[webkit-changes] [213443] trunk/LayoutTests
Title: [213443] trunk/LayoutTests Revision 213443 Author carlo...@webkit.org Date 2017-03-05 23:38:24 -0800 (Sun, 05 Mar 2017) Log Message Unreviewed GTK+ gardening. Rebaseline fast/css/css2-system-fonts.html after r213267. * platform/gtk/fast/css/css2-system-fonts-expected.txt: Modified Paths trunk/LayoutTests/ChangeLog trunk/LayoutTests/platform/gtk/fast/css/css2-system-fonts-expected.txt Diff Modified: trunk/LayoutTests/ChangeLog (213442 => 213443) --- trunk/LayoutTests/ChangeLog 2017-03-06 05:51:57 UTC (rev 213442) +++ trunk/LayoutTests/ChangeLog 2017-03-06 07:38:24 UTC (rev 213443) @@ -1,3 +1,9 @@ +2017-03-05 Carlos Garcia Campos+ +Unreviewed GTK+ gardening. Rebaseline fast/css/css2-system-fonts.html after r213267. + +* platform/gtk/fast/css/css2-system-fonts-expected.txt: + 2017-03-05 Simon Fraser Avoid backing store for layers with empty text nodes in a few more cases Modified: trunk/LayoutTests/platform/gtk/fast/css/css2-system-fonts-expected.txt (213442 => 213443) --- trunk/LayoutTests/platform/gtk/fast/css/css2-system-fonts-expected.txt 2017-03-06 05:51:57 UTC (rev 213442) +++ trunk/LayoutTests/platform/gtk/fast/css/css2-system-fonts-expected.txt 2017-03-06 07:38:24 UTC (rev 213443) @@ -1,10 +1,10 @@ This tests platform specific system font styles. If any of the styles appear in monospace the test fails. -caption: normal normal normal 16px/17px 'Liberation Sans' -icon: normal normal normal 16px/17px 'Liberation Sans' -menu: normal normal normal 16px/17px 'Liberation Sans' -message-box: normal normal normal 16px/17px 'Liberation Sans' -small-caption: normal normal normal 16px/17px 'Liberation Sans' -status-bar: normal normal normal 16px/17px 'Liberation Sans' --webkit-mini-control: normal normal normal 16px/17px 'Liberation Sans' --webkit-small-control: normal normal normal 16px/17px 'Liberation Sans' --webkit-control: normal normal normal 16px/17px 'Liberation Sans' +caption: normal normal normal normal 16px/17px 'Liberation Sans' +icon: normal normal normal normal 16px/17px 'Liberation Sans' +menu: normal normal normal normal 16px/17px 'Liberation Sans' +message-box: normal normal normal normal 16px/17px 'Liberation Sans' +small-caption: normal normal normal normal 16px/17px 'Liberation Sans' +status-bar: normal normal normal normal 16px/17px 'Liberation Sans' +-webkit-mini-control: normal normal normal normal 16px/17px 'Liberation Sans' +-webkit-small-control: normal normal normal normal 16px/17px 'Liberation Sans' +-webkit-control: normal normal normal normal 16px/17px 'Liberation Sans' ___ webkit-changes mailing list webkit-changes@lists.webkit.org https://lists.webkit.org/mailman/listinfo/webkit-changes
[webkit-changes] [213442] trunk/Source/WebKit2
Title: [213442] trunk/Source/WebKit2 Revision 213442 Author beid...@apple.com Date 2017-03-05 21:51:57 -0800 (Sun, 05 Mar 2017) Log Message Remove new files accidentally added in r213441 Unreviewed. * UIProcess/WebLoadingProviderProxy.cpp: Removed. * UIProcess/WebLoadingProviderProxy.h: Removed. * UIProcess/WebLoadingProviderProxy.messages.in: Removed. Modified Paths trunk/Source/WebKit2/ChangeLog Removed Paths trunk/Source/WebKit2/UIProcess/WebLoadingProviderProxy.cpp trunk/Source/WebKit2/UIProcess/WebLoadingProviderProxy.h trunk/Source/WebKit2/UIProcess/WebLoadingProviderProxy.messages.in Diff Modified: trunk/Source/WebKit2/ChangeLog (213441 => 213442) --- trunk/Source/WebKit2/ChangeLog 2017-03-06 02:53:31 UTC (rev 213441) +++ trunk/Source/WebKit2/ChangeLog 2017-03-06 05:51:57 UTC (rev 213442) @@ -1,5 +1,15 @@ 2017-03-05 Brady Eidson+Remove new files accidentally added in r213441 + +Unreviewed. + +* UIProcess/WebLoadingProviderProxy.cpp: Removed. +* UIProcess/WebLoadingProviderProxy.h: Removed. +* UIProcess/WebLoadingProviderProxy.messages.in: Removed. + +2017-03-05 Brady Eidson + Rename CustomProtocolFoo to LegacyCustomProtocolFoo. https://bugs.webkit.org/show_bug.cgi?id=169188 Deleted: trunk/Source/WebKit2/UIProcess/WebLoadingProviderProxy.cpp (213441 => 213442) --- trunk/Source/WebKit2/UIProcess/WebLoadingProviderProxy.cpp 2017-03-06 02:53:31 UTC (rev 213441) +++ trunk/Source/WebKit2/UIProcess/WebLoadingProviderProxy.cpp 2017-03-06 05:51:57 UTC (rev 213442) @@ -1,32 +0,0 @@ -/* - * Copyright (C) 2017 Apple Inc. All rights reserved. - * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions - * are met: - * 1. Redistributions of source code must retain the above copyright - *notice, this list of conditions and the following disclaimer. - * 2. Redistributions in binary form must reproduce the above copyright - *notice, this list of conditions and the following disclaimer in the - *documentation and/or other materials provided with the distribution. - * - * THIS SOFTWARE IS PROVIDED BY APPLE INC. AND ITS CONTRIBUTORS ``AS