- 20 Feb, 2018 40 commits
-
-
Koji Ishii authored
Similar to CL:923925, this patch captures the 3 default SkFontMgr to see if they are the same in crash dumps. Bug: 409784 Change-Id: I9ca760dfbb28f7d23caca05d3311bca05be7d1c1 Reviewed-on: https://chromium-review.googlesource.com/923744Reviewed-by:
Dominik Röttsches <drott@chromium.org> Commit-Queue: Koji Ishii <kojii@chromium.org> Cr-Commit-Position: refs/heads/master@{#537744}
-
Sasha Morrissey authored
Deselect elements and cancel a touchscreen drag if the drag is initiated in the files list padding area, below the list. Bug: 760552 Cq-Include-Trybots: master.tryserver.chromium.linux:closure_compilation Change-Id: I163ebe6c16f7295af3feb25c56a9abfd99d483fc Reviewed-on: https://chromium-review.googlesource.com/923619Reviewed-by:
Naoki Fukino <fukino@chromium.org> Commit-Queue: Sasha Morrissey <sashab@chromium.org> Cr-Commit-Position: refs/heads/master@{#537743}
-
Naoki Fukino authored
Depending on DPI settings, there is a chance to show unnecessary scroll bar on directory tree rows. As we don't need to scroll labels of directory tree rows, we can just overflow property to hidden. Bug: 787720 Test: Manually tested on sentry. Cq-Include-Trybots: master.tryserver.chromium.linux:closure_compilation Change-Id: I6735847daaa707244620e1039455848f4030ee45 Reviewed-on: https://chromium-review.googlesource.com/926024Reviewed-by:
Tatsuhisa Yamaguchi <yamaguchi@chromium.org> Commit-Queue: Naoki Fukino <fukino@chromium.org> Cr-Commit-Position: refs/heads/master@{#537742}
-
Makoto Shimazu authored
The crashes means |request_| and |legacy_request_| may not be set correctly. This change is for checking if the assumption is true. Bug: 810685 Change-Id: Ie104f656972ff550ca9a5f4d8c30ed266a561d66 Reviewed-on: https://chromium-review.googlesource.com/926023 Commit-Queue: Makoto Shimazu <shimazu@chromium.org> Reviewed-by:
Matt Falkenhagen <falken@chromium.org> Cr-Commit-Position: refs/heads/master@{#537741}
-
Yoshifumi Inoue authored
This patch add |const| to invariant member variables in |TypingCommand| for improving readability. Change-Id: Ieb9e63d02cb092754a50d7b450515da24fe7e411 Reviewed-on: https://chromium-review.googlesource.com/925985Reviewed-by:
Yoichi Osato <yoichio@chromium.org> Commit-Queue: Yoshifumi Inoue <yosin@chromium.org> Cr-Commit-Position: refs/heads/master@{#537740}
-
Yoshifumi Inoue authored
This patch changes |DeleteSelectionCommand| to hold |DeleteSelectionOptions| to simplify constructors for improving code health. This is a follow-up of the patch[1]. [1] http://crrev.com/c/925125 Introduce DeleteSelectionOptions for DeleteSelectionCommand construction Change-Id: I415e167c24e2c789a3b3cda76361a7a70d2a898e Reviewed-on: https://chromium-review.googlesource.com/925762Reviewed-by:
-
Kinuko Yasuda authored
So that we can set up necessary throttles and resource/request contexts when Prefetch needs to create additional requests, e.g. for fetching certificate for Signed HTTP Exchanges. (The usage example of the throttles, resource_context and request_context_getter would look like: https://chromium-review.googlesource.com/c/chromium/src/+/920565) Bug: 803776 Change-Id: I71b1e7e8bb1178ecae89b4d9e37b60df658eba32 Reviewed-on: https://chromium-review.googlesource.com/923467Reviewed-by:
Tsuyoshi Horo <horo@chromium.org> Commit-Queue: Kinuko Yasuda <kinuko@chromium.org> Cr-Commit-Position: refs/heads/master@{#537738}
-
Jan Krcal authored
