1175 * rendering/RenderTableRow.cpp:
1176 (WebCore::RenderTableRow::addChild):
1177 * rendering/RenderTableSection.cpp:
1178 (WebCore::RenderTableSection::addChild):
1179 * rendering/updating/RenderTreeBuilder.cpp:
1180 (WebCore::RenderTreeBuilder::insertChildToRenderTable):
1181 (WebCore::RenderTreeBuilder::insertChildToRenderTableSection):
1182 (WebCore::RenderTreeBuilder::insertChildToRenderTableRow):
1183 * rendering/updating/RenderTreeBuilder.h:
1184 * rendering/updating/RenderTreeBuilderTable.cpp:
1185 (WebCore::RenderTreeBuilder::Table::insertChild):
1186 * rendering/updating/RenderTreeBuilderTable.h:
1188 2018-01-09 Zalan Bujtas <zalan@apple.com>
1190 [RenderTreeBuilder] Transition Render*::addChild() calls to RenderTreeBuilder::insertChildToRender*()
1191 https://bugs.webkit.org/show_bug.cgi?id=181407
1192 <rdar://problem/36361176>
1194 Reviewed by Antti Koivisto.
1196 Eventually all mutation will be going through the RenderTreeBuilder.
1198 No change in functionality.
1200 * rendering/RenderGrid.cpp:
1201 (WebCore::RenderGrid::addChild):
1202 * rendering/RenderMultiColumnFlow.cpp:
1203 (WebCore::RenderMultiColumnFlow::processPossibleSpannerDescendant):
1204 * rendering/RenderRubyRun.cpp:
1205 (WebCore::RenderRubyRun::rubyBaseSafe):
1206 * rendering/updating/RenderTreeBuilderBlockFlow.cpp:
1207 (WebCore::RenderTreeBuilder::BlockFlow::insertChild):
1208 * rendering/updating/RenderTreeBuilderMultiColumn.cpp:
1209 (WebCore::RenderTreeBuilder::MultiColumn::createFragmentedFlow):
1210 * rendering/updating/RenderTreeBuilderRuby.cpp:
1211 (WebCore::RenderTreeBuilder::Ruby::findOrCreateParentForChild):
1213 2018-01-09 Philippe Normand <pnormand@igalia.com>
1215 [GStreamer] Ensure SleepDisabler is not held by pages in page cache
1216 https://bugs.webkit.org/show_bug.cgi?id=180197
1218 Reviewed by Eric Carlson.
1220 The sleep disabler is now checked again whenever the media player
1221 playback state or other characteristic has changed in the media
1224 * html/HTMLMediaElement.cpp:
1225 (WebCore::HTMLMediaElement::mediaPlayerPlaybackStateChanged):
1226 Check if the sleep disabler state needs to be updated or not.
1227 (WebCore::HTMLMediaElement::mediaPlayerCharacteristicChanged): Ditto.
1228 (WebCore::HTMLMediaElement::shouldDisableSleep const): Enable
1229 sleep disabler for GTK and WPE ports.
1231 2018-01-09 Philippe Normand <pnormand@igalia.com>
1233 [GStreamer] Missing notifications to player client
1234 https://bugs.webkit.org/show_bug.cgi?id=181432
1236 Reviewed by Carlos Garcia Campos.
1238 We now properly emit playback state changes and characteristic
1239 updates to the client when the pipeline state changes and when
1240 audio/video tracks changes are detected.
1242 * platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp:
1243 (WebCore::MediaPlayerPrivateGStreamer::playbackPosition const):
1244 This debug message appears too often in logs. Demote.
1245 (WebCore::MediaPlayerPrivateGStreamer::durationMediaTime const): Ditto.
1246 (WebCore::MediaPlayerPrivateGStreamer::notifyPlayerOfVideo): Emit
1247 charasteristicChanged notification.
1248 (WebCore::MediaPlayerPrivateGStreamer::notifyPlayerOfAudio): Ditto.
1249 (WebCore::MediaPlayerPrivateGStreamer::updateStates): Keep track
1250 of old and current pipeline state.
1251 * platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.h:
1254 2018-01-09 Youenn Fablet <youenn@apple.com>
1256 Add CSP support to service workers
1257 https://bugs.webkit.org/show_bug.cgi?id=181385
1259 Reviewed by Chris Dumez.
1261 Covered by rebased tests.
1263 Added recovery of CSP information from WorkerScriptLoader.
1264 Added plumbing to pass the CSP information to Service Workers.
1265 Did not add persistency support for the CSP information as this requires changing the SQL database schema.
1266 This will be done in a follow-up.
1268 * workers/WorkerScriptLoader.cpp:
1269 (WebCore::WorkerScriptLoader::loadAsynchronously):
1270 (WebCore::WorkerScriptLoader::didReceiveResponse):
1271 * workers/WorkerScriptLoader.h:
1272 (WebCore::WorkerScriptLoader::contentSecurityPolicy const):
1273 * workers/service/SWClientConnection.cpp:
1274 (WebCore::SWClientConnection::finishedFetchingScript):
1275 (WebCore::SWClientConnection::failedFetchingScript):
1276 * workers/service/SWClientConnection.h:
1277 * workers/service/ServiceWorkerContainer.cpp:
1278 (WebCore::ServiceWorkerContainer::addRegistration):
1279 (WebCore::ServiceWorkerContainer::jobFailedWithException):
1280 (WebCore::ServiceWorkerContainer::jobResolvedWithRegistration):
1281 (WebCore::ServiceWorkerContainer::jobFinishedLoadingScript):
1282 * workers/service/ServiceWorkerContainer.h:
1283 * workers/service/ServiceWorkerContextData.cpp:
1284 (WebCore::ServiceWorkerContextData::isolatedCopy const):
1285 * workers/service/ServiceWorkerContextData.h:
1286 (WebCore::ServiceWorkerContextData::encode const):
1287 (WebCore::ServiceWorkerContextData::decode):
1288 * workers/service/ServiceWorkerFetchResult.h:
1289 (WebCore::ServiceWorkerFetchResult::encode const):
1290 (WebCore::ServiceWorkerFetchResult::decode):
1291 * workers/service/ServiceWorkerGlobalScope.cpp:
1292 (WebCore::ServiceWorkerGlobalScope::create):
1293 * workers/service/ServiceWorkerGlobalScope.h:
1294 * workers/service/ServiceWorkerJob.cpp:
1295 (WebCore::ServiceWorkerJob::notifyFinished):
1296 * workers/service/ServiceWorkerJobClient.h:
1297 * workers/service/context/ServiceWorkerThread.cpp:
1298 (WebCore::ServiceWorkerThread::ServiceWorkerThread):
1299 (WebCore::ServiceWorkerThread::createWorkerGlobalScope):
1300 * workers/service/server/RegistrationDatabase.cpp:
1301 (WebCore::v1RecordsTableSchema):
1302 (WebCore::RegistrationDatabase::importRecords):
1303 * workers/service/server/SWServer.cpp:
1304 (WebCore::SWServer::updateWorker):
1305 (WebCore::SWServer::installContextData):
1306 * workers/service/server/SWServer.h:
1307 * workers/service/server/SWServerJobQueue.cpp:
1308 (WebCore::SWServerJobQueue::scriptFetchFinished):
1309 * workers/service/server/SWServerWorker.cpp:
1310 (WebCore::SWServerWorker::SWServerWorker):
1311 (WebCore::m_contentSecurityPolicy):
1312 (WebCore::SWServerWorker::contextData const):
1313 * workers/service/server/SWServerWorker.h:
1315 2018-01-09 Antoine Quint <graouts@apple.com>
1317 Provide a method to obtain a CSS value from a RenderStyle by CSSPropertyID
1318 https://bugs.webkit.org/show_bug.cgi?id=181429
1320 Reviewed by Antti Koivisto.
1322 The KeyframeEffect.getKeyframes() method from Web Animations requires that we return all styles applied at a given
1323 animation effect keyframe. When we parse keyframes, we create RenderStyle objects that are stored in a KeyframeList,
1324 as well as a list of CSSPropertyIDs that are animated. In order to provide the list of animated properties and their
1325 values when calling getKeyframes(), we need to be able to read back from the RenderStyle we create. As a first step
1326 towards this goal, we factor some code in ComputedStyleExtractor::propertyValue() out into a new method
1327 ComputedStyleExtractor::valueForPropertyinStyle() which takes in an explicit RenderStyle.
1329 No test as this is simply refactoring that doesn't change or add behavior.
1331 * css/CSSComputedStyleDeclaration.cpp:
1332 (WebCore::ComputedStyleExtractor::propertyValue):
1333 (WebCore::ComputedStyleExtractor::valueForPropertyinStyle):
1334 * css/CSSComputedStyleDeclaration.h:
1336 2018-01-09 Youenn Fablet <youenn@apple.com>
1338 SWClientConnection should not keep references to service worker jobs
1339 https://bugs.webkit.org/show_bug.cgi?id=181381
1341 Reviewed by Chris Dumez.
1343 Difficult to test determiniscally but corresponding crash log should no longer happen in debug builds.
1345 Stopped passing ServiceWorkerJob references from ServiceWorkerContainer (potentially in service worker thread) to SWClientConnection (main thread).
1346 Instead pass job identifiers and related data to the main thread.
1348 Minor refactoring to use ServiceWorkerJobIdentifier instead of ServiceWorkerJobDataIdentifier which contains more data than needed.
1350 * workers/service/SWClientConnection.cpp:
1351 (WebCore::SWClientConnection::scheduleJob):
1352 (WebCore::SWClientConnection::failedFetchingScript):
1353 (WebCore::SWClientConnection::postTaskForJob):
1354 (WebCore::SWClientConnection::jobRejectedInServer):
1355 (WebCore::SWClientConnection::registrationJobResolvedInServer):
1356 (WebCore::SWClientConnection::unregistrationJobResolvedInServer):
1357 (WebCore::SWClientConnection::startScriptFetchForServer):
1358 (WebCore::SWClientConnection::clearPendingJobs):
1359 (WebCore::SWClientConnection::finishedFetchingScript): Deleted.
1360 * workers/service/SWClientConnection.h:
1361 * workers/service/ServiceWorkerContainer.cpp:
1362 (WebCore::ServiceWorkerContainer::scheduleJob):
1363 (WebCore::ServiceWorkerContainer::startScriptFetchForJob):
1364 (WebCore::ServiceWorkerContainer::jobFinishedLoadingScript):
1365 (WebCore::ServiceWorkerContainer::jobFailedLoadingScript):
1366 * workers/service/ServiceWorkerContainer.h:
1367 * workers/service/server/SWServer.cpp:
1368 (WebCore::SWServer::rejectJob):
1369 (WebCore::SWServer::resolveRegistrationJob):
1370 (WebCore::SWServer::resolveUnregistrationJob):
1371 (WebCore::SWServer::startScriptFetch):
1372 * workers/service/server/SWServer.h:
1374 2018-01-09 Carlos Garcia Campos <cgarcia@igalia.com>
1376 REGRESSION(r224460): Text fields sometimes get "messed up"
1377 https://bugs.webkit.org/show_bug.cgi?id=181115
1379 Reviewed by Zalan Bujtas.
1381 Add the scroll position to paint offset before computing the clip rectangle. Before r224460, scroll position was
1382 substracted after the clip rectangle was computed.
1384 * rendering/LayoutState.cpp:
1385 (WebCore::LayoutState::computeClipRect):
1387 2018-01-09 Ali Juma <ajuma@chromium.org>
1389 Implement VisualViewport API events
1390 https://bugs.webkit.org/show_bug.cgi?id=179386
1392 Reviewed by Frédéric Wang.
1394 Implement the events (resize and scroll) defined by the Visual Viewport API
1395 (https://wicg.github.io/visual-viewport/#events).
1397 This is behind the VisualViewportAPI experimental feature flag.
1399 In order to detect when events need to be fired, change the computation of
1400 Visual Viewport attributes to happen whenever the layout viewport is updated
1401 rather than only on-demand.
1403 Tests: fast/visual-viewport/resize-event-fired-window-resized.html
1404 fast/visual-viewport/resize-event-fired.html
1405 fast/visual-viewport/scroll-event-fired.html
1408 (WebCore::Document::addListenerTypeIfNeeded):
1409 Add support for tracking resize event listeners.
1411 * dom/DocumentEventQueue.cpp:
1412 (WebCore::DocumentEventQueue::enqueueOrDispatchScrollEvent):
1413 (WebCore::DocumentEventQueue::enqueueScrollEvent):
1414 Factored out of enqueueOrDispatchScrollEvent so that this logic can be reused
1415 for Visual Viewport scroll events.
1416 (WebCore::DocumentEventQueue::enqueueResizeEvent):
1417 (WebCore::DocumentEventQueue::pendingEventTimerFired):
1418 * dom/DocumentEventQueue.h:
1419 * page/FrameView.cpp:
1420 (WebCore::FrameView::updateLayoutViewport):
1421 * page/VisualViewport.cpp:
1422 (WebCore::VisualViewport::addEventListener):
1423 (WebCore::layoutIfNonNull):
1424 (WebCore::VisualViewport::offsetLeft const):
1425 Remove attribute computation logic since this now happens during update().
1426 (WebCore::VisualViewport::offsetTop const): Ditto.
1427 (WebCore::VisualViewport::pageLeft const): Ditto.
1428 (WebCore::VisualViewport::pageTop const): Ditto.
1429 (WebCore::VisualViewport::width const): Ditto.
1430 (WebCore::VisualViewport::height const): Ditto.
1431 (WebCore::VisualViewport::scale const):
1432 (WebCore::VisualViewport::update):
1433 Added. Computes all of the Visual Viewport attributes and determines
1434 whether events need to be fired.
1435 (WebCore::VisualViewport::enqueueResizeEvent):
1436 (WebCore::VisualViewport::enqueueScrollEvent):
1437 (WebCore::getFrameViewAndLayoutIfNonNull): Deleted.
1438 * page/VisualViewport.h:
1440 2018-01-09 Yacine Bandou <yacine.bandou_ext@softathome.com>
1442 [EME] Add the CENC initData support in ClearKey CDM
1443 https://bugs.webkit.org/show_bug.cgi?id=180081
1445 Reviewed by Xabier Rodriguez-Calvar.
1447 Add the "cenc" initDataType support in ClearKey CDM.
1448 Parse the CENC initData and extract the KIDs by following the W3C spec
1449 https://www.w3.org/TR/eme-initdata-cenc/#common-system
1451 Tests: media/encrypted-media/clearKey/clearKey-message-cenc-event.html
1452 media/encrypted-media/clearKey/clearKey-message-cenc-event-mse.html
1454 * platform/encryptedmedia/clearkey/CDMClearKey.cpp:
1455 (WebCore::extractKeyidsLocationFromCencInitData):
1456 (WebCore::isCencInitData):
1457 (WebCore::extractKeyidsFromCencInitData):
1458 (WebCore::CDMPrivateClearKey::supportsInitDataType const):
1459 (WebCore::CDMPrivateClearKey::supportsInitData const):
1460 (WebCore::CDMInstanceClearKey::requestLicense):
1462 2018-01-09 Zan Dobersek <zdobersek@igalia.com>
1464 [Cairo] Pass state values directly to Cairo operations
1465 https://bugs.webkit.org/show_bug.cgi?id=181389
1467 Reviewed by Carlos Garcia Campos.
1469 Instead of passing reference to the GraphicsContextState object to
1470 various Cairo operations, only pass the required state values. This
1471 makes it explicit what state values are used in these operations, at the
1472 expense of some long parameter lists, but this will be better addressed
1473 by future refactoring of this code into more concise functions.
1475 No new tests -- no change in functionality.
1477 * platform/graphics/cairo/CairoOperations.cpp:
1478 (WebCore::Cairo::drawGlyphsShadow):
1479 (WebCore::Cairo::dashedLineCornerWidthForStrokeWidth):
1480 (WebCore::Cairo::dashedLinePatternWidthForStrokeWidth):
1481 (WebCore::Cairo::drawGlyphs):
1482 (WebCore::Cairo::drawRect):
1483 (WebCore::Cairo::drawLine):
1484 (WebCore::Cairo::drawEllipse):
1485 * platform/graphics/cairo/CairoOperations.h:
1486 * platform/graphics/cairo/FontCairo.cpp:
1487 (WebCore::FontCascade::drawGlyphs):
1488 * platform/graphics/cairo/GraphicsContextCairo.cpp:
1489 (WebCore::GraphicsContext::drawRect):
1490 (WebCore::GraphicsContext::drawLine):
1491 (WebCore::GraphicsContext::drawEllipse):
1493 2018-01-09 Ryosuke Niwa <rniwa@webkit.org>
1495 Release assert in addResourceTiming when a cache resource is requested during style recalc
1496 https://bugs.webkit.org/show_bug.cgi?id=181137
1497 <rdar://problem/35666574>
1499 Reviewed by Simon Fraser.
1501 Make the dispatching of resourcetimingbufferfull event asynchronous to avoid dispatching it
1502 synchronously during a style resolution when CachedResourceLoader::requestImage requests
1503 a previously loaded image.
1505 We now schedule a timer when the resource timing buffer becomes full, and dispatch the event
1506 when the timer fires. Meanwhile, we have a backup buffer to which additional resource timing
1507 entries would be added. Once the event is dispatched, we refill the buffer exposed to author
1508 scripts. When refilling the buffer results in it becoming full again, we keep repeating the
1509 process of firing resourcetimingbufferfull and re-filling the buffer until either we stop
1510 making progress (i.e. the script didn't increase the number of empty entires in the buffer)
1511 or the backup buffer (at the time we started this process) becomes empty.
1513 Also fixed a bug that we were firing resourcetimingbufferfull event when the last entry that
1514 fits within the buffer size was added instead of when an entry is being added to an already
1515 full buffer. To make this work, the patch introduces m_resourceTimingBufferFullFlag,
1516 representing the concept "resource timing buffer full" flag in the resource timing specification.
1518 Test: http/tests/performance/performance-resource-timing-resourcetimingbufferfull-crash.html
1520 * page/Performance.cpp:
1521 (WebCore::Performance::Performance):
1522 (WebCore::Performance::clearResourceTimings):
1523 (WebCore::Performance::setResourceTimingBufferSize):
1524 (WebCore::Performance::addResourceTiming):
1525 (WebCore::Performance::resourceTimingBufferFullTimerFired):
1526 * page/Performance.h:
1528 2018-01-08 Chris Nardi <csnardi1@gmail.com>
1530 ::first-letter incorrectly selects grapheme pairs
1531 https://bugs.webkit.org/show_bug.cgi?id=181315
1533 Reviewed by Darin Adler.
1535 Grapheme pairs were incorrectly selected by ::first-letter. This
1536 change fixes their selection.
1538 Tests: Updated fast/css/first-letter-punctuation.html and imported
1539 LayoutTests/imported/w3c/web-platform-tests/css/css-pseudo-4/first-letter-004.html
1541 * rendering/updating/RenderTreeBuilderFirstLetter.cpp:
1542 (WebCore::isPunctuationForFirstLetter):
1543 (WebCore::shouldSkipForFirstLetter):
1544 (WebCore::RenderTreeBuilder::FirstLetter::createRenderers):
1546 2018-01-08 Darin Adler <darin@apple.com>
1548 Special list-item counter starts from an incorrect number for ::before and ::after
1549 https://bugs.webkit.org/show_bug.cgi?id=181084
1551 Reviewed by Zalan Bujtas.
1553 Test: fast/css/counters/counter-list-item.html
1555 * Sources.txt: Removed CounterDirectives.cpp.
1556 * WebCore.xcodeproj/project.pbxproj: Ditto.
1558 * css/CSSComputedStyleDeclaration.cpp:
1559 (WebCore::counterToCSSValue): Updated for changes to the CounterDirectives struct.
1560 * css/StyleBuilderCustom.h:
1561 (WebCore::StyleBuilderCustom::applyInheritCounter): Ditto.
1562 (WebCore::StyleBuilderCustom::applyValueCounter): Ditto.
1564 * html/HTMLLIElement.cpp:
1565 (WebCore::HTMLLIElement::parseValue): Call setExplicitValue(std::nullopt) instead
1566 of clearExplicitValue since we are using std::optional now.
1568 * rendering/RenderCounter.cpp:
1569 (WebCore::listItemCounterDirectives): Added. Computes the counter directives that
1570 express the effects on the list-item counter from list item and list elements.
1571 Used something as close to what the CSS 3 draft says as possible. This uses a
1572 negative increment when creating a list to counteract the positive increment done
1573 by a list element, except in the case of an unordered list. This is where the bug
1574 fix actually lies. Also fixed handling of reversed ordered lists at the same time.
1575 (WebCore::planCounter): Refactored to use the function above. Also changed the
1576 code to pay attention to both the counter directives and the implicit ones from
1577 list item and list elements, getting as close as possible to what the specification
1580 * rendering/RenderListItem.cpp:
1581 (WebCore::RenderListItem::RenderListItem): Since we are using std::optional and no
1582 longer using bit fields, simplified the constructor for each list item.
1583 (WebCore::RenderListItem::calcValue const): Deleted.
1584 (WebCore::RenderListItem::updateValueNow const): Merged in all the code from the
1585 old calcValue function, but it is also simpler now since m_value is std::optional.
1586 (WebCore::RenderListItem::updateValue): Updated to use std::optional.
1587 (WebCore::RenderListItem::setExplicitValue): Ditto.
1588 (WebCore::RenderListItem::clearExplicitValue): Deleted.
1589 (WebCore::RenderListItem::updateListMarkerNumbers): Updated to use std::optional.
1590 (WebCore::RenderListItem::isInReversedOrderedList const): Added. This is used by
1591 the counter code so it can decrement instead of incrementing.
1593 * rendering/RenderListItem.h: Updated to use std::optional. Also marked functions
1594 final instead of override and initialized m_notInList after making it not be a
1597 * rendering/style/CounterDirectives.cpp: Removed.
1598 * rendering/style/CounterDirectives.h: Removed most of the CounterDirectives
1599 class and replaced it with a struct with two std::optional. Added an addClamped
1600 function so the counter code can share it with the addIncrementValue function.
1601 If we want to make a faster version that doesn't use double, we can come back
1602 and do that. Also moved the == function to the header since the implementation
1605 * rendering/style/StyleAllInOne.cpp: Removed CounterDirectives.cpp.
1607 * rendering/style/StyleRareNonInheritedData.cpp:
1608 (WebCore::StyleRareNonInheritedData::StyleRareNonInheritedData): Updated to
1609 use std::make_unique directly instead of using a clone function.
1611 2018-01-08 Wenson Hsieh <wenson_hsieh@apple.com>
1613 [Attachment Support] Expose HTMLAttachmentElement.uniqueIdentifier to bindings
1614 https://bugs.webkit.org/show_bug.cgi?id=181420
1615 <rdar://problem/36365827>
1617 Reviewed by Alex Christensen.
1619 Add a new IDL definition for uniqueIdentifier on HTMLAttachmentElement. This allows clients to relate attachment
1620 elements in the DOM to _WKAttachments delivered via Objective-C SPI.
1622 Adjusted existing tests in WKAttachmentTests.
1624 * html/HTMLAttachmentElement.idl:
1626 2018-01-08 Don Olmstead <don.olmstead@sony.com>
1628 AccessibilityARIAGrid does not compile when accessibility is disabled
1629 https://bugs.webkit.org/show_bug.cgi?id=181418
1630 <rdar://problem/36365398>
1632 Reviewed by Chris Fleizach.
1634 No new tests. No change in behavior.
1636 * accessibility/AccessibilityARIAGrid.cpp:
1638 2018-01-08 Commit Queue <commit-queue@webkit.org>
1640 Unreviewed, rolling out r226532 and r226540.
1641 https://bugs.webkit.org/show_bug.cgi?id=181422
1643 jessie says basic browsing does not seem to work (Requested by
1644 alexchristensen on #webkit).
1646 Reverted changesets:
1648 "Add CSP support to service workers"
1649 https://bugs.webkit.org/show_bug.cgi?id=181385
1650 https://trac.webkit.org/changeset/226532
1652 "SWClientConnection should not keep references to service
1654 https://bugs.webkit.org/show_bug.cgi?id=181381
1655 https://trac.webkit.org/changeset/226540
1657 2018-01-08 Wenson Hsieh <wenson_hsieh@apple.com>
1659 [Attachment Support] Expose file name and content type of WKAttachment
1660 https://bugs.webkit.org/show_bug.cgi?id=181390
1661 <rdar://problem/36336837>
1663 Reviewed by Tim Horton.
1665 Add new structs in AttachmentTypes.h and refactor HTMLAttachmentElement::requestData to requestInfo. See below
1666 for more details. Augmented existing API tests in WKAttachmentTests.
1668 * html/AttachmentTypes.h:
1670 Add a struct to represent a snapshot of information about an attachment element.
1672 * html/HTMLAttachmentElement.cpp:
1673 (WebCore::HTMLAttachmentElement::requestInfo):
1674 (WebCore::HTMLAttachmentElement::requestData): Deleted.
1675 * html/HTMLAttachmentElement.h:
1677 Change requestData to requestInfo. Instead of fetching and delivering data via callback, requestInfo returns an
1680 * rendering/RenderThemeIOS.mm:
1681 (WebCore::RenderAttachmentInfo::addLine):
1682 (WebCore::RenderAttachmentInfo::buildWrappedLines):
1683 (WebCore::RenderAttachmentInfo::buildSingleLine):
1684 (WebCore::RenderAttachmentInfo::RenderAttachmentInfo):
1685 (WebCore::RenderThemeIOS::attachmentBaseline const):
1686 (WebCore::paintAttachmentIcon):
1687 (WebCore::paintAttachmentText):
1688 (WebCore::paintAttachmentProgress):
1689 (WebCore::attachmentBorderPath):
1690 (WebCore::RenderThemeIOS::paintAttachment):
1691 (WebCore::AttachmentInfo::addLine): Deleted.
1692 (WebCore::AttachmentInfo::buildWrappedLines): Deleted.
1693 (WebCore::AttachmentInfo::buildSingleLine): Deleted.
1694 (WebCore::AttachmentInfo::AttachmentInfo): Deleted.
1696 Rename AttachmentInfo to RenderAttachmentInfo to resolve a name conflict with the new AttachmentInfo struct.
1698 2018-01-08 Saam Barati <sbarati@apple.com>
1700 Speculative build fix after r226600. We only use clflush on x86 and the `asm volatile` syntax is not available in the Windows build.
1702 No new tests because this is a build fix.
1705 (WebCore::Comment::clflushReadLength):
1707 2018-01-08 Michael Saboff <msaboff@apple.com>
1709 Add a DOM gadget for Spectre testing
1710 https://bugs.webkit.org/show_bug.cgi?id=181351
1712 Reviewed by Saam Barati.
1714 This change is used to test Spectre mitigations.
1716 Added a side data array to the Comment DOM node to test for Spectre issues in
1717 the DOM layer. This additional functionality is disabled by default and must
1718 be enabled through the JSC option "enableSpectreGadgets".
1721 (WebCore::Comment::Comment):
1722 (WebCore::Comment::setReadLength):
1723 (WebCore::Comment::charCodeAt):
1724 (WebCore::Comment::clflushReadLength):
1727 * page/RuntimeEnabledFeatures.cpp:
1728 (WebCore::RuntimeEnabledFeatures::spectreGadgetsEnabled const):
1729 * page/RuntimeEnabledFeatures.h:
1731 2018-01-08 Said Abou-Hallawa <sabouhallawa@apple.com>
1733 A canvas should not be tainted if it draws a data URL SVGImage with a <foreignObject>
1734 https://bugs.webkit.org/show_bug.cgi?id=180301
1736 Reviewed by Dean Jackson.
1738 Don't taint the canvas if it draws a data URL SVGImage with a <foreignObject>.
1739 There should not be a cross-origin data leak in this case.
1741 Tests: svg/as-image/svg-canvas-data-url-svg-with-feimage-not-tainted.html
1742 svg/as-image/svg-canvas-data-url-svg-with-foreign-object-not-tainted.html
1743 svg/as-image/svg-canvas-data-url-svg-with-image-not-tainted.html
1745 * html/ImageBitmap.cpp:
1746 (WebCore::taintsOrigin):
1747 * html/canvas/CanvasRenderingContext.cpp:
1748 (WebCore::CanvasRenderingContext::wouldTaintOrigin):
1750 2018-01-08 Don Olmstead <don.olmstead@sony.com>
1752 Simplify platform checks in Graphics Context
1753 https://bugs.webkit.org/show_bug.cgi?id=181344
1755 Reviewed by Alex Christensen.
1757 No new tests. No change in behavior.
1759 * platform/graphics/ANGLEWebKitBridge.h:
1760 * platform/graphics/GLContext.h:
1761 * platform/graphics/GraphicsContext3D.h:
1762 * platform/graphics/OpenGLESShims.h:
1763 * platform/graphics/opengl/GraphicsContext3DOpenGLCommon.cpp:
1764 (WebCore::GraphicsContext3D::checkVaryingsPacking const):
1766 2018-01-08 Zalan Bujtas <zalan@apple.com>
1768 [RenderTreeBuilder] Move SVG addChild logic to RenderTreeBuilder
1769 https://bugs.webkit.org/show_bug.cgi?id=181405
1770 <rdar://problem/36360476>
1772 Reviewed by Antti Koivisto.
1774 This is about moving the code, no cleanup and/or normalization (unfortunately it also means
1775 some temporary changes).
1777 No change in functionality.
1780 * WebCore.xcodeproj/project.pbxproj:
1781 * rendering/svg/RenderSVGContainer.cpp:
1782 (WebCore::RenderSVGContainer::addChild):
1783 * rendering/svg/RenderSVGInline.cpp:
1784 (WebCore::RenderSVGInline::addChild):
1785 * rendering/svg/RenderSVGRoot.cpp:
1786 (WebCore::RenderSVGRoot::addChild):
1787 * rendering/svg/RenderSVGText.cpp:
1788 (WebCore::RenderSVGText::addChild):
1789 * rendering/updating/RenderTreeBuilder.cpp:
1790 (WebCore::RenderTreeBuilder::RenderTreeBuilder):
1791 (WebCore::RenderTreeBuilder::insertChildToSVGContainer):
1792 (WebCore::RenderTreeBuilder::insertChildToSVGInline):
1793 (WebCore::RenderTreeBuilder::insertChildToSVGRoot):
1794 (WebCore::RenderTreeBuilder::insertChildToSVGText):
1795 * rendering/updating/RenderTreeBuilder.h:
1796 (WebCore::RenderTreeBuilder::SVGBuilder):
1797 * rendering/updating/RenderTreeBuilderSVG.cpp: Added.
1798 (WebCore::RenderTreeBuilder::SVG::SVG):
1799 (WebCore::RenderTreeBuilder::SVG::insertChild):
1800 * rendering/updating/RenderTreeBuilderSVG.h: Added.
1802 2018-01-08 John Wilander <wilander@apple.com>
1804 Storage Access API: Remove access for all frames under a page when the page is closed
1805 https://bugs.webkit.org/show_bug.cgi?id=181398
1806 <rdar://problem/36357879>
1808 Reviewed by Alex Christensen.
1810 No new tests. Discussed with Alex Christensen and we concluded that
1811 both a layout test and an API test would require a lot of work and
1812 we have existing tests for clearing out storage access for frames.
1814 * platform/network/NetworkStorageSession.h:
1815 * platform/network/cf/NetworkStorageSessionCFNet.cpp:
1816 (WebCore::NetworkStorageSession::removeStorageAccessForAllFramesOnPage):
1818 2018-01-08 Matt Lewis <jlewis3@apple.com>
1820 Unreviewed, rolling out r226385.
1822 The test introduced with this was a flaky since being added.
1826 "AX: when invert colors is on, double-invert certain media
1827 elements in UserAgentStyleSheet"
1828 https://bugs.webkit.org/show_bug.cgi?id=168447
1829 https://trac.webkit.org/changeset/226385
1831 2018-01-08 Youenn Fablet <youenn@apple.com>
1833 SWClientConnection should not keep references to service worker jobs
1834 https://bugs.webkit.org/show_bug.cgi?id=181381
1836 Reviewed by Chris Dumez.
1838 Difficult to test determiniscally but corresponding crash log should no longer happen in debug builds.
1840 Stopped passing ServiceWorkerJob references from ServiceWorkerContainer (potentially in service worker thread) to SWClientConnection (main thread).
1841 Instead pass job identifiers and related data to the main thread.
1843 Minor refactoring to use ServiceWorkerJobIdentifier instead of ServiceWorkerJobDataIdentifier which contains more data than needed.
1845 * workers/service/SWClientConnection.cpp:
1846 (WebCore::SWClientConnection::scheduleJob):
1847 (WebCore::SWClientConnection::failedFetchingScript):
1848 (WebCore::SWClientConnection::postTaskForJob):
1849 (WebCore::SWClientConnection::jobRejectedInServer):
1850 (WebCore::SWClientConnection::registrationJobResolvedInServer):
1851 (WebCore::SWClientConnection::unregistrationJobResolvedInServer):
1852 (WebCore::SWClientConnection::startScriptFetchForServer):
1853 (WebCore::SWClientConnection::clearPendingJobs):
1854 (WebCore::SWClientConnection::finishedFetchingScript): Deleted.
1855 * workers/service/SWClientConnection.h:
1856 * workers/service/ServiceWorkerContainer.cpp:
1857 (WebCore::ServiceWorkerContainer::scheduleJob):
1858 (WebCore::ServiceWorkerContainer::startScriptFetchForJob):
1859 (WebCore::ServiceWorkerContainer::jobFinishedLoadingScript):
1860 (WebCore::ServiceWorkerContainer::jobFailedLoadingScript):
1861 * workers/service/ServiceWorkerContainer.h:
1862 * workers/service/server/SWServer.cpp:
1863 (WebCore::SWServer::rejectJob):
1864 (WebCore::SWServer::resolveRegistrationJob):
1865 (WebCore::SWServer::resolveUnregistrationJob):
1866 (WebCore::SWServer::startScriptFetch):
1867 * workers/service/server/SWServer.h:
1869 2018-01-08 Wenson Hsieh <wenson_hsieh@apple.com>
1871 Copying, pasting, and then deleting an attachment element breaks attachment data requests
1872 https://bugs.webkit.org/show_bug.cgi?id=181365
1873 <rdar://problem/36340647>
1875 Reviewed by Tim Horton.
1877 Currently, copying and pasting an attachment element within the same document and then deleting backwards to
1878 remove the pasted attachment element causes the original attachment element to be inaccessible via SPI. This is
1879 because there are now two different attachment elements with the same unique identifier, such that Document,
1880 which keeps a map of all unique attachment identifiers to attachment elements, will lose track of the original
1883 To fix this, we ensure that attachment elements should always have unique identifiers when they are inserted
1884 into the document. We make several small adjustments to accomplish this:
1886 1. First, refactor HTMLAttachmentElement's unique identifier so that it no longer depends on the value of the
1887 "webkitattachmentid" attribute, and is instead just a member of HTMLAttachmentElement that is not exposed to
1888 DOM bindings. This means setting and querying an attachment element's uniqueIdentifier can be done without
1889 triggering any side effects, such as layout or mutation events.
1891 2. Next, make "webkitattachmentid" a temporary attribute similar to "webkitattachmentpath" and
1892 "webkitattachmentbloburl", so that it is added only when generating a markup fragment for editing, and
1893 removed upon deserialization.
1895 3. Lastly, shift the responsibility of assigning a unique identifier to an attachment away from places where we
1896 create attachment elements, and instead have Document enforce this when an attachment element is inserted.
1898 Tests: WKAttachmentTests.InsertAndRemoveDuplicateAttachment
1899 WKAttachmentTests.InsertDuplicateAttachmentAndUpdateData
1902 (WebCore::Document::didInsertAttachmentElement):
1904 Assign the unique identifier of an attachment element that has been inserted. If the identifier already tracks
1905 an existing attachment element in the document or is missing, reassign the identifier to a new value.
1907 * editing/cocoa/WebContentReaderCocoa.mm:
1908 (WebCore::createFragmentForImageAttachment):
1909 (WebCore::replaceRichContentWithAttachments):
1910 (WebCore::WebContentReader::readFilePaths):
1912 Remove calls to setUniqueIdentifier here, since Document will assign a unique identifier upon insertion.
1914 * editing/markup.cpp:
1915 (WebCore::StyledMarkupAccumulator::appendCustomAttributes):
1916 (WebCore::createFragmentFromMarkup):
1918 Set the attachment's unique identifier to the value of the "webkitattachmentid" attribute. When moving existing
1919 attachments around in the DOM without duplication, this ensures that the attachment will be removed and
1920 reinserted in the document without triggering removal and insertion client delegate methods.
1922 When pasting an attachment element that has the same identifier as an existing attachment, we let Document
1923 realize that the attachment identifier already exists, and reassign it to a unique value.
1925 * html/HTMLAttachmentElement.cpp:
1926 (WebCore::HTMLAttachmentElement::uniqueIdentifier const): Deleted.
1927 (WebCore::HTMLAttachmentElement::setUniqueIdentifier): Deleted.
1928 * html/HTMLAttachmentElement.h:
1930 2018-01-08 Zalan Bujtas <zalan@apple.com>
1932 [RenderTreeBuilder] Move RenderBlockFlow addChild logic to RenderTreeBuilder
1933 https://bugs.webkit.org/show_bug.cgi?id=181348
1934 <rdar://problem/36328117>
1936 Reviewed by Antti Koivisto.
1938 This is about moving the code, no cleanup and/or normalization (unfortunately it also means
1939 some temporary changes).
1941 No change in functionality.
1944 * WebCore.xcodeproj/project.pbxproj:
1945 * rendering/RenderBlockFlow.cpp:
1946 (WebCore::RenderBlockFlow::addChild):
1947 * rendering/updating/RenderTreeBuilder.cpp:
1948 (WebCore::RenderTreeBuilder::RenderTreeBuilder):
1949 (WebCore::RenderTreeBuilder::insertChildToRenderBlockFlow):
1950 * rendering/updating/RenderTreeBuilder.h:
1951 (WebCore::RenderTreeBuilder::blockFlowBuilder):
1952 * rendering/updating/RenderTreeBuilderBlockFlow.cpp: Added.
1953 (WebCore::RenderTreeBuilder::BlockFlow::BlockFlow):
1954 (WebCore::RenderTreeBuilder::BlockFlow::insertChild):
1955 * rendering/updating/RenderTreeBuilderBlockFlow.h: Added.
1957 2018-01-08 Youenn Fablet <youenn@apple.com>
1959 Add CSP support to service workers
1960 https://bugs.webkit.org/show_bug.cgi?id=181385
1962 Reviewed by Chris Dumez.
1964 Covered by rebased tests.
1966 Added recovery of CSP information from WorkerScriptLoader.
1967 Added plumbing to pass the CSP information to Service Workers.
1968 Added persistency support for the CSP information.
1970 * workers/WorkerScriptLoader.cpp:
1971 (WebCore::WorkerScriptLoader::loadAsynchronously):
1972 (WebCore::WorkerScriptLoader::didReceiveResponse):
1973 * workers/WorkerScriptLoader.h:
1974 (WebCore::WorkerScriptLoader::contentSecurityPolicy const):
1975 * workers/service/SWClientConnection.cpp:
1976 (WebCore::SWClientConnection::finishedFetchingScript):
1977 (WebCore::SWClientConnection::failedFetchingScript):
1978 * workers/service/SWClientConnection.h:
1979 * workers/service/ServiceWorkerContainer.cpp:
1980 (WebCore::ServiceWorkerContainer::addRegistration):
1981 (WebCore::ServiceWorkerContainer::jobFailedWithException):
1982 (WebCore::ServiceWorkerContainer::jobResolvedWithRegistration):
1983 (WebCore::ServiceWorkerContainer::jobFinishedLoadingScript):
1984 * workers/service/ServiceWorkerContainer.h:
1985 * workers/service/ServiceWorkerContextData.cpp:
1986 (WebCore::ServiceWorkerContextData::isolatedCopy const):
1987 * workers/service/ServiceWorkerContextData.h:
1988 (WebCore::ServiceWorkerContextData::encode const):
1989 (WebCore::ServiceWorkerContextData::decode):
1990 * workers/service/ServiceWorkerFetchResult.h:
1991 (WebCore::ServiceWorkerFetchResult::encode const):
1992 (WebCore::ServiceWorkerFetchResult::decode):
1993 * workers/service/ServiceWorkerGlobalScope.cpp:
1994 (WebCore::ServiceWorkerGlobalScope::create):
1995 * workers/service/ServiceWorkerGlobalScope.h:
1996 * workers/service/ServiceWorkerJob.cpp:
1997 (WebCore::ServiceWorkerJob::notifyFinished):
1998 * workers/service/ServiceWorkerJobClient.h:
1999 * workers/service/context/ServiceWorkerThread.cpp:
2000 (WebCore::ServiceWorkerThread::ServiceWorkerThread):
2001 (WebCore::ServiceWorkerThread::createWorkerGlobalScope):
2002 * workers/service/server/RegistrationDatabase.cpp:
2003 (WebCore::v1RecordsTableSchema):
2004 (WebCore::RegistrationDatabase::openSQLiteDatabase):
2005 (WebCore::RegistrationDatabase::doPushChanges):
2006 (WebCore::RegistrationDatabase::importRecords):
2007 * workers/service/server/SWServer.cpp:
2008 (WebCore::SWServer::updateWorker):
2009 (WebCore::SWServer::installContextData):
2010 * workers/service/server/SWServer.h:
2011 * workers/service/server/SWServerJobQueue.cpp:
2012 (WebCore::SWServerJobQueue::scriptFetchFinished):
2013 * workers/service/server/SWServerWorker.cpp:
2014 (WebCore::SWServerWorker::SWServerWorker):
2015 (WebCore::m_contentSecurityPolicy):
2016 (WebCore::SWServerWorker::contextData const):
2017 * workers/service/server/SWServerWorker.h:
2019 2018-01-08 Youenn Fablet <youenn@apple.com>
2021 Use no-cache fetch mode when loading main documents with location.reload()
2022 https://bugs.webkit.org/show_bug.cgi?id=181285
2024 Reviewed by Alex Christensen.
2026 Covered by rebased tests.
2028 Start to translate cache policy used for navigation as FetchOptions::Cache.
2029 This allows ensuring service workers receive the right cache mode when intercepting navigation loads.
2030 To not change current navigation behavior, ReturnCacheDataElseLoad and ReturnCacheDataDontLoad still trigger default fetch cache mode.
2032 For Reload and ReloadExpiredOnly frame load types, using no-cache mode is more efficient than reload mode,
2033 as a conditional request will be sent if possible. This applies to location.reload which is consistent with other browsers.
2034 Keep reload mode for ReloadFromOrigin.
2036 * loader/DocumentLoader.cpp:
2037 (WebCore::toFetchOptionsCache):
2038 (WebCore::DocumentLoader::loadMainResource):
2039 * loader/FrameLoader.cpp:
2040 (WebCore::FrameLoader::loadFrameRequest):
2041 (WebCore::FrameLoader::loadURL):
2042 (WebCore::FrameLoader::load):
2043 (WebCore::FrameLoader::reload):
2044 (WebCore::FrameLoader::defaultRequestCachingPolicy):
2045 (WebCore::FrameLoader::loadDifferentDocumentItem):
2046 * loader/NavigationScheduler.cpp:
2048 2018-01-08 Youenn Fablet <youenn@apple.com>
2050 Stop exposing fetch and extendable events to window
2051 https://bugs.webkit.org/show_bug.cgi?id=181325
2053 Reviewed by Chris Dumez.
2055 Covered by updated tests.
2057 Marked FetchEvent and ExtendableEvent as visible in ServiceWorker environments only.
2058 Moved related Internals testing routines to ServiceWorkerInternals.
2060 * testing/Internals.cpp:
2061 (WebCore::Internals::waitForFetchEventToFinish): Deleted.
2062 (WebCore::Internals::createBeingDispatchedFetchEvent): Deleted.
2063 * testing/Internals.h:
2064 * testing/Internals.idl:
2065 * testing/ServiceWorkerInternals.cpp:
2066 (WebCore::ServiceWorkerInternals::waitForFetchEventToFinish):
2067 (WebCore::ServiceWorkerInternals::createBeingDispatchedFetchEvent):
2068 * testing/ServiceWorkerInternals.h:
2069 * testing/ServiceWorkerInternals.idl:
2070 * workers/service/ExtendableEvent.idl:
2071 * workers/service/FetchEvent.idl:
2073 2018-01-08 Antti Koivisto <antti@apple.com>
2075 REGRESSION (r219145): Toggling layer borders on a static document no longer works immediately
2076 https://bugs.webkit.org/show_bug.cgi?id=176260
2077 <rdar://problem/34219966>
2079 Reviewed by Simon Fraser.
2081 Optimization reveled bugs in debug indicator painting.
2083 Test: compositing/debug-borders-dynamic.html
2085 * platform/graphics/ca/GraphicsLayerCA.cpp:
2086 (WebCore::GraphicsLayerCA::commitLayerChangesBeforeSublayers):
2087 (WebCore::GraphicsLayerCA::updateDebugIndicators):
2088 (WebCore::GraphicsLayerCA::updateDebugBorder): Deleted.
2090 - Rename to indicate this is not just about debug borders.
2091 - Trigger display so repaint counters get painted. This helper is only called when the indicators change.
2093 * platform/graphics/ca/GraphicsLayerCA.h:
2094 * rendering/RenderLayerCompositor.cpp:
2095 (WebCore::RenderLayerCompositor::cacheAcceleratedCompositingFlags):
2097 Ensure we do compositing update on debug border change even when there is no layout.
2099 2018-01-08 Zalan Bujtas <zalan@apple.com>
2101 [RenderTreeBuilder] Move RenderInline addChild logic to RenderTreeBuilder
2102 https://bugs.webkit.org/show_bug.cgi?id=181336
2103 <rdar://problem/36324693>
2105 Reviewed by Antti Koivisto.
2107 This is about moving the code, no cleanup and/or normalization (unfortunately it also means
2108 some temporary changes).
2110 No change in functionality.
2113 * WebCore.xcodeproj/project.pbxproj:
2114 * rendering/RenderInline.cpp:
2115 (WebCore::RenderInline::addChild):
2116 (WebCore::RenderInline::addChildIgnoringContinuation):
2117 (WebCore::RenderInline::childBecameNonInline):
2118 (WebCore::nextContinuation): Deleted.
2119 (WebCore::RenderInline::continuationBefore): Deleted.
2120 (WebCore::newChildIsInline): Deleted.
2121 (WebCore::RenderInline::cloneAsContinuation const): Deleted.
2122 (WebCore::RenderInline::splitInlines): Deleted.
2123 (WebCore::RenderInline::splitFlow): Deleted.
2124 (WebCore::canUseAsParentForContinuation): Deleted.
2125 (WebCore::RenderInline::addChildToContinuation): Deleted.
2126 * rendering/RenderInline.h:
2127 * rendering/updating/RenderTreeBuilder.cpp:
2128 (WebCore::RenderTreeBuilder::RenderTreeBuilder):
2129 (WebCore::RenderTreeBuilder::insertChildToRenderInline):
2130 (WebCore::RenderTreeBuilder::insertChildToRenderInlineIgnoringContinuation):
2131 (WebCore::RenderTreeBuilder::splitFlow):
2132 * rendering/updating/RenderTreeBuilder.h:
2133 (WebCore::RenderTreeBuilder::inlineBuilder):
2134 * rendering/updating/RenderTreeBuilderInline.cpp: Added.
2135 (WebCore::canUseAsParentForContinuation):
2136 (WebCore::nextContinuation):
2137 (WebCore::continuationBefore):
2138 (WebCore::cloneAsContinuation):
2139 (WebCore::newChildIsInline):
2140 (WebCore::inFlowPositionedInlineAncestor):
2141 (WebCore::RenderTreeBuilder::Inline::Inline):
2142 (WebCore::RenderTreeBuilder::Inline::insertChild):
2143 (WebCore::RenderTreeBuilder::Inline::insertChildToContinuation):
2144 (WebCore::RenderTreeBuilder::Inline::insertChildIgnoringContinuation):
2145 (WebCore::RenderTreeBuilder::Inline::splitFlow):
2146 (WebCore::RenderTreeBuilder::Inline::splitInlines):
2147 * rendering/updating/RenderTreeBuilderInline.h: Added.
2149 2018-01-08 Zalan Bujtas <zalan@apple.com>
2151 [RenderTreeBuilder] Move RenderBlock addChild logic to RenderTreeBuilder
2152 https://bugs.webkit.org/show_bug.cgi?id=181319
2153 <rdar://problem/36313464>
2155 Reviewed by Antti Koivisto.
2157 This is about moving the code, no cleanup and/or normalization (unfortunately it also means
2158 some temporary changes).
2160 No change in functionality.
2163 * WebCore.xcodeproj/project.pbxproj:
2164 * rendering/RenderBlock.cpp:
2165 (WebCore::RenderBlock::addChild):
2166 (WebCore::RenderBlock::addChildIgnoringContinuation):
2167 (WebCore::RenderBlock::childBecameNonInline):
2168 (WebCore::RenderBlock::continuationBefore): Deleted.
2169 (WebCore::RenderBlock::addChildToContinuation): Deleted.
2170 (WebCore::getInlineRun): Deleted.
2171 (WebCore::RenderBlock::makeChildrenNonInline): Deleted.
2172 * rendering/RenderBlock.h:
2173 * rendering/RenderBox.cpp:
2174 (WebCore::markBoxForRelayoutAfterSplit): Deleted.
2175 (WebCore::RenderBox::splitAnonymousBoxesAroundChild): Deleted.
2176 * rendering/RenderBox.h:
2177 * rendering/RenderRubyBase.cpp:
2178 (WebCore::RenderRubyBase::moveChildren):
2179 (WebCore::RenderRubyBase::moveBlockChildren):
2180 * rendering/RenderTable.cpp:
2181 (WebCore::RenderTable::addChild):
2182 * rendering/RenderTableRow.cpp:
2183 (WebCore::RenderTableRow::addChild):
2184 * rendering/RenderTableSection.cpp:
2185 (WebCore::RenderTableSection::addChild):
2186 * rendering/updating/RenderTreeBuilder.cpp:
2187 (WebCore::markBoxForRelayoutAfterSplit):
2188 (WebCore::getInlineRun):
2189 (WebCore::RenderTreeBuilder::RenderTreeBuilder):
2190 (WebCore::RenderTreeBuilder::insertChildToRenderBlock):
2191 (WebCore::RenderTreeBuilder::insertChildToRenderBlockIgnoringContinuation):
2192 (WebCore::RenderTreeBuilder::makeChildrenNonInline):
2193 (WebCore::RenderTreeBuilder::splitAnonymousBoxesAroundChild):
2194 * rendering/updating/RenderTreeBuilder.h:
2195 (WebCore::RenderTreeBuilder::blockBuilder):
2196 * rendering/updating/RenderTreeBuilderBlock.cpp: Added.
2197 (WebCore::continuationBefore):
2198 (WebCore::RenderTreeBuilder::Block::Block):
2199 (WebCore::RenderTreeBuilder::Block::insertChild):
2200 (WebCore::RenderTreeBuilder::Block::insertChildToContinuation):
2201 (WebCore::RenderTreeBuilder::Block::insertChildIgnoringContinuation):
2202 * rendering/updating/RenderTreeBuilderBlock.h: Added.
2204 2018-01-08 Jeremy Jones <jeremyj@apple.com>
2206 Standard controls sometimes say video is in pip when it isnt.
2207 https://bugs.webkit.org/show_bug.cgi?id=181095
2208 rdar://problem/36182687
2210 Reviewed by Eric Carlson.
2212 * html/HTMLMediaElement.cpp:
2213 (WebCore::HTMLMediaElement::exitFullscreen):
2215 2018-01-08 Jer Noble <jer.noble@apple.com>
2217 REGRESSION: [iOS] ASSERTION FAILED: !node.isConnected() in WebCore::notifyNodeInsertedIntoDocument
2218 https://bugs.webkit.org/show_bug.cgi?id=181091
2220 Reviewed by Darin Adler.
2222 Move the work previously performed in insertedIntoAncestor() into didFinishInsertingNode().
2224 * html/HTMLMediaElement.cpp:
2225 (WebCore::HTMLMediaElement::prepareForLoad):
2227 2018-01-08 Zan Dobersek <zdobersek@igalia.com>
2229 [Cairo] Use isolated fill and stroke source containers
2230 https://bugs.webkit.org/show_bug.cgi?id=181386
2232 Reviewed by Carlos Garcia Campos.
2234 Generate fill and stroke source objects upon invocation of various Cairo
2235 operations, initializing necessary data from the GraphicsContextState
2238 Cairo::FillSource and Cairo::StrokeSource structs are introduced, both
2239 being default-constructible as well as providing a constructor that
2240 accepts a GraphicsContextState object from which the appropriate
2241 resources are created.
2243 The FillSource and StrokeSource objects are then passed to
2244 PlatformContextCairo's prepareForFilling() and prepareForStroking()
2245 methods. Here the helper prepareCairoContextSource() function is now
2246 invoked with cairo_pattern_t objects as pattern or gradient sources, or
2247 the source color if neither cairo_pattern_t object is specified.
2249 The FillSource and StrokeSource constructors mimic the previous behavior
2250 of prepareCairoContextSource(). In case the source is a Pattern object,
2251 a cairo_pattern_t object is created from that. In case of FillSource,
2252 we also retrieve pattern size, transform and repetition information. In
2253 case the source os a Gradient object, we create a 'base' cairo_pattern_t
2254 object for a completely opaque alpha channel. Additionally, if the alpha
2255 value on the state is not 1, we create an alpha-adjusted cairo_pattern_t
2256 that is potentially used for any filling or stroking operation that has
2257 to preserve transparency. If neither Pattern or Gradient objects are set
2258 on the GraphicsContextState, we default to the current fill or stroke
2261 Overall, there's no change in behavior, this is simply a refactoring
2262 that enables us to construct Cairo objects for filling and stroking
2263 sources at the time of Cairo operation dispatch, instead of pulling down
2264 GraphicsContextState deeper into the Cairo-specific code.
2266 No new tests -- no change in functionality.
2268 * platform/graphics/cairo/CairoOperations.cpp:
2269 (WebCore::Cairo::drawPathShadow):
2270 (WebCore::Cairo::fillCurrentCairoPath):
2271 (WebCore::Cairo::FillSource::FillSource):
2272 (WebCore::Cairo::StrokeSource::StrokeSource):
2273 (WebCore::Cairo::fillRect):
2274 (WebCore::Cairo::fillRectWithRoundedHole):
2275 (WebCore::Cairo::fillPath):
2276 (WebCore::Cairo::strokeRect):
2277 (WebCore::Cairo::strokePath):
2278 (WebCore::Cairo::drawGlyphs):
2279 * platform/graphics/cairo/CairoOperations.h:
2280 * platform/graphics/cairo/FontCairo.cpp:
2281 (WebCore::FontCascade::drawGlyphs):
2282 * platform/graphics/cairo/GraphicsContextCairo.cpp:
2283 (WebCore::GraphicsContext::fillPath):
2284 (WebCore::GraphicsContext::strokePath):
2285 (WebCore::GraphicsContext::fillRect):
2286 (WebCore::GraphicsContext::strokeRect):
2287 (WebCore::GraphicsContext::fillRectWithRoundedHole):
2288 * platform/graphics/cairo/PlatformContextCairo.cpp:
2289 (WebCore::prepareCairoContextSource):
2290 (WebCore::PlatformContextCairo::prepareForFilling):
2291 (WebCore::PlatformContextCairo::prepareForStroking):
2292 (WebCore::PlatformContextCairo::clipForPatternFilling):
2293 * platform/graphics/cairo/PlatformContextCairo.h:
2295 2018-01-08 Youenn Fablet <youenn@apple.com>
2297 navigator.onLine does not work inside service workers
2298 https://bugs.webkit.org/show_bug.cgi?id=181079
2299 <rdar://problem/36178606>
2301 Reviewed by Darin Adler.
2303 Test: http/wpt/service-workers/online.https.html
2305 Added support for onLine by reusing a similar implementation as regular workers.
2306 Added ServiceWorkerInternals as an interface for an object exposed as self.internals in WTR.
2307 This object has currently one method to trigger change in the online/offline status.
2308 This allows writing a test for the onLine feature.
2310 Note that self.internals is inserted asynchronously after the script was evaluated.
2311 When writing a worker script using self.internals, one must make sure to use self.internals when initialized.
2312 online-worker.js for instance makes use of self.internals in a postMessage callback.
2315 * DerivedSources.make:
2316 * WebCore.xcodeproj/project.pbxproj:
2317 * bindings/js/WorkerScriptController.h:
2318 * dom/ScriptExecutionContext.h:
2319 * testing/ServiceWorkerInternals.cpp: Added.
2320 (WebCore::ServiceWorkerInternals::ServiceWorkerInternals):
2321 (WebCore::ServiceWorkerInternals::setOnline):
2322 * testing/ServiceWorkerInternals.h: Added.
2323 * testing/ServiceWorkerInternals.idl: Added.
2324 * testing/js/WebCoreTestSupport.cpp:
2325 (WebCoreTestSupport::setupNewlyCreateServiceWorker):
2326 * testing/js/WebCoreTestSupport.h:
2327 * workers/service/context/SWContextManager.cpp:
2328 (WebCore::SWContextManager::registerServiceWorkerThreadForInstall):
2329 (WebCore::SWContextManager::startedServiceWorker):
2330 * workers/service/context/SWContextManager.h:
2331 (WebCore::SWContextManager::setServiceWorkerCreationCallback):
2332 (WebCore::SWContextManager::workerByID):
2333 * workers/service/context/ServiceWorkerThread.cpp:
2334 (WebCore::ServiceWorkerThread::ServiceWorkerThread):
2335 * workers/service/context/ServiceWorkerThreadProxy.cpp:
2336 (WebCore::ServiceWorkerThreadProxy::ServiceWorkerThreadProxy):
2337 (WebCore::ServiceWorkerThreadProxy::~ServiceWorkerThreadProxy):
2338 (WebCore::ServiceWorkerThreadProxy::networkStateChanged):
2339 (WebCore::ServiceWorkerThreadProxy::notifyNetworkStateChange):
2340 * workers/service/context/ServiceWorkerThreadProxy.h:
2342 2018-01-08 Zan Dobersek <zdobersek@igalia.com>
2344 [Cairo] Contain shadow blur requirement state in a separate object
2345 https://bugs.webkit.org/show_bug.cgi?id=181380
2347 Reviewed by Carlos Garcia Campos.
2349 Instead of picking up the necessary state parameters from a
2350 GraphicsContextState object, store the necessary parameters in a helper
2351 ShadowBlurUsage struct. Mimicking the mustUseShadowBlur() function that
2352 is being removed, values of the shadow color, shadow blur, and shadow
2353 transform ignorance are stored there. Additionally, the required()
2354 method accepts a PlatformContextCairo object through which it can
2355 retrieve the current CTM and determine whether it's an identity, finally
2356 deciding whether shadow blur can or cannot be ignored.
2358 Goal of this change is to limit usage of GraphicsContextState directly
2359 in operations implemented inside the Cairo namespace. Instead, the
2360 state parameters should be passed directly, or an equivalent but limited
2361 state object should be constructed for invocation of such operations.
2363 This is likely only an intermediate solution. It's possible it will be
2364 replaced by a more complete shadow state struct that would be used for
2365 any operation that is required to draw shadows, if necessary.
2367 No new tests -- no change in functionality.
2369 * platform/graphics/cairo/CairoOperations.cpp:
2370 (WebCore::Cairo::drawGlyphsShadow):
2371 (WebCore::Cairo::ShadowBlurUsage::ShadowBlurUsage):
2372 (WebCore::Cairo::ShadowBlurUsage::required const):
2373 (WebCore::Cairo::fillRectWithRoundedHole):
2374 (WebCore::Cairo::drawGlyphs):
2375 (WebCore::Cairo::mustUseShadowBlur): Deleted.
2376 * platform/graphics/cairo/CairoOperations.h:
2377 * platform/graphics/cairo/FontCairo.cpp:
2378 (WebCore::FontCascade::drawGlyphs):
2379 * platform/graphics/cairo/GraphicsContextCairo.cpp:
2380 (WebCore::GraphicsContext::fillRectWithRoundedHole):
2382 2018-01-08 Philippe Normand <pnormand@igalia.com>
2384 [GStreamer][Soup] Remove dead getCreateOrReadBuffer code path
2385 https://bugs.webkit.org/show_bug.cgi?id=181376
2387 Reviewed by Carlos Garcia Campos.
2389 This code path is no longer used now that the GStreamer HTTP
2390 source element uses data coming directly from the network process.
2394 * loader/MediaResourceLoader.cpp:
2395 (WebCore::MediaResource::getOrCreateReadBuffer): Deleted.
2396 * loader/MediaResourceLoader.h:
2397 * loader/SubresourceLoader.h:
2398 * loader/cache/CachedRawResource.h:
2399 * loader/cache/CachedRawResourceClient.h:
2400 (WebCore::CachedRawResourceClient::finishedTimingForWorkerLoad):
2401 (WebCore::CachedRawResourceClient::getOrCreateReadBuffer): Deleted.
2402 * loader/cache/CachedResource.h:
2403 (WebCore::CachedResource::getOrCreateReadBuffer): Deleted.
2404 * loader/soup/CachedRawResourceSoup.cpp: Removed.
2405 * loader/soup/SubresourceLoaderSoup.cpp: Removed.
2406 * platform/graphics/PlatformMediaResourceLoader.h:
2407 (WebCore::PlatformMediaResourceClient::loadFinished):
2408 (WebCore::PlatformMediaResourceClient::getOrCreateReadBuffer): Deleted.
2409 * platform/graphics/gstreamer/WebKitWebSourceGStreamer.cpp:
2410 (CachedResourceStreamingClient::getOrCreateReadBuffer): Deleted.
2411 * platform/network/ResourceHandleClient.h:
2412 (WebCore::ResourceHandleClient::getOrCreateReadBuffer): Deleted.
2413 * platform/network/soup/ResourceHandleSoup.cpp:
2414 (WebCore::ResourceHandle::ensureReadBuffer):
2416 2018-01-08 Zan Dobersek <zdobersek@igalia.com> and Carlos Garcia Campos <cgarcia@igalia.com>
2418 [GStreamer] use FastMalloc-based GstAllocator
2419 https://bugs.webkit.org/show_bug.cgi?id=165793
2421 Reviewed by Philippe Normand.
2423 Add GstAllocatorFastMalloc, a GstAllocator implementation using fast malloc. It's only used when fast malloc is
2424 enabled and can be disabled using an environment variable for debugging purposes.
2426 * platform/GStreamer.cmake:
2427 * platform/graphics/gstreamer/GStreamerUtilities.cpp:
2428 (WebCore::initializeGStreamer):
2429 * platform/graphics/gstreamer/GstAllocatorFastMalloc.cpp: Added.
2430 (gstMemoryFastMallocNew):
2431 (gstAllocatorFastMallocAlloc):
2432 (gstAllocatorFastMallocFree):
2433 (gstAllocatorFastMallocMemMap):
2434 (gstAllocatorFastMallocMemUnmap):
2435 (gstAllocatorFastMallocMemCopy):
2436 (gstAllocatorFastMallocMemShare):
2437 (gstAllocatorFastMallocMemIsSpan):
2438 (gst_allocator_fast_malloc_class_init):
2439 (gst_allocator_fast_malloc_init):
2440 * platform/graphics/gstreamer/GstAllocatorFastMalloc.h: Added.
2442 2018-01-07 Ms2ger <Ms2ger@igalia.com>
2444 Implement createImageBitmap(ImageBitmap)
2445 https://bugs.webkit.org/show_bug.cgi?id=181287
2447 Reviewed by Darin Adler.
2449 Tests: imported/w3c/web-platform-tests/2dcontext/imagebitmap/createImageBitmap-drawImage.html
2450 imported/w3c/web-platform-tests/2dcontext/imagebitmap/createImageBitmap-invalid-args.html
2452 * html/ImageBitmap.cpp:
2453 (WebCore::ImageBitmap::createPromise):
2455 2018-01-07 Ryosuke Niwa <rniwa@webkit.org>
2457 Reduce the precision of "high" resolution time to 1ms
2458 https://bugs.webkit.org/show_bug.cgi?id=180910
2459 <rdar://problem/36085943>
2461 Reviewed by Saam Barati.
2463 Reduced the high prevision time's resolution to 1ms, the same precision as Date.now().
2465 Also fixed the bug in fillRTCStats that we weren't reducing the time resolution in RTCStats dictionaries.
2467 * Modules/mediastream/libwebrtc/LibWebRTCMediaEndpoint.cpp:
2468 (WebCore::fillRTCStats):
2469 * page/Performance.cpp:
2470 (WebCore::Performance::reduceTimeResolution):
2472 2018-01-06 Simon Fraser <simon.fraser@apple.com>
2474 Possible crash computing event regions
2475 https://bugs.webkit.org/show_bug.cgi?id=181368
2476 rdar://problem/34847081
2478 Reviewed by Zalan Bujtas.
2480 Don't trigger layout in Element::absoluteEventHandlerBounds(), since this can run arbirary script
2481 which might delete elements or re-enter Document::absoluteRegionForEventTargets().
2483 It's OK to not trigger layout, because if layout is dirty, the next layout will update event regions again.
2485 Add a LayoutDisallowedScope to check that Document::absoluteRegionForEventTargets() doesn't
2486 trigger layout, and move the check for LayoutDisallowedScope::isLayoutAllowed() from Document::updateLayout()
2487 to LayoutContext::layout(), since some layouts don't happen via the former (e.g. the one being removed here).
2489 The test checks that the assertion does not fire. I was not able to get a reliable test for any crash.
2491 Test: fast/events/event-handler-regions-layout.html
2494 (WebCore::Document::updateLayout):
2495 (WebCore::Document::absoluteRegionForEventTargets):
2497 (WebCore::Element::absoluteEventHandlerBounds):
2498 * page/LayoutContext.cpp:
2499 (WebCore::LayoutContext::layout):
2500 * rendering/LayoutDisallowedScope.h: Move the #ifdefs around to avoid defining the enum twice.
2501 (WebCore::LayoutDisallowedScope::LayoutDisallowedScope):
2502 (WebCore::LayoutDisallowedScope::isLayoutAllowed):
2504 2018-01-06 Simon Fraser <simon.fraser@apple.com>
2506 Crash under RenderLayer::scrollTo() with marquee
2507 https://bugs.webkit.org/show_bug.cgi?id=181349
2508 rdar://problem/36190168
2510 Reviewed by Zalan Bujtas.
2512 Don't call updateWidgetPositions() synchonously during RenderLayer scrolling, because it
2513 can run arbitrary script which may trigger destruction of this RenderLayer.
2515 Instead, queue up updateWidgetPositions() on a zero-delay timer.
2517 Under some circumstances this may allow a paint to occur before the widgets have been
2518 updated (which could be fixed with a more invasive change), but in practice I saw no
2519 painting issues with plug-ins or iframes inside overflow scroll, in WebKit or LegacyWebKit.
2521 Test: fast/scrolling/marquee-scroll-crash.html
2523 * page/FrameView.cpp:
2524 (WebCore::FrameView::FrameView):
2525 (WebCore::FrameView::updateWidgetPositions):
2526 (WebCore::FrameView::scheduleUpdateWidgetPositions):
2527 (WebCore::FrameView::updateWidgetPositionsTimerFired):
2529 * rendering/RenderLayer.cpp:
2530 (WebCore::RenderLayer::scrollTo):
2532 2018-01-05 Dean Jackson <dino@apple.com>
2534 Accurately clip copyTexImage2D and copyTexSubImage2D
2535 https://bugs.webkit.org/show_bug.cgi?id=181356
2536 <rdar://problem/35083877>
2538 Reviewed by Eric Carlson.
2540 The code to make sure copyTexSubImage2D and copyTexImage2D will not try to read
2541 out of bounds had a bad bug introduced here:
2542 https://bugs.webkit.org/show_bug.cgi?id=51421
2544 With appropriate parameters, it would produce a rectangle with
2545 negative dimensions. Most GL drivers just ignored this, but some
2548 Test: fast/canvas/webgl/copy-tex-image-and-sub-image-2d-bad-input.html
2550 * html/canvas/WebGLRenderingContextBase.cpp:
2551 (WebCore::clip2D): Reimplement this in a more sane manner, and use
2552 checked arithmetic while here.
2553 * html/canvas/WebGLRenderingContextBase.h:
2554 (WebCore::clip1D): Deleted.
2555 (WebCore::clip2D): Deleted.
2557 2018-01-06 Antti Koivisto <antti@apple.com>
2559 Use WeakPtr for RenderTreePosition::m_nextSibling
2560 https://bugs.webkit.org/show_bug.cgi?id=181363
2562 Reviewed by Zalan Bujtas.
2564 For safety. In most cases it is null and won't cause us to instantiate WeakReferences for
2567 * rendering/updating/RenderTreePosition.cpp:
2568 (WebCore::RenderTreePosition::computeNextSibling):
2569 * rendering/updating/RenderTreePosition.h:
2570 (WebCore::RenderTreePosition::RenderTreePosition):
2571 (WebCore::RenderTreePosition::nextSibling const):
2573 2018-01-05 David Kilzer <ddkilzer@apple.com>
2575 Re-enable -Wcast-qual in WebCore for Apple ports
2576 <https://webkit.org/b/177895>
2577 <rdar://problem/34960830>
2579 Reviewed by Joseph Pecoraro.
2581 * Configurations/Base.xcconfig:
2582 (WARNING_CFLAGS): Remove FIXME and add -Wcast-qual back to
2585 * crypto/mac/SerializedCryptoKeyWrapMac.mm:
2586 (WebCore::createAndStoreMasterKey):
2587 - Use checked_cf_cast<SecACLRef>().
2589 * editing/cocoa/DataDetection.mm:
2590 (WebCore::detectItemAtPositionWithRange):
2591 - Manually cast CFTypeRef to DDResultRef until
2592 DDResultGetTypeID() is available as SPI.
2594 * platform/gamepad/mac/HIDGamepad.cpp:
2595 (WebCore::HIDGamepad::initElementsFromArray):
2596 - Use checked_cf_cast<IOHIDElementRef>().
2598 * platform/graphics/avfoundation/objc/MediaSampleAVFObjC.mm:
2599 (WebCore::MediaSampleAVFObjC::createImageSample):
2600 (WebCore::CMSampleBufferIsRandomAccess):
2601 (WebCore::CMSampleBufferIsNonDisplaying):
2602 (WebCore::MediaSampleAVFObjC::createNonDisplayingCopy const):
2603 - Use checked_cf_cast<CFMutableDictionaryRef>() and
2604 checked_cf_cast<CFDictionaryRef>().
2606 * platform/graphics/cocoa/IOSurface.h:
2607 (WebCore::IOSurface::asLayerContents):
2608 - Use reinterpret_cast<id>() to cast from IOSurfaceRef to id.
2610 * platform/graphics/cocoa/WebCoreDecompressionSession.mm:
2611 (WebCore::WebCoreDecompressionSession::getFirstVideoFrame):
2612 (WebCore::WebCoreDecompressionSession::automaticDequeue):
2613 (WebCore::WebCoreDecompressionSession::imageForTime):
2614 (WebCore::WebCoreDecompressionSession::getDecodeTime):
2615 (WebCore::WebCoreDecompressionSession::getPresentationTime):
2616 (WebCore::WebCoreDecompressionSession::getDuration):
2617 - Use checked_cf_cast<CMSampleBufferRef>().
2619 * platform/graphics/Font.h:
2620 (WebCore::Font::m_kernedCFStringAttributes):
2621 (WebCore::Font::m_nonKernedCFStringAttributes):
2622 - Change type from RetainPtr<CFDictionaryRef> to
2623 RetainPtr<CFMutableDictionaryRef> since that's what they are.
2624 * platform/graphics/mac/SimpleFontDataCoreText.cpp:
2625 (WebCore::Font::getCFStringAttributes const):
2626 - Replace local `mutableAttributes` variable with
2627 `attributesDictionary.get()` since it returns the correct type
2630 * platform/ios/wak/WAKView.mm:
2631 (-[WAKView _initWithViewRef:]):
2633 * platform/ios/wak/WKView.mm:
2634 (_WKViewClearSuperview):
2636 (WKViewNextSibling):
2637 - Use static_cast<WKViewRef>(const_cast<void*>()) to convert
2638 const void* variable to WKViewRef.
2640 * platform/mac/PasteboardMac.mm:
2641 (WebCore::flipImageSpec):
2642 (WebCore::setDragImageImpl):
2643 - Use const_cast<> to remove 'const' modifier from
2644 unsigned char pointers. This regressed while -Wcast-qual was
2645 disabled for WebCore.
2647 * platform/mac/SSLKeyGeneratorMac.mm:
2648 (WebCore::signedPublicKeyAndChallengeString):
2649 - Use checked_cf_cast<SecACLRef>().
2651 * platform/mediastream/mac/RealtimeIncomingVideoSourceCocoa.cpp:
2652 (WebCore::RealtimeIncomingVideoSourceCocoa::OnFrame):
2653 - Use checked_cf_cast<CFMutableDictionaryRef>().
2655 * platform/network/cf/SocketStreamHandleImplCFNet.cpp:
2656 (WebCore::copyCONNECTProxyResponse):
2657 - Use checked_cf_cast<CFHTTPMessageRef>().
2659 * platform/network/cocoa/ResourceResponseCocoa.mm:
2660 (WebCore::ResourceResponse::platformCertificateInfo const):
2661 - Use checked_cf_cast<SecTrustRef>().
2663 * platform/network/mac/CertificateInfoMac.mm:
2664 (WebCore::CertificateInfo::containsNonRootSHA1SignedCertificate const):
2665 (WebCore::CertificateInfo::dump const):
2666 - Use checked_cf_cast<SecCertificateRef>().
2668 * testing/cocoa/WebArchiveDumpSupport.mm:
2669 (WebCoreTestSupport::createCFURLResponseFromResponseData):
2670 - Use checked_cf_cast<>() for CFMutable* types.
2672 2018-01-05 John Wilander <wilander@apple.com>
2674 Storage Access API: Refactor to make naming accurate and explicit, simplify access table, and prepare for access removal for page
2675 https://bugs.webkit.org/show_bug.cgi?id=181357
2676 <rdar://problem/36331031>
2678 Reviewed by Alex Christensen.
2680 No new tests. The only changed functionality that isn't covered
2681 by existing tests is cross-origin iframes in the same partition
2682 should be handled as already having access. This cannot be
2683 tested in layout tests since they don't support subdomains.
2685 This change does the following:
2686 - Changes function and message names to reflect how this feature
2687 was eventually implemented, i.e. access per frame.
2688 - Makes it explicit that the UI process is only involved in
2689 granting storage access and not removing storage access.
2690 The latter is done directly by the web process.
2691 - Simplifies the network process' entry map since only needs to
2692 be able to give access to one domain in one frame at a time.
2693 Access goes away on frame navigation so there can only be one
2694 domain at a time per frame. Also, the map now uses pageIDs as
2695 main keys to prepare for efficient access removal for all
2696 frames under a page.
2697 - Fixes a bug in so that a cross-origin iframe with the same
2698 partition as the top frame correctly is handled as already
2701 * platform/network/NetworkStorageSession.h:
2702 * platform/network/cf/NetworkStorageSessionCFNet.cpp:
2703 (WebCore::NetworkStorageSession::cookieStoragePartition const):
2704 The only change here is the changed named of the call to
2705 NetworkStorageSession::hasStorageAccessForFrame().
2706 (WebCore::NetworkStorageSession::hasStorageAccessForFrame const):
2707 (WebCore::NetworkStorageSession::grantStorageAccessForFrame):
2708 (WebCore::NetworkStorageSession::removeStorageAccessForFrame):
2709 (WebCore::NetworkStorageSession::isStorageAccessGranted const): Deleted.
2710 (WebCore::NetworkStorageSession::setStorageAccessGranted): Deleted.
2711 (WebCore::NetworkStorageSession::removeStorageAccess): Deleted.
2713 2018-01-05 Youenn Fablet <youenn@apple.com>
2715 Implement Cache API partitioning based on ClientOrigin
2716 https://bugs.webkit.org/show_bug.cgi?id=181240
2718 Reviewed by Alex Christensen.
2720 Covered by updated tests.
2722 Previously, cache storage was partitioned according the origin of the client, represented as a String.
2723 We now partition according both client and top origins, represented as a ClientOrigin
2725 Minor refactoring to use more makePendingActivity.
2726 Added support for IPC serialization of ClientOrigin.
2727 Added SecurityOriginData::toString which is used by WebKit2 Cache Storage implementation.
2729 * Modules/cache/CacheStorageConnection.cpp:
2730 (WebCore::CacheStorageConnection::open):
2731 (WebCore::CacheStorageConnection::retrieveCaches):
2732 * Modules/cache/CacheStorageConnection.h:
2733 (WebCore::CacheStorageConnection::clearMemoryRepresentation):
2734 (WebCore::CacheStorageConnection::doOpen):
2735 (WebCore::CacheStorageConnection::doRetrieveCaches):
2736 * Modules/cache/DOMCacheStorage.cpp:
2737 (WebCore::DOMCacheStorage::origin const):
2738 (WebCore::DOMCacheStorage::retrieveCaches):
2739 (WebCore::DOMCacheStorage::open):
2740 (WebCore::DOMCacheStorage::remove):
2741 * Modules/cache/DOMCacheStorage.h:
2742 * Modules/cache/WorkerCacheStorageConnection.cpp:
2743 (WebCore::WorkerCacheStorageConnection::doOpen):
2744 (WebCore::WorkerCacheStorageConnection::doRetrieveCaches):
2745 * Modules/cache/WorkerCacheStorageConnection.h:
2746 * page/ClientOrigin.h:
2747 (WebCore::ClientOrigin::isolatedCopy const):
2748 (WebCore::ClientOrigin::encode const):
2749 (WebCore::ClientOrigin::decode):
2750 * page/SecurityOriginData.cpp:
2751 (WebCore::SecurityOriginData::toString const):
2752 (WebCore::SecurityOriginData::debugString const): Deleted.
2753 * page/SecurityOriginData.h:
2754 (WebCore::SecurityOriginData::debugString const):
2755 * testing/Internals.cpp:
2756 (WebCore::Internals::clearCacheStorageMemoryRepresentation):
2758 2018-01-05 Wenson Hsieh <wenson_hsieh@apple.com>
2760 [Attachment Support] Add a way to write blob data to a file URL from the UI process
2761 https://bugs.webkit.org/show_bug.cgi?id=181236
2763 Reviewed by Brady Eidson.
2765 Add support for writing a blob to a designated file path. See comments below for more detail. No new tests, as
2766 there change in behavior yet. See part 2: https://bugs.webkit.org/show_bug.cgi?id=181199.
2768 * page/DragController.cpp:
2769 (WebCore::DragController::dragAttachmentElement):
2770 * platform/PromisedBlobInfo.h:
2772 Remove PromisedBlobData entirely. This was added with the premise of having the web process deliver blob data to
2773 the UI process. However, the new approach I'm taking just has the UI process tell the network process to write
2774 a blob to a given location, so a data structure to deliver blob data over IPC is no longer necessary.
2776 (WebCore::PromisedBlobData::hasData const): Deleted.
2777 (WebCore::PromisedBlobData::hasFile const): Deleted.
2778 (WebCore::PromisedBlobData::operator bool const): Deleted.
2779 (WebCore::PromisedBlobData::fulfills const): Deleted.
2780 * platform/network/BlobRegistryImpl.cpp:
2781 (WebCore::BlobRegistryImpl::populateBlobsForFileWriting):
2783 Introduce a new helper to build a list of blob data for file writing.
2785 (WebCore::writeFilePathsOrDataBuffersToFile):
2787 Introduce a new static helper to write blob data (a list of file paths and data buffers) to a given file handle.
2788 Automatically closes the given file handle upon exit.
2790 (WebCore::BlobRegistryImpl::writeBlobsToTemporaryFiles):
2791 (WebCore::BlobRegistryImpl::writeBlobToFilePath):
2793 Pull out common logic in writeBlobsToTemporaryFiles and writeBlobToFilePath into helper methods (see above), and
2794 refactor both methods to use the helpers.
2796 * platform/network/BlobRegistryImpl.h:
2798 2018-01-05 Alex Christensen <achristensen@webkit.org>
2800 Forbid < and > in URL hosts
2801 https://bugs.webkit.org/show_bug.cgi?id=181308
2802 <rdar://problem/36012757>
2804 Reviewed by Tim Horton.
2806 https://url.spec.whatwg.org/#forbidden-host-code-point does not include these characters yet, but I think it should.
2807 Firefox fails to parse URLs with < or > in the host. Chrome percent encodes them. Safari needs to do something.
2808 The web platform tests are unclear on this case, and they will need to be updated with the specification.
2809 They do show a change in behavior, though.
2811 * platform/URLParser.cpp:
2812 Add < and > to the list of forbidden host code points.
2814 2018-01-05 Eric Carlson <eric.carlson@apple.com>
2816 [MediaStream] Add Mac screen capture source
2817 https://bugs.webkit.org/show_bug.cgi?id=181333
2818 <rdar://problem/36323219>
2820 Reviewed by Dean Jackson.
2822 * SourcesCocoa.txt: Add ScreenDisplayCaptureSourceMac.mm.
2824 * WebCore.xcodeproj/project.pbxproj: Ditto.
2826 * platform/cocoa/CoreVideoSoftLink.cpp: Declare new constants used.
2827 * platform/cocoa/CoreVideoSoftLink.h:
2829 * platform/mediastream/mac/DisplayCaptureManagerCocoa.cpp:
2830 (WebCore::displayReconfigurationCallBack): Call refreshCaptureDevices.
2831 (WebCore::DisplayCaptureManagerCocoa::~DisplayCaptureManagerCocoa): Unregister for display
2832 reconfiguration callbacks.
2833 (WebCore::DisplayCaptureManagerCocoa::captureDevices): Register for display reconfigrations.
2834 (WebCore::DisplayCaptureManagerCocoa::refreshCaptureDevices): Use CGActiveDisplayList to
2835 get list of active screens.
2836 (WebCore::DisplayCaptureManagerCocoa::screenCaptureDeviceWithPersistentID): Validate screen
2837 ID, return CaptureDevice.
2838 * platform/mediastream/mac/DisplayCaptureManagerCocoa.h:
2840 * platform/mediastream/mac/RealtimeMediaSourceCenterMac.cpp:
2841 (WebCore::VideoCaptureSourceFactoryMac::createVideoCaptureSource): Deal with screen capture
2844 Implement Mac screen capture with CGDisplayStream.
2845 * platform/mediastream/mac/ScreenDisplayCaptureSourceMac.h: Added.
2846 (WebCore::ScreenDisplayCaptureSourceMac::DisplaySurface::~DisplaySurface):
2847 (WebCore::ScreenDisplayCaptureSourceMac::DisplaySurface::operator=):
2848 (WebCore::ScreenDisplayCaptureSourceMac::DisplaySurface::ioSurface const):
2849 * platform/mediastream/mac/ScreenDisplayCaptureSourceMac.mm: Added.
2850 (WebCore::roundUpToMacroblockMultiple):
2851 (WebCore::ScreenDisplayCaptureSourceMac::updateDisplayID):
2852 (WebCore::ScreenDisplayCaptureSourceMac::create):
2853 (WebCore::ScreenDisplayCaptureSourceMac::ScreenDisplayCaptureSourceMac):
2854 (WebCore::ScreenDisplayCaptureSourceMac::~ScreenDisplayCaptureSourceMac):
2855 (WebCore::ScreenDisplayCaptureSourceMac::createDisplayStream):
2856 (WebCore::ScreenDisplayCaptureSourceMac::startProducingData):
2857 (WebCore::ScreenDisplayCaptureSourceMac::stopProducingData):
2858 (WebCore::ScreenDisplayCaptureSourceMac::sampleBufferFromPixelBuffer):
2859 (WebCore::ScreenDisplayCaptureSourceMac::pixelBufferFromIOSurface):
2860 (WebCore::ScreenDisplayCaptureSourceMac::generateFrame):
2861 (WebCore::ScreenDisplayCaptureSourceMac::startDisplayStream):
2862 (WebCore::ScreenDisplayCaptureSourceMac::applySize):
2863 (WebCore::ScreenDisplayCaptureSourceMac::applyFrameRate):
2864 (WebCore::ScreenDisplayCaptureSourceMac::commitConfiguration):
2865 (WebCore::ScreenDisplayCaptureSourceMac::displayWasReconfigured):
2866 (WebCore::ScreenDisplayCaptureSourceMac::displayReconfigurationCallBack):
2867 (WebCore::ScreenDisplayCaptureSourceMac::frameAvailable):
2869 2018-01-05 Don Olmstead <don.olmstead@sony.com>
2871 [curl] Can't load file:// URL with a URL fragment identifier
2872 https://bugs.webkit.org/show_bug.cgi?id=181170
2874 Reviewed by Alex Christensen.
2876 No new tests. No change in behavior.
2878 * platform/network/curl/CurlRequest.cpp:
2879 (WebCore::CurlRequest::invokeDidReceiveResponseForFile):
2881 2018-01-05 Don Olmstead <don.olmstead@sony.com>
2883 TextCodec uses std::array but does not include it
2884 https://bugs.webkit.org/show_bug.cgi?id=181340
2886 Reviewed by Alex Christensen.
2888 No new tests. No change in behavior.
2890 * platform/text/TextCodec.h:
2892 2018-01-05 Said Abou-Hallawa <sabouhallawa@apple.com>
2894 SVGAnimatedListPropertyTearOff::synchronizeWrappersIfNeeded() should do nothing if the property is not animating
2895 https://bugs.webkit.org/show_bug.cgi?id=181316
2896 <rdar://problem/36147545>
2898 Reviewed by Simon Fraser.
2900 This is a speculative change to fix a crash which appeared after r226065.
2901 The crash is very intermittent and sometimes very hard to reproduce. The
2902 basic code analysis did not show how this crash can even happen.
2904 * svg/SVGAnimatedTypeAnimator.h:
2905 (WebCore::SVGAnimatedTypeAnimator::resetFromBaseValues): For SVG property
2906 with two values, e.g. <SVGAngleValue, SVGMarkerOrientType>, we need to
2907 detach the wrappers of the animated property if the animated values are
2908 going to change. This is similar to what we did in resetFromBaseValue().
2910 * svg/properties/SVGAnimatedListPropertyTearOff.h:
2911 (WebCore::SVGAnimatedListPropertyTearOff::synchronizeWrappersIfNeeded):
2913 2018-01-05 Matt Lewis <jlewis3@apple.com>
2915 Unreviewed, rolling out r226401.
2917 This caused timeouts on multiple platforms.
2921 "Implement Cache API partitioning based on ClientOrigin"
2922 https://bugs.webkit.org/show_bug.cgi?id=181240
2923 https://trac.webkit.org/changeset/226401
2925 2018-01-05 Dan Bernstein <mitz@apple.com>
2927 Fixed the build following AppKit API deprecations in a recent SDKs
2929 * platform/mac/PasteboardMac.mm:
2930 (WebCore::setDragImageImpl): Suppressed deprecation warnings.
2931 * platform/mac/WidgetMac.mm:
2932 (WebCore::Widget::paint): Ditto.
2934 2018-01-05 Joseph Pecoraro <pecoraro@apple.com>
2936 ServiceWorkers: Enable UserTiming / ResourceTiming
2937 https://bugs.webkit.org/show_bug.cgi?id=181297
2938 <rdar://problem/36307306>
2940 Reviewed by Youenn Fablet.
2942 Tests: http/tests/workers/service/service-worker-resource-timing.https.html
2943 http/tests/workers/service/service-worker-user-timing.https.html
2945 * loader/ResourceTiming.cpp:
2946 (WebCore::ResourceTiming::ResourceTiming):
2947 We used to clear extra NetworkLoadMetrics data early on. However,
2948 for Workers we want to pass the complete NetworkLoadMetrics to
2949 the Worker so that a Worker inspector has access to it.
2951 * page/PerformanceResourceTiming.cpp:
2952 (WebCore::PerformanceResourceTiming::PerformanceResourceTiming):
2953 Instead move the clearing of extra data to here, when the NetworkLoadMetrics
2954 have finally settled into being used only for a performance entry.
2956 2018-01-04 Philippe Normand <pnormand@igalia.com>
2958 [EME][GStreamer] Fix wrong ifdef
2959 https://bugs.webkit.org/show_bug.cgi?id=181289
2961 Reviewed by Alex Christensen.
2963 * platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp:
2964 (WebCore::MediaPlayerPrivateGStreamer::handleMessage): Remove the
2965 ENCRYPTED_MEDIA ifdef from the VIDEO_TRACK ifdef block. Both have
2966 nothing to do together.
2968 2018-01-05 Fujii Hironori <Hironori.Fujii@sony.com>
2970 [Cairo] Canvas: Path::clear should clear its transform
2971 https://bugs.webkit.org/show_bug.cgi?id=181320
2973 Reviewed by Carlos Garcia Campos.
2975 Path of Cairo port has its cairo context. Path::clear() didn't
2976 clear the transform matrix of the context.
2978 Test: fast/canvas/reset-scaling-by-height-change.html
2980 * platform/graphics/cairo/PathCairo.cpp:
2981 (WebCore::Path::clear): Reset the transform matrix of Path.
2983 2018-01-04 Devin Rousso <webkit@devinrousso.com>
2985 Web Inspector: replace HTMLCanvasElement with CanvasRenderingContext for instrumentation logic
2986 https://bugs.webkit.org/show_bug.cgi?id=180770
2988 Reviewed by Joseph Pecoraro.
2990 No change in functionality.
2992 * html/HTMLCanvasElement.h:
2993 * html/HTMLCanvasElement.cpp:
2994 (WebCore::HTMLCanvasElement::createContext2d):
2995 (WebCore::HTMLCanvasElement::createContextWebGL):
2996 (WebCore::HTMLCanvasElement::createContextWebGPU):
2997 (WebCore::HTMLCanvasElement::createContextBitmapRenderer):
2998 (WebCore::HTMLCanvasElement::reset):
2999 (WebCore::HTMLCanvasElement::paint):
3000 (WebCore::HTMLCanvasElement::setImageBuffer const):
3001 (WebCore::HTMLCanvasElement::addObserver): Deleted.
3002 (WebCore::HTMLCanvasElement::removeObserver): Deleted.
3003 (WebCore::HTMLCanvasElement::cssCanvasClients): Deleted.
3004 (WebCore::HTMLCanvasElement::notifyObserversCanvasChanged): Deleted.
3005 * html/OffscreenCanvas.h:
3006 * html/canvas/CanvasRenderingContext.h:
3007 * html/canvas/CanvasRenderingContext.cpp:
3008 * html/canvas/CanvasRenderingContext2D.h:
3009 * html/canvas/CanvasRenderingContext2D.cpp:
3010 (WebCore::CanvasRenderingContext2D::create):
3011 * html/canvas/CanvasRenderingContext2DBase.h:
3012 * html/canvas/ImageBitmapRenderingContext.h:
3013 * html/canvas/ImageBitmapRenderingContext.cpp:
3014 (WebCore::ImageBitmapRenderingContext::create):
3015 * html/canvas/WebGL2RenderingContext.h:
3016 * html/canvas/WebGL2RenderingContext.cpp:
3017 (WebCore::WebGL2RenderingContext::create):
3018 * html/canvas/WebGLRenderingContext.h:
3019 * html/canvas/WebGLRenderingContext.cpp:
3020 (WebCore::WebGLRenderingContext::create):
3021 * html/canvas/WebGLRenderingContextBase.h:
3022 * html/canvas/WebGLRenderingContextBase.cpp:
3023 (WebCore::WebGLRenderingContextBase::create):
3024 * html/canvas/WebGPURenderingContext.cpp:
3025 (WebCore::WebGPURenderingContext::create):
3026 Instead of adding didCreateCanvasRenderingContext calls at the construction sites of each
3027 context, we can make the constructors private and force the usage of static `create` functions.
3028 This way, we have access to the fully constructed object and have a guaranteed path for creation.
3030 * html/CanvasBase.h:
3031 * html/CanvasBase.cpp:
3032 (WebCore::CanvasBase::~CanvasBase):
3033 (WebCore::CanvasBase::renderingContext const):
3034 (WebCore::CanvasBase::addObserver):
3035 (WebCore::CanvasBase::removeObserver):
3036 (WebCore::CanvasBase::notifyObserversCanvasChanged):
3037 (WebCore::CanvasBase::notifyObserversCanvasResized):
3038 (WebCore::CanvasBase::notifyObserversCanvasDestroyed):
3039 (WebCore::CanvasBase::cssCanvasClients const):
3040 * Modules/mediastream/CanvasCaptureMediaStreamTrack.h:
3041 * Modules/mediastream/CanvasCaptureMediaStreamTrack.cpp:
3042 (WebCore::CanvasCaptureMediaStreamTrack::Source::canvasDestroyed):
3043 (WebCore::CanvasCaptureMediaStreamTrack::Source::canvasResized):
3044 (WebCore::CanvasCaptureMediaStreamTrack::Source::canvasChanged):
3045 * css/CSSCanvasValue.h:
3046 Move the CanvasObserver class to CanvasBase so that it can also be used for OffscreenCanvas.
3048 * inspector/InspectorInstrumentation.h:
3049 (WebCore::InspectorInstrumentation::didChangeCSSCanvasClientNodes):
3050 (WebCore::InspectorInstrumentation::didCreateCanvasRenderingContext):
3051 (WebCore::InspectorInstrumentation::didChangeCanvasMemory):
3052 (WebCore::InspectorInstrumentation::recordCanvasAction):
3053 (WebCore::InspectorInstrumentation::didFinishRecordingCanvasFrame):
3054 (WebCore::InspectorInstrumentation::didEnableExtension):
3055 (WebCore::InspectorInstrumentation::didCreateProgram):
3056 (WebCore::InspectorInstrumentation::willDeleteProgram):
3057 (WebCore::InspectorInstrumentation::isShaderProgramDisabled):
3058 (WebCore::InspectorInstrumentation::consoleStartRecordingCanvas):
3059 (WebCore::InspectorInstrumentation::didCreateCSSCanvas): Deleted.
3060 * inspector/InspectorInstrumentation.cpp:
3061 (WebCore::InspectorInstrumentation::consoleStartRecordingCanvasImpl):
3062 (WebCore::InspectorInstrumentation::didChangeCSSCanvasClientNodesImpl):
3063 (WebCore::InspectorInstrumentation::didCreateCanvasRenderingContextImpl):
3064 (WebCore::InspectorInstrumentation::didChangeCanvasMemoryImpl):
3065 (WebCore::InspectorInstrumentation::didFinishRecordingCanvasFrameImpl):
3066 (WebCore::InspectorInstrumentation::didEnableExtensionImpl):
3067 (WebCore::InspectorInstrumentation::didCreateProgramImpl):
3068 (WebCore::InspectorInstrumentation::didCreateCSSCanvasImpl): Deleted.
3070 * inspector/agents/InspectorCanvasAgent.h:
3071 * inspector/agents/InspectorCanvasAgent.cpp:
3072 (WebCore::InspectorCanvasAgent::enable):
3073 (WebCore::InspectorCanvasAgent::requestNode):
3074 (WebCore::InspectorCanvasAgent::requestContent):
3075 (WebCore::InspectorCanvasAgent::requestCSSCanvasClientNodes):
3076 (WebCore::contextAsScriptValue):
3077 (WebCore::InspectorCanvasAgent::resolveCanvasContext):
3078 (WebCore::InspectorCanvasAgent::startRecording):
3079 (WebCore::InspectorCanvasAgent::stopRecording):
3080 (WebCore::InspectorCanvasAgent::updateShader):
3081 (WebCore::InspectorCanvasAgent::frameNavigated):
3082 (WebCore::InspectorCanvasAgent::didChangeCSSCanvasClientNodes):
3083 (WebCore::InspectorCanvasAgent::didCreateCanvasRenderingContext):
3084 (WebCore::InspectorCanvasAgent::didChangeCanvasMemory):
3085 (WebCore::InspectorCanvasAgent::recordCanvasAction):
3086 (WebCore::InspectorCanvasAgent::canvasDestroyed):
3087 (WebCore::InspectorCanvasAgent::didFinishRecordingCanvasFrame):
3088 (WebCore::InspectorCanvasAgent::consoleStartRecordingCanvas):
3089 (WebCore::InspectorCanvasAgent::didEnableExtension):
3090 (WebCore::InspectorCanvasAgent::didCreateProgram):
3091 (WebCore::InspectorCanvasAgent::canvasRecordingTimerFired):
3092 (WebCore::InspectorCanvasAgent::clearCanvasData):
3093 (WebCore::InspectorCanvasAgent::unbindCanvas):
3094 (WebCore::InspectorCanvasAgent::findInspectorCanvas):
3095 (WebCore::InspectorCanvasAgent::unbindProgram):
3096 (WebCore::InspectorCanvasAgent::didCreateCSSCanvas): Deleted.
3098 * inspector/InspectorCanvas.h:
3099 * inspector/InspectorCanvas.cpp:
3100 (WebCore::InspectorCanvas::create):
3101 (WebCore::InspectorCanvas::InspectorCanvas):
3102 (WebCore::InspectorCanvas::canvasElement):
3103 (WebCore::InspectorCanvas::resetRecordingData):
3104 (WebCore::InspectorCanvas::recordAction):
3105 (WebCore::InspectorCanvas::buildObjectForCanvas):
3106 (WebCore::InspectorCanvas::getCanvasContentAsDataURL):
3107 (WebCore::InspectorCanvas::buildInitialState):
3108 (WebCore::InspectorCanvas::~InspectorCanvas): Deleted.
3110 * inspector/InspectorShaderProgram.h:
3111 * inspector/InspectorShaderProgram.cpp:
3112 (WebCore::InspectorShaderProgram::context const):
3114 * page/PageConsoleClient.cpp:
3115 (WebCore::PageConsoleClient::record):
3116 (WebCore::PageConsoleClient::recordEnd):
3120 (WebCore::Document::getCSSCanvasElement):
3121 (WebCore::Document::nameForCSSCanvasElement const):
3122 We have no reason to save the CSS canvas name for each InspectorCanvas object, so instead we
3123 can just query for the name based on the CanvasRenderingContext's HTMLCanvasElement (assuming
3124 it is not an OffscreenCanvas) when we need it.
3126 2018-01-04 Chris Fleizach <cfleizach@apple.com>
3128 AX: Implement updated CSS3 Speech for 'speak' and 'speak-as' properties
3129 https://bugs.webkit.org/show_bug.cgi?id=180361
3131 Reviewed by Zalan Bujtas.
3133 Change speak -> speakAs, and allow a combination of properties.
3135 Tests: Updated accessibility/mac/css-speech-speak.html
3137 * accessibility/AccessibilityObject.h:
3138 (WebCore::AccessibilityObject::speakAsProperty const):
3139 (WebCore::AccessibilityObject::speakProperty const): Deleted.
3140 * accessibility/AccessibilityRenderObject.cpp:
3141 (WebCore::AccessibilityRenderObject::speakAsProperty const):
3142 (WebCore::AccessibilityRenderObject::speakProperty const): Deleted.
3143 * accessibility/AccessibilityRenderObject.h:
3144 * accessibility/ios/WebAccessibilityObjectWrapperIOS.mm:
3145 (-[WebAccessibilityObjectWrapper accessibilitySpeechHint]):
3146 * accessibility/mac/WebAccessibilityObjectWrapperBase.h:
3147 * accessibility/mac/WebAccessibilityObjectWrapperBase.mm:
3148 (-[WebAccessibilityObjectWrapperBase baseAccessibilitySpeechHint]):
3149 * accessibility/mac/WebAccessibilityObjectWrapperMac.mm:
3150 (-[WebAccessibilityObjectWrapper accessibilityAttributeValue:]):
3151 * css/CSSComputedStyleDeclaration.cpp:
3152 (WebCore::speakAsToCSSValue):
3153 (WebCore::ComputedStyleExtractor::propertyValue):
3154 * css/CSSPrimitiveValueMappings.h:
3155 (WebCore::CSSPrimitiveValue::CSSPrimitiveValue):
3156 (WebCore::CSSPrimitiveValue::operator ESpeakAs const):
3157 (WebCore::CSSPrimitiveValue::operator ESpeak const): Deleted.
3158 * css/CSSProperties.json:
3159 * css/StyleBuilderConverter.h:
3160 (WebCore::StyleBuilderConverter::convertSpeakAs):
3161 * css/parser/CSSParserFastPaths.cpp:
3162 (WebCore::CSSParserFastPaths::isValidKeywordPropertyAndValue):
3163 (WebCore::CSSParserFastPaths::isKeywordPropertyID):
3164 * css/parser/CSSPropertyParser.cpp:
3165 (WebCore::consumeSpeakAs):
3166 (WebCore::CSSPropertyParser::parseSingleValue):
3167 * rendering/style/RenderStyle.h:
3168 (WebCore::RenderStyle::speakAs const):
3169 (WebCore::RenderStyle::setSpeakAs):
3170 (WebCore::RenderStyle::initialSpeakAs):
3171 (WebCore::RenderStyle::speak const): Deleted.
3172 (WebCore::RenderStyle::setSpeak): Deleted.
3173 (WebCore::RenderStyle::initialSpeak): Deleted.
3174 * rendering/style/RenderStyleConstants.h:
3175 (WebCore::operator| ):
3176 (WebCore::operator|= ):
3177 * rendering/style/StyleRareInheritedData.cpp:
3178 (WebCore::StyleRareInheritedData::StyleRareInheritedData):
3179 (WebCore::StyleRareInheritedData::operator== const):
3180 * rendering/style/StyleRareInheritedData.h:
3182 2018-01-04 Brian Burg <bburg@apple.com>
3184 Web Inspector: Capture Element Screenshot looks fuzzy
3185 https://bugs.webkit.org/show_bug.cgi?id=175734
3186 <rdar://problem/33803377>
3188 Reviewed by Joseph Pecoraro and Simon Fraser.
3190 Screenshots taken by Web Inspector were being downscaled from the
3191 internal size to the logical size, causing them to be blurry when
3192 later upscaled to the internal size.
3194 Replace ScaleBehavior { Scaled, Unscaled } with PreserveResolution { No, Yes }.
3195 This is a lot less confusing to read both inside ImageBuffer and at its use sites.
3197 Remove unused CoordinateSystem argument for ImageBuffer::toDataURL,
3198 and replace it with PreserveResolution. Plumb PreserveResolution into toCFData
3199 so that PreserveResolution::Yes will preserve the internal size of
3200 the image buffer, just as it does in other methods that take PreserveResolution.
3202 At the use site in InspectorPageAgent, always request PreserveResolution::Yes snapshots
3203 when taking an element screenshot. For now, keep using downscaled (smaller)
3204 snapshots when capturing canvas previews, as the previews are not full-size.
3206 Test: inspector/page/hidpi-snapshot-size.html
3208 * html/HTMLCanvasElement.cpp:
3209 (WebCore::HTMLCanvasElement::makePresentationCopy):
3210 (WebCore::HTMLCanvasElement::copiedImage const):
3211 * html/canvas/CanvasRenderingContext2DBase.cpp:
3212 (WebCore::CanvasRenderingContext2DBase::createPattern):
3213 * inspector/agents/InspectorPageAgent.cpp:
3214 (WebCore::InspectorPageAgent::snapshotNode):
3215 (WebCore::InspectorPageAgent::snapshotRect):
3216 * page/TextIndicator.cpp:
3217 (WebCore::takeSnapshot):
3218 * platform/DragImage.cpp:
3219 (WebCore::createDragImageFromSnapshot):
3220 * platform/graphics/BitmapImage.cpp:
3221 (WebCore::BitmapImage::drawPattern):
3222 * platform/graphics/ImageBuffer.h:
3223 * platform/graphics/cairo/ImageBufferCairo.cpp:
3224 (WebCore::ImageBuffer::sinkIntoImage):
3225 (WebCore::ImageBuffer::copyImage const):
3226 (WebCore::ImageBuffer::toDataURL const):
3227 * platform/graphics/cg/ImageBufferCG.cpp:
3228 (WebCore::createBitmapImageAfterScalingIfNeeded):
3229 (WebCore::ImageBuffer::copyImage const):
3230 (WebCore::ImageBuffer::sinkIntoImage):
3231 (WebCore::ImageBuffer::toDataURL const):
3232 (WebCore::ImageBuffer::toData const):
3233 (WebCore::ImageBuffer::toCFData const):
3234 * platform/graphics/gtk/ImageBufferGtk.cpp:
3235 (WebCore::ImageBuffer::toDataURL const):
3236 * platform/graphics/win/ImageBufferDirect2D.cpp:
3237 (WebCore::ImageBuffer::copyImage const):
3238 (WebCore::ImageBuffer::sinkIntoImage):
3239 (WebCore::ImageBuffer::toDataURL const):
3240 * svg/graphics/SVGImage.cpp:
3241 (WebCore::SVGImage::drawPatternForContainer):
3243 2018-01-04 John Wilander <wilander@apple.com>
3245 Storage Access API: Turn feature on by default in Settings.yaml
3246 https://bugs.webkit.org/show_bug.cgi?id=181298
3247 <rdar://problem/36302506>
3249 Reviewed by Brent Fulgham.
3251 No new tests. This is just a feature settings change.
3253 * page/Settings.yaml:
3255 2018-01-04 Zalan Bujtas <zalan@apple.com>
3257 WebContent process crashes while loading https://www.classicspecs.com
3258 https://bugs.webkit.org/show_bug.cgi?id=181290
3259 <rdar://problem/36225906>
3261 Reviewed by Simon Fraser.
3263 Floats can overhang multiple blocks (they are called intruding floats).
3264 Each block keeps track of such intruding floats. When an overhanging float box is destroyed,
3265 we need to deregister it from all those blocks. We do it by walking up the ancestor block chain
3266 and check if the parent (grandparent etc) block still contains this float. Once we find the topmost block,
3267 we start deregistering it by traversing back on the descendant blocks.
3268 Normally we do it in RenderElement::takeChildInternal right before the box is getting detached.
3269 However in certain cases (like when the float's parent happens to be an anonymous wrapper)
3270 by the time we get to ::takeChildInternal the subtree is already detached and we can't access all the
3272 This patch ensure that the floating box is still attached during de-registration.
3274 Test: fast/block/float/crash-when-intruding-float-has-anonymous-parent-and-detach.html
3276 * rendering/RenderObject.cpp:
3277 (WebCore::RenderObject::removeFromParentAndDestroyCleaningUpAnonymousWrappers):
3279 2018-01-04 Eric Carlson <eric.carlson@apple.com>
3281 [MediaStream] Add Mock screen capture source
3282 https://bugs.webkit.org/show_bug.cgi?id=181291
3283 <rdar://problem/36298164>
3285 Reviewed by Dean Jackson.
3287 Tests: http/tests/media/media-stream/get-display-media-prompt.html
3288 GetDisplayMediaTest.BasicPrompt
3289 GetDisplayMediaTest.Constraints
3291 * Modules/mediastream/MediaDevices.cpp:
3292 (WebCore::MediaDevices::MediaDevices): Add static_assert to ensure MediaDevices::DisplayCaptureSurfaceType
3293 and RealtimeMediaSourceSettings::DisplaySurfaceType values are equivalent.
3294 (WebCore::MediaDevices::getSupportedConstraints): Remove bogus code.
3295 * Modules/mediastream/MediaDevices.h: Add DisplayCaptureSurfaceType.
3296 * Modules/mediastream/MediaDevices.idl: Ditto.
3298 * Modules/mediastream/MediaStreamTrack.cpp:
3299 (WebCore::MediaStreamTrack::getSettings const): Add a FIXME.
3300 * Modules/mediastream/MediaStreamTrack.h: Add displaySurface and logicalSurface.
3302 * Modules/mediastream/MediaTrackSupportedConstraints.h: Remove displaySurface and logicalSurface.
3303 * Modules/mediastream/MediaTrackSupportedConstraints.idl:
3305 * SourcesCocoa.txt: Add DisplayCaptureManagerCocoa.cpp and DisplayCaptureSourceCocoa.cpp.
3307 * WebCore.xcodeproj/project.pbxproj: Ditto.
3309 * platform/mediastream/CaptureDevice.h:
3310 (WebCore::CaptureDevice::encode const): Add.
3311 (WebCore::CaptureDevice::decode):
3313 * platform/mediastream/RealtimeMediaSourceCenter.cpp:
3314 (WebCore::RealtimeMediaSourceCenter::getMediaStreamDevices): Include display capture "devices".
3315 (WebCore::RealtimeMediaSourceCenter::validateRequestConstraints): Deal with display capture devices.
3316 (WebCore::RealtimeMediaSourceCenter::captureDeviceWithPersistentID): Ditto.
3317 * platform/mediastream/RealtimeMediaSourceCenter.h:
3319 * platform/mediastream/RealtimeMediaSourceSettings.h:
3320 (WebCore::RealtimeMediaSourceSettings::displaySurface const): Return a DisplaySurfaceType.
3321 (WebCore::RealtimeMediaSourceSettings::setDisplaySurface): Take a DisplaySurfaceType.
3323 * platform/mediastream/mac/DisplayCaptureManagerCocoa.cpp:
3324 (WebCore::DisplayCaptureManagerCocoa::singleton):
3325 (WebCore::DisplayCaptureManagerCocoa::~DisplayCaptureManagerCocoa):
3326 (WebCore::DisplayCaptureManagerCocoa::captureDevices):
3327 (WebCore::DisplayCaptureManagerCocoa::screenCaptureDeviceWithPersistentID):
3328 (WebCore::DisplayCaptureManagerCocoa::captureDeviceWithPersistentID):
3329 * platform/mediastream/mac/DisplayCaptureManagerCocoa.h:
3331 * platform/mediastream/mac/DisplayCaptureSourceCocoa.cpp: Added.
3332 (WebCore::DisplayCaptureSourceCocoa::DisplayCaptureSourceCocoa):
3333 (WebCore::DisplayCaptureSourceCocoa::~DisplayCaptureSourceCocoa):
3334 (WebCore::DisplayCaptureSourceCocoa::capabilities const):
3335 (WebCore::DisplayCaptureSourceCocoa::settings const):
3336 (WebCore::DisplayCaptureSourceCocoa::settingsDidChange):
3337 (WebCore::DisplayCaptureSourceCocoa::startProducingData):
3338 (WebCore::DisplayCaptureSourceCocoa::stopProducingData):
3339 (WebCore::DisplayCaptureSourceCocoa::elapsedTime):
3340 (WebCore::DisplayCaptureSourceCocoa::applyFrameRate):
3341 (WebCore::DisplayCaptureSourceCocoa::emitFrame):
3342 * platform/mediastream/mac/DisplayCaptureSourceCocoa.h:
3344 * platform/mediastream/mac/RealtimeMediaSourceCenterMac.cpp:
3345 (WebCore::RealtimeMediaSourceCenterMac::displayCaptureDeviceManager): New.
3346 * platform/mediastream/mac/RealtimeMediaSourceCenterMac.h:
3348 * platform/mock/MockRealtimeMediaSource.cpp:
3349 (WebCore::deviceMap): Add screen capture "devices".
3350 (WebCore::MockRealtimeMediaSource::displayDevices): New.
3351 * platform/mock/MockRealtimeMediaSource.h:
3353 * platform/mock/MockRealtimeMediaSourceCenter.cpp: Clean up includes.
3354 * platform/mock/MockRealtimeMediaSourceCenter.h:
3356 * platform/mock/MockRealtimeVideoSource.cpp:
3357 (WebCore::MockRealtimeVideoSource::MockRealtimeVideoSource): Mock two screen devices.
3358 (WebCore::MockRealtimeVideoSource::updateSettings): Deal with mock screens.
3359 (WebCore::MockRealtimeVideoSource::initializeCapabilities): Ditto.
3360 (WebCore::MockRealtimeVideoSource::initializeSupportedConstraints): Ditto.
3361 (WebCore::MockRealtimeVideoSource::drawText): Ditto.
3362 (WebCore::MockRealtimeVideoSource::generateFrame): Ditto.
3363 * platform/mock/MockRealtimeVideoSource.h:
3364 (WebCore::MockRealtimeVideoSource::mockCamera const):
3365 (WebCore::MockRealtimeVideoSource::mockScreen const):
3367 2018-01-04 Youenn Fablet <youenn@apple.com>
3369 FetchResponse should set its internal response text encoding name
3370 https://bugs.webkit.org/show_bug.cgi?id=181284
3372 Reviewed by Alex Christensen.
3374 Covered by rebased test.
3376 * Modules/fetch/FetchResponse.cpp:
3377 (WebCore::FetchResponse::create): Set response text encoding based on content type charset.
3379 2018-01-04 John Wilander <wilander@apple.com>
3381 Storage Access API: Remove JavaScript confirm() prompt from Document::requestStorageAccess()
3382 https://bugs.webkit.org/show_bug.cgi?id=181276
3383 <rdar://problem/36290463>
3385 Reviewed by Alex Christensen.
3387 No new tests. Existing test expectations updated.
3390 (WebCore::Document::requestStorageAccess):
3392 2018-01-04 Carlos Garcia Campos <cgarcia@igalia.com>
3394 [GTK] Issues with Ahem's ex / x-height
3395 https://bugs.webkit.org/show_bug.cgi?id=180581
3397 Reviewed by Michael Catanzaro.
3399 Get the x-height value from the TT_OS2 table if available.
3401 Fixes: fast/text/break-word-pre-wrap.html
3402 imported/w3c/web-platform-tests/css/css-shapes-1/shape-outside/values/shape-outside-shape-arguments-000.html
3404 * platform/graphics/freetype/SimpleFontDataFreeType.cpp:
3405 (WebCore::Font::platformInit):
3407 2018-01-04 Philippe Normand <pnormand@igalia.com>
3409 Unreviewed, GTK build fix attempt after r226357
3411 * platform/graphics/gstreamer/GStreamerUtilities.h: The
3412 GST_BUFFER_DTS_OR_PTS macro was added in GStreamer 1.8 but old
3413 versions of Debian might not have this release yet.
3415 2018-01-04 Youenn Fablet <youenn@apple.com>
git://git.webkit.org / WebKit-https.git / blob