IS'' - * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, - * THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR - * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL APPLE INC. OR ITS CONTRIBUTORS - * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR - * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF - * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS - * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN - * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) - * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF - * THE POSSIBILITY OF SUCH DAMAGE. - */ - -#include "config.h" -#include "WebLoadingProviderProxy.h" - -namespace WebKit { - - -} // namespace WebKit Deleted: trunk/Source/WebKit2/UIProcess/WebLoadingProviderProxy.h (213441 => 213442) --- trunk/Source/WebKit2/UIProcess/WebLoadingProviderProxy.h 2017-03-06 02:53:31 UTC (rev 213441) +++ trunk/Source/WebKit2/UIProcess/WebLoadingProviderProxy.h 2017-03-06 05:51:57 UTC (rev 213442) @@ -1,31 +0,0 @@ -/* - * Copyright (C) 2017 Apple Inc. All rights reserved. - * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions - * are met: - * 1. Redistributions of source code must retain the above copyright - *notice, this list of conditions and the following disclaimer. - * 2. Redistributions in binary form must reproduce the above copyright - *notice, this list of conditions and the following disclaimer in the - *documentation and/or other materials provided with the distribution. - * - * THIS SOFTWARE IS PROVIDED BY APPLE INC. AND ITS CONTRIBUTORS ``AS IS'' - * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, - * THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR - * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL APPLE INC. OR ITS CONTRIBUTORS - * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR - * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF - * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS - * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN - * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) - * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF - * THE POSSIBILITY OF SUCH DAMAGE. - */ - -#pragma once - -namespace WebKit { - - -} // namespace WebKit Deleted: trunk/Source/WebKit2/UIProcess/WebLoadingProviderProxy.messages.in (213441 => 213442) ---
[webkit-changes] [213440] trunk
Title: [213440] trunk Revision 213440 Author simon.fra...@apple.com Date 2017-03-05 18:48:20 -0800 (Sun, 05 Mar 2017) Log Message Avoid backing store for layers with empty text nodes in a few more cases https://bugs.webkit.org/show_bug.cgi?id=169185 Reviewed by Dan Bernstein. Source/WebCore: In hasPaintingNonLayerDescendants() we can check whether the RenderText's linesBoundingBox() is empty to avoid backing store in a few more cases. Also use containsOnlyWhitespace() rather than isAllCollapsibleWhitespace(), because there's no need for backing store for non-selectable whitespace text. Covered by existing tests. * rendering/RenderLayer.cpp: (WebCore::RenderLayer::calculateClipRects): LayoutTests: inline-block-no-backing.html enhanced to have a layer with non-collapsible whitespace (an ). Rebaselined other tests. * compositing/backing/inline-block-no-backing-expected.txt: * compositing/backing/inline-block-no-backing.html: * compositing/iframes/page-cache-layer-tree-expected.txt: * css3/blending/blend-mode-clip-accelerated-blending-canvas-expected.txt: * platform/ios-simulator-wk2/compositing/backing/inline-block-no-backing-expected.txt: * platform/ios-simulator-wk2/compositing/iframes/page-cache-layer-tree-expected.txt: Modified Paths trunk/LayoutTests/ChangeLog trunk/LayoutTests/compositing/backing/inline-block-no-backing-expected.txt trunk/LayoutTests/compositing/backing/inline-block-no-backing.html trunk/LayoutTests/compositing/iframes/page-cache-layer-tree-expected.txt trunk/LayoutTests/css3/blending/blend-mode-clip-accelerated-blending-canvas-expected.txt trunk/LayoutTests/platform/ios-simulator-wk2/compositing/backing/inline-block-no-backing-expected.txt trunk/LayoutTests/platform/ios-simulator-wk2/compositing/iframes/page-cache-layer-tree-expected.txt trunk/Source/WebCore/ChangeLog trunk/Source/WebCore/rendering/RenderLayer.cpp Diff Modified: trunk/LayoutTests/ChangeLog (213439 => 213440) --- trunk/LayoutTests/ChangeLog 2017-03-06 01:37:53 UTC (rev 213439) +++ trunk/LayoutTests/ChangeLog 2017-03-06 02:48:20 UTC (rev 213440) @@ -1,3 +1,21 @@ +2017-03-05 Simon Fraser+ +Avoid backing store for layers with empty text nodes in a few more cases +https://bugs.webkit.org/show_bug.cgi?id=169185 + +Reviewed by Dan Bernstein. + +inline-block-no-backing.html enhanced to have a layer with non-collapsible whitespace (an ). + +Rebaselined other tests. + +* compositing/backing/inline-block-no-backing-expected.txt: +* compositing/backing/inline-block-no-backing.html: +* compositing/iframes/page-cache-layer-tree-expected.txt: +* css3/blending/blend-mode-clip-accelerated-blending-canvas-expected.txt: +* platform/ios-simulator-wk2/compositing/backing/inline-block-no-backing-expected.txt: +* platform/ios-simulator-wk2/compositing/iframes/page-cache-layer-tree-expected.txt: + 2017-03-05 Chris Dumez Using in causes following to get swallowed Modified: trunk/LayoutTests/compositing/backing/inline-block-no-backing-expected.txt (213439 => 213440) --- trunk/LayoutTests/compositing/backing/inline-block-no-backing-expected.txt 2017-03-06 01:37:53 UTC (rev 213439) +++ trunk/LayoutTests/compositing/backing/inline-block-no-backing-expected.txt 2017-03-06 02:48:20 UTC (rev 213440) @@ -1,4 +1,4 @@ - + (GraphicsLayer (anchor 0.00 0.00) @@ -7,7 +7,7 @@ (GraphicsLayer (bounds 800.00 600.00) (contentsOpaque 1) - (children 3 + (children 4 (GraphicsLayer (position 18.00 18.00) (bounds 160.00 320.00) @@ -14,7 +14,7 @@ (drawsContent 1) (children 2 (GraphicsLayer - (position 10.00 10.00) + (position 18.00 10.00) (bounds 120.00 120.00) ) (GraphicsLayer @@ -51,6 +51,20 @@ ) ) ) +(GraphicsLayer + (position 570.00 18.00) + (bounds 160.00 320.00) + (children 2 +(GraphicsLayer + (position 10.00 10.00) + (bounds 120.00 120.00) +) +(GraphicsLayer + (position 10.00 154.00) + (bounds 120.00 120.00) +) + ) +) ) ) ) Modified: trunk/LayoutTests/compositing/backing/inline-block-no-backing.html (213439 => 213440) --- trunk/LayoutTests/compositing/backing/inline-block-no-backing.html 2017-03-06 01:37:53 UTC (rev 213439) +++ trunk/LayoutTests/compositing/backing/inline-block-no-backing.html 2017-03-06 02:48:20 UTC (rev 213440) @@ -42,6 +42,10 @@ + + Modified: trunk/LayoutTests/compositing/iframes/page-cache-layer-tree-expected.txt (213439 => 213440) --- trunk/LayoutTests/compositing/iframes/page-cache-layer-tree-expected.txt 2017-03-06 01:37:53 UTC
[webkit-changes] [213439] trunk/Source/WebCore
Title: [213439] trunk/Source/WebCore Revision 213439 Author cdu...@apple.com Date 2017-03-05 17:37:53 -0800 (Sun, 05 Mar 2017) Log Message Drop unnecessary backslash in ResourceResponse::sanitizeSuggestedFilename() https://bugs.webkit.org/show_bug.cgi?id=169066 Reviewed by Sam Weinig. Drop unnecessary backslash in ResourceResponse::sanitizeSuggestedFilename(). * platform/network/ResourceResponseBase.cpp: (WebCore::ResourceResponseBase::sanitizeSuggestedFilename): Modified Paths trunk/Source/WebCore/ChangeLog trunk/Source/WebCore/platform/network/ResourceResponseBase.cpp Diff Modified: trunk/Source/WebCore/ChangeLog (213438 => 213439) --- trunk/Source/WebCore/ChangeLog 2017-03-06 01:36:30 UTC (rev 213438) +++ trunk/Source/WebCore/ChangeLog 2017-03-06 01:37:53 UTC (rev 213439) @@ -1,5 +1,17 @@ 2017-03-05 Chris Dumez+Drop unnecessary backslash in ResourceResponse::sanitizeSuggestedFilename() +https://bugs.webkit.org/show_bug.cgi?id=169066 + +Reviewed by Sam Weinig. + +Drop unnecessary backslash in ResourceResponse::sanitizeSuggestedFilename(). + +* platform/network/ResourceResponseBase.cpp: +(WebCore::ResourceResponseBase::sanitizeSuggestedFilename): + +2017-03-05 Chris Dumez + Using in causes following to get swallowed https://bugs.webkit.org/show_bug.cgi?id=163552 Modified: trunk/Source/WebCore/platform/network/ResourceResponseBase.cpp (213438 => 213439) --- trunk/Source/WebCore/platform/network/ResourceResponseBase.cpp 2017-03-06 01:36:30 UTC (rev 213438) +++ trunk/Source/WebCore/platform/network/ResourceResponseBase.cpp 2017-03-06 01:37:53 UTC (rev 213439) @@ -231,7 +231,7 @@ ResourceResponse response(URL(ParsedURLString, "http://example.com/"), String(), -1, String()); response.setHTTPStatusCode(200); -String escapedSuggestedFilename = String(suggestedFilename).replace('\\', "").replace('\"', "\\\""); +String escapedSuggestedFilename = String(suggestedFilename).replace('\\', "").replace('"', "\\\""); String value = makeString("attachment; filename=\"", escapedSuggestedFilename, '"'); response.setHTTPHeaderField(HTTPHeaderName::ContentDisposition, value); return response.suggestedFilename(); ___ webkit-changes mailing list webkit-changes@lists.webkit.org https://lists.webkit.org/mailman/listinfo/webkit-changes
[webkit-changes] [213438] trunk