This CL enables writing on-demand favicons if all other icons are expired (and have a different type). The motivation is to be able to fetch on-demand favicons for bookmarked URLs that have an (expired) favicon from sync. Fetching higher-res icons for bookmarks from the Google favicon server is currently being implemented for iOS bookmarks UI. For the approach to work, the CL changes the MergeFavicon function to set all merged icons as expired, notably all favicons from sync. A follow-up CL will make LargeIconService:: GetLargeIconOrFallbackStyleFromGoogleServerSkippingLocalCache() check CanSetOnDemandFavicons() before issuing the request (and also rename the function to GetLargeIconFromGoogleServer()). Bug: 799447 Change-Id: I59f4e16c1e6e8a8bd754e122c98de93cd258bfea Reviewed-on: https://chromium-review.googlesource.com/852256 Commit-Queue: Jan Krcal <jkrcal@chromium.org> Reviewed-by:
Sylvain Defresne <sdefresne@chromium.org> Reviewed-by:
Peter Kotwicz <pkotwicz@chromium.org> Cr-Commit-Position: refs/heads/master@{#537737}
-
Kent Tamura authored
This follows a DOM specification change; https://github.com/whatwg/dom/pull/566 - createElement() and createElementNS() don't add "is" content attribute - cloneNode() should refer "is value" instead of "is" content attribute value. This CL adds a data member for "is value" to ElementRareData, and initialize it in "create an element" implementation. Note that new failure for serializing-html-fragments.html is expected. It will be fixed for crbug.com/813675. Bug: 807871 Change-Id: Ie7c5f6282dd1a84a3d159b89d0d4efd72b0ddee0 Reviewed-on: https://chromium-review.googlesource.com/925901Reviewed-by:
Yoshifumi Inoue <yosin@chromium.org> Commit-Queue: Kent Tamura <tkent@chromium.org> Cr-Commit-Position: refs/heads/master@{#537736}
-
Daniel Cheng authored
Bug: 803620 Change-Id: I6c28067a10d812faf815a04360a701aee741abc5 Reviewed-on: https://chromium-review.googlesource.com/925541Reviewed-by:
Yuki Shiino <yukishiino@chromium.org> Reviewed-by:
Kentaro Hara <haraken@chromium.org> Commit-Queue: Daniel Cheng <dcheng@chromium.org> Cr-Commit-Position: refs/heads/master@{#537735}
-
Takashi Toyoshima authored
We are not exposing |forPreload| boolean flag to FrameFetchContext any more. So, the major concern for the TODO has already been resolved. Bug: 675883 Change-Id: I56c33ab41dac89185a679250f71fe6a8a48d3e69 Reviewed-on: https://chromium-review.googlesource.com/923771 Commit-Queue: Takashi Toyoshima <toyoshim@chromium.org> Reviewed-by:
Yutaka Hirano <yhirano@chromium.org> Cr-Commit-Position: refs/heads/master@{#537734}
-
Pavol Marko authored
On user session start, only request the chrome process to be restarted if site isolation policies effectively change between the sign-in screen and the user session. The DeviceLoginScreenSitePerProcess and DeviceLoginScreenIsolateOrigins policies map to --site-per-process and --isolate-origins command-line flags passed to the sign-in screen chrome process by session manager (see CL:924421). The user session should respect SitePerProcess/IsolateOrigins user policies instead. Replace special handling of the site isolation flags with placing them between --policy-flags-begin and --policy-flags-end sentinels. This makes sure that chrome is restarted exactly when the user session flags don't match the login screen flags. Background: chrome decides if it should be restarted by comparing the parts of the current command line with parts of the target command line. Specifically, the parts between flags and policy sentinels are compared. Also, evaluate the user policies mapped through user Profile prefs instead of local_state. Reason: When login_manager starts chrome with site isolation flags, the prefs are set by ChromeCommandLinePrefStore. Bug: 800117 Test: browser_tests --gtest_filter=*SiteIsolationFlagHandlingTest* Change-Id: Ic11f8d687ab55ccd08928df7655896c19936cd34 Reviewed-on: https://chromium-review.googlesource.com/924147 Commit-Queue: Pavol Marko <pmarko@chromium.org> Reviewed-by:
Maksim Ivanov <emaxx@chromium.org> Reviewed-by:
Xiyuan Xia <xiyuan@chromium.org> Cr-Commit-Position: refs/heads/master@{#537733}
-
Dominic Mazzoni authored
Select-to-speak on Android wants to know what nodes in the accessibility tree contain an image. We can't rely on exposing this in the class name or role, because sometimes an image might be inside a button, link, or other leaf-level UI element. Add a new custom flag "AccessibilityNodeInfo.hasImage" indicating if it has an image. BUG=813116 Change-Id: Ia4d3779eecc534ebf997c655a817c5fd06a22d5a Reviewed-on: https://chromium-review.googlesource.com/923207 Commit-Queue: Dominic Mazzoni <dmazzoni@chromium.org> Reviewed-by:
Katie Dektar <katie@chromium.org> Cr-Commit-Position: refs/heads/master@{#537732}
-
Menglu Huang authored
Details in doc: https://docs.google.com/document/d/1QaAZgmYuEImL1wAYJQxPcbex-fPLVd3rx8oMMKIgZX8/edit# Bug: 808267 Cq-Include-Trybots: master.tryserver.chromium.mac:ios-simulator-cronet;master.tryserver.chromium.mac:ios-simulator-full-configs Change-Id: I55fc27fcf31120a7c9d39e8953f72a39dd6394a7 Reviewed-on: https://chromium-review.googlesource.com/917309 Commit-Queue: Menglu Huang <huangml@chromium.org> Reviewed-by:
Sergey Berezin <sergeyberezin@chromium.org> Cr-Commit-Position: refs/heads/master@{#537731}
-
Kent Tamura authored
We should just call Attr::Clone(). Bug: 812089, 812105 Change-Id: I163769e134a52d82b88834ae29083a01a3480e95 Reviewed-on: https://chromium-review.googlesource.com/925865Reviewed-by:
-
Yoshifumi Inoue authored
This patch changes |CompositeEditCommand::DeleteSelection()| to utilize |DeleteSelectionOptions| to avoid |bool| parameters for improving code health. Change-Id: I9bdf8f37dbd0b272fc4eb9e43101ad8b5e9e61bc Reviewed-on: https://chromium-review.googlesource.com/923615 Commit-Queue: Yoshifumi Inoue <yosin@chromium.org> Reviewed-by:
-
Yoshifumi Inoue authored
This patch introduces |DeleteSelectionOptions| to avoid |bool| parameters of |DeleteSelectionCommand| construction for improving code health. Following patch will utilize |DeleteSelectionOptions| for |CompositeEditCommand::DeleteSelection()|[1] and change member variables in |DeleteSelectionCommand| class. [1] http://crrev.com/c/923615 Utilize DeleteSelectionOptions in CompositeEditCommand: :DeleteSelection() Change-Id: I84d036d109199287df250409ce99901e7bb1e589 Reviewed-on: https://chromium-review.googlesource.com/925125 Commit-Queue: Yoshifumi Inoue <yosin@chromium.org> Reviewed-by:
-
Yoshifumi Inoue authored
This patch marks "memory-cache.html" flaky. NOTRY=true TBR=alken@chromium.org, horo@chromium.org, shimazu@chromium.org Bug: 813468 Change-Id: I8c49b37caca7d328eb31fb65fa2ae6ab074a9636 Reviewed-on: https://chromium-review.googlesource.com/925871 Commit-Queue: Yoshifumi Inoue <yosin@chromium.org> Reviewed-by:
-
Kent Tamura authored
Now the test always fails on Windows 10 bot. We need to do rebaseline to see if the test is flaky or not. TBR=keishi@chromium.org Bug: 674466 Change-Id: I72567da52e59ad413eed68410147a42bb00dc7d2 Reviewed-on: https://chromium-review.googlesource.com/925941Reviewed-by:
Kent Tamura <tkent@chromium.org> Commit-Queue: Kent Tamura <tkent@chromium.org> Cr-Commit-Position: refs/heads/master@{#537726}
