Update Dockerfile #12
Closed
This pull request is big! We’re only showing the most recent 250 commits.
Commits on Mar 15, 2024
-
Enable PrefetchRedirects by default.
1% stable data looks good; per best practices, will request ramp to 100% once this CL reaches a release build. Bug: 40872301 Change-Id: I81bd24993fc8b84af4234ac87a974079379d3a4b Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5372832 Auto-Submit: Jeremy Roman <[email protected]> Reviewed-by: Rakina Zata Amni <[email protected]> Commit-Queue: Rakina Zata Amni <[email protected]> Commit-Queue: Jeremy Roman <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273205}
-
Enable the 'writingsuggestions' feature attribute by default. Intent to Ship: https://groups.google.com/a/chromium.org/g/blink-dev/c/NhWn64yxB5Q Bug: 41486518 Change-Id: I80973773eed85f3c1b7f903b7a541c839da971b4 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5370351 Reviewed-by: Sanket Joshi <[email protected]> Reviewed-by: Dan Clark <[email protected]> Commit-Queue: Stephanie Zhang <[email protected]> Reviewed-by: Domenic Denicola <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273214}
-
webnn: Use ASSIGN_OR_RETURN in //services/webnn/tflite
Replaces most uses of RETURN_IF_ERROR with this macro, which improves readability. Specifically, it allows for replacing many uses of "auto" (previously used to hide overly-verbose expected<T, String> types) with the type of the assigned variable after unwrapping the expected<T, String>. See https://google.github.io/styleguide/cppguide.html#Type_deduction This CL contains no functional changes Bug: 324541572 Change-Id: I33d67d37f493193c578a13a2414b1e4e0cfa6716 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5371913 Commit-Queue: Austin Sullivan <[email protected]> Auto-Submit: Austin Sullivan <[email protected]> Reviewed-by: ningxin hu <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273217}
-
Pass synthetic Sec-Purpose header to shouldOverrideUrlLoading
When Prerender2 is enabled on WebView, the embedder will need to be able to distinguish prerender navigation vs other navigation. To enable that without a WebView API change, we pass `Sec-Purpose` header to `WebResourceRequest` that `shouldOverrideURLLoading` takes as its `request` argument. [1] [1] https://developer.android.com/reference/android/webkit/WebViewClient#shouldOverrideUrlLoading(android.webkit.WebView,%20android.webkit.WebResourceRequest) Bug: 41490450 Change-Id: Ia181730f01d40ed699fb3441e3dba59ce0414770 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5348108 Commit-Queue: Kouhei Ueno <[email protected]> Reviewed-by: Rakina Zata Amni <[email protected]> Reviewed-by: Hiroki Nakagawa <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273219}
-
Remove FixedFlatSet use in PolicyWatcherTest.
MakeFixedFlatSet is only for building compile time maps. As such, there are restrictions on when it can be used: in particular, the set itself must be constructed with a constant expression, which means any variables used in the initializer must also be constexpr variables or be constant-initialized variables. A forward-declared extern variable is neither of these. Rather than trying to make the initializer a constant expression, just use base::flat_map: since this is test-only code, there is no implication on production code anyway. Bug: 41486252 Change-Id: Ic8555b23e95b28bafdb036a0eb5fcccb2634aebf Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5372938 Auto-Submit: Daniel Cheng <[email protected]> Reviewed-by: Joe Downing <[email protected]> Commit-Queue: Joe Downing <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273225}
-
[URL] Support empty hosts in relative URLs for non-special base URLs
Update Chromium's handling of relative URLs by allowing them to have empty hosts (like "///path") when used with non-special base URLs. Before: > const url = new URL("///path", "git://host/a") ; => throws an Invalid URL error. After: > const url = new URL("///path", "git://host/a") ; => `url' is valid. url.href is "git:///path". This change basically implements "host state" in URL Standard: https://url.spec.whatwg.org/#host-state > 2.1. If url is special and buffer is the empty string, host-missing > validation error, return failure. Prior CLs already implemented this, but was not implemented correctly in resolving relative URLs for non-special base URLs. While the WPT URL tests have a good coverage for these scenarios, I also added unit tests for additional assurance. Bug: 40063064 Change-Id: I46786955d5db27d29864e80a259ce4076500cb33 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5364453 Commit-Queue: Hayato Ito <[email protected]> Reviewed-by: Adam Rice <[email protected]> Reviewed-by: Daniel Cheng <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273227} -
webnn: Migrated unit tests for input and constant interfaces into WPT
This CL adds WPT tests for 1) `input(name, descriptor)` interface, 2) `constant(descriptor, bufferView)` interface, and removes these unit tests 1) `MLGraphBuilderTest.InputTest`, 2) `MLGraphBuilderTest.ConstantTest`. Bug: 327337525, 327337526, 328026885 Change-Id: I95cba29fc189854650ba7ad3bdece4c2d1e26390 Cq-Include-Trybots: luci.chromium.try:win11-blink-rel Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5365030 Commit-Queue: Feng Dai <[email protected]> Reviewed-by: Austin Sullivan <[email protected]> Reviewed-by: ningxin hu <[email protected]> Auto-Submit: Feng Dai <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273229}
-
Update fieldtrial_testing_config.json to be matched with the experime…
…nt setting Bug: 40874535 Change-Id: I756ece84e4c9fb407db6bb5743e275e31dd385a0 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5369017 Reviewed-by: Yoshisato Yanagisawa <[email protected]> Auto-Submit: Minoru Chikamune <[email protected]> Commit-Queue: Minoru Chikamune <[email protected]> Reviewed-by: Shunya Shishido <[email protected]> Commit-Queue: Shunya Shishido <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273231}
-
Reland "text-box-trim: remove inline box tests (tentatively)"
This is a reland of commit b42e5ff text-box-trim-half-leading-block-box-001.html's reference file was set to inline box's by mistake, and this reland fixes the issue. Original change's description: > text-box-trim: remove inline box tests (tentatively) > > The specification is not updated yet, and it is not that clear to us > which edges could be cut in these cases. > Since we do not plan to implement the text-line-edge? property for > now, I'd like to remove these tests to avoid confusion. > > Bug: 40254880 > Change-Id: I08af3d3806e4e49a9e46de0f85271fb995259bbf > Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5360144 > Commit-Queue: Lingqi Chi <[email protected]> > Reviewed-by: Koji Ishii <[email protected]> > Cr-Commit-Position: refs/heads/main@{#1271417} Bug: 40254880 Change-Id: Ie2c1821c37c838f81bccbe51c6d15864f8550462 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5366868 Commit-Queue: Lingqi Chi <[email protected]> Reviewed-by: Koji Ishii <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273240}
