Upgrade from Chromium 126 to Chromium 127 #23829

…7.0.6533.17.

Chromium change: https://chromium.googlesource.com/chromium/src/+/e3f9c0f85cb27216d031a76538c83289f2f79e7e commit e3f9c0f85cb27216d031a76538c83289f2f79e7e Author: Karol Sygiet <sygiet@google.com> Date: Wed May 15 08:21:38 2024 +0000 Remove undecryptable passwords from the login db If getting(decrypting) passwords failed and Chrome has access to encryption it means that the issue is with the data itself(eg. item might have been encrypted using a different key that is currently used). Encryption key comes from the OS and if it decided to change it there is no way for Chrome to restore it. Meaning that the item will never be decryptable again. If undecryptable passwords are present in the login database, functionality of PWM is very limited. To address this if Chrome detects that this entries exist and feature flag responsible for this is enabled, an existing function(currently only called if user starts syncing) will be called to remove them. Bug: 40286735

This feature flag has been deleted upstream and enabled by default. All references to it have been removed from brave, as this flag was being also enabled on brave. There used to be a switch that allowed disabling DoH, but this won't be supported anymore. Chromium change: https://chromium.googlesource.com/chromium/src/+/c40b0bab6871a54b741268bc0911f136c4a3ecd5 commit c40b0bab6871a54b741268bc0911f136c4a3ecd5 Author: Andrew Williams <awillia@chromium.org> Date: Tue May 14 17:49:04 2024 +0000 [DoH] Remove kDnsOverHttps feature flag Removes kDnsOverHttps and its feature params. I couldn't think of a compelling reason for why we would still need this since the functionality is launched for all platforms (since iOS doesn't use code in chrome/ AFAICT, and since we don't support Chrome for Fuchsia anymore). Also, we still have the kDnsOverHttpsUpgrade flag if we want to disable that functionality, and we can also disable the DoH auto-upgrade behavior on a per-provider basis. I don't think keeping this feature flag around as a killswitch makes sense either, since using it would have security/privacy ramifications for users that specifically opted in to the protections of DoH. Bug: b:338033843

The flags being removed by this change are now all part of `kPartitionConnectionsByNetworkIsolationKey`, which is also enabled by default, like the flags that are being deleted. Chromium change: https://chromium.googlesource.com/chromium/src/+/d7956b801d61b2b5046ed59a336645a7e4aa0b7d commit d7956b801d61b2b5046ed59a336645a7e4aa0b7d Author: Abhi Patel <abhipatel@google.com> Date: Tue May 14 01:46:45 2024 +0000 Removing NAK-related features no longer needed. Removing the following features: * SplitHostCacheByNetworkIsolationKey * PartitionHttpServerPropertiesByNetworkIsolationKey * PartitionNelAndReportingByNetworkIsolationKey * PartitionSSLSessionsByNetworkIsolationKey In most cases, the feature being removed was replaced with `PartitionConnectionsByNetworkIsolationKey`. R=abhipatel@chromium.org Bug: 338071856

`resetPageVisibilityForTesting` has substituted `setPageVisibilityForTesting`. Chromium change: https://chromium.googlesource.com/chromium/src/+/eead53722d334312dad1ff63325c0c12d706f02a commit eead53722d334312dad1ff63325c0c12d706f02a Author: dpapad <dpapad@chromium.org> Date: Tue May 14 00:52:17 2024 +0000 Settings: Add resetPageVisibilityForTesting() helper. Whenever a `loadTimeData.overrideValues('isGuest': ...)` call is made in tests the `pageVisibility` object needs to be updated, otherwise it can hold stale values. As a result of that, `routes` also held stale values, since it depends on `pageVisibility`. This is in preparation of deploying resetRouterForTesting() and resetPageVisibilityForTesting() in tests that need so. Bug: 339570198

The tab search caption button was removed from its pre-refresh location on Windows. We need to continue to ensure that it appears on the right side of the frame on Windows, after the tab strip. Chromium change: https://chromium.googlesource.com/chromium/src/+/283209314632cfebd470b57f071a8b6549060360 commit 283209314632cfebd470b57f071a8b6549060360 Author: Taylor Bergquist <tbergquist@chromium.org> Date: Tue May 14 21:13:20 2024 +0000 Clean up pre-Refresh tab search caption buttons on Windows. Also remove now-unused plumbing from BrowserNonClientFrameView. #chrome-refresh-2023-cleanup Bug: 335903961

The security team at Brave verified that dropping support for this flag is OK, as we've already stubbed out the privacy sandbox at a higher level. Chromium change: https://chromium.googlesource.com/chromium/src/+/ef5c23b961d4ea3b0dc0ad11d56c166a88372e5b commit ef5c23b961d4ea3b0dc0ad11d56c166a88372e5b Author: Alex Turner <alexmt@chromium.org> Date: Wed May 15 18:23:27 2024 +0000 Remove aggregatable reports padding feature This feature has been enabled by default since M119 and can now be removed as we should not need the capability to disable this. Bug: 40280536

Stub out PrivateAggregationManager::GetManager to achieve the same effect. Chromium change: https://chromium.googlesource.com/chromium/src/+/7dd428af03d905e4796d1b7d16a7ea6b46447c5d commit 7dd428af03d905e4796d1b7d16a7ea6b46447c5d Author: Alex Turner <alexmt@chromium.org> Date: Wed May 15 19:53:26 2024 +0000 Remove Private Aggregation context ID enhancement feature This functionality has been enabled since M119 and can now be removed, as we should not need the ability to disable this. Bug: 40281717

Chromium change: https://source.chromium.org/chromium/chromium/src/+/a132d3713b07b5e78f3c7a0d4b1b0fc9401f4214 Stop accepting fragment as raw string in SettingsLauncher API No behavior change. Before this CL, the SettingsLauncher API used to take fragments as 3 different types. 1) Fragment class 2) SettingsFragment enum 3) String for class name of the fragment 2 and 3 are used when the calling code is modularized but not the corresponding fragment. 3 is less safe than 2 (think features breaking because the fragment got renamed or changed packages). So this CL drops 3 and adopts only 2. Long term, it would be good to unify more, i.e. either - Modularize all settings fragments and only use 1. - Only use 2. But that work is beyond this CL. Change-Id: I1d2c99aa60bc0393f514a9e10d895185be399196

