summaryrefslogtreecommitdiffstats
path: root/third_party/WebKit/LayoutTests/fast/canvas
Commit message (Collapse)AuthorAgeFilesLines
* Use ImageDecoder for createImageBitmap(HTMLImageElement, premultiplyAlpha=false)xidachen2016-03-2419-4061/+337
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This is specifically for the case of premultiplyAlpha=false. Following is an example. Given an image with only one pixel with value of (255, 0, 0, 26), and we load the this image either as a Blob or a HTMLImageElement and call createImageBitmap with premultiplyAlpha=false. the ImageBitmap::copyBitmapData will return a pixel value of (197, 0, 0, 26) which is totally wrong. The reason it fails is that the SkImage that is extracted from ImageDecoder is always in kPremul alphaType even if "AlphaNotPremultiplied" is given to the ImageDecoder. With this CL: https://codereview.chromium.org/1820733004/, the extracted SkImage now has a kUnpremul alphaType. ImageBitmap should adapt to this change as well. So this CL make correction for the case of createImageBitmap from Blob and HTMLImageElement so that copyBitmapData will succeed with the SkImage extracted from ImageDecoder. We have also updated corresponding layout tests. In particular, we change an existing image's alpha channel from 128 to 26. In this case, if the ImageBitmap constructor handles premultiplyAlpha=false by first premultiply and then un-premultiply, we will see a large loss comparing with input. Here we can see that the color is very close (tolerance of 2) to its input image. Additionally, we added several new layout tests to test the cases of createImageBitmap(ImageBitmap) where the parameter ImageBitmap is created from various kinds of sources. 2 layout tests are deleted under fast/canvas/. The reason is that these layout tests uses 2d canvas to display an ImageBitmap, and it cannot distinguish an ImageBitmap created with premultiplyAlpha= false with an ImageBitmap created with true. Now that we have a large set of layout tests under fast/canvas/webgl, these 2 tests are redundant. Review URL: https://codereview.chromium.org/1809603004 Cr-Commit-Position: refs/heads/master@{#383004}
* Implemented/Tested line-drawing functions in OffscreenCanvas 2D Contextxlai2016-03-222-0/+85
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The new Web APIs that are implemented are: * OffscreenCanvasRendering2D.beginPath() * OffscreenCanvasRendering2D.fill(optional CanvasFillRule) * OffscreenCanvasRendering2D.fill(Path2D, optional CanvasFillRule) * OffscreenCanvasRendering2D.stroke() * OffscreenCanvasRendering2D.stroke(Path2D) * OffscreenCanvasRendering2D.lineWidth * OffscreenCanvasRendering2D.lineCap * OffscreenCanvasRendering2D.lineJoin * OffscreenCanvasRendering2D.miterLimit * OffscreenCanvasRendering2D.setLineDash(array); * OffscreenCanvasRendering2D.getLineDash * OffscreenCanvasRendering2D.lineDashOffset A subset of existing Web APIs shared with non-Offscreen canvas from CanvasPathMethods are also tested, including: arc, lineTo, bezierCurveTo, moveTo, closePath. BUG=563856 Review URL: https://codereview.chromium.org/1806943005 Cr-Commit-Position: refs/heads/master@{#382584}
* Implement canvas tainting for canvas filtersajuma2016-03-181-0/+119
| | | | | | | | | | | | | | | | | | | | This implements a whitelist for canvas filters. Any non-whitelisted filter taints the canvas. To start with, CSS shorthand filters are whitelisted, along with SVG filters that consist only of the following effects: 1) SourceGraphic 2) SourceAlpha 3) FillPaint 4) StrokePaint 5) feColorMatrix 6) feMerge BUG=571048 Review URL: https://codereview.chromium.org/1783013003 Cr-Commit-Position: refs/heads/master@{#381961}
* Add ColorspaceConversion in ImageBitmapOptionsxidachen2016-03-158-41/+89
| | | | | | | | | | | | This option is specifically for createImageBitmap from a Blob or a HTMLImageElement. This CL adds support for this option, starting with Blob. Layout tests are added to ensure the correctness. BUG=592005 Review URL: https://codereview.chromium.org/1763043002 Cr-Commit-Position: refs/heads/master@{#381224}
* Make OffscreenCanvasRenderingContext2D renderable on a workerjunov2016-03-1410-0/+219
| | | | | | | | | | | | | | | | | | | This change makes 2D contexts renderable on workers. It only implements the bare minimum interfaces required to draw rectangles on a worker and bring the results to screen via ImageBitmap transfer. The new Web APIs that are implemented are: * OffscreenCanvasRendering2D.fillStyle * OffscreenCanvasRendering2D.strokeStyle * OffscreenCanvasRendering2D.fillRect * OffscreenCanvasRendering2D.strokeRect * OffscreenCanvasRendering2D.clearRect * OffscreenCanvas.transferToImageBitmap BUG=563856, 563832 Review URL: https://codereview.chromium.org/1775153002 Cr-Commit-Position: refs/heads/master@{#381085}
* Canvas2d: Implement rerouting event by hit region's control.jinho.bang2016-03-121-0/+267
| | | | | | | | | | | | | | | According to the spec[1], should support rerouting event if there is a control on hit region. This CL is also includes fixing two spec issues[2][3] and test. [1] https://html.spec.whatwg.org/multipage/scripting.html#canvas-mouseevent-rerouting-steps [2] https://github.com/whatwg/html/issues/547 [3] https://github.com/whatwg/html/issues/548 BUG=579076,579614,582792 Review URL: https://codereview.chromium.org/1654653002 Cr-Commit-Position: refs/heads/master@{#380870}
* Remove non-used resource files under Layouttests/fast/canvas/webgl/xidachen2016-03-0848-3770/+0
| | | | | | | | | | | | | A few weeks ago, there is an awesome CL here: https://codereview.chromium.org/1601093008 that removed duplicate WebGL layout tests. However, it appears that the resources that are associated with them were left over. This CL cleans up the non-used resource files. Review URL: https://codereview.chromium.org/1777483004 Cr-Commit-Position: refs/heads/master@{#379939}
* Add layout tests for tex(Sub)Image2D for ImageBitmapxidachen2016-03-0817-0/+4254
| | | | | | | | | | | | | | | | | | | | | At this moment, there is an extensive coverage of this testing hosted on KhronosGroup/WebGL. However, the problem is that we cannot have coverage of the case for ImageBitmap constructor to take a "default" value for some of the constructor options such as premultiplyAlpha. The reason is that the option "default" is implementation-specific, and the results could vary for different browsers. In here, we add layout tests in chromium code base to have more test cases than the ones hosted on KhronosGroup/WebGL. Also, the KhronosGroup/WebGL testing suite tests tex(Sub)Image2D for 38 different kinds of pixel formats such as RGBA + UNSIGNED_SHORT_4444. Here we test the basic one which is RGBA + UNSIGNED_BYTE, we can always extend that in the future. Review URL: https://codereview.chromium.org/1767113003 Cr-Commit-Position: refs/heads/master@{#379879}
* Add rendering context and rendering 2D context to OffscreenCanvasxlai2016-03-086-27/+114
| | | | | | | | | | | | This patch implements the getContext() function in OffscreenCanvas. It also implements the skeleton of OffscreenCanvasRenderingContext2D, an object type returned by getContext("2d"). BUG=563826 Review URL: https://codereview.chromium.org/1748163003 Cr-Commit-Position: refs/heads/master@{#379740}
* Add "premultiply" in ImageBitmapOptions enum PremultiplyAlphaxidachen2016-03-044-12/+1054
| | | | | | | | | | | | | | | | | At this moment, there are two options of "none" and "default", this CL adds "premultiply". The code path for "premultiply" and "default" is the same so we can avoid large changes of any existing C++ code. Layout tests has been created to reflect this additional option. Locally I have added this option to all the conformance tests for ImageBitmap hosted by KhronosGroup/WebGL, and they all pass. Once this CL landed, there will be a pull request to update those tests. BUG=591713 Review URL: https://codereview.chromium.org/1756963003 Cr-Commit-Position: refs/heads/master@{#379302}
* Perform null check in ImageBitmapxidachen2016-02-294-3/+22
| | | | | | | | | | | | | | | | The ImageBitmap class lacks null check. In particular, the cropImage() method creates a SkSurface and this CL does null check for that. Also, in several other methods in this class, it calls m_image->setOriginClean() and this CL does null check for m_image as well. Moreover, in the case where m_image is null, the promise will be rejected. After this null check, the crash reported in the bug goes away. BUG=588980 Review URL: https://codereview.chromium.org/1740633003 Cr-Commit-Position: refs/heads/master@{#378228}
* Implement ImageBitmap options premultiplyAlpha for ImageDataxidachen2016-02-272-102/+2
| | | | | | | | | | | | | | | | | | | With premultiplyAlpha=true, the implementation of this function creates a premultiplied SkSurface and grab a snapshot from it. To prevent any data loss, we should try to create a SkImage directly from the raw unpremultiplied data. There are several tricky cases here. First, we need to take care of the case when the crop region is not the same as the original ImageData size, where we need to malloc a temporary buffer and copy the data over. Also, we need to take care of the case when flipY is set to be true, because we need to only flip the area that is inside the crop region. BUG=589131 Review URL: https://codereview.chromium.org/1680173003 Cr-Commit-Position: refs/heads/master@{#378080}
* Rename OffscreenCanvasTemp to OffscreenCanvasxidachen2016-02-259-38/+38
| | | | | | | | | | | In a previous CL here: https://codereview.chromium.org/1727813002/, we re-named OffScreenCanvas to OffscreenCanvasTemp. This CL changes it to OffscreenCanvas which is complied with the proposal here: https://wiki.whatwg.org/wiki/OffscreenCanvas Review URL: https://codereview.chromium.org/1733763003 Cr-Commit-Position: refs/heads/master@{#377572}
* Rename OffScreenCanvas to OffscreenCanvasTemp (eventually OffscreenCanvas)xidachen2016-02-249-38/+38
| | | | | | | | | | | | | When we start it, this was named OffScreenCanvas. The whatwg proposal here: https://wiki.whatwg.org/wiki/OffscreenCanvas actually names it OffscreenCanvas. We cannot directly change OffScreenCanvas to OffscreenCanvas because windows and Mac are case-insensitive. So in this CL we change it to a temp name and in a follow up CL this will be changed to OffscreenCanvas. Review URL: https://codereview.chromium.org/1727813002 Cr-Commit-Position: refs/heads/master@{#377340}
* Update canvas filter when canvas fill or stroke style changesajuma2016-02-184-0/+68
| | | | | | | | | | When the canvas fill or stroke style changes, filters need to be rebuilt so that their FillPaint and StrokePaint values match the updated fill and stroke. Review URL: https://codereview.chromium.org/1708603002 Cr-Commit-Position: refs/heads/master@{#376169}
* Canvas2d: The control in hit region should be a canvas fallback element.jinho.bang2016-02-131-0/+91
| | | | | | | | | | | | | According to the hit region spec[1], the control parameter should be a supported interactive canvas fallback element. [1] https://html.spec.whatwg.org/multipage/scripting.html#supported-interactive-canvas-fallback-element BUG=328961 Review URL: https://codereview.chromium.org/1692183002 Cr-Commit-Position: refs/heads/master@{#375312}
* Reland of: Revise the code for premultiplyAlpha=false for ↵xidachen2016-02-122-803/+6
| | | | | | | | | | | | | | | | createImageBitmap(HTMLCanvasElement) It turns out that HTMLCanvasElement is always premultiplied in Blink, so we cannot simply use a decoder to re-decode it to a un-premultiplied format. The strategy this CL takes is that after cropping the HTMLCanvasElement into a SkImage, we convert the SkImage into a un-premultiplied format if that's what the option specifies. PS#1 causes compilation error, PS#4 should work. Review URL: https://codereview.chromium.org/1694793002 Cr-Commit-Position: refs/heads/master@{#375212}
* This change adds the ImageBitmapRenderingContext componentjunov2016-02-126-0/+87
| | | | | | | | | | | | | | of the OffscreenCanvas[1] feature proposal. It is currently hidden behind the 'experimental canvas features' runtime flag. [1] https://wiki.whatwg.org/wiki/OffscreenCanvas BUG=563861 Review URL: https://codereview.chromium.org/1598923002 Cr-Commit-Position: refs/heads/master@{#375198}
* Revert of Revise the code for premultiplyAlpha=false for ↵xidachen2016-02-122-6/+803
| | | | | | | | | | | | | | | | | | | | | | | | | | | createImageBitmap(HTMLCanvasElement) (patchset #1 id:1 of https://codereview.chromium.org/1694793002/ ) Reason for revert: Cause compilation error https://build.chromium.org/p/chromium.mac/builders/Mac%20Builder/builds/37356/steps/compile/logs/stdio Original issue's description: > Revise the code for premultiplyAlpha=false for createImageBitmap(HTMLCanvasElement) > > It turns out that HTMLCanvasElement is always premultiplied in Blink, so > we cannot simply use a decoder to re-decode it to a un-premultiplied format. > The strategy this CL takes is that after cropping the HTMLCanvasElement > into a SkImage, we convert the SkImage into a un-premultiplied format if > that's what the option specifies. TBR=junov@chromium.org # Skipping CQ checks because original CL landed less than 1 days ago. NOPRESUBMIT=true NOTREECHECKS=true NOTRY=true Review URL: https://codereview.chromium.org/1693863002 Cr-Commit-Position: refs/heads/master@{#375193}
* Revise the code for premultiplyAlpha=false for ↵xidachen2016-02-122-803/+6
| | | | | | | | | | | | | | createImageBitmap(HTMLCanvasElement) It turns out that HTMLCanvasElement is always premultiplied in Blink, so we cannot simply use a decoder to re-decode it to a un-premultiplied format. The strategy this CL takes is that after cropping the HTMLCanvasElement into a SkImage, we convert the SkImage into a un-premultiplied format if that's what the option specifies. Review URL: https://codereview.chromium.org/1694793002 Cr-Commit-Position: refs/heads/master@{#375191}
* Auto-rebaseline for r374703Walter Korman2016-02-101-2/+2
| | | | | | | | | | | https://chromium.googlesource.com/chromium/src/+/0c7e2f7bc BUG=581355 TBR=fmalita@chromium.org Review URL: https://codereview.chromium.org/1687093005 . Cr-Commit-Position: refs/heads/master@{#374775}
* Remove SK_SUPPORT_LEGACY_HQ_DOWNSAMPLINGfmalita2016-02-101-2/+4
| | | | | | | | | | | | | | | | | | | | | | | Switches Skia's software HQ downscaling algorithm from Mitchell to mipmaps. This is a good idea because 1) it aligns the results with Ganesh (which already uses mipmaps for downscaling) 2) mipmaps are cheaper/more efficient (to produce and to cache) 3) yields (arguably) sharper results, particularly when scaling to half size - which appears to be a common technique used for Retina-ready pages. BUG=583478,562162 R=reed@google.com,senorblanco@chromium.org Review URL: https://codereview.chromium.org/1511113004 Cr-Commit-Position: refs/heads/master@{#374703}
* Implement ImageBitmap options premultiplyAlpha for HTMLVideoElementxidachen2016-02-092-43/+125
| | | | | | | | | | | | | | | | In this CL, we implement the premultiplyAlpha option for ImageBitmap in the case of HTMLVideoElement. This CL modifies the existing layout test to test all the cases of cobmining flipY and premultiplyAlpha. Several cases are commentted out in the layout test because skia doesn't yet support draw an umpremul SkImage. BUG=249382 Review URL: https://codereview.chromium.org/1677383002 Cr-Commit-Position: refs/heads/master@{#374511}
* Update canvas filter when svg filter in DOM changesajuma2016-02-096-0/+118
| | | | | | | | | | | | | | When a canvas filter includes a reference to an svg filter, the canvas filter needs to be rebuilt whenever the svg filter changes. This CL makes the canvas context get notified of such changes, so that the filter gets rebuilt before its next use. BUG=581135 Review URL: https://codereview.chromium.org/1640203003 Cr-Commit-Position: refs/heads/master@{#374507}
* Implement ImageBitmap options premultiplyAlphaxidachen2016-02-082-582/+333
| | | | | | | | | | | | | | | | In this CL, we implement the option of premultiplyAlpha for ImageBitmap. At this moment, it works for createImageBitmap from HTMLImageElement, HTMLCanvasElement and Blob. The cases of from ImageBitmap, HTMLVideoElemtn and ImageData will be solved in future CL(s). This CL also includes modifies a current layout tests, which tests the combination of ImageBitmap options "flipY" and premultiplyAlpha BUG=249382 Review URL: https://codereview.chromium.org/1609763002 Cr-Commit-Position: refs/heads/master@{#374132}
* Reland Simplify CanvasAsyncBlobCreator by removing ContextObserver and testsxlai2016-02-0414-355/+0
| | | | | | | | | | | | | | | | | | | | | | | After we confirm to use idle-tasks implementation in toBlob, the cancellation mechanism for async thread implementation is no longer useful. First, compositor thread is definitely alive when users open Chrome browser; so the alternative code path for PNG image encoding when compositor thread is disabled would be unreachable in real-world situations; as a result, PNG image encoding happens in idle tasks only. Second, JPEG and WEBP image formats do not have progressive encoding at this moment and thus cancellation mechanism is not useful to them either; once their progressive encoding implementations are completed, they would be moved over to idle-tasks implementation too. This patch removes ContextObserver and CanvasAsyncBlobCreatorTest; it also moves all layout tests that use toBlob on PNG image formats to virtual/threaded as idle tasks require compositor thread to be alive. There is no need to perform tests on toBlob (png) in scenarios when compositor thread is not available. BUG=581574 Review URL: https://codereview.chromium.org/1669713002 Cr-Commit-Position: refs/heads/master@{#373567}
* Implementing option imageOrientation for createImageBitmap(ImageData)xidachen2016-02-034-0/+4060
| | | | | | | | | | | | | | | | | | | | | | | This CL implements the ImageBitmap option imageOrientation for the method createImageBitmap(ImageData) only. For createImageBitmap from other sources, there will be separate CLs to handle them. The whatwg wiki for the ImageBitmap options is here: https://wiki.whatwg.org/wiki/ImageBitmap_Options In the case of ImageData, the option none/topLeft should behave the same, which is "Do not change orientation". This is because ImageData does not contain any image media meta data. Similarly, For the option of flipY/topLeft, the behavior should be flip the image vertically. This CL also includes a layout test, which re-uses the existing test with the additional parameter of "flipY", and currently the layout test is for the ImageData as source only. BUG=249382 Review URL: https://codereview.chromium.org/1631733003 Cr-Commit-Position: refs/heads/master@{#373274}
* Revert of Simplify CanvasAsyncBlobCreator by removing ContextObserver and ↵schenney2016-02-0312-0/+323
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | related tests (patchset #3 id:40001 of https://codereview.chromium.org/1653253003/ ) Reason for revert: This patch appears to be causing Document leaks. See, for example: https://build.chromium.org/p/chromium.webkit/builders/WebKit%20Linux%20Leak/builds/17250 Original issue's description: > Simplify CanvasAsyncBlobCreator by removing ContextObserver and related tests > > After we confirm to use idle-tasks implementation in toBlob, the cancellation > mechanism for async thread implementation is no longer useful. First, compositor > thread is definitely alive when users open Chrome browser; so the alternative > code path for PNG image encoding when compositor thread is disabled would be > unreachable in real-world situations; as a result, PNG image encoding happens > in idle tasks only. Second, JPEG and WEBP image formats do not have progressive > encoding at this moment and thus cancellation mechanism is not useful to them > either; once their progressive encoding implementations are completed, they > would be moved over to idle-tasks implementation too. > > This patch removes ContextObserver and CanvasAsyncBlobCreatorTest; it also moves > all layout tests that use toBlob on PNG image formats to virtual/threaded as > idle tasks require compositor thread to be alive. There is no need to perform > tests on toBlob (png) in scenarios when compositor thread is not available. > > BUG=581574 > > Committed: https://crrev.com/d3991457ab5fcb905f2a8a07657e26b16b2974fa > Cr-Commit-Position: refs/heads/master@{#373068} TBR=junov@chromium.org,xlai@chromium.org # Skipping CQ checks because original CL landed less than 1 days ago. NOPRESUBMIT=true NOTREECHECKS=true NOTRY=true BUG=581574 Review URL: https://codereview.chromium.org/1660993002 Cr-Commit-Position: refs/heads/master@{#373250}
* Simplify CanvasAsyncBlobCreator by removing ContextObserver and related testsxlai2016-02-0212-323/+0
| | | | | | | | | | | | | | | | | | | | | | | After we confirm to use idle-tasks implementation in toBlob, the cancellation mechanism for async thread implementation is no longer useful. First, compositor thread is definitely alive when users open Chrome browser; so the alternative code path for PNG image encoding when compositor thread is disabled would be unreachable in real-world situations; as a result, PNG image encoding happens in idle tasks only. Second, JPEG and WEBP image formats do not have progressive encoding at this moment and thus cancellation mechanism is not useful to them either; once their progressive encoding implementations are completed, they would be moved over to idle-tasks implementation too. This patch removes ContextObserver and CanvasAsyncBlobCreatorTest; it also moves all layout tests that use toBlob on PNG image formats to virtual/threaded as idle tasks require compositor thread to be alive. There is no need to perform tests on toBlob (png) in scenarios when compositor thread is not available. BUG=581574 Review URL: https://codereview.chromium.org/1653253003 Cr-Commit-Position: refs/heads/master@{#373068}
* Add a close() method to ImageBitmapxidachen2016-02-022-0/+103
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This is not in the ImageBitmap spec yet, but it is part of the OffScreenCanvas proposal. In the proposal, the words about this method are: // It's crucial that there be a way to explicitly dispose of ImageBitmaps // since they refer to potentially large graphics resources. So this method is very similar to ImageData::dispose(), which explicitly dispose any resources that ImageBitmap is associated with. At this moement, ImageBitmap holds a RefPtr to a StaticBitmapImage, so the close() method clears the RefPtr. Since this method is not in the ImageBitmap spec yet, we should hide it behind a RuntimeEnabled flag. A layout test is also included in this CL. In particular, the width and height of the ImageBitmap should be both 0 after calling the close() method. The layout test also includes cases that calls close() to a already closed ImageBitmap, or calls close() to a neutered ImageBitmap. In both cases, the close() operation should be noop. Now that we have three operations for ImageBitmap: structured clone (SK), transfering (T) and close (C), we should test the behavior of combining them. In my previous CLs: https://codereview.chromium.org/1647083003/, https://codereview.chromium.org/1652553003/, the cases of calling SK--T and T--SK have been tested. In this CL, the layout test contains cases for calling SK--C, T--C, C--SK and C--T. BUG=580649 Review URL: https://codereview.chromium.org/1631763004 Cr-Commit-Position: refs/heads/master@{#373050}
* Add a new layout test case to canvas-ImageBitmap-structured-clone.htmlxidachen2016-02-023-110/+97
| | | | | | | | | | | | | | | We should test the behavior of combing the action of structured clone and transfer. This CL adds a new layout test case, which applies transferring an ImageBitmap to worker after structured clone is applied to the same ImageBitmap. The expected behavior is that transferring would succeed because structured cloning does not neuter the ImageBitmap. A separate CL will be filed to test the case of applying structured clone to a already neutered ImageBitmap (by transferring). Review URL: https://codereview.chromium.org/1647083003 Cr-Commit-Position: refs/heads/master@{#372870}
* Add a new layout test case to canvas-ImageBitmap-transferable.htmlxidachen2016-02-012-7/+15
| | | | | | | | | | | In this CL, we add a new layout test case that applies structured clone to a neutered ImageBitmap after it was transfered to a worker. The expected behavior is to throw an exception to user notifying that the ImageBitmap has been neutered. Review URL: https://codereview.chromium.org/1652553003 Cr-Commit-Position: refs/heads/master@{#372676}
* Canvas2d: addHitRegion() should take clipping region into account.jinho.bang2016-01-271-0/+126
| | | | | | | | | | | | Current implementation ignores the clipping region when calling addHitRegion(). In the past, we had no way to get the current clip path from canvas state stack in Blink side but it's been possible since display list 2d canvas feature. BUG=387057 Review URL: https://codereview.chromium.org/1630683002 Cr-Commit-Position: refs/heads/master@{#371832}
* Change to idle-task implementations in toBlobxlai2016-01-274-3/+3
| | | | | | | | | | | | | | | | | | | | This is a combined revert of https://codereview.chromium.org/1491653002 and https://codereview.chromium.org/1614173002. One more additional edit is the addition of "testRunner.dumpAsText();" in canvas-toBlob-defaultpng.html and canvas-toBlob-toDataURL-race.js. Reason for this revert is that we found out idle-task performs much better (has less negative impacts on other activities) than threaded implementation on majority of machines. BUG=581574 Review URL: https://codereview.chromium.org/1637163002 Cr-Commit-Position: refs/heads/master@{#371806}
* Make ImageBitmap support structured clonexidachen2016-01-233-0/+182
| | | | | | | | | | | | | | | ImageBitmap is already transferable, but at this moment we cannot use postMessage(ImageBitmap) via structured clone. This CL will enable it. This CL also includes a layout test and sends an ImageBitmap to a worker, the worker sends the ImageBitmap back to main, and does check that the underline pixel data survives the round trip. BUG=579539 Review URL: https://codereview.chromium.org/1616213003 Cr-Commit-Position: refs/heads/master@{#371155}
* Fix assertion failure in CanvasRenderingContext2D::compositedDrawajuma2016-01-224-0/+46
| | | | | | | | | | When a canvas has a filter but that filter is undefined (either because it refers to a non-existent filter, or because that filter is defined later), compositedDraw shouldn't be called. Review URL: https://codereview.chromium.org/1606363002 Cr-Commit-Position: refs/heads/master@{#370994}
* Removal all idle-task implementations in toBlobxlai2016-01-225-3/+5
| | | | | | | | BUG=564218 Review URL: https://codereview.chromium.org/1614173002 Cr-Commit-Position: refs/heads/master@{#370908}
* Use BlobCallback in toBlob instead of FileCallbackxlai2016-01-222-0/+31
| | | | | | | | | | Recently there's a spec change: https://github.com/whatwg/html/pull/391. BUG=568603 Review URL: https://codereview.chromium.org/1609313002 Cr-Commit-Position: refs/heads/master@{#370846}
* Use correct size for unattached canvases when building canvas filtersajuma2016-01-212-0/+36
| | | | | | | | | | | | The equivalent bug for attached canvases was fixed in http://crrev.com/370047, but unattached canvases were ignored there. BUG=561045 Review URL: https://codereview.chromium.org/1613193002 Cr-Commit-Position: refs/heads/master@{#370719}
* Remove duplicated WebGL layout tests.zmo2016-01-20200-22450/+0
| | | | | | | | | | | | | | | They are already covered in WebGL conformance tests. Note that I only remove those who have exact corresponding tests in conformance suite. BUG= TEST=layout tests R=bajones@chromium.org,kbr@chromium.org Review URL: https://codereview.chromium.org/1601093008 Cr-Commit-Position: refs/heads/master@{#370472}
* This CL put [Exposed=(Window,Worker)] in the ImageBitmap.idlxidachen2016-01-192-2/+11
| | | | | | | | | | | | | | | | Along with that, some of the layout tests results has to be updated. Also, a layout test has been updated to show that the worker is able to get width&height for an ImageBitmap. The thread of Intent to ship ImageBitmap can be found here: https://groups.google.com/a/chromium.org/d/topic/blink-dev/jRzvehX9U1U/discussion BUG=563984 Review URL: https://codereview.chromium.org/1486663002 Cr-Commit-Position: refs/heads/master@{#370179}
* Modify createImageBitmap-blob-in-worker to check race conditionsxidachen2016-01-193-22/+71
| | | | | | | | | | | | | | | In this CL, we modify the behavior of the layout test for createImageBitmap from a blob in worker in this way: As soon as we postMessage the blob to worker, both the main thread and the worker thread start createImageBitmap(blob) which means that the decoding will happen at the same time. We call createImageBitmap(blob) in the worker thread for only once but in the main thread we call it five times. The expected result of this is that the created ImageBitmap from the main thread and from the worker thread should have the same pixel data as the original blob. Review URL: https://codereview.chromium.org/1574393009 Cr-Commit-Position: refs/heads/master@{#370124}
* Fix errors in fast/canvas/toBlob/canvas-toBlob-defaultpng.htmlnolan.robin.cao2016-01-191-2/+3
| | | | | | | | | | | | The dw and dh parameters passed to drawImage should be 300x150, which is the default size of canvas elment. Otherwise, the image will be scaled when drawing. This is why the test fails. BUG=577687 Review URL: https://codereview.chromium.org/1600603002 Cr-Commit-Position: refs/heads/master@{#370070}
* Add layout tests for canvas filters with specified widths and heightsajuma2016-01-186-0/+120
| | | | | | | | | | | This adds layout tests that use hidpi and CSS scaling. This is a follow-up to http://crrev.com/370047 BUG=561045 Review URL: https://codereview.chromium.org/1606673002 Cr-Commit-Position: refs/heads/master@{#370054}
* Use an up-to-date canvas size when building canvas filtersajuma2016-01-182-0/+24
| | | | | | | | | | | | | | | | | When canvas filters specify a filter primitive subregion (e.g. by specifying a width or height), filter building depends on having an up to date canvas size; otherwise, the crop rect computed for the filter is incorrect. This CL directly passes the canvas size to filter building code, rather than having the filter builder use a size from the layout tree, since the latter might be stale. BUG=561045 Review URL: https://codereview.chromium.org/1583103007 Cr-Commit-Position: refs/heads/master@{#370047}
* Add more layout test cases to createImageBitmapxidachen2016-01-144-0/+67
| | | | | | | | | | | | | | Comparing Firefox's layout tests against Chrome's, we find that there are two cases that is currently missing in our code base. 1). createImageBitmap from a neutered ImageBitmap. 2). createImageBitmap from a Blob object that is created by canvas.toBlob This CL adds these two test cases. Review URL: https://codereview.chromium.org/1589923002 Cr-Commit-Position: refs/heads/master@{#369443}
* Forward fill and stroke styles from 2d canvas to canvas filtersajuma2016-01-1212-0/+256
| | | | | | | | | | | This forwards the fill and stroke style from 2d canvas to the FillPaint and StrokePaint inputs of canvas filters. BUG=502877 Review URL: https://codereview.chromium.org/1543593002 Cr-Commit-Position: refs/heads/master@{#368726}
* Add a origin clean flag in ImageBitmap classxidachen2016-01-111-1/+0
| | | | | | | | | | | | | | | This CL adds a origin clean flag, such that in the case when the source of the ImageBitmap contains cross-origin content, we can simply set this flag instead of reject the promise. BUG=569779 Committed: https://crrev.com/10b9b4435e25fb8ede2122482426ae81c7980630 Cr-Commit-Position: refs/heads/master@{#368595} Review URL: https://codereview.chromium.org/1532473002 Cr-Commit-Position: refs/heads/master@{#368681}
* Canvas2d: Fix incorrect alignment of hit region when applied css/zoom.jinho.bang2016-01-117-159/+341
| | | | | | | | | | | | This change is fixing incorrect alignment of hit region on canvas when applied css transform/width/height/padding/border or device pinch-zoom or high DPI. Also, this CL adds/rewrites related tests. BUG=328961,497750,519603 Review URL: https://codereview.chromium.org/1553373002 Cr-Commit-Position: refs/heads/master@{#368607}
* Revert of Add a origin clean flag in ImageBitmap class (patchset #14 ↵treib2016-01-111-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | id:260001 of https://codereview.chromium.org/1532473002/ ) Reason for revert: Seems to have broken Oilpan builds: https://build.chromium.org/p/chromium.webkit/buildstatus?builder=WebKit%20Mac%20Oilpan&number=28566 https://build.chromium.org/p/chromium.webkit/buildstatus?builder=WebKit%20Linux%20Oilpan%20Builder&number=100 Original issue's description: > Add a origin clean flag in ImageBitmap class > > This CL adds a origin clean flag, such that in the case when the source > of the ImageBitmap contains cross-origin content, we can simply set this > flag instead of reject the promise. > > BUG=569779 > > Committed: https://crrev.com/10b9b4435e25fb8ede2122482426ae81c7980630 > Cr-Commit-Position: refs/heads/master@{#368595} TBR=junov@chromium.org,xidachen@chromium.org # Skipping CQ checks because original CL landed less than 1 days ago. NOPRESUBMIT=true NOTREECHECKS=true NOTRY=true BUG=569779 Review URL: https://codereview.chromium.org/1577783003 Cr-Commit-Position: refs/heads/master@{#368598}
generated by cgit v1.1 at 2025-03-23 21:30:34 +0000