-
Expose TestController to Ash and Lacros through a common interface
This CL reduces the need to #ifdef CHROMEOS_ASH/LACROS in tests by exposing a common interface for the TestController. See app_install_navigation_throttle_browsertest.cc for new usage of the interface. Bug: b/324511698 Change-Id: I3c4bde714d8531bc67ed024994009152d293c756 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5048292 Reviewed-by: Hidehiko Abe <[email protected]> Reviewed-by: David Tseng <[email protected]> Commit-Queue: Alan Cutter <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273242}
-
Update the warning text of deprecated registerRouter() call
We'd like to remove InstallEvent.registerRouter() soon. Our plan is remove it from M125. Before doing that, we give one milestone period as a final removal announcement. InstallEvent.registerRouter() is an old API, this was updated to addRoutes() based on the Origin Trial feedback. UseCounter shows nearly zero, we expect it's very safe to remove it. https://chromestatus.com/metrics/feature/timeline/popularity/4589 Bug: 329285464 Change-Id: I8fd6846690a16303f294461060175a01eeab2465 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5369018 Reviewed-by: Kouhei Ueno <[email protected]> Commit-Queue: Kouhei Ueno <[email protected]> Commit-Queue: Shunya Shishido <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273243}
-
Notify FSP Service observers on shutdown
- file_system_provider::Service is a keyed service while OneDriveIntegrationServiceAsh isn't. On OneDriveIntegrationServiceAsh destruction, the observer tries to remove itself from the FSP service which is no longer there, so it crashes. - Changed file_system_provider::Service to notify its observer when the shutting down starts. Bug: b:329305132 Change-Id: Ifeee5be1e7591c5c1c32d8a7e2711c1c4d85d6b6 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5366056 Auto-Submit: Aya Elsayed <[email protected]> Reviewed-by: Erik Chen <[email protected]> Reviewed-by: Peter Marshall <[email protected]> Commit-Queue: Erik Chen <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273247}
-
Add NotCondition support to ServiceWorkerRouterEvaluator
This CL adds evaluation logic on the not condition to ServiceWorker router evaluator. Explainer: https://github.com/WICG/service-worker-static-routing-api/blob/main/final-form.md Design Doc: https://docs.google.com/document/d/1B2D8w_2M4j9CZC1VccBOAhfCdiRP3yPCSV5UKKDhU70/ Change-Id: Ie281a2b099beb50ad75b8cdb223a5414c49161a0 Bug: 328565554 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5368785 Auto-Submit: Yoshisato Yanagisawa <[email protected]> Reviewed-by: Shunya Shishido <[email protected]> Reviewed-by: Minoru Chikamune <[email protected]> Commit-Queue: Yoshisato Yanagisawa <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273248}
-
Shared Storage: Pass FrameTreeNodeId to DevTools Shared Storage events
We recently changed Shared Storage event notifications sent to DevTools so that they would pass a `GlobalRenderFrameHostId ` to the DevTools `content::StorageHandler` instead of the DevTools frame token (https://crrev.com/c/5357793). But this inadvertently introduced a behavior change, as the the `GlobalRenderFrameHostId` currently changes on certain navigations, unlike the DevTools frame token, and moreover the former should eventually change on every navigation with the new "RenderDocument" feature. The end result would still be the same for the current use case of Shared Storage event notifications, as DevTools would still recover the same frame token despite the changed routing id, but this new behavior makes testing more difficult. Hence, we switch to passing `FrameTreeNodeId` instead, as this id will not change upon navigation. Bug: 328823983 Change-Id: I13a9b61f37fb2a4bb38af908ba7809ae0f9be483 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5370400 Reviewed-by: Rakina Zata Amni <[email protected]> Commit-Queue: Cammie Smith Barnes <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273249}
-
Gardening: disable flaky test on Linux MSAN
Bug: 329174385 Change-Id: I62fad4b7ce376eb76cf7c234ca45142671b58ce0 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5367786 Reviewed-by: Takashi Sakamoto <[email protected]> Commit-Queue: Mingyu Lei <[email protected]> Reviewed-by: Alan Cutter <[email protected]> Auto-Submit: Mingyu Lei <[email protected]> Owners-Override: Mingyu Lei <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273251}
-
text-box-trim: remove the logic of trimming leading from inline boxes
Trimming inline boxes will be determined by line-box-edge but we are focusing on "text-box-trim" + "text-box-edge" first. So, this CL removes the logic of trimming inline boxes for now. Bug: 40254880 Change-Id: Ib7effb57cd9927996ed33f746bdf8d273334fde9 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5366784 Reviewed-by: Koji Ishii <[email protected]> Commit-Queue: Lingqi Chi <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273254}
-
siso: Add more actions to large compile list for asan builds
Bug: b/311312613 Change-Id: I9954b48974741dcb9a20d22aa6f7f1d9a620dff4 Cq-Include-Trybots: luci.chromium.try:linux_chromium_asan_siso_rel_ng Cq-Include-Trybots: luci.chromium.try:linux_chromium_tsan_siso_rel_ng Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5373983 Auto-Submit: Fumitoshi Ukai <[email protected]> Reviewed-by: Junji Watanabe <[email protected]> Commit-Queue: Junji Watanabe <[email protected]> Reviewed-by: Takuto Ikuta <[email protected]> Commit-Queue: Fumitoshi Ukai <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273255}
-
environment_provider: Remove fields from EnvironmentProvider
This CL removes fields from EnvironemtnProvider. This is a part of the work to resolve the dependency with EnvironmentProvider. Bug: 328682979 Change-Id: Icd525370fb7854c4c045256ce995e73e20459516 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5355411 Commit-Queue: Eriko Kurimoto <[email protected]> Reviewed-by: Igor <[email protected]> Reviewed-by: Hidehiko Abe <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273259}
-
environment_provider: Remove dependency from BrowserManager/Launcher
This CL makes EnvironmentProvider singleton and remove the dependency from BrowserManager and BrowserLauncher. It also moves some APIs in EnvironmentProvider to crosapi_util anonymous space for APIs which are independent from chrome/*. This is a preparation to move EnvironmentProvider APIs to chromeos/*. This does not change the behavior. Bug: 328682979 Change-Id: I03820bd74d378788861831486d8b0fdd6397b40a Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5355055 Reviewed-by: Hidehiko Abe <[email protected]> Commit-Queue: Eriko Kurimoto <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273262}
-
Revert "[pdf-a11y]: be more assertive in resetting the plugin seriali…