Chromium change: https://source.chromium.org/chromium/chromium/src/+/8607f4c52155209d87fe2beb46c680bafcb68d13 Split SearchActivityUtils part 2: isolate Client logic (and tests). There's no functional changes. The code is simply moved around. Even static method continue to be static to simplify review. Bug: 329663295

Chromium change: https://source.chromium.org/chromium/chromium/src/+/8d32d8bc17cad5c37c8e5869501e1e8e3283bc47 Android: Remove remnants of ProfileAndroid ProfileAndroid was merged with Profile, so just use Profile directly everywhere. Bug: b/339851174

@rules

Chromium change: https://source.chromium.org/chromium/chromium/src/+/64ce0d2b147c9be9cc4024a2157fe37b82fa07d7 Android: Simplify test runners by inlining hooks into //base and //content The hooks abstraction isn't necessary in these contexts, and make it harder to understand what is going on. Also removes a couple of unnecessary @rules Bug: None

This feature flag is being removed. Confirmed that the current default behavior as implemented via kLocationProviderManager is acceptable. Chromium change: https://source.chromium.org/chromium/chromium/src/+/cc04eb8474feb1d62287c0866af29351aa762186 commit cc04eb8474feb1d62287c0866af29351aa762186 Author: alvinji <alvinji@chromium.org> Date: Wed May 15 23:06:21 2024 +0000 geolocation: Add kLocationProviderManager feature flag This CL adds `kLocationProviderManager` feature flag and associated parameters to enable flexible mode configuration of the LocationProviderManager (implementation to follow). The configuration will determine which location provider is used as the source for Geolocation API. Change-Id: I98a2a25f7b05002fb6a1244fde5091a96bad2756 Bug: 333294295

Chromium change: https://source.chromium.org/chromium/chromium/src/+/68c336db7df055d4066dcb4d196ff3c699c5caee commit 68c336db7df055d4066dcb4d196ff3c699c5caee Author: Chris Fredrickson <cfredric@chromium.org> Date: Thu May 16 21:40:31 2024 +0000 Add CHECKs for 3PCD invariants in CookieSettingsBase Change-Id: Ic66622fb6fff7efb02a3154c015856f7be9cad66

…bled Chromium change: https://source.chromium.org/chromium/chromium/src/+/05baea744f4111671b9e3b9baf28f8c1727f78c5 commit 05baea744f4111671b9e3b9baf28f8c1727f78c5 Author: Alison Gale <agale@chromium.org> Date: Wed May 15 17:38:31 2024 +0000 Remove references to launched kChromeRefreshSecondary2023 feature Also cleaned up some test-only references to kChromeRefresh2023 when they were adjacent. #chrome-refresh-2023-cleanup Bug: 335904039

…3Enabled() Chromium change: https://source.chromium.org/chromium/chromium/src/+/05baea744f4111671b9e3b9baf28f8c1727f78c5 commit 05baea744f4111671b9e3b9baf28f8c1727f78c5 Author: Alison Gale <agale@chromium.org> Date: Wed May 15 17:38:31 2024 +0000 Remove references to launched kChromeRefreshSecondary2023 feature Also cleaned up some test-only references to kChromeRefresh2023 when they were adjacent. #chrome-refresh-2023-cleanup Bug: 335904039

Since `kAttributionReportingTriggerContextId` is now deleted upstream, stub out AggregatableTriggerConfig to achieve the same effect (approach verified with Brave security team). Chromium change: https://chromium.googlesource.com/chromium/src/+/b2a7eed342dce867df33aece233e838210a5f3b0 commit b2a7eed342dce867df33aece233e838210a5f3b0 Author: Nan Lin <linnan@chromium.org> Date: Mon May 20 12:45:08 2024 +0000 Remove feature kAttributionReportingTriggerContextId The feature was default enabled in M121. Change-Id: I5bd76749ac97eb1e5e9eebfc77c190f7104f1afa Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5548744 Reviewed-by: Andrew Paseltiner <apaseltiner@chromium.org> Commit-Queue: Nan Lin <linnan@chromium.org> Cr-Commit-Position: refs/heads/main@{#1303274}

This change is of no consequence for our use, as the value passed is now the default behaviour. Chromium change: https://chromium.googlesource.com/chromium/src/+/b19527f6ad4ee90e5c13ee8f34e9b8e9f9219cb8 commit b19527f6ad4ee90e5c13ee8f34e9b8e9f9219cb8 Author: Andrew Williams <awillia@chromium.org> Date: Fri May 17 14:05:14 2024 +0000 [Cookies] Make the "Block Truncated Cookies" behavior permanent This functionality has been enabled by default for 8 milestones and we haven't received any breakage reports, so we should remove the ability to opt out of this behavior. OBSOLETE_HISTOGRAM[Cookie.TruncatingCharacterInCookieString]=No longer needed since we now block truncated cookies unconditionally Bug: 40191620

The token used for subsitution to make the brave-implementation class a friend class got deleted. This change uses a new function for the substitution. Chromium change: https://chromium.googlesource.com/chromium/src/+/37faa2783f201f8dc302fc00282e3cfc21ebc3ef commit 37faa2783f201f8dc302fc00282e3cfc21ebc3ef Author: Traian Captan <tcaptan@chromium.org> Date: Mon May 20 22:33:56 2024 +0000 Fix print-only object bitmap-image loading An object element with image data that is not rendered on screen will not initially load the image. This is because of step 2 of the spec algorithm which states that "if the element is not being rendered, then jump to the step below labeled fallback", skipping over step 3.5 to actually "Fetch request". [1] If the object element is to be displayed in printing mode, the bitmap image loading will be kicked off by the post layout call back `HTMLPlugInElement::CustomStyleForLayoutObject`. This will be too late for the first print rendering, however it will render on subsequent print attempts if they are made. The proposed fix for this issue is to follow the pattern in `Document::WillPrintSoon`, which gets called before printing, to ensure that bitmap images in all object elements in the doc are loaded before printing by performing a print-mode layout that kicks off the style or layout dependent loading. `PrintRenderFrameHelper::PrintRequestedPages` is modified to also call `WillPrintSoon` so that printing from the system print dialog works as expected. This matches `PrintRenderFrameHelper::RequestPrintPreview` which calls it for the print preview path. `WebFrameTestProxy::StartTest` was modified to call WillPrintSoon` so that wpt tests for print only will load the images before dumping the print rendering pixels and match the system print dialog and print preview paths. [1]https://html.spec.whatwg.org/multipage/iframe-embed-object.html#the-object-element Bug: 41477900