-
Hajime Hoshi authored
This is part of efforts to replace default timer task runners with other appropriate task runners in the renderer. Bug: 624694, 804694 Change-Id: Ibe59f2d49e197f7badc2832482de7849c4e2da09 Reviewed-on: https://chromium-review.googlesource.com/923875Reviewed-by:
Alexander Timin <altimin@chromium.org> Reviewed-by:
-
v8-autoroll authored
Summary of changes available at: https://chromium.googlesource.com/v8/v8/+log/5f4eca57..74f87f4e Please follow these instructions for assigning/CC'ing issues: https://github.com/v8/v8/wiki/Triaging%20issues Please close rolling in case of a roll revert: https://v8-roll.appspot.com/ This only works with a Google account. CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel;master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel;master.tryserver.chromium.android:android_optional_gpu_tests_rel TBR=hablich@chromium.org,machenbach@chromium.org,kozyatinskiy@chromium.org,sergiyb@chromium.org Change-Id: I236161eebb2d6f4cc45a07e7d73aae632f0c6e15 Reviewed-on: https://chromium-review.googlesource.com/925623Reviewed-by:
v8 autoroll <v8-autoroll@chromium.org> Commit-Queue: v8 autoroll <v8-autoroll@chromium.org> Cr-Commit-Position: refs/heads/master@{#537724}
-
tanvir.rizvi authored
Inside a table, if there are two empty list items, deleting the second list deletes the entire table. In an empty list the deep equivalent for first position and its next position comes as same. I think that FirstInSpecialElement was for the hidden elements like span, br etc, therefore this CL do not consider list element for a special element category. Bug: 731913 Change-Id: I03a6ec7559a5271954c85561d9c27db75a921452 Reviewed-on: https://chromium-review.googlesource.com/921301Reviewed-by:
-
Hajime Hoshi authored
This is part of efforts to replace default timer task runners with other appropriate task runners in the renderer. Bug: 624694, 804694 Change-Id: I4115f9c706c735c73bdb14f7243c0e8142293028 Reviewed-on: https://chromium-review.googlesource.com/924101Reviewed-by:
-
Kouhei Ueno authored
This CL introduces *experimental* CBORReader::ReadDataItemHeader() method which allows streaming consuming of a large CBOR-encoded array. Since this API exposes CBORReader::DataItemHeader, which should conceptually be encapsulated inside CBORReader, the method should not be used widely, and should be replaced by event-based CBORReader API. The method is really only for short-term use from SignedExchangeParser, which definitely requires streaming parsing, but is currently unclear if it will continue to use CBOR encoding in long term. Bug: 803774, 811717 Change-Id: If37c70b034afcdecbe6e7793c95b4e1f60eee9d0 Reviewed-on: https://chromium-review.googlesource.com/920723 Commit-Queue: Kouhei Ueno <kouhei@chromium.org> Reviewed-by:
Balazs Engedy <engedy@chromium.org> Cr-Commit-Position: refs/heads/master@{#537721}
-
Morten Stenshorne authored
An out-of-flow positioned box with hypothetically static (aka. automatic) position needs to update its layer's static position also when it's inside a (non-atomic) inline. This fixes one layout test, but is also a step on the way to fixing another one: fast/block/positioning/table-cell-static-position.html Bug: 813525 Cq-Include-Trybots: master.tryserver.chromium.linux:linux_layout_tests_layout_ng Change-Id: Ic4665a5371f2b2c306682f51e1d492205be6c719 Reviewed-on: https://chromium-review.googlesource.com/925601Reviewed-by:
Aleks Totic <atotic@chromium.org> Reviewed-by:
Koji Ishii <kojii@chromium.org> Commit-Queue: Morten Stenshorne <mstensho@chromium.org> Cr-Commit-Position: refs/heads/master@{#537720}
-
Morten Stenshorne authored
https://chromium-review.googlesource.com/c/chromium/src/+/920776 caused regressions. Strings dumped from LayoutObject::DumpLayoutObject() used to escape newlines, non-ASCII characters and other special characters, but not after this CL. The code depended on a special std::ostream& operator<<(std::ostream& out, const String& string) overload, which took care of this string conversion. Since we no longer use std::ostream, introduce add String::EncodeForDebugging() and have Node::ToString() (and the aforementioned operator<< overload) use that. Added a unit test so that we don't accidentally break this again. Change-Id: I62172bf1ebd12da1fdb71f78768e17aa937ea5cb Reviewed-on: https://chromium-review.googlesource.com/925271Reviewed-by:
Xiaocheng Hu <xiaochengh@chromium.org> Commit-Queue: Morten Stenshorne <mstensho@chromium.org> Cr-Commit-Position: refs/heads/master@{#537719}
-
Megumi Hattori authored
This reverts commit b5cae0bd. Reason for revert: Settings button and close button on MessageCenterView are missing. Original change's description: > Implement scroll shadow on MessageCenterView. > > When MessageCenterView can be scrolled, the 2dp scroll shadow is > displayed at the bottom. > > The shadow is displayed over notifications on MessageCenterView. > > If MessageCenterView is all scrolled, the shadow is hidden. > > BUG=779880 > > Change-Id: I48fdb3e97c57075114ba4210cea40039c295a6d7 > Reviewed-on: https://chromium-review.googlesource.com/923244 > Commit-Queue: Megumi Hattori <megumihattori@google.com> > Reviewed-by: Yoshiki Iguchi <yoshiki@chromium.org> > Reviewed-by: Tetsui Ohkubo <tetsui@chromium.org> > Cr-Commit-Position: refs/heads/master@{#537711} TBR=yoshiki@chromium.org,tetsui@chromium.org,maajid@chromium.org,megumihattori@google.com Change-Id: Ib697e58df63ccda943eb4caf920620787bf31577 No-Presubmit: true No-Tree-Checks: true No-Try: true Bug: 779880 Reviewed-on: https://chromium-review.googlesource.com/925926Reviewed-by:
Tetsui Ohkubo <tetsui@chromium.org> Commit-Queue: Tetsui Ohkubo <tetsui@chromium.org> Cr-Commit-Position: refs/heads/master@{#537718}
-
Darren Shen authored
This patch adds the clear() method to inline & declared StylePropertyMap Spec: https://drafts.css-houdini.org/css-typed-om-1/#dom-stylepropertymap-clear Bug: 813653 Change-Id: Ib56f60fa59dc8d241e7fb1a7695fce2bea094026 Reviewed-on: https://chromium-review.googlesource.com/925761 Commit-Queue: Darren Shen <shend@chromium.org> Reviewed-by:
nainar <nainar@chromium.org> Cr-Commit-Position: refs/heads/master@{#537717}
-
Yoshifumi Inoue authored
This patch gets rid of extra namespace for jumbo build from "VisibleUnitsWordTest.cpp" to simplify source code for improving code health. [1] http://crrev.com/c/923723 [jumbo] avoid helper function collisions in VisibleUnits*Test.cpp Change-Id: I35e5f45f458e370000aeaf8f161e26c76f7f5152 Reviewed-on: https://chromium-review.googlesource.com/925922Reviewed-by:
-
Koji Ishii authored
Following bot results are included. 3839 3841 3843 3845 3846 50 lines were removed and 51 lines were deflaked by consecutive results since 3770. TBR=eae@chromium.org NOTRY=true Bug: 591099 Cq-Include-Trybots: master.tryserver.chromium.linux:linux_layout_tests_layout_ng Change-Id: I73fdf6b07145a907d4025838718f30cc3eadf045 Reviewed-on: https://chromium-review.googlesource.com/923666 Commit-Queue: Koji Ishii <kojii@chromium.org> Reviewed-by:
-
Yuki Shiino authored
The current implementation is running ResizeObserver's callback functions even when they're NOT wrapper-traced. This patch makes it sure that the observed Element is performing wrapper-tracing otherwise ResizeObserver does not invoke an untraced callback function. The attached test case causes crash with the current implementation, but not with this patch. Bug: 809784 Change-Id: If2e66d388ab4f4410ca8c472176fbdee5a957f08 Reviewed-on: https://chromium-review.googlesource.com/924104Reviewed-by:
Michael Lippautz <mlippautz@chromium.org> Cr-Commit-Position: refs/heads/master@{#537714}
-
Kinuko Yasuda authored
Example failures: https://ci.chromium.org/buildbot/chromium.mac/Mac10.10%20Tests/29357 https://ci.chromium.org/buildbot/chromium.mac/Mac10.11%20Tests/23562 https://ci.chromium.org/buildbot/chromium.mac/Mac10.12%20Tests/10337 https://ci.chromium.org/buildbot/chromium.webkit/WebKit%20Win7/60528 TBR=hiroshige@chromium.org Bug: 813704 Change-Id: I2632b5acd189b9a203700c8b5b1f57fc449aba93 No-Try: True Reviewed-on: https://chromium-review.googlesource.com/925870Reviewed-by:
Kinuko Yasuda <kinuko@chromium.org> Commit-Queue: Kinuko Yasuda <kinuko@chromium.org> Cr-Commit-Position: refs/heads/master@{#537713}
-
Kinuko Yasuda authored
TBR=klausw@chromium.org NOTRY=true Bug: 813697 Change-Id: Ibef66c36488eb83052c0b8da91e232c9be832020 Reviewed-on: https://chromium-review.googlesource.com/925867Reviewed-by:
-
Megumi Hattori authored
When MessageCenterView can be scrolled, the 2dp scroll shadow is displayed at the bottom. The shadow is displayed over notifications on MessageCenterView. If MessageCenterView is all scrolled, the shadow is hidden. BUG=779880 Change-Id: I48fdb3e97c57075114ba4210cea40039c295a6d7 Reviewed-on: https://chromium-review.googlesource.com/923244 Commit-Queue: Megumi Hattori <megumihattori@google.com> Reviewed-by:
Yoshiki Iguchi <yoshiki@chromium.org> Reviewed-by:
-
Naina Raisinghani authored
This patch allows CSSTransformValues items to be mutated/appended. Spec issue here: https://github.com/w3c/css-houdini-drafts/issues/690 Bug: 545318 Change-Id: I841704e1f2ec8db0f27ae254a0949d4f9b46c8f7 Reviewed-on: https://chromium-review.googlesource.com/925882Reviewed-by:
Darren Shen <shend@chromium.org> Commit-Queue: nainar <nainar@chromium.org> Cr-Commit-Position: refs/heads/master@{#537710}
-
Amirhossein Simjour authored
PermissionDialogController uses PermissionAppModalDialogView if the flag is enabled. Since the goal is to remove the current implementation based on AlertDialogs, I didn't introduce common interface between PermissionAppModalDialogView and PermissionDialogView. BUG=811134 Change-Id: Ia3f5e072eccf4b25800ae430b8af10d263344409 Reviewed-on: https://chromium-review.googlesource.com/916551 Commit-Queue: Amirhossein Simjour <asimjour@chromium.org> Reviewed-by:
Dominick Ng <dominickn@chromium.org> Reviewed-by:
Ted Choc <tedchoc@chromium.org> Reviewed-by:
Becky Zhou <huayinz@chromium.org> Cr-Commit-Position: refs/heads/master@{#537709}
-
Matt Falkenhagen authored
I believe this code is never reached since Mojo Blobs is now enabled by default. R=kinuko Change-Id: I31ff07484290ab932a3569c9e10e00316a5bafd0 Reviewed-on: https://chromium-review.googlesource.com/925863Reviewed-by:
-
https://webrtc.googlesource.com/src.git/+log/2b6f13508ec1..500874effdad $ git log 2b6f13508..500874eff --date=short --no-merges --format='%ad %ae %s' Created with: roll-dep src/third_party/webrtc BUG=None The AutoRoll server is located here: https://webrtc-chromium-roll.skia.org Documentation for the AutoRoller is here: https://skia.googlesource.com/buildbot/+/master/autoroll/README.md If the roll is causing failures, please contact the current sheriff, who should be CC'd on the roll, and stop the roller if necessary. CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_chromium_archive_rel_ng;master.tryserver.chromium.mac:mac_chromium_archive_rel_ng;master.tryserver.chromium.win:win-msvc-dbg TBR=webrtc-chromium-sheriffs-robots@google.com Change-Id: I518c2f64cbf962dcdf4bbedab6d0ee2dfb06c682 Reviewed-on: https://chromium-review.googlesource.com/925687Reviewed-by:
webrtc-chromium-autoroll <webrtc-chromium-autoroll@skia-buildbots.google.com.iam.gserviceaccount.com> Commit-Queue: webrtc-chromium-autoroll <webrtc-chromium-autoroll@skia-buildbots.google.com.iam.gserviceaccount.com> Cr-Commit-Position: refs/heads/master@{#537707}
-
Matt Falkenhagen authored
Back in issue 413094, we decided to skip service worker for fetches from Pepper plugins with "private permission" for security purposes. The motivation was that Pepper plugins without "private permission" could be assumed to enfore the same-origin policy. However, the spec has since mandated skipping service workers for the request for the plugin itself, i.e., the target URL of an EMBED or OBJECT element. This patch makes two changes: 1) Requests *for* the target URL of EMBED or OBJECT element that load a Pepper plugin skip service workers. This aligns with recent patches to skip all EMBED or OBJECT element requests: r537245 skipped them for embedded HTML content, and r537386 skipped them for MIME handler plugins. The code change is in ppb_nacl_private_impl.cc::CreateWebURLRequest. This stops the requests for the manifest and pexe from being intercepted by the service worker. 2) Requests *from* any Pepper plugin skip service workers. Previously, we only skipped if the plugin had private permission, now we skip regardless of permission. The code change is in url_request_info_util.cc::CreateWebURLRequest. One thing I'm not so sure about is PepperPluginInstanceImpl::Navigate which apparently does a navigation in a frame. This is also changed to to skip service workers (by changing the utility function CreateWebURLRequest), but it's unclear whether that's needed. You might ask why we don't change the service worker interception code to just skip plugins, instead of changing the plugin callsites. This is because at the SW interception site, we don't know whether the request came from a plugin or not: for the manifest request, the RequestContextType is "INTERNAL", and the ResourceType is "SUBRESOURCE". It's also worth nothing that NetworkService/S13nServiceWorker already skip the service worker for all these requests, since we don't hook in at the URLRequestJob level anymore. In NS/S13nSW, we likely won't need to set skip service worker at these callsites. R=kinuko TBR=bradnelson Bug: 771933,413094 Cq-Include-Trybots: master.tryserver.chromium.linux:linux_mojo Change-Id: I09db0eda46f2e7d9372495a6205f5cb0026de6c7 Reviewed-on: https://chromium-review.googlesource.com/923663 Commit-Queue: Matt Falkenhagen <falken@chromium.org> Reviewed-by:
Raymes Khoury <raymes@chromium.org> Cr-Commit-Position: refs/heads/master@{#537706}
-
Alice Boxhall authored
Idle tasks are too difficult to use in layout tests, and are also not implemented anywhere other than Chrome. Bug: 792783 Change-Id: I802c2a9cdd292138ff68fefd3f69b71e0f6e2e98 Reviewed-on: https://chromium-review.googlesource.com/923469 Commit-Queue: Alice Boxhall <aboxhall@chromium.org> Reviewed-by:
-