…zer" This reverts commit 3c23a93. Reason for revert: Breaking arc.Print.container_p. See b/326733014 Original change's description: > [pdf-a11y]: be more assertive in resetting the plugin serializer > > The PDF a11y tree and serializer have a tenuous stitch to the ancestor > nodes. In particular, any subtree clears in the ancestry nodes require > the serializer to be cleared. If not, potential bugs like the one > referenced occur where descendants in PDF are not re-sent. > > In the past, we dealt with this by always serializing the entire PDF > a11y tree i.e. > https://crrev.com/c/5274721 > > Though this "works", it always means shoving a lot of data into an IPC > to the browser. Consider a 200 page PDF. > > This change attempts to cover more of the cases that may arise in which > we want to reset the plugin serializer: > - calls to AXObjectCacheImpl::Reset. These come from requests from the > browser to re-send all data. > - update->node_id_to_clear > 0 during > AXObjectCacheImpl::SerializeDirtyObjectsAndEvents. THis change makes > it so that we reset the plugin serializer for *any* node clearing (its > subtree) even if the update contains no embedded objects. > > [email protected] > > Bug: 326733014 > Test: cq > Change-Id: I8adaae9456ddc134ad8204154bf711b5dfaef387 > Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5347803 > Reviewed-by: Aaron Leventhal <[email protected]> > Commit-Queue: David Tseng <[email protected]> > Cr-Commit-Position: refs/heads/main@{#1269165} Bug: 326733014, 326733014 Change-Id: I1d5dc79618e678e1a567f5babd082557d705fe50 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5372768 Commit-Queue: Eriko Kurimoto <[email protected]> Owners-Override: Eriko Kurimoto <[email protected]> Bot-Commit: Rubber Stamper <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273264}
-
[rbe] Rebase //build/config/unsafe_buffers_paths.txt with rbe_exec_root
Rewrapper's -inputs flag is based on -exec_root. When `rbe_exec_root` GN arg is specified, `-inputs=build/config/unsafe_buffers_paths.txt` doesn't work as expected. This CL rebases the rewrapper inputs path with rbe_exec_root. Bug: b/329458056 Fixed: b/329458056 Change-Id: I08772dbe5ae9e3a9e50cb8cd4f7e7e9c06bad18d Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5367053 Reviewed-by: Michael Savigny <[email protected]> Commit-Queue: Junji Watanabe <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273266}
-
Revert "[pdf-a11y]: enable tree consistency checks for plugin tree"
This reverts commit a6f948e. Reason for revert: Breaking arc.Print.container_p. See b/326733014 Original change's description: > [pdf-a11y]: enable tree consistency checks for plugin tree > > [email protected] > > Bug: 326131114 > Test: cq > Change-Id: I03720dea3d63c9afc7ea5089d90d30676c3a161c > Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5323727 > Reviewed-by: Aaron Leventhal <[email protected]> > Commit-Queue: David Tseng <[email protected]> > Cr-Commit-Position: refs/heads/main@{#1266161} Bug: 326131114, 326733014 Change-Id: I520295122fb9ed00ccd16b10ccf866513ee3242b Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5371655 Reviewed-by: Jun Ishiguro <[email protected]> Owners-Override: Eriko Kurimoto <[email protected]> Commit-Queue: Eriko Kurimoto <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273267}
-
Gardening: mark canvas-blend-image.html as slow on Win
A wrong test was marked as slow in https://crrev.com/c/5371650, this CL fixes that. Bug: 1081534 Change-Id: I42502a2d3057fc3e20ce060b14e9335fc40bd517 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5371656 Auto-Submit: Mingyu Lei <[email protected]> Commit-Queue: Mingyu Lei <[email protected]> Reviewed-by: Takashi Sakamoto <[email protected]> Owners-Override: Mingyu Lei <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273268}
-
[local compile hints] Generate local compile hints multiple times
Previously, we generated local compile hints only when the page turned interactive. It's possible that the user navigates away before that, so in that case we didn't get any compile hints. This CL makes us compile hints earlier, and when the page turns interactive, we generate them again and replace the data. We also add a histogram for observing how many navigations now create compile hints data vs. how many would have created it in the original version. Bug: 40286622 Change-Id: Iec1bd1ec26a5fe870c456d1977109912cc560482 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5344969 Reviewed-by: Mike West <[email protected]> Reviewed-by: Charlie Reis <[email protected]> Reviewed-by: Chris Harrelson <[email protected]> Reviewed-by: Kouhei Ueno <[email protected]> Commit-Queue: Marja Hölttä <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273270}
-
[//media] Rename variables to match names of base::Features
Followup to [1]. Has no functional impact, but eliminates chance that someone in the future incorrectly renames either of the base::Features back to the variable name. [1] https://chromium-review.googlesource.com/c/chromium/src/+/5349219 Bug: 328540028 Change-Id: Ib179cd81890ca8b8ccd75ad1805d29ab8cac80fe Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5372043 Reviewed-by: Frank Liberato <[email protected]> Commit-Queue: Colin Blundell <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273272}
-
Revert "[pdf-a11y]: further simplify PluginAXTreeActionTargetAdapter"
This reverts commit 7503f75. Reason for revert: Breaking arc.Print.container_p. See b/326733014 Original change's description: > [pdf-a11y]: further simplify PluginAXTreeActionTargetAdapter > > Following comments from > https://crrev.com/c/5302041 > - simplify the PluginAXTreeActionTargetAdapter interface by combining > the two methods. GetFromId is not strictly needed as the only > production callsite used it immediately before calling the other > method, CreateActionTarget. > - addresses one nit > > Bug: 326131114 > Test: cq > Change-Id: I4bbf1cf5365c4857422bb46b0d6cf56ba28fbd57 > Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5322047 > Reviewed-by: Lei Zhang <[email protected]> > Commit-Queue: David Tseng <[email protected]> > Reviewed-by: Dave Tapuska <[email protected]> > Cr-Commit-Position: refs/heads/main@{#1265511} Bug: 326131114, 326733014 Change-Id: I455da818a6a540413a2b43cda66e53ce90184cdf Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5369301 Owners-Override: Eriko Kurimoto <[email protected]> Reviewed-by: Jun Ishiguro <[email protected]> Commit-Queue: Eriko Kurimoto <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273273}
-
Revert "[pdf-a11y]: allow for null/detached AXObject(s) backing the P…