Chromium change: https://source.chromium.org/chromium/chromium/src/+/949087e7c888b1fb0bdfee27c833e8e01e6e2c7b Delete tab suggestions This feature never launched. There were considerations for reusing this code for declutter, but that is no longer relevant. See: https://chromium-review.googlesource.com/c/chromium/src/+/5539862 Delete everything and cleanup tests, histograms, resources. OBSOLETE_HISTOGRAMS=Tab suggestions feature was removed. Bug: 341296315

Chromium change: https://source.chromium.org/chromium/chromium/src/+/c83f07b9fd2335e108fb151cbab2493877d924f8 commit c83f07b9fd2335e108fb151cbab2493877d924f8 Author: Nan Lin <linnan@chromium.org> Date: Mon Apr 8 00:17:38 2024 +0000 Refactor GetNullAggregatableReports This CL moves non-randomized logic to //components/attribution_reporting to allow better test coverage. This also allows supporting null aggregatable reports in interop tests in a follow up CL. Bug: 332589874

The UmaHistogramEnumeration functions perform static assertion checks on their types using uintmax_t, but we've added to the types using negative numbers so we need to allow for those. Chromium change: https://source.chromium.org/chromium/chromium/src/+/1be087741feee0d0558405ffb0ac12bec48059ec commit 1be087741feee0d0558405ffb0ac12bec48059ec Author: Brian White <bcwhite@chromium.org> Date: Tue May 28 18:51:44 2019 +0000 Have histogram-functions take native char* names. Without this, the compiler emits ctor/dtor code to convert constant strings to std::string making the call about 100x larger in size and almost 2x the size of a caching macro instantiation. According to the Android binary-size try-bot, saves about 4K... Specifics: -4,096 bytes main lib size -4,738 bytes normalized apk size InstallSize: -4,122 bytes APK size -4,122 bytes Estimated installed size (Android Go) -4,122 bytes Estimated installed size InstallBreakdown (-4,066 bytes): +28 bytes unwind_cfi (dev and canary only) size -4,096 bytes Native code size +2 bytes Package metadata size Bug: 94484

The base class assumes that this feature is enabled and ends up dereferencing a null pointer when it isn't enabled. Chromium change: https://source.chromium.org/chromium/chromium/src/+/34d560511d3ee6c870e91d68e0b103bfe8e435fc commit 888646e7a0766ac48b215e5b852746b84a10071f Author: Joseph Park <josephjoopark@chromium.org> Date: Mon May 20 16:24:32 2024 +0000 Add Translate action item Bug: 323962066

`NOTREACHED()` has been marked as `[[noreturn]]`, and therefore will break in many cases when building the code. The migration variant of the macro is being used upstream to allow for a gradual migration of each callsite. Chromium change: https://chromium.googlesource.com/chromium/src/+/61831cb39155842c781f63214a7862bfd23e3390 commit 61831cb39155842c781f63214a7862bfd23e3390 Author: Peter Boström <pbos@chromium.org> Date: Wed May 22 21:32:53 2024 +0000 Reland "Make NOTREACHED() without parameters [[noreturn]]" This is a reland of commit fdeac78ccd92e2cc578d8cfa04219599520138b4 Original change's description: > Make NOTREACHED() without parameters [[noreturn]] > > Previous callers have been migrated to NOTREACHED_IN_MIGRATION(). This > also renames the few NOTREACHED_NORETURN() callers in base/ to > NOTREACHED() to verify that it's still working (no missing-return). > > Let's see how many times this needs to be reverted and relanded. :) > > Bug: 40580068 > Change-Id: Ib8155e461ea9d0853c9508e28fc45389114da48c > Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5548908 > Code-Coverage: findit-for-me@appspot.gserviceaccount.com <findit-for-me@appspot.gserviceaccount.com> > Commit-Queue: Lei Zhang <thestig@chromium.org> > Owners-Override: Lei Zhang <thestig@chromium.org> > Auto-Submit: Peter Boström <pbos@chromium.org> > Commit-Queue: Peter Boström <pbos@chromium.org> > Reviewed-by: Lei Zhang <thestig@chromium.org> > Cr-Commit-Position: refs/heads/main@{#1303439} Bug: 40580068

This feature flag has been deleted upstream but all code it was guarding was also being guarded by `kAggregationServiceMultipleCloudProviders`, which is still being disabled by default. Chromium change: https://chromium.googlesource.com/chromium/src/+/503d03785d45cfa5b13e7080f781702e9a7f29e9 commit 503d03785d45cfa5b13e7080f781702e9a7f29e9 Author: Alex Turner <alexmt@chromium.org> Date: Wed May 22 15:08:46 2024 +0000 Remove Private Aggregation multi-cloud feature This feature has been enabled by default since M121 and can now be removed as we should not need the capability to disable it. Note that the aggregation service multi-cloud feature is kept as the feature has a FeatureParam that's still in use. Bug: 40281610

This change adjusts the way `SearchEngineChoiceService` instances are being created during tests, to use either the browser factory for it in tests, or to use the global browser local state, as local state is now required to be provided during instantiation. Chromium change: https://chromium.googlesource.com/chromium/src/+/411d1c6892720ae439c253236640569fbc1039c5 commit 411d1c6892720ae439c253236640569fbc1039c5 Author: Jack Yammine <jyammine@google.com> Date: Wed May 22 14:22:08 2024 +0000 [waffle]Fix crash caused by passing the wrong pref service to a function https://crrev.com/c/5538557 uses the `IsMetricsReportingEnabled` function by passing the profile pref service to it while the local pref service should be passed. Enable test that was disabled because of this issue. Fixed: 331645729, b:341643240