Title: [213438] trunk Revision 213438 Author cdu...@apple.com Date 2017-03-05 17:36:30 -0800 (Sun, 05 Mar 2017) Log Message Using in causes following to get swallowed https://bugs.webkit.org/show_bug.cgi?id=163552 Reviewed by Sam Weinig. Source/WebCore: As per the HTML specification [1], when finding a "form" tag in the "in body" insertion mode, we should insert an HTML element for the token, and, if there is no template element on the stack of open elements, set the form element pointer to point to the element created. We were missing the "if there is no template element on the stack of open elements" check and setting the form element pointer unconditionally. This patch fixes the issue. [1] https://html.spec.whatwg.org/multipage/syntax.html#parsing-main-inbody:form-element-pointer-2 Test: fast/parser/form-after-template.html * html/parser/HTMLConstructionSite.cpp: (WebCore::HTMLConstructionSite::insertHTMLFormElement): LayoutTests: Add layout test coverage. * fast/parser/form-after-template-expected.html: Added. * fast/parser/form-after-template.html: Added. Modified Paths trunk/LayoutTests/ChangeLog trunk/Source/WebCore/ChangeLog trunk/Source/WebCore/html/parser/HTMLConstructionSite.cpp Added Paths trunk/LayoutTests/fast/parser/form-after-template-expected.html trunk/LayoutTests/fast/parser/form-after-template.html Diff Modified: trunk/LayoutTests/ChangeLog (213437 => 213438) --- trunk/LayoutTests/ChangeLog 2017-03-05 20:31:06 UTC (rev 213437) +++ trunk/LayoutTests/ChangeLog 2017-03-06 01:36:30 UTC (rev 213438) @@ -1,3 +1,15 @@ +2017-03-05 Chris Dumez+ +Using in causes following to get swallowed +https://bugs.webkit.org/show_bug.cgi?id=163552 + +Reviewed by Sam Weinig. + +Add layout test coverage. + +* fast/parser/form-after-template-expected.html: Added. +* fast/parser/form-after-template.html: Added. + 2017-03-04 Myles C. Maxfield Update CSSFontSelector's matching algorithm to understand ranges Added: trunk/LayoutTests/fast/parser/form-after-template-expected.html (0 => 213438) --- trunk/LayoutTests/fast/parser/form-after-template-expected.html (rev 0) +++ trunk/LayoutTests/fast/parser/form-after-template-expected.html 2017-03-06 01:36:30 UTC (rev 213438) @@ -0,0 +1,9 @@ + + + +Test that the form following a template element properly gets parsed. + +Submit + + + Added: trunk/LayoutTests/fast/parser/form-after-template.html (0 => 213438) --- trunk/LayoutTests/fast/parser/form-after-template.html (rev 0) +++ trunk/LayoutTests/fast/parser/form-after-template.html 2017-03-06 01:36:30 UTC (rev 213438) @@ -0,0 +1,10 @@ + + + +Test that the form following a template element properly gets parsed. + + +Submit + + + Modified: trunk/Source/WebCore/ChangeLog (213437 => 213438) --- trunk/Source/WebCore/ChangeLog 2017-03-05 20:31:06 UTC (rev 213437) +++ trunk/Source/WebCore/ChangeLog 2017-03-06 01:36:30 UTC (rev 213438) @@ -1,3 +1,26 @@ +2017-03-05 Chris Dumez + +Using in causes following to get swallowed +https://bugs.webkit.org/show_bug.cgi?id=163552 + +Reviewed by Sam Weinig. + +As per the HTML specification [1], when finding a "form" tag in the "in body" +insertion mode, we should insert an HTML element for the token, and, if there +is no template element on the stack of open elements, set the form element +pointer to point to the element created. + +We were missing the "if there is no template element on the stack of open +elements" check and setting the form element pointer unconditionally. +This patch fixes the issue. + +[1] https://html.spec.whatwg.org/multipage/syntax.html#parsing-main-inbody:form-element-pointer-2 + +Test: fast/parser/form-after-template.html + +* html/parser/HTMLConstructionSite.cpp: +(WebCore::HTMLConstructionSite::insertHTMLFormElement): + 2017-03-04 Myles C. Maxfield Update CSSFontSelector's matching algorithm to understand ranges Modified: trunk/Source/WebCore/html/parser/HTMLConstructionSite.cpp (213437 => 213438) --- trunk/Source/WebCore/html/parser/HTMLConstructionSite.cpp 2017-03-05 20:31:06 UTC (rev 213437) +++ trunk/Source/WebCore/html/parser/HTMLConstructionSite.cpp 2017-03-06 01:36:30 UTC (rev 213438) @@ -476,10 +476,14 @@ void HTMLConstructionSite::insertHTMLFormElement(AtomicHTMLToken&& token, bool isDemoted) { auto element = createHTMLElement(token); -m_form = (element.get()); -m_form->setDemoted(isDemoted); -attachLater(currentNode(), *m_form); -m_openElements.push(HTMLStackItem::create(*m_form, WTFMove(token))); +auto& formElement = downcast(element.get()); +// If there is no template element on the stack of open elements, set the +// form element pointer to
[webkit-changes] [213437] trunk/Source/WebCore
Title: [213437] trunk/Source/WebCore Revision 213437 Author aes...@apple.com Date 2017-03-05 12:31:06 -0800 (Sun, 05 Mar 2017) Log Message Add support for canvas.toBlob https://bugs.webkit.org/show_bug.cgi?id=148878 Reviewed by Tim Horton. Address additional review feedback received after r213412 landed. * html/HTMLCanvasElement.cpp: (WebCore::HTMLCanvasElement::toBlob): Passed null to the blob callback if blob conversion fails for any reason. * platform/graphics/cg/ImageBufferCG.cpp: (WebCore::cgImage): Removed an unnecessary local variable. Modified Paths trunk/Source/WebCore/ChangeLog trunk/Source/WebCore/html/HTMLCanvasElement.cpp trunk/Source/WebCore/platform/graphics/cg/ImageBufferCG.cpp Diff Modified: trunk/Source/WebCore/ChangeLog (213436 => 213437) --- trunk/Source/WebCore/ChangeLog 2017-03-05 20:14:02 UTC (rev 213436) +++ trunk/Source/WebCore/ChangeLog 2017-03-05 20:31:06 UTC (rev 213437) @@ -238,6 +238,21 @@ Add support for canvas.toBlob https://bugs.webkit.org/show_bug.cgi?id=148878 + +Reviewed by Tim Horton. + +Address additional review feedback received after r213412 landed. + +* html/HTMLCanvasElement.cpp: +(WebCore::HTMLCanvasElement::toBlob): Passed null to the blob callback if blob conversion +fails for any reason. +* platform/graphics/cg/ImageBufferCG.cpp: +(WebCore::cgImage): Removed an unnecessary local variable. + +2017-03-03 Andy Estes+ +Add support for canvas.toBlob +https://bugs.webkit.org/show_bug.cgi?id=148878 Reviewed by Tim Horton. Modified: trunk/Source/WebCore/html/HTMLCanvasElement.cpp (213436 => 213437) --- trunk/Source/WebCore/html/HTMLCanvasElement.cpp 2017-03-05 20:14:02 UTC (rev 213436) +++ trunk/Source/WebCore/html/HTMLCanvasElement.cpp 2017-03-05 20:31:06 UTC (rev 213437) @@ -504,8 +504,11 @@ #if USE(CG) if (auto imageData = getImageData()) { +RefPtr blob; Vector blobData = data(*imageData, encodingMIMEType, quality); -callback->scheduleCallback(context, Blob::create(WTFMove(blobData), encodingMIMEType)); +if (!blobData.isEmpty()) +blob = Blob::create(WTFMove(blobData), encodingMIMEType); +callback->scheduleCallback(context, WTFMove(blob)); return { }; } #endif @@ -512,8 +515,11 @@ makeRenderingResultsAvailable(); +RefPtr blob; Vector blobData = buffer()->toData(encodingMIMEType, quality); -callback->scheduleCallback(context, Blob::create(WTFMove(blobData), encodingMIMEType)); +if (!blobData.isEmpty()) +blob = Blob::create(WTFMove(blobData), encodingMIMEType); +callback->scheduleCallback(context, WTFMove(blob)); return { }; } Modified: trunk/Source/WebCore/platform/graphics/cg/ImageBufferCG.cpp (213436 => 213437) --- trunk/Source/WebCore/platform/graphics/cg/ImageBufferCG.cpp 2017-03-05 20:14:02 UTC (rev 213436) +++ trunk/Source/WebCore/platform/graphics/cg/ImageBufferCG.cpp 2017-03-05 20:31:06 UTC (rev 213437) @@ -610,8 +610,7 @@ if (!dataProvider) return nullptr; -auto image = adoptCF(CGImageCreate(source.width(), source.height(), 8, 32, 4 * source.width(), sRGBColorSpaceRef(), kCGBitmapByteOrderDefault | dataAlphaInfo, dataProvider.get(), 0, false, kCGRenderingIntentDefault)); -return image; +return adoptCF(CGImageCreate(source.width(), source.height(), 8, 32, 4 * source.width(), sRGBColorSpaceRef(), kCGBitmapByteOrderDefault | dataAlphaInfo, dataProvider.get(), 0, false, kCGRenderingIntentDefault)); } String dataURL(const ImageData& source, const String& mimeType, std::optional quality) ___ webkit-changes mailing list webkit-changes@lists.webkit.org https://lists.webkit.org/mailman/listinfo/webkit-changes
[webkit-changes] [213436] trunk
Title: [213436] trunk Revision 213436 Author mmaxfi...@apple.com Date 2017-03-05 12:14:02 -0800 (Sun, 05 Mar 2017) Log Message Update CSSFontSelector's matching algorithm to understand ranges https://bugs.webkit.org/show_bug.cgi?id=168892 Reviewed by Jon Lee. Source/WebCore: This patch migrates the font selection algorithm out of FontCacheCoreText and into its own file which can be shared among all ports. It then migrates our web font selection algorithm to use it. This patch doesn't actually change the parsing rules; it just changes the internal machinery for how fonts get selected. Therefore, this patch simply makes zero-length ranges from the discrete values the parser emits, and passes those zero-length ranges to the range-based font selection routine. This means that this patch doesn't actually change the results of the font selection algorithm. One of the inputs to the new font selection algorithm is font-stretch, which previously was not parsed inside @font-face blocks or the CSS Font Loading API. This patch therefore adds parsing support and modifies the existing tests for these pieces to expect parsing to work. Because the font selection algorithm itself is shared between installed fonts and webfonts, this patch doesn't add any additional tests for it (because it is already covered under existing tests). No new tests because there is no behavior change. * CMakeLists.txt: * WebCore.xcodeproj/project.pbxproj: Add new file for the font selection algorithm. * css/CSSFontFace.cpp: (WebCore::CSSFontFace::calculateStretch): Used on @font-face blocks and the CSS Font Loading API. (WebCore::CSSFontFace::setStretch): Fill out the previously stubbed function. * css/CSSFontFace.h: Add the range member variable to CSSFontFaces. * css/CSSFontFaceSet.cpp: (WebCore::CSSFontFaceSet::ensureLocalFontFacesForFamilyRegistered): Now that we care about font-stretch, we need to look it up for local fonts too. This current approach of an extra FontSelectionValue hanging around with a FontTraitsMask is very ugly and will be removed soon (https://bugs.webkit.org/show_bug.cgi?id=168889 and https://bugs.webkit.org/show_bug.cgi?id=168890). (WebCore::computeFontStretch): Used on @font-face blocks and the CSS Font Loading API. (WebCore::CSSFontFaceSet::matchingFaces): Educate about font-stretch. (WebCore::CSSFontFaceSet::fontFace): Migrate to the new font-selection algorithm. (WebCore::fontFaceComparator): Deleted. * css/CSSFontFaceSet.h: * css/CSSFontSelector.cpp: (WebCore::CSSFontSelector::addFontFaceRule): Educate about font-stretch. (WebCore::CSSFontSelector::fontRangesForFamily): Ditto. * css/FontFace.cpp: (WebCore::FontFace::setStretch): Ditto. (WebCore::FontFace::stretch): Ditto. * css/parser/CSSPropertyParser.cpp: (WebCore::CSSPropertyParser::parseFontFaceDescriptor): Ditto. * platform/graphics/FontCache.h: Ditto. * platform/graphics/FontDescription.h: * platform/graphics/FontSelectionAlgorithm.cpp: Added. (WebCore::FontSelectionAlgorithm::filterCapability): (WebCore::FontSelectionAlgorithm::indexOfBestCapabilities): (WebCore::fontSelectionRequestForTraitsMask): (WebCore::initialFontSelectionCapabilitiesForTraitsMask): (WebCore::fontSelectionCapabilitiesForTraitsMask): * platform/graphics/FontSelectionAlgorithm.h: Added. Taken from FontCacheCoreText.cpp. (WebCore::FontSelectionValue::FontSelectionValue): (WebCore::FontSelectionValue::operator float): (WebCore::FontSelectionValue::rawValue): (WebCore::FontSelectionValue::maximumValue): (WebCore::FontSelectionValue::minimumValue): (WebCore::FontSelectionValue::operator+): (WebCore::FontSelectionValue::operator-): (WebCore::FontSelectionValue::operator*): (WebCore::FontSelectionValue::operator/): (WebCore::FontSelectionValue::operator==): (WebCore::FontSelectionValue::operator!=): (WebCore::FontSelectionValue::operator<): (WebCore::FontSelectionValue::operator<=): (WebCore::FontSelectionValue::operator>): (WebCore::FontSelectionValue::operator>=): (WebCore::FontSelectionRange::isValid): (WebCore::FontSelectionRange::expand): (WebCore::FontSelectionRange::includes): (WebCore::FontSelectionRequest::operator==): (WebCore::FontSelectionRequest::operator!=): (WebCore::FontSelectionRequestKey::FontSelectionRequestKey): (WebCore::FontSelectionRequestKey::isHashTableDeletedValue): (WebCore::FontSelectionRequestKey::operator==): (WebCore::FontSelectionRequestKeyHash::hash): (WebCore::FontSelectionRequestKeyHash::equal): (WebCore::FontSelectionCapabilities::expand): (WebCore::FontSelectionAlgorithm::FontSelectionAlgorithm): (WebCore::FontSelectionAlgorithm::iterateActiveCapabilitiesWithReturn): (WebCore::FontSelectionAlgorithm::iterateActiveCapabilities): * platform/graphics/cocoa/FontCacheCoreText.cpp: Moved to FontSelectionAlgorithm. (WebCore::stretchFromCoreTextTraits): (WebCore::FontDatabase::capabilitiesForFontDescriptor): (WebCore::findClosestFont): (WebCore::calculateFontSelectionRequest): (WebCore::platformFontLookupWithFamily):
[webkit-changes] [213435] trunk/Source/WebCore
Title: [213435] trunk/Source/WebCore Revision 213435 Author simon.fra...@apple.com Date 2017-03-05 11:14:12 -0800 (Sun, 05 Mar 2017) Log Message Make some RenderLayer tree traversal in RenderLayerBacking more generic https://bugs.webkit.org/show_bug.cgi?id=169177 Reviewed by Zalan Bujtas. The real goal of this patch is reduce the number of callers of RenderLayerBacking::isPaintDestinationForDescendantLayers() to one. To achieve that, have the setContentsVisible() logic (which is really just about the CSS 'visibility' property) do its own tree traversal which just consults layer.hasVisibleContent(). So make descendantLayerPaintsIntoAncestor() a generic traversal function which walks descendant layers which may paint into the target layer. The "Visible" in the name reflects the fact that it can bypass a subtree for a layer with !hasVisibleDescendant(). * rendering/RenderLayerBacking.cpp: (WebCore::RenderLayerBacking::updateAfterDescendants): (WebCore::traverseVisibleNonCompositedDescendantLayers): (WebCore::RenderLayerBacking::isPaintDestinationForDescendantLayers): (WebCore::RenderLayerBacking::hasVisibleNonCompositedDescendants): (WebCore::descendantLayerPaintsIntoAncestor): Deleted. * rendering/RenderLayerBacking.h: Modified Paths trunk/Source/WebCore/ChangeLog trunk/Source/WebCore/rendering/RenderLayerBacking.cpp trunk/Source/WebCore/rendering/RenderLayerBacking.h Diff Modified: trunk/Source/WebCore/ChangeLog (213434 => 213435) --- trunk/Source/WebCore/ChangeLog 2017-03-05 11:10:10 UTC (rev 213434) +++ trunk/Source/WebCore/ChangeLog 2017-03-05 19:14:12 UTC (rev 213435) @@ -1,3 +1,26 @@ +2017-03-05 Simon Fraser+ +Make some RenderLayer tree traversal in RenderLayerBacking more generic +https://bugs.webkit.org/show_bug.cgi?id=169177 + +Reviewed by Zalan Bujtas. + +The real goal of this patch is reduce the number of callers of +RenderLayerBacking::isPaintDestinationForDescendantLayers() to one. To achieve that, +have the setContentsVisible() logic (which is really just about the CSS 'visibility' property) +do its own tree traversal which just consults layer.hasVisibleContent(). So +make descendantLayerPaintsIntoAncestor() a generic traversal function which walks +descendant layers which may paint into the target layer. The "Visible" in the name +reflects the fact that it can bypass a subtree for a layer with !hasVisibleDescendant(). + +* rendering/RenderLayerBacking.cpp: +(WebCore::RenderLayerBacking::updateAfterDescendants): +(WebCore::traverseVisibleNonCompositedDescendantLayers): +(WebCore::RenderLayerBacking::isPaintDestinationForDescendantLayers): +(WebCore::RenderLayerBacking::hasVisibleNonCompositedDescendants): +(WebCore::descendantLayerPaintsIntoAncestor): Deleted. +* rendering/RenderLayerBacking.h: + 2017-03-04 Simon Fraser Clarify some terminology in RenderLayerBacking Modified: trunk/Source/WebCore/rendering/RenderLayerBacking.cpp (213434 => 213435) --- trunk/Source/WebCore/rendering/RenderLayerBacking.cpp 2017-03-05 11:10:10 UTC (rev 213434) +++ trunk/Source/WebCore/rendering/RenderLayerBacking.cpp 2017-03-05 19:14:12 UTC (rev 213435) @@ -1104,7 +1104,7 @@ updateDrawsContent(isSimpleContainer); -m_graphicsLayer->setContentsVisible(m_owningLayer.hasVisibleContent() || isPaintDestinationForDescendantLayers()); +m_graphicsLayer->setContentsVisible(m_owningLayer.hasVisibleContent() || hasVisibleNonCompositedDescendants()); if (m_scrollingLayer) { m_scrollingLayer->setContentsVisible(renderer().style().visibility() == VISIBLE); m_scrollingLayer->setUserInteractionEnabled(renderer().style().pointerEvents() != PE_NONE); @@ -1948,7 +1948,9 @@ return true; } -static bool descendantLayerPaintsIntoAncestor(RenderLayer& parent) +enum class LayerTraversal { Continue, Stop }; + +static LayerTraversal traverseVisibleNonCompositedDescendantLayers(RenderLayer& parent, std::function layerFunc) { // FIXME: We shouldn't be called with a stale z-order lists. See bug 85512. parent.updateLayerListsIfNeeded(); @@ -1959,40 +1961,75 @@ if (auto* normalFlowList = parent.normalFlowList()) { for (auto* childLayer : *normalFlowList) { -if (!compositedWithOwnBackingStore(*childLayer) && (childLayer->isVisuallyNonEmpty() || descendantLayerPaintsIntoAncestor(*childLayer))) -return true; +if (compositedWithOwnBackingStore(*childLayer)) +continue; + +if (layerFunc(*childLayer) == LayerTraversal::Stop) +return LayerTraversal::Stop; + +if (traverseVisibleNonCompositedDescendantLayers(*childLayer, layerFunc) == LayerTraversal::Stop) +return LayerTraversal::Stop; } } if
[webkit-changes] [213434] trunk/LayoutTests
Title: [213434] trunk/LayoutTests Revision 213434 Author commit-qu...@webkit.org Date 2017-03-05 03:10:10 -0800 (Sun, 05 Mar 2017) Log Message [GTK] Two file reset tests are failing in the bots since they were added in r213042 https://bugs.webkit.org/show_bug.cgi?id=169126 Patch by Carlos Garcia Camposon 2017-03-05 Reviewed by Michael Catanzaro. They are failing because the GTK+ event sender is not firing the second UIHelper.activateAt() after the change event. In one case this causes that the reset button is not clicked, and in the other one the file input is not unfocused. This is because the change event is emitted before the mouse up, and the GTK+ event sender ignores mouse down events when the button is already down to not confuse Xvfb. So, we can easily fix theses tests by using a timeout to ensure the next UIHelper.activateAt() happens in a different run loop iteration, after the mouse up. * fast/forms/file/file-input-reset-using-open-panel.html: * fast/forms/file/file-reset-in-change-using-open-panel.html: Modified Paths trunk/LayoutTests/ChangeLog trunk/LayoutTests/fast/forms/file/file-input-reset-using-open-panel.html trunk/LayoutTests/fast/forms/file/file-reset-in-change-using-open-panel.html Diff Modified: trunk/LayoutTests/ChangeLog (213433 => 213434) --- trunk/LayoutTests/ChangeLog 2017-03-05 08:37:34 UTC (rev 213433) +++ trunk/LayoutTests/ChangeLog 2017-03-05 11:10:10 UTC (rev 213434) @@ -1,3 +1,20 @@ +2017-03-05 Carlos Garcia Campos + +[GTK] Two file reset tests are failing in the bots since they were added in r213042 +https://bugs.webkit.org/show_bug.cgi?id=169126 + +Reviewed by Michael Catanzaro. + +They are failing because the GTK+ event sender is not firing the second UIHelper.activateAt() after the change +event. In one case this causes that the reset button is not clicked, and in the other one the file input is not +unfocused. This is because the change event is emitted before the mouse up, and the GTK+ event sender ignores +mouse down events when the button is already down to not confuse Xvfb. So, we can easily fix theses tests by +using a timeout to ensure the next UIHelper.activateAt() happens in a different run loop iteration, after the +mouse up. + +* fast/forms/file/file-input-reset-using-open-panel.html: +* fast/forms/file/file-reset-in-change-using-open-panel.html: + 2017-03-04 Commit Queue Unreviewed, rolling out r213379. Modified: trunk/LayoutTests/fast/forms/file/file-input-reset-using-open-panel.html (213433 => 213434) --- trunk/LayoutTests/fast/forms/file/file-input-reset-using-open-panel.html 2017-03-05 08:37:34 UTC (rev 213433) +++ trunk/LayoutTests/fast/forms/file/file-input-reset-using-open-panel.html 2017-03-05 11:10:10 UTC (rev 213434) @@ -23,12 +23,15 @@ file.addEventListener("change", function() { document.getElementById("console").innerHTML = file.value + " is selected."; -UIHelper.activateAt(reset.offsetLeft + reset.offsetWidth / 2, reset.offsetTop + reset.offsetHeight / 2).then(function() { -// Move a cursor out of the reset button. -UIHelper.activateAt(reset.offsetLeft + reset.offsetWidth + 10, reset.offsetTop + reset.offsetHeight + 10).then(function() { -testRunner.notifyDone(); +// The change event can be fired before the mouse up, so use a timeout here to ensure the next mouse down happens after the pending mouse up. +window.setTimeout(function() { +UIHelper.activateAt(reset.offsetLeft + reset.offsetWidth / 2, reset.offsetTop + reset.offsetHeight / 2).then(function() { +// Move a cursor out of the reset button. +UIHelper.activateAt(reset.offsetLeft + reset.offsetWidth + 10, reset.offsetTop + reset.offsetHeight + 10).then(function() { +testRunner.notifyDone(); +}); }); -}); +}, 0); }); openFilesInElement(file, ["foo.txt"]); Modified: trunk/LayoutTests/fast/forms/file/file-reset-in-change-using-open-panel.html (213433 => 213434) --- trunk/LayoutTests/fast/forms/file/file-reset-in-change-using-open-panel.html 2017-03-05 08:37:34 UTC (rev 213433) +++ trunk/LayoutTests/fast/forms/file/file-reset-in-change-using-open-panel.html 2017-03-05 11:10:10 UTC (rev 213434) @@ -15,9 +15,12 @@ function handleChange() { file1.value = ''; -UIHelper.activateAt(1, 1).then(function() { -testRunner.notifyDone(); -}); +// The change event can be fired before the mouse up, so use a timeout here to ensure the next mouse down happens after the pending mouse up. +window.setTimeout(function() { +UIHelper.activateAt(1, 1).then(function() { +testRunner.notifyDone(); +}); +}, 0); } if (window.testRunner)
[webkit-changes] [213433] trunk/Source/JavaScriptCore
Title: [213433] trunk/Source/_javascript_Core Revision 213433 Author utatane@gmail.com Date 2017-03-05 00:37:34 -0800 (Sun, 05 Mar 2017) Log Message [JSCOnly][GTK] Suppress warnings on return type in B3 and WASM https://bugs.webkit.org/show_bug.cgi?id=168869 Reviewed by Keith Miller. * b3/B3Width.h: * wasm/WasmSections.h: Modified Paths trunk/Source/_javascript_Core/ChangeLog trunk/Source/_javascript_Core/b3/B3Width.h trunk/Source/_javascript_Core/wasm/WasmSections.h Diff Modified: trunk/Source/_javascript_Core/ChangeLog (213432 => 213433) --- trunk/Source/_javascript_Core/ChangeLog 2017-03-05 07:41:25 UTC (rev 213432) +++ trunk/Source/_javascript_Core/ChangeLog 2017-03-05 08:37:34 UTC (rev 213433) @@ -1,3 +1,13 @@ +2017-03-05 Yusuke Suzuki+ +[JSCOnly][GTK] Suppress warnings on return type in B3 and WASM +https://bugs.webkit.org/show_bug.cgi?id=168869 + +Reviewed by Keith Miller. + +* b3/B3Width.h: +* wasm/WasmSections.h: + 2017-03-04 Csaba Osztrogonác [ARM] Unreviewed buildfix after r213376. Modified: trunk/Source/_javascript_Core/b3/B3Width.h (213432 => 213433) --- trunk/Source/_javascript_Core/b3/B3Width.h 2017-03-05 07:41:25 UTC (rev 213432) +++ trunk/Source/_javascript_Core/b3/B3Width.h 2017-03-05 08:37:34 UTC (rev 213433) @@ -30,6 +30,11 @@ #include "B3Bank.h" #include "B3Type.h" +#if COMPILER(GCC) && ASSERT_DISABLED +#pragma GCC diagnostic push +#pragma GCC diagnostic ignored "-Wreturn-type" +#endif // COMPILER(GCC) && ASSERT_DISABLED + namespace JSC { namespace B3 { enum Width : int8_t { @@ -103,5 +108,9 @@ } // namespace WTF +#if COMPILER(GCC) && ASSERT_DISABLED +#pragma GCC diagnostic pop +#endif // COMPILER(GCC) && ASSERT_DISABLED + #endif // ENABLE(B3_JIT) Modified: trunk/Source/_javascript_Core/wasm/WasmSections.h (213432 => 213433) --- trunk/Source/_javascript_Core/wasm/WasmSections.h 2017-03-05 07:41:25 UTC (rev 213432) +++ trunk/Source/_javascript_Core/wasm/WasmSections.h 2017-03-05 08:37:34 UTC (rev 213433) @@ -27,6 +27,11 @@ #if ENABLE(WEBASSEMBLY) +#if COMPILER(GCC) && ASSERT_DISABLED +#pragma GCC diagnostic push +#pragma GCC diagnostic ignored "-Wreturn-type" +#endif // COMPILER(GCC) && ASSERT_DISABLED + namespace JSC { namespace Wasm { #define FOR_EACH_WASM_SECTION(macro) \ @@ -81,4 +86,8 @@ } } // namespace JSC::Wasm +#if COMPILER(GCC) && ASSERT_DISABLED +#pragma GCC diagnostic pop +#endif // COMPILER(GCC) && ASSERT_DISABLED + #endif // ENABLE(WEBASSEMBLY) ___ webkit-changes mailing list webkit-changes@lists.webkit.org https://lists.webkit.org/mailman/listinfo/webkit-changes