…DF plugin container" This reverts commit 930c285. Reason for revert: Breaking arc.Print.container_p. See b/326733014 Original change's description: > [pdf-a11y]: allow for null/detached AXObject(s) backing the PDF plugin container > > In > https://crrev.com/c/5302041 > > we attach the plugin a11y tree directly using the WebAXObject for the > PDF's plugin container element through UpdateDependentObjects. > > This method may be called even when accessibility is off, so it is > reasonable to have a null/detached WebAXObject in production. > > Example calltsack > pdf::PdfAccessibilityTree::UpdateDependentObjects > pdf::PdfAccessibilityTree::MaybeHandleAccessibilityChange > pdf::PdfAccessibilityTree::PdfAccessibilityTree > pdf::PdfViewWebPluginClient::CreateAccessibilityDataHandler > > [email protected] > > Fixed: 326720339 > Test: cq > Change-Id: Ic99aadbf037964297587fa0ae843b34ec8ba45f1 > Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5322522 > Reviewed-by: Lei Zhang <[email protected]> > Commit-Queue: Lei Zhang <[email protected]> > Cr-Commit-Position: refs/heads/main@{#1264906} Bug: 326733014 Change-Id: I9d02963662d33adde24f80f940d57f03ffed4eb6 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5369603 Commit-Queue: Eriko Kurimoto <[email protected]> Owners-Override: Eriko Kurimoto <[email protected]> Reviewed-by: Jun Ishiguro <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273274}
-
Revert "[pdf-a11y]: move PluginAXTreeSource|Serializer to AXObject"
This reverts commit d976056. Reason for revert: Breaking arc.Print.container_p. See b/326733014 Original change's description: > [pdf-a11y]: move PluginAXTreeSource|Serializer to AXObject > > This change continues the work described in > go/chrome-pdf-a11y-stabilization: > - moves (most) plugin tree state and logic into Blink AXObject > - simplifies dependency into and out of RenderAccessibilityImpl > - makes the plugin tree a pure AXTreeSource > - revamps RenderAccessibilityImpl -> plugin tree relationship as purely one for actions (via a new adapter interface) > - numerous cleanups, simplifications > > Bug: 326131114 > Test: cq > > Change-Id: Ifa4f7a94cd7e259948873cff7606be3f336ffe96 > Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5302041 > Commit-Queue: David Tseng <[email protected]> > Reviewed-by: Lei Zhang <[email protected]> > Reviewed-by: Dave Tapuska <[email protected]> > Reviewed-by: Aaron Leventhal <[email protected]> > Reviewed-by: Chris Harrelson <[email protected]> > Cr-Commit-Position: refs/heads/main@{#1264786} Bug: 326131114, 326733014 Change-Id: Icc1b035e59d2418f947d65aec52ee754e1056d51 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5369604 Owners-Override: Eriko Kurimoto <[email protected]> Reviewed-by: Jun Ishiguro <[email protected]> Commit-Queue: Eriko Kurimoto <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273279}
-
Add new print scaling option: kCenterShrinkToFitPaper
When printing HTML (using Blink), we'd use kSourceSize when printing to PDF, and kFitToPrintableArea when printing to a printer. The latter doesn't really match what we do. We actually ignore the printable area, and instead use the entire paper. As such, kFitToPaper would be slightly more correct, but that option, just like kFitToPrintableArea, expects the page to be scaled up if it's smaller than the paper. What printing (to a printer) with Blink actually does, is to center the page on paper. If the page size is larger than the paper, it's scaled down and centered. If it's smaller, it's just centered. Therefore, add kCenterShrinkToFitPaper, to avoid confusion (and to allow for adding support for the other options in the future, if that ever becomes a priority). No behavior changes intended. Change-Id: I4e86b492c8330e173535d97eaa6dc931f3e52b7c Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5349263 Commit-Queue: Arthur Sonzogni <[email protected]> Auto-Submit: Morten Stenshorne <[email protected]> Reviewed-by: Lei Zhang <[email protected]> Reviewed-by: Alan Screen <[email protected]> Reviewed-by: Arthur Sonzogni <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273281}
-
IWYU: usage of WebNNBufferImpl requires include in context_impl.cc
Bug: 40939315 Change-Id: I3a87b4f021281668f8083fedca7be0b9e06562ea Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5369297 Reviewed-by: Andrew Moylan <[email protected]> Commit-Queue: José Dapena Paz <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273283}
-
cryptohome: Add UMA event when Downloads folder is already migrated
BUG=b:329533315 Change-Id: Ib8ce45c5f0c0d7ed68dcb97a849ab3787c15e7f2 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5366737 Commit-Queue: Howard Yang <[email protected]> Auto-Submit: François Degros <[email protected]> Reviewed-by: Howard Yang <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273288}
-
Bundz: support custom parameters for model execution session creation
Integrating with the updated optimization guide (https://crrev.com/c/5346730) to pass the top_k and temperature for model execution session creation. Bug: 324165372 Change-Id: I55e18062092d07067a8f49679ca866af2f138c24 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5359219 Reviewed-by: Fergal Daly <[email protected]> Commit-Queue: Mingyu Lei <[email protected]> Reviewed-by: Takashi Toyoshima <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273290}
-
Autofill flattens forms across frames and routes all iframe forms into their enclosing main frame form (as detailed in form_forest.h). In case of iframes with multiple forms (multiple <form> elements or both form and formless elements), this causes a preview problem because we currently route every single filling call by itself, and every filling call clears the preview of anything that could've happened before, which means that in the described case only the last form that is sent to the renderer for filling will actually be previewed. To overcome this, we make the agent fill a list of fields instead, so that we can pass all concerned fields of a single frame in a single function call. Precisely, this CL: - Removes FormData::FillData - Adds FormFieldData::FillData::host_form_id - Updates AutofillAgent::ApplyFormAction to take a list of fields instead of a form and updates its implementation accordingly. - Renames [AutofillAgent|form_util]::ApplyFormAction to ApplyFieldsAction, reflecting the shift in perspective. A follow-up CL will ensure that every frame gets a single fill call per autofill operation. Bug: 40943206, 41489087 Change-Id: I5ddcba8c014fca4a3bfec9b20b97fd47301c7524 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5331893 Reviewed-by: Arthur Sonzogni <[email protected]> Commit-Queue: Jihad Hanna <[email protected]> Reviewed-by: Christoph Schwering <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273292}
-
[histograms] Extend WebCore.Scripts.V8CompileHintsStatus
The histogram was recently fixed in https://chromium-review.googlesource.com/c/chromium/src/+/5344415 Bug: b/40286622 Change-Id: If07ed6108b640e792f418f6fb872b630ed3d892f Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5361438 Commit-Queue: Marja Hölttä <[email protected]> Reviewed-by: Chris Harrelson <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273293}
-
[infra] Add more compile targets to __use_large_b289968566 for ci/Lin…
…ux MSan Builder Some local fallback compiles cause build failures for OOM on the builder. This CL runs those large compiles on a large worker to avoid the remote/local OOM. This CL also increases the timeout to 4m because the clang/cxx/b289968566/exit-137 actions sometimes fail for remote timeout. (exit code 142) Bug: 329698344 Cq-Include-Trybots: luci.chromium.try:linux_chromium_asan_siso_rel_ng Cq-Include-Trybots: luci.chromium.try:linux_chromium_tsan_siso_rel_ng Change-Id: I3a9f9c3d16a762350e2b38e5507d910cf63715b0 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5372769 Commit-Queue: Junji Watanabe <[email protected]> Commit-Queue: Takuto Ikuta <[email protected]> Auto-Submit: Junji Watanabe <[email protected]> Reviewed-by: Takuto Ikuta <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273296}
-
webnn: Support Slice in //services/webnn/tflite
The data type of starts and sizes is signed integer for tflite SLICE builtin operator. Bug: 1273291 Change-Id: Ib72fce03e810760c8b92b32972823f9d459f65e1 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5369386 Reviewed-by: Reilly Grant <[email protected]> Reviewed-by: ningxin hu <[email protected]> Commit-Queue: Junwei Fu <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273298}
-
Check the timeout within the loop body
While changing window bounds ChromeDriver conducts a poll and wait loop bound by a timeout. Before this commit the timeout expiration check was done in the loop condition. This however was error prone because under high load the timeout could expire before the first loop iteration. This commit moves this timeout check into the loop body. Bug: chromedriver:4642 Change-Id: I66fea03c6e0846c08b2ba90474d526454a3299dc Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5366899 Auto-Submit: Vladimir Nechaev <[email protected]> Reviewed-by: Maksim Sadym <[email protected]> Commit-Queue: Maksim Sadym <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273300}
-
Fix bug of not running disabled tests with board suffix
Originally, the logic of processing the board suffix ("@" + board name like "tast_test_name@board") was in `template` of the section and worked only for not running tests, so it doesn't work well for `chrome_disabled_tast_tests`, that runs the "disabled" tests. This CL moves the logic to chromeos/BUILD.gn, so that it can work well even for `chrome_disabled_tast_tests`. Bug: b/329170253 Test: None Change-Id: I46be2b613b3dcd791f3747756aab88d98724b23d Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5363957 Auto-Submit: Yoshiki Iguchi <[email protected]> Commit-Queue: Yoshiki Iguchi <[email protected]> Reviewed-by: Takuto Ikuta <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273304} -
[//gpu] Introduce ClientSharedImage::GetTextureTarget()
This method will be *the* way to for clients to get the texture target to use for a given SharedImage. It has the following conceptual logic: On Mac, the platform-specific texture target is needed iff the SharedImage is using a native buffer (this is because when not using ANGLE-Metal, it is necessary to use GL_TEXTURE_RECTANGLE_ARB as the texture target with IOSurfaces). On all other platforms, the platform-specific texture target is needed iff using external sampling, as external sampling requires the EXTERNAL target. We will change all other ClientSharedImage::GetTextureTarget* variants to call this one under a killswitch, and then change all clients to call this method directly once the killswitch has rolled out safely. We also intend to move the logic for GetPlatformSpecificTextureTarget() in here once the dust has settled on the client rollout as well as conversions of service-side usage of that function to instead go through ClientSharedImage static functions. Bug: 41494843 Change-Id: I634c2050cef02ae734972ca2abf06ef883955d91 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5372103 Reviewed-by: Vasiliy Telezhnikov <[email protected]> Commit-Queue: Colin Blundell <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273305}
-
[Autofill] Set very high refill time limit
The following two tests are flaky because a refill doesn't happen: - AutofillInteractiveTestDynamicForm.* - AutofillAcrossIframesTest_Dynamic.RefillDynamicFormWithNewFrame This is probably due to very slow bots where action that triggers the refill takes more than one minute. This CL therefore increases the refill time limit from to 1 hour. Bug: 40933655 Change-Id: I7fbc57af8a23c98b3f8898c1d7acd51f502ea69b Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5372113 Commit-Queue: Florian Leimgruber <[email protected]> Reviewed-by: Florian Leimgruber <[email protected]> Auto-Submit: Christoph Schwering <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273307}
-
gardening: Re-enable dlp.DataLeakPreventionRulesListFilesUSB.ash_blocked
Verified locally the test has now fixed by crrev.com/c/5369218. Bug: 329323601 Change-Id: I0090957824b75dc235121d36796b1ece6033571d Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5371398 Commit-Queue: Eriko Kurimoto <[email protected]> Reviewed-by: Jun Ishiguro <[email protected]> Reviewed-by: Sergey Poromov <[email protected]> Auto-Submit: Eriko Kurimoto <[email protected]> Commit-Queue: Jun Ishiguro <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273310}
-
Renew the deadline of Cras.StreamAddError
BUG=b:41497440 TEST=cq passes Change-Id: Ifff403781dcc95ea048ad5a7a9259248d76c8979 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5366913 Reviewed-by: Yu-Hsuan Hsu <[email protected]> Reviewed-by: Li-Yu Yu <[email protected]> Commit-Queue: Baili Deng <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273315}
-
Enable DanglingPointerDetector on Windows (dcheck && !official)
The DanglingPointerDetector is currently enabled by default on Linux, and there are specific bots for other platforms. This patches adds Windows by default. Follow-ups: - Cleanup the win-rel config. - Iterate for every other platforms Benefits: - No more CQ vs CI differences (See chromium:327473315) - Increased coverage. Include-Ci-Only-Tests: true Bug: chromium:328104161, chromium:327473315 Change-Id: I15fad09ed44150ab007852bb398bd22f16256bd5 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5345629 Commit-Queue: Arthur Sonzogni <[email protected]> Reviewed-by: danakj <[email protected]> Owners-Override: danakj <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273325}
-
[Cronet] Modify request/stream's equality check
While our documentation does not guarantee that the request/stream passed to the callbacks will be the same as the one created by the user, our users have assumed this is the case. Eg https://github.com/google/ExoPlayer/blob/release-v2/extensions/cronet/src/main/java/com/google/android/exoplayer2/ext/cronet/CronetDataSource.java compares the request objects in their callbacks. This cl changes the equals method to, in addition to the original equality check, consider two wrappers with same backend to be equal. Bug: b/324583507 Change-Id: I266228a802b89778ae5a91d9be3e0c6e32a584b1 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5307640 Reviewed-by: Etienne Dechamps <[email protected]> Reviewed-by: Stefano Duo <[email protected]> Commit-Queue: Chidera Olibie <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273331}
-
Fix button in download infobar
- fix accessibility identifier that was set to the wrong button - do not update the button on dismissal Fixed: 329433709, 329143765 Change-Id: I1fc17914eb8cd309461473e2008c9f3e16f62e5b Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5372445 Auto-Submit: Olivier Robin <[email protected]> Reviewed-by: Quentin Pubert <[email protected]> Reviewed-by: Gauthier Ambard <[email protected]> Commit-Queue: Gauthier Ambard <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273332}
-
[PEPC] Add restrictions to word-spacing for PEPC
Bug: 1462930 Change-Id: I30edd87589dd222cd069c33cffeb5b4d4716671d Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5352970 Reviewed-by: Anders Hartvoll Ruud <[email protected]> Commit-Queue: Andy Paicu <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273334}
-
Implement constructor in RTCEncodedAudioFrame
This is inline with the algorithm described in w3c/webrtc-encoded-transform@main...palak8669:webrtc-encoded-transform:constructor_frame Bug: 14709 Change-Id: Ie9a94fb061c981f0748afcc5752390dc5a3f83f1 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5300593 Reviewed-by: Guido Urdaneta <[email protected]> Commit-Queue: Palak Agarwal <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273342}
-
Fix erroneous declaration of AutoWritableMemory::writers
The current declaration of AutoWritableMemory<T>::writers leads to different instances of |writers| for each T. However, all instances of base::ProtectedMemory are declared within one section, making an instance necessary which is shared across all Ts. Bug: 1521272, 1522964 Change-Id: If3f416d2c5141f22485b419b09a79c515bbb0a2a Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5285375 Reviewed-by: Kyle Charbonneau <[email protected]> Reviewed-by: Peter Kasting <[email protected]> Commit-Queue: Andre Kempe <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273344}
-
VTClasses: Add :active-view-transition-types and remove args from a-v-t
This patch implements the resolution on w3c/csswg-drafts#9972 (comment) Note that we increase the bitfield size on node to 32bits from 16bits. Bug: 329439849 [email protected], [email protected] Change-Id: Ibff7c68b779815398257380e04ac93ab0a239091 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5370900 Auto-Submit: Vladimir Levin <[email protected]> Reviewed-by: Noam Rosenthal <[email protected]> Commit-Queue: Noam Rosenthal <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273354}
-
Reland "[DevTools] Enable feature AutofillTestFormWithDevtools by def…
…ault" This is a reland of commit 0978b49 Original change's description: > [DevTools] Enable feature AutofillTestFormWithDevtools by default > > Bug: chromium:326381145 > Change-Id: I6352f30ddfcb202c80ec29c972adfbbe8daf4b37 > Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5349481 > Commit-Queue: Wolfgang Beyer <[email protected]> > Reviewed-by: Bruno Braga <[email protected]> > Reviewed-by: Danil Somsikov <[email protected]> > Reviewed-by: Christoph Schwering <[email protected]> > Cr-Commit-Position: refs/heads/main@{#1272056} Bug: chromium:326381145 Change-Id: I505473a2314a06ed762b33ea151f0875b78969f7 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5372469 Reviewed-by: Bruno Braga <[email protected]> Reviewed-by: Christoph Schwering <[email protected]> Reviewed-by: Danil Somsikov <[email protected]> Commit-Queue: Wolfgang Beyer <[email protected]> Auto-Submit: Wolfgang Beyer <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273355}
-
[iOS] Wait for History Sync Opt-In capabilities in advance
The History Sync Opt-In requires capabilities to load with the correct UI. Capabilities fetching are currently done as the screen loads with a spinner. This CL allows the opt-in screen to load instantly by waiting for capabilities in previous views while reusing existing spinners. Demo can be found in the bug entry. Additional testing (existing for FRE but disabled) requires updates to ChromeEarlGrey done in follow-up CL. Bug: 328398906 Change-Id: I3d68edfcbb085904e652fe392f52d936364fab59 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5352948 Reviewed-by: Nohemi Fernandez <[email protected]> Reviewed-by: Jérôme Lebel <[email protected]> Auto-Submit: Duong Dac <[email protected]> Commit-Queue: Duong Dac <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273362}
-
[//gpu] Add Feature to enable using CSI::GetTextureTarget() everywhere
This CL adds a base::Feature to control all of the different ClientSharedImage::GetTextureTarget*(...) variants calling through to ClientSharedImage::GetTextureTarget(), which is designed to handle all cases correctly. We also adjust the one client-side case wherein code should actually change based on using ClientSharedImage::GetTextureTarget() to proactively make that change if the base::Feature is enabled. For all other clients of these functions the change will be transparent - post-rollout of the Feature as a killswitch, we can simply sweep through the codebase changing all of these callsites to invoke ClientSharedImage::GetTextureTarget() directly. This CL leaves the base::Feature disabled in the event that enabling it causes problems and gets reverted. The next CL will enable it. Bug: 41494843 Change-Id: I9485905ae80dfd2530c93b225670f88db2d9089e Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5371991 Commit-Queue: Colin Blundell <[email protected]> Reviewed-by: Kentaro Hara <[email protected]> Reviewed-by: Vasiliy Telezhnikov <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273365}
-
Remove constexpr from defaulted functions.
clang now diagnoses when a constexpr is set on a default function in c++20 mode. Bug: 329313649 Change-Id: Ibf86a531a5c993a9aec71f6f83964bf530dd5736 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5370767 Reviewed-by: Florian Leimgruber <[email protected]> Commit-Queue: Nico Weber <[email protected]> Owners-Override: Nico Weber <[email protected]> Reviewed-by: Nico Weber <[email protected]> Reviewed-by: Rouslan Solomakhin <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273374}
-
Make GetRelatedWebsiteSets fail gracefully rather than crashing
Bug: 329530175 Change-Id: Ic91af1e1162094e7e461d193bdc02e713ff50418 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5370262 Commit-Queue: Shuran Huang <[email protected]> Reviewed-by: Danil Somsikov <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273375}
-
Add documentation about mocking methods with multiple move-only args.
Bug: 326047091 Change-Id: I0acbdb6da7919bed46ca0f81d34707331fa004ed Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5368248 Reviewed-by: Peter Kasting <[email protected]> Commit-Queue: Peter Kasting <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273382}
-
[base][ios] Fix the init value of g_was_hardware_model_name_called
g_was_hardware_model_name_called was introduced by https://chromium-review.googlesource.com/c/chromium/src/+/5326905 but the initial value was wrong by mistake. This CL fixes it. Bug: 327599938 Change-Id: I6efa2e33dcc5801c2573259f6ba0427f45557ded Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5374284 Commit-Queue: Peter Kasting <[email protected]> Auto-Submit: Asami Doi <[email protected]> Reviewed-by: Peter Kasting <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273383}
-
[UniversalInstall]: Add flag for backend implementation.
This should facilitate easier testing. Bug: None Change-Id: I93729421ccebeb93aae70c0fa1840ed24170cc68 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5374661 Reviewed-by: Ella Ge <[email protected]> Commit-Queue: Ella Ge <[email protected]> Auto-Submit: Finnur Thorarinsson <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273389}
-
Adjust game dynamics for more flexibility
Bug: 329674305 Change-Id: I175c97d15fd1ad41c5f6af269b5fad49d5ced9ad Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5366896 Reviewed-by: Edward Jung <[email protected]> Commit-Queue: Julian Pastarmov <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273396}
-
Add content/browser/navigation_transitions/OWNERS
[email protected] Change-Id: I0ee1443b462003e63e16399dd152932062d85d93 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5370901 Reviewed-by: Rakina Zata Amni <[email protected]> Commit-Queue: Vladimir Levin <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273399}
-
[layout] Remove BlockLengthUnresolvable in tables.
Attempt to resolve the min-length instead. Bug: 328572265 Change-Id: Iab8cc0874caf864c140b0d1e7b5ccd0fefef78e4 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5360098 Reviewed-by: David Baron <[email protected]> Commit-Queue: David Baron <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273400}
-
Fix duplicate Play/GMS Cronet providers
Bug: b:329440572 Change-Id: I6bcd914ca7d3e0525eb02c52d708e8d4a9fde2b8 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5374213 Commit-Queue: Etienne Dechamps <[email protected]> Commit-Queue: Stefano Duo <[email protected]> Auto-Submit: Etienne Dechamps <[email protected]> Reviewed-by: Stefano Duo <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273401}
-
[FedCM] Ensure that we reset FedCM state between tests
The web_test runner already resets a lot of state between tests; add FedCM to the list. This fixes fecdm-login-status-unknown.html when not run on its own. Bug: 329477540 Change-Id: I13d99ec5f95ecab1b5abd97adf2852cc151bd6f8 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5372545 Commit-Queue: Christian Biesinger <[email protected]> Reviewed-by: Peter Beverloo <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273406}
-
Reland "Add system-ui as font to substitute when font from PlatformDa…
…ta is null" This is a reland of commit 06cd1e9 Add last resort check of matched substitute font. Original change's description: > Add system-ui as font to substitute when font from PlatformData is null > > For some web fonts for which we use FreeType backend (for instance some > colour fonts), `CtFont` from `FontPlatformData` is null. For these fonts > FontCache::PlatformFallbackFontForCharacter should be still able to find > a substitute font. > Use default standard font from user settings as the font to substitute > from in `CTFontCreateUIFontForLanguage` when CT font from `PlatformData` > is null. > > Bug: 328316695 > Change-Id: I7f5b7f3340eec99ef8b4192ae0f763a9bfbbb8e1 > Cq-Include-Trybots: luci.chromium.try:mac13.arm64-blink-rel,mac13-blink-rel,mac12.0-blink-rel,mac12.0.arm64-blink-rel,mac11.0-blink-rel,mac11.0.arm64-blink-rel,mac10.15-blink-rel > Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5352176 > Reviewed-by: Dominik Röttsches <[email protected]> > Commit-Queue: Munira Tursunova <[email protected]> > Cr-Commit-Position: refs/heads/main@{#1272169} Bug: 328316695 Change-Id: I2b1f9f240530f4671c3ad3c6ef9a076942586dd8 Cq-Include-Trybots: luci.chromium.try:mac13.arm64-blink-rel,mac13-blink-rel,mac12.0-blink-rel,mac12.0.arm64-blink-rel,mac11.0-blink-rel,mac11.0.arm64-blink-rel,mac10.15-blink-rel,mac_chromium_10.15_rel_ng,mac_chromium_compile_rel_ng Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5366416 Reviewed-by: Dominik Röttsches <[email protected]> Commit-Queue: Munira Tursunova <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273408}
-
WebGPU: Add D3D shader model to adapter info developer features
Test: https://webgpureport.org/ Change-Id: I2ed2a81a0e0036a9c8e9fc1cd274dd0c4c9b73b2 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5366599 Reviewed-by: Corentin Wallez <[email protected]> Commit-Queue: Fr <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273409}
-
Remove FixedFlatSet use in BrowserUserEducationServiceBrowserTest.
MakeFixedFlatSet is only for building compile time maps. As such, there are restrictions on when it can be used: in particular, the set itself must be constructed with a constant expression, which means any variables used in the initializer must also be constexpr variables or be constant-initialized variables. A forward-declared extern variable is neither of these. Rather than trying to make the initializer a constant expression, just use base::flat_map: since this is test-only code, there is no implication on production code anyway. Bug: 41486252 Change-Id: Icca68cb000bf95223d5a8db0260e130ce36eaea8 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5372939 Reviewed-by: Mickey Burks <[email protected]> Auto-Submit: Daniel Cheng <[email protected]> Commit-Queue: Mickey Burks <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273410}
-
views: Mark InkDropContainerView as DecorativeView
InkDropContainerView is used to locate InkDrop. In the new BoxLayout algorithm, the InkDropContainerView is taken into account and the vertical axis is applied to it. Mark it as DecorativeView to skip layout. [email protected] Fixed: 329032155 Change-Id: Iec85375f0e2720c0b1b5f3af715a343288d1efd1 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5371394 Reviewed-by: Keren Zhu <[email protected]> Commit-Queue: Weidong Liu <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273411}
-
Add plumbing for GetProductInfoForUrl for webui
This is similar to the existing *ForCurrentUrl api but allows the url to be specified. With the recent caching changes that disallow queries for urls that have not been visited, this is safe to expose to web ui. Bug: b:315378610 Change-Id: If43287b782799854ae0dcccd7b38a469de47494c Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5370356 Reviewed-by: Ayman Almadhoun <[email protected]> Reviewed-by: Yue Zhang <[email protected]> Commit-Queue: Matthew Jones <[email protected]> Reviewed-by: danakj <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273412}
-
Revert "Add some debug logs to SelectActionMenuHelper."
This reverts commit 96899be. Reason for revert: Don't want these logs in stable release. Original change's description: > Add some debug logs to SelectActionMenuHelper. > > There's currently an issue where IDS_ACTIONBAR_SHARE and > IDS_ACTIONBAR_WEB_SEARCH aren't being translated properly. It's unclear > whether this is a build issue or a context issue. The bug is only > reproducible on release builds so we can submit this CL, check the > output of the logs on canary and revert the CL at the end of the week. > > Bug: 41486995 > Change-Id: I6c5fdfebd92751fd22daf795ecea5200f9e2cd98 > Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5361597 > Reviewed-by: Jinsuk Kim <[email protected]> > Commit-Queue: Alex Mitra <[email protected]> > Cr-Commit-Position: refs/heads/main@{#1272299} Bug: 41486995 Change-Id: I9cbed4bd1db5f71fa88c1ad271a0a270fd53b1fc Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5374365 Auto-Submit: Alex Mitra <[email protected]> Reviewed-by: Peter Conn <[email protected]> Commit-Queue: Jinsuk Kim <[email protected]> Reviewed-by: Jinsuk Kim <[email protected]> Bot-Commit: Rubber Stamper <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273416}
-
[A11y] Wire up Prefers Reduced Motion to the ChromeOS setting.
This uses the ChromeOS ReducedAnimations system pref to determine whether to tell animations to reduce motion. Blink uses this value, so it works in web content. BUG=b:260208248 TEST=https://webkit.org/blog-files/prefers-reduced-motion/prm.htm Change-Id: If8add8f65fa246abb049ff3f4d2037f8be00d7c8 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5361534 Reviewed-by: Anastasia Helfinstein <[email protected]> Commit-Queue: Tzarial Kuznia <[email protected]> Reviewed-by: Mitsuru Oshima <[email protected]> Reviewed-by: Sean Kau <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273417}
-
WebGL and PaintCanvasVideoRenderer: Don't discard color space (reland)
This fixes several WPT sub-tests, but the full tests still have several failures and have been marked as flakey on several platforms. Disable the two WPT tests of this functionality while we fix all of the sub-tests. Behavior change: When UNPACK_COLORSPACE_CONVERSION_WEBGL is BROWSER_DEFAULT_WEBGL, this fixes uploads from HTMLVideoElement and VideoFrame to match the WebGL spec and convert to the unpackColorSpace. If it is NONE, then preserve the existing Chromium behavior of dropping the color space information. This is not what the spec says to do but it is what all other browsers do. This will not change the behavior of using Rec709 video (far and away the most common video color space) with an unpackColorSpace of srgb (the default and also far and away the most common value). For other combinations, this is likely to change behavior. Code change: In PaintCanvasVideoRenderer::UpdateLastImage we sometimes use the right color space (e.g, when creating the SharedImage) and sometimes don't (e.g, when specifying the SkImageInfo for that image), resulting in color space information mostly being dropped on the floor. Fix this by using the same color space information in all of PaintCanvasVideoRenderer. Use the existing compatibility helper function GetVideoFrameRGBColorSpacePreferringSRGB, but move it into VideoFrame for use elsewhere. In WebGLRenderingContextBase::TexImageHelperMediaVideoFrame, fix the places where we would fail to specify the SkColorSpace in an SkImageInfo. Replace unpack_color_space_is_srgb with needs_rgb_color_conversion, because that is what is being checked. For HTMLVideoElement and VideoFrame, if UNPACK_COLORSPACE_CONVERSION_WEBGL is NONE then preserve the existing behavior of dropping the color space information. Bug: 40233404, 1341235 Change-Id: Ice25c1ab18b85e1f7acd3977bc83c2938584d9e8 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5349357 Commit-Queue: ccameron chromium <[email protected]> Reviewed-by: Philip Jägenstedt <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273423}
-
[FedCM] Don't send SameSite=Strict cookies for FedCM requests
See w3c-fedid/FedCM#320 (comment) This is behind the off-by-default "FedCmSameSiteNone" feature. Bug: 329145816 Change-Id: I6408255a01118cd5ac4d0d0263a34051796dc301 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5366009 Reviewed-by: John Abd-El-Malek <[email protected]> Reviewed-by: Philip Rogers <[email protected]> Commit-Queue: Christian Biesinger <[email protected]> Reviewed-by: Nicolás Peña <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273426}
-
[UniversalInstall]: Ensure dialog is suppressed after redirect.
When we redirect to the install/create shortcut dialog, we should make sure that the Universal Install dialog can not open as a result of the posted task being executed (so as to not show the Universal Install dialog on top of the Install dialog). Bug: None Change-Id: I66a6b40d104cf41fe38a15dd54b0d45dd74c9faa Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5374581 Reviewed-by: Glenn Hartmann <[email protected]> Commit-Queue: Finnur Thorarinsson <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273427}
-
WebGPU: Add Vulkan driver version to adapter info developer features
Test: https://webgpureport.org/ Change-Id: I3c23bad24a386fbe9dd259e0cdf30ad1d667158b Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5374310 Reviewed-by: Corentin Wallez <[email protected]> Commit-Queue: Fr <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273429}
-
time: base::StringPiece -> std::string_view
Bug: 691162 Change-Id: I0d8e77b92ec09664938da1e86944c7246ab26f05 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5353084 Reviewed-by: Peter Kasting <[email protected]> Reviewed-by: Daniel Cheng <[email protected]> Commit-Queue: Peter Kasting <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273430}
-
Disable TabGridDialogTest#testSwipeToDismiss_Dialog as it is failing …
…on android-12l-x64-dbg-tests Bug: 329707221 Change-Id: I0825927a12ba29ba21ce08ad9a3f056abe529cf2 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5374367 Reviewed-by: Igor Kraskevich <[email protected]> Owners-Override: Alex Mitra <[email protected]> Commit-Queue: Alex Mitra <[email protected]> Auto-Submit: Alex Mitra <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273432}
-
[DirectSockets] Add counters for the API
Bug: 329830245 Change-Id: Ie63208e7169bad206a379a41abd67fb6d2d0bdc3 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5372656 Reviewed-by: Simon Hangl <[email protected]> Commit-Queue: Andrew Rayskiy <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273434}
-
[ios] Add switch to go back to old form mutations observer
There were no changes done on the old observer even if the block of code looks new here. This is a kill switch to use the old mutations observer if something bad happens with the new version. Logic in sendFormMutationMessagesAfterDelay() can remain as is as the new version should be safe. Bug: 325415935 Change-Id: I99fb5f31268d6622e9adca93b1494b21c75c8dd5 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5372288 Commit-Queue: Vincent Boisselle <[email protected]> Reviewed-by: Ernesto Izquierdo Clua <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273439}
-
[iOS] Mark testDownloadDisplayedPDF as flaky.
testDownloadDisplayedPDF is intermittently failing on some bots; mariking it FLAKY. Bug: 329834941 Change-Id: Ia165d4a001d88ff99def60010844105bba53e1fa Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5372277 Reviewed-by: Vincent Boisselle <[email protected]> Auto-Submit: Mark Cogan <[email protected]> Commit-Queue: Mark Cogan <[email protected]> Cr-Commit-Position: refs/heads/main@{#1273440}
Commits on May 1, 2024
Commits on May 4, 2024
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.