Chromium change: https://source.chromium.org/chromium/chromium/src/+/d96f6e3da6fd136dce36911d1fe1140c42f3fdfb Break down library_loader and suppliers into their own target Broke down `library_loader` and `suppliers` java classes into their own build target which resulted in breaking down its dependencies as well. The goal is to remove Cronet's class deletion logic after compiling https://source.chromium.org/chromium/chromium/src/+/main:components/cronet/android/BUILD.gn;l=935;drc=c039aed598121229887f47d965638d89d41c487b;bpv=0;bpt=0. Java classes were broken down into targets where relevant classes live in the same target, this was done on a best-effort basis and some might not be the most optimal structure. Consumers can create their own `java_group` and create their own set of dependencies as they wish. The reason is that adding unrelated classes to the same target leads to irregular dependency such as `tasks_java` in which a java package is depending on classes outside its package and everything is one big cyclic dependency. Changed the jni headers as they are reliant on the target name. Removed the dependency between ContextUtil and ApiHelper classes as it is trivial to do so. Bug: 326512877

…Coordinator Chromium change: https://source.chromium.org/chromium/chromium/src/+/42b4616046112d454bb06bc82a872b625d483b36 Remove getLastUsedRegularProfile from TabSwitcherActionMenuCoordinator. Bug: 40254448

Disable the IsolateSandboxedIframes feature flag as it was affecting chrome-untrusted://line-chart-display Wallet browser tests. Entered brave/brave-browser#38771 to address this in a follow-up. Chromium change: https://source.chromium.org/chromium/chromium/src/+/b2c49e9ef800bc606041fc9356e665d11869cb30 commit b2c49e9ef800bc606041fc9356e665d11869cb30 Author: W. James MacLean <wjmaclean@chromium.org> Date: Wed May 15 19:54:12 2024 +0000 Enable IsolateSandboxedIframes by default. This CL enables the IsolateSandboxedIframes feature by default. Bug: 40082497

Now that IsChromeRefreshEnabled() returns true, the toolbar always has three children one of which is the container_view. Chromium change: https://source.chromium.org/chromium/chromium/src/+/05baea744f4111671b9e3b9baf28f8c1727f78c5 commit 05baea744f4111671b9e3b9baf28f8c1727f78c5 Author: Alison Gale <agale@chromium.org> Date: Wed May 15 17:38:31 2024 +0000 Remove references to launched kChromeRefreshSecondary2023 feature Also cleaned up some test-only references to kChromeRefresh2023 when they were adjacent. #chrome-refresh-2023-cleanup Bug: 335904039

Chromium change: https://chromium.googlesource.com/chromium/src/+/11e9b39a46f3986257c86a0914b7ca4ef3ab6ed6 commit 11e9b39a46f3986257c86a0914b7ca4ef3ab6ed6 Author: Caroline Rising <corising@chromium.org> Date: Thu May 23 22:32:01 2024 +0000 Clean up SidePanelPinning flag and unused code from core side panel logic This deletes the side panel combobox, side panel toolbar button, and side panel toolbar container that are no longer used since side panel pinning launched to stable. Still to do: fix tests, remove other uses of SidePanelPinning flag, remove now unused parts of SidePanelEntry (title, icon, etc) which are replaced by ActionItem data. Bug: 318835260, 318834948, 318835911

This change corrects a few build failures relating to changes in the underlying inclusion files that result in a given override being broken.

This function has been removed as `RenderFrame` already has an appropriate way to retrieve this information. Chromium change: https://chromium.googlesource.com/chromium/src/+/12bfde02d4db8717c763cc0be3e3f7d7c0248bc2 commit 12bfde02d4db8717c763cc0be3e3f7d7c0248bc2 Author: Christoph Schwering <schwering@google.com> Date: Fri May 24 14:47:25 2024 +0000 [Autofill] Remove RenderFrame::ElementBoundsInWindow() RenderFrame::ElementBoundsInWindow() is a special case of RenderFrame::ConvertViewportToWindow(). Autofill is the only caller of both functions. This CL - removes ElementBoundsInWindow() - converts the out-parameter of ConvertViewportToWindow() into a return type. Bug: 40100455

This feature flag is now called `kTrackingProtectionContentSettingUbControl`, as some other similarly grouped flags are being added. Chromium change: https://chromium.googlesource.com/chromium/src/+/81e61608f94b70a4f1d228c832a8e27b91ac1705 commit 81e61608f94b70a4f1d228c832a8e27b91ac1705 Author: Fiona Macintosh <fmacintosh@google.com> Date: Thu May 23 19:35:30 2024 +0000 Rename kTrackingProtectionContentSetting to be specific to UserBypass Change-Id: Ib181791efd3e9d1aa5441c1a8464df1dd63a6f27

This change only affects the way we refer to the gn target path. Chromium change: https://chromium.googlesource.com/chromium/src/+/4b76feb60f21d832821abd809bedb0b49b816ccc commit 4b76feb60f21d832821abd809bedb0b49b816ccc Author: Erik Chen <erikchen@chromium.org> Date: Tue May 28 10:41:25 2024 +0000 Modularize //chrome/browser/themes This CL has no intended behavior change. This CL moves all files in //chrome/browser/themes into a standalone BUILD.gn file. There is 1 circular dependency onto //chrome/browser/extensions, expressed as an allow_circular_includes_from inside of //chrome/browser/extensions/BUILD.gn. This CL also: * Moves a small bit of logic into ThemeInstalledInfoBarDelegate to finish encapsulating dependencies onto //chrome/browser/ui. * Fixes some IWYU issues. * Modularizes //chrome/browser/ui/frame/. Change-Id: I518d8c5fa0b2401899a6b95c19b05c22e13f68fe Bug: 343037853, 341594886

This only affects the inclusion of a single header. Chromium change: https://chromium.googlesource.com/chromium/src/+/74660652ad627e748c7a871d3227791879c47514 commit 74660652ad627e748c7a871d3227791879c47514 Author: Kartar Singh <kartarsingh@google.com> Date: Wed May 29 09:34:30 2024 +0000 Setup //components/input and move web_keyboard_event related code there. This CL sets up a new component `input` which can be depended upon by browser and viz. The requirement is part of InputVizard project[1] where touch input handling is being moved to Viz on Android. Earlier we had been refactoring input related bits from browser classes and placing them in //content/common/input but since Viz cannot depend on content the input code is being moved into a component of it's own[2]. Apart from setting up the component we are moving web_keyboard_event related code in the component. In the future CLs we will be moving more input related code here. P.S. `$ git cl format` has re-ordered some of the headers, they weren't intentionally re-ordered. [1] https://docs.google.com/document/d/1mcydbkgFCO_TT9NuFE962L8PLJWT2XOfXUAPO88VuKE [2] Google internal discussion thread g/chrome-input/wWPmYmVkFWY/IqCZNJr-AAAJ Bug: b/340182114

This function has been moved into a new file, and a corresponding overriding file must be created to re-apply the substitution. Chromium change: https://chromium.googlesource.com/chromium/src/+/99defa39a5dee2988c8f0ba87fef142b41a625ed commit 99defa39a5dee2988c8f0ba87fef142b41a625ed Author: rbpotter <rbpotter@chromium.org> Date: Tue May 28 20:04:07 2024 +0000 What's New: Remove page on ChromeOS Due to differences in the ChromeOS update process (specifically, browser and OS updating together) vs other platforms, What's New will not be launching on ChromeOS any time soon. Instead of including all the page code on ChromeOS but having the page disabled there, remove most of the code on ChromeOS platforms. Leave limited functions in whats_new_util to determine e.g. whether the page is enabled, to avoid excessive #if statements in code outside of c/b/ui/webui/whats_new. This CL refactors code from whats_new_util to whats_new_fetcher, so that the latter can hold network code that is not needed on CrOS platforms. Code making direct network requests is stubbed out in tests as this causes tests to flake (i.e. hard to test). Low-Coverage-Reason: HARD_TO_TEST Bug: 339711041

…tings Chromium change: https://source.chromium.org/chromium/chromium/src/+/f1c26c11b4c54f337d667b20d33d7cee2dcb92c9 commit f1c26c11b4c54f337d667b20d33d7cee2dcb92c9 Author: Florian Leimgruber <fleimgruber@google.com> Date: Thu May 23 10:28:29 2024 +0000 Add PLUS_ADDRESS_SETTING model type The model type mirrors PLUS_ADDRESS. In particular, as described in the design doc, the model type is not tied to any user selectable type. Bug: b/342089839

Chromium change: https://source.chromium.org/chromium/chromium/src/+/337f5100c5a83dd193b52a64debac0f4c5bd628c commit 337f5100c5a83dd193b52a64debac0f4c5bd628c Author: Rainhard Findling <rainhard@chromium.org> Date: Tue May 21 17:59:58 2024 +0000 CBD dialog: extract clearButton to element interface This button is used often in tests. Extract it to the element interface to simplify the code required to use this button. Bug: 40283307

…erification methods Chromium change: https://source.chromium.org/chromium/chromium/src/+/bdc8576287c548b3d131437532794cb980b792c4 Remove default implementations of SafetyNetApiHandler app verification methods The default implementations were added so we could land implementations in clank internal without breaking the build. That's now done, so we no longer need the defaults. Bug: 341790041

Chromium change: https://source.chromium.org/chromium/chromium/src/+/f46caa514f421ef199f1370042147b309728fc8f JNI Zero: Require @JniType conversions functions to appear in .h files This has the following advantages: * Missing conversions will be a compile-time error rather than link-time. * Allows us to define a default implementations that maps a pointer type to its dereferenced type. * Allows us to define a default implementations for std::optional<T>. Bug: 334905258

Chromium change: https://source.chromium.org/chromium/chromium/src/+/d723ecf9b19de66f0cff1aa2758d8d16682676e1 commit d723ecf9b19de66f0cff1aa2758d8d16682676e1 Author: weidongliu <weidongliu@chromium.org> Date: Wed May 22 13:38:37 2024 +0000 part2: GetHeightForWidth default uses CalculatePreferredSize(SizeBounds) In order to ensure implementation consistency and avoid oversights, we should unify CalculatePreferredSize() and GetHeightForWidth(). Replace them with the CalculatePreferredSize(SizeBounds) overload. third_party_cookies_label_wrapper_ Use the SetPreferredSize API to set the preferred width. But this causes problems with the migration of GetHeightForWidth. Bug: 40232718

Chromium change: https://source.chromium.org/chromium/chromium/src/+/4164a6fe77902aac286464c517dc20931ea5571d commit 4164a6fe77902aac286464c517dc20931ea5571d Author: Junji Watanabe <jwata@google.com> Date: Fri May 24 03:33:49 2024 +0000 build: Rename rbe_* GN args/variables with reclient_* This CL renames the following GN args/variables. - rbe_bin_dir -> reclient_bin_dir - rbe_cfg_dir -> reclient_cfg_dir - rbe_cros_cc_wrapper -> reclient_cros_cc_wrapper - rbe_*_cfg_file -> reclient_*_cfg_file When an old GN arg is specified, it will print a warning message. Bug: 342270134

This type has now to be instantiated with the `.type` argument being provided during construction, as the most approppriate ctor variant. Chromium change: https://chromium.googlesource.com/chromium/src/+/8ba69cd6d9ae8900d32a67dabb1e36b14e9f5273 commit 8ba69cd6d9ae8900d32a67dabb1e36b14e9f5273 Author: Dirk Pranke <dpranke@google.com> Date: Thu May 30 22:12:14 2024 +0000 Remove 0-arg (default) constructor for views::Widget::InitParams. We are refactoring Views to change the default ownership model of Widgets from NATIVE_WIDGET_OWNS_WIDGET to CLIENT_OWNS_WIDGET. One of the first steps in this path is to ensure that every user is specifying the ownership directly, so we can then target them for changes more easily. At this point, I think all of the last usages are gone. This CL removes the default constructor, ensuring that users will have to specify the ownership type going forward. Eventually we'll probably change the default to CLIENT_OWNS_WIDGET, but we probably won't do that for quite a while, when we are fairly confident that CLIENT_OWNS_WIDGET is stable. Bug: 3396919005

Chromium change: https://chromium.googlesource.com/chromium/src/+/1a3a63173de17e612af8c0db7367cf39b9a0be18 commit 1a3a63173de17e612af8c0db7367cf39b9a0be18 Author: Alison Gale <agale@chromium.org> Date: Fri May 31 16:35:27 2024 +0000 Delete kChromeRefresh2023 and kChromeRefreshSecondary2023 flags #chrome-refresh-2023-cleanup Fixed: 335903780

Chromium change: https://chromium.googlesource.com/chromium/src/+/d97c5a73e614939c32a7fd4c93ea6f37033bd7d6 commit 1a3a63173de17e612af8c0db7367cf39b9a0be18 Author: Alison Gale <agale@chromium.org> Date: Fri May 31 16:35:27 2024 +0000 Delete kChromeRefresh2023 and kChromeRefreshSecondary2023 flags #chrome-refresh-2023-cleanup Fixed: 335903780

This change is of no consequence for the existing override, as it was not making use of these extra default args. Chromium change: https://chromium.googlesource.com/chromium/src/+/d603973306633a07c602428c1f786cb54c7f9b87 commit d603973306633a07c602428c1f786cb54c7f9b87 Author: Nicolas Dossou-Gbete <dgn@chromium.org> Date: Fri May 31 15:11:25 2024 +0000 🧇 Stop populating the current DSE in choice screen list This is now irrelevant since we only show the choice screen when Google is the DSE, and it is in the prepopulated lists already. This code was intended in the past to allow including custom search engines in the list, but now adds unnecessary complexity OBSOLETE_HISTOGRAM[Search.ChoiceDebug.IsDefaultProviderAddedToChoices]=Non-prepopulated engines support removed. Bug: b:343357518

These overrides and the associated unit test are no longer relevant now that ChromeRefresh2023 is active. Chromium change: https://chromium.googlesource.com/chromium/src/+/e5f1b76c9cb27f5972a1d414596e69030e40f6e9 commit e5f1b76c9cb27f5972a1d414596e69030e40f6e9 Author: stluong <stluong@chromium.org> Date: Tue Jun 4 19:17:09 2024 +0000 Remove IsChromeRefresh2023() from c/b/ui/views/tabs #chrome-refresh-2023-cleanup Bug: 335903961

Upstream changes were affecting the layout of BraveTooltipView, obscuring the button container. Chromium change: https://source.chromium.org/chromium/chromium/src/+/75dcfb263c5e6766f4993c30cead306bb114b312 commit 75dcfb263c5e6766f4993c30cead306bb114b312 Author: weidongliu <weidongliu@chromium.org> Date: Fri May 3 01:54:10 2024 +0000 use CalculatePreferredSize(SizeBounds) in Views. In order to ensure implementation consistency and avoid oversights, we should unify CalculatePreferredSize() and GetHeightForWidth(). Replace them with the CalculatePreferredSize(SizeBounds) overload. Clean up previously missed and recently added. Bug: 40232718

Chromium change: https://source.chromium.org/chromium/chromium/src/+/4be096753311e6a49438896c8fd8b47d31a941d6 commit 4be096753311e6a49438896c8fd8b47d31a941d6 Author: Kristian Monsen <kristianm@google.com> Date: Fri May 31 06:24:43 2024 +0000 Send registration requests Implementing sending registration requests when starting a new session. Bug: 41495201

Chromium change: https://source.chromium.org/chromium/chromium/src/+/6f978d9d8ecc89a8bed761e1b15a7b8d6d4b34a1 commit 6f978d9d8ecc89a8bed761e1b15a7b8d6d4b34a1 Author: danakj <danakj@chromium.org> Date: Mon Jun 3 17:05:51 2024 +0000 Make base/hash bounds-safe wrt unsafe-buffer-usage Remove all #pragma allow_unsafe_buffers in base/hash/ and fix the corresponding errors. Mark deprecated pointer+size functions as UNSAFE_BUFFER_USAGE and convert places that have warnings enabled to the span version. NaCl code does not warn and is excluded from this change. It's not clear how the code is unit tested, so leaving it to be done separately if at all. It will eventually all be deprecated and removed. Also remove dead code in md5_constexpr_unittest.cc Change-Id: I8c17aa89981aa84abd71a76f8492da3e9a71cc78

Chromium change: https://chromium.googlesource.com/chromium/src/+/8648eb0f5be751114ae8eb44416b049257862ade commit 8648eb0f5be751114ae8eb44416b049257862ade Author: Alison Gale <agale@chromium.org> Date: Wed May 29 21:43:24 2024 +0000 Remove references to IsChromeRefresh2023 in chrome/browser/ui/tabs #chrome-refresh-2023-cleanup Bug: 335903961

Chromium change: https://chromium.googlesource.com/chromium/src/+/d97c5a73e614939c32a7fd4c93ea6f37033bd7d6 commit d97c5a73e614939c32a7fd4c93ea6f37033bd7d6 Author: stluong <stluong@chromium.org> Date: Sat Jun 1 00:11:57 2024 +0000 Remove Chrome Refresh Flags from Tab Groups #chrome-refresh-2023 Low-Coverage-Reason: LARGE_SCALE_REFACTOR This CL is part of Chrome Refresh flag clean up and isn't introducing new logic. Bug: 335903961

The non-chromium-refresh-2023 path for LayoutProvider::GetCornerRadiusMetric has been deleted via https://crrev.com/c/5546799 Fixed by setting our radius directly instead of querying to GetCornerRadiusMetric() as kOmniboxExpandedRadius is mapped to more general token ShapeSysTokens::kMedium. Chromium change: https://chromium.googlesource.com/chromium/src/+/99fe5ddf4ecd908fd52d1d03565176f2f481c79e commit 99fe5ddf4ecd908fd52d1d03565176f2f481c79e Author: Dirk Pranke <dpranke@google.com> Date: Fri May 17 21:33:00 2024 +0000 Remove non-ChromeRefresh2023 code paths from some //ui/views files. The 2023 Chrome Refresh UI is on by default now and we won't be turning it off again, so it's safe to start deleting code paths that are now dead. This CL removes the //ui/views/ references that are not in //ui/views/controls (a different CL will remove those). Bug: 326156510

… GetCookieSettingInternal Force GetCookieSetting to use a first-party context when calling into GetCookieSettingInternal, rather than assuming a third-party context. Chromium change: https://source.chromium.org/chromium/chromium/src/+/1eca8080b750b1b3e3067cfb7209163b9026de8a commit 1eca8080b750b1b3e3067cfb7209163b9026de8a Author: Chris Fredrickson <cfredric@chromium.org> Date: Fri May 31 15:00:57 2024 +0000 Rewrite GetCookieSetting to stop assuming a same-site context Of the 3 non-test callers, one creates a synthetic first-party context, one creates a synthetic third-party context, and one is ambiguous. This updates the first-party and ambiguous callers to use a new override that accepts a SiteForCookies explicitly; and updates the existing method to assume a third-party context, rather than a first-party context. Bug: 343827271

Chromium change: https://source.chromium.org/chromium/chromium/src/+/9a6b6c3edba4305b3a31660c0eba7454d1084209 Reland "[android] Enable shipping both snapshots on android" This is a reland of commit dbe657ecce09220eee9c16d8cfd2e585349ca368 Relanding without changes, the android-x86 build issues are fixed in https://chromium-review.googlesource.com/c/chromium/src/+/5526637 Original change's description: > [android] Enable shipping both snapshots on android > > Reland of crrev.com/c/3219567. We want to experiment with this again, > potentially limiting it to high-end Android only (depending on impact). > > Binary-Size: Increase is temporary for experiment. > Bug: 40200623, 40539769 > Change-Id: Ide7e11ab9b9c1fcf9c42cb97df9e831dbd0cb6e7 > Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5518514 > Auto-Submit: Leszek Swirski <leszeks@chromium.org> > Commit-Queue: Andrew Grieve <agrieve@chromium.org> > Reviewed-by: Andrew Grieve <agrieve@chromium.org> > Cr-Commit-Position: refs/heads/main@{#1297053} Binary-Size: Increase is temporary for experiment. Bug: 40200623, 40539769

Chromium change: https://source.chromium.org/chromium/chromium/src/+/d864e62dddc14e305f0a274f6058ab082ad2b99a [SurfacePolish][CleanUp] Clean up surface polish code for MVT (Part 5) This CL represents part 5 of the cleanup for the surface polish code in the MVT, focusing on refining the code associated most visited tiles layout. The update removes the grid layout from most visited tiles and renames the most visited tiles carousel layout to simply most visited tiles layout. Bug: 331667743

…kground Now that ChromeRefresh2023 is active, ProfileMenuViewBase uses identity_info_color_callback_ to draw its background which results in unwanted changes to the menu UI. This commit switches us back to continue using the profile_background_container_ to set the background as was done pre-refresh. Chromium change: https://source.chromium.org/chromium/chromium/src/+/a8756e741953659ac2fe3feeaed08a2aafe75d46 commit a8756e741953659ac2fe3feeaed08a2aafe75d46 Author: Joseph Park <josephjoopark@chromium.org> Date: Mon Jun 3 17:39:05 2024 +0000 [chrome-refresh-2023-cleanup] Clean up GM3 flags under cbui/profiles/ Bug: 335903961

As close button has more padding, it seems favicon is too close to the left edge of the tab left border comppared with close button. Give additional left padding to make both visible with same space from tab border. See https://www.github.com/brave/brave-browser/issues/30469. BraveTab::GetInsets() is updated as upstream Tab::extra_padding_before_content_ is deleted.

This flag had already been gutted in upstream and was entirely no-op. Chromium change: https://chromium.googlesource.com/chromium/src/+/7307ce05e3efcaaefa6273f1ec0cb31769c520f9 commit 7307ce05e3efcaaefa6273f1ec0cb31769c520f9 Author: Yi Gu <yigu@chromium.org> Date: Wed Jun 5 15:23:30 2024 +0000 [FedCM] Remove unused flag FedCmWithoutThirdPartyCookies Bug: None

The previous variant for `GetCookieSetting` has been removed, and now only the second variant that requires `SiteForCookies` can be used. This change makes use of this new variant, while passing a default value to the new argument. Chromium change: https://chromium.googlesource.com/chromium/src/+/a2def2379534e95558ba0c54dd7efb0582a898e1 commit a2def2379534e95558ba0c54dd7efb0582a898e1 Author: Chris Fredrickson <cfredric@chromium.org> Date: Wed Jun 5 19:20:27 2024 +0000 Remove uses of deprecated GetCookieSettings override This override uses an implicit SiteForCookies value. Really, the caller should be responsible for supplying this value, to avoid CookieSettingsBase making incorrect assumptions about whether the context is same-site or cross-site. This CL was uploaded by git cl split. Fixed: 343827271

These functions don't need to be mocked in the context menu mocks anymore, as the item has been removed. Chromium change: https://chromium.googlesource.com/chromium/src/+/ceb042766ebe602dd8c40f135110a4cc71c8b298 commit ceb042766ebe602dd8c40f135110a4cc71c8b298 Author: Ramin Halavati <rhalavati@chromium.org> Date: Wed Jun 5 05:12:39 2024 +0000 Remove PDF OCR context menu. PDF OCR will be always enabled and the context menu is no more needed. AX-Relnotes: PDF OCR context menu is removed. Bug: 341856144

Chromium change: https://source.chromium.org/chromium/chromium/src/+/0ba68daa433fbaf91e18225d6c84c35208d8bceb commit 0ba68daa433fbaf91e18225d6c84c35208d8bceb Author: Will Harris <wfh@chromium.org> Date: Tue Jun 4 21:51:01 2024 +0000 Remove deprecated app-bound metrics implementation These metrics are logged by the App-Bound key provider, which is now enabled by default, so do not need their own separate logging. This CL moves the extended logging as well as the OSCrypt.AppBoundEncryption.SupportLevel metric to the provider. All other histograms previously logged are removed, along with the prefs. OBSOLETE_HISTOGRAMS=OSCrypt.AppBoundEncryption.PathValidation.* superseded by OSCrypt.AppBoundProvider.* BUG=40227925

Chromium change: https://source.chromium.org/chromium/chromium/src/+/3e435173f223d45b8b117c8c0de554bd22cfd3f3 commit 3e435173f223d45b8b117c8c0de554bd22cfd3f3 Author: Shibalik Mohapatra <shibalik@chromium.org> Date: Tue May 21 19:03:36 2024 +0000 [chrome-refresh-2023-cleanup] cleanup for toolbar directory 1/n. #chrome-refresh-2023-cleanup Change-Id: I40f7e848ac379116fab02f0b96fb56ac2766f1ac

This feature causes UI issues with bottom controls. Chromium change: https://source.chromium.org/chromium/chromium/src/+/ed4add002d0322cb110f30ce41bdd58145a2bee6 [Tab Groups] Launch Android Tab Group Stable IDs Launch AndroidTabGroupStableIds to 100%. See launch/4309143 for metrics. Bug: 340916880

Chromium change: https://source.chromium.org/chromium/chromium/src/+/d864e62dddc14e305f0a274f6058ab082ad2b99a [SurfacePolish][CleanUp] Clean up surface polish code for MVT (Part 5) This CL represents part 5 of the cleanup for the surface polish code in the MVT, focusing on refining the code associated most visited tiles layout. The update removes the grid layout from most visited tiles and renames the most visited tiles carousel layout to simply most visited tiles layout. Bug: 331667743

Chromium change: https://source.chromium.org/chromium/chromium/src/+/085bda27b9f5c36b81ea18d0574bb51dffe12846 commit 085bda27b9f5c36b81ea18d0574bb51dffe12846 Author: Alison Gale <agale@chromium.org> Date: Mon May 20 18:48:29 2024 +0000 Remove uses of IsChromeRefresh2023 from browser/ui/color #chrome-refresh-2023-cleanup Bug: 335903961

This feature is related to Chromium's NTP, but we have its own implementaion. Chromium change: https://source.chromium.org/chromium/chromium/src/+/6c5b63627d835e9fede52f137482583edaa2baac [MagicStack] Enable Magic Stack and Price change module by default. In this CL: 1. we clean up three feature flags: MAGIC_STACK_ANDROID, PRICE_CHANGE_MODULE and SegmentationPlatformAndroidHomeModuleRanker. 2. Clean up tests for showing single tab cards, and replaced with showing the magic stack. Bug: 41485537

As upstream started to use browser action for each side panel, our panel items also should have their panel action items.

This is merely a path change, which does affect a few inclusion paths and patches. Chromium change: https://chromium.googlesource.com/chromium/src/+/f4096571c202d7aa7e5c9d9b8dc9edebbbeb631e commit f4096571c202d7aa7e5c9d9b8dc9edebbbeb631e Author: Erik Chen <erikchen@chromium.org> Date: Fri Jun 7 05:33:08 2024 +0000 Move sidepanel C++ code into a single directory. This CL is a refactor with no intended behavior change. This CL was generated with 4 steps: (1) deleting redundant OWNER files (2) moving all files from chrome/browser/ui/side_panel to chrome/browser/ui/views/side_panel and updating BUILD.gn files. (3) Updating all includes with the command: git grep -l 'chrome\/browser\/ui\/side_panel' | xargs sed -i \ 's/chrome\/browser\/ui\/side_panel/chrome\/browser\/ui\/views\/side_panel/g' (4) Updating all header preprocessor macros with the command: git grep -l 'CHROME_BROWSER_UI_SIDE_PANEL' | xargs sed -i \ 's/CHROME_BROWSER_UI_SIDE_PANEL/CHROME_BROWSER_UI_VIEWS_SIDE_PANEL/g' Bug: 345499793

Chromium change: https://chromium.googlesource.com/chromium/src/+/1ce8e544967fc8b9f3a0fa0589f43417c83b1216 commit 1ce8e544967fc8b9f3a0fa0589f43417c83b1216 Author: Caroline Rising <corising@chromium.org> Date: Thu Jun 6 14:39:21 2024 +0000 Remove all references to side panel pinning flag This flag is being removed since it has been launched and enabled by default for almost 2 months. Bug: 318835260

The only place affected has been replaced with a check that seems to be the equivalent of what was being done, which was checking the tab count. Chromium change: https://chromium.googlesource.com/chromium/src/+/f29b089cb45db2917edafd595acbe6a5b1423225 commit f29b089cb45db2917edafd595acbe6a5b1423225 Author: Shibalik Mohapatra <shibalik@chromium.org> Date: Fri Jun 7 16:38:00 2024 +0000 [TabStripModelRefactor] Add interface for contents data. Change-Id: Iff6cbca8ec71958e8125bd1c713a64bd4e6c948c

Per Simon, this change doesn't affect us. Chromium change: https://chromium.googlesource.com/chromium/src/+/106c77b970f3d050b24521df3bab67aef382145a commit 106c77b970f3d050b24521df3bab67aef382145a Author: Caroline Rising <corising@chromium.org> Date: Thu Jun 6 16:32:16 2024 +0000 Remove name/icon from SidePanelEntry and remove unneeded side panel content wrapper. Name and icon are no longer needed in SidePanelEntry since the name/icon from the feature's ActionItem is used instead after the side panel pinning feature launch. Chrome refresh 2023 also moved the side panel header into the border already so the side panel's content wrapper is no longer needed. Bug: 318835260

This simplifies the way strings are passed into `Extecute` and other methods. Chromium change: https://chromium.googlesource.com/chromium/src/+/e63b8343567073e5dee4285a0b59549a1589e4fb commit e63b8343567073e5dee4285a0b59549a1589e4fb Author: Evan Stade <estade@chromium.org> Date: Fri Jun 7 15:55:23 2024 +0000 Sql: change SQL script parameter type from char* to cstring_view And some other related updates. GetUniqueStatement is updated to take a base::cstring_view, which is a null-terminated string that can be either a char[] or std::string. GetCachedStatement should very rarely be dynamically constructed, as it should almost always be statically known. Thus it has less need to take a `std::string`. However, taking a `char*` has not prevented callers from using a `string` with `c_str()`, whether or not the string was necessary or appropriate. Therefore we might as well update GetCachedStatement as well for consistency. Bug: 341639215

Chromium change: https://chromium.googlesource.com/chromium/src/+/a484a844ce3a9a8377fbce9ed9ca29c1b2455153 commit a484a844ce3a9a8377fbce9ed9ca29c1b2455153 Author: Shibalik Mohapatra <shibalik@chromium.org> Date: Thu Jun 6 22:22:00 2024 +0000 [TabStripModelRefactor] Add method for moving multiple tabs to simplify MoveSelectedTo and MoveGroupTo without breaking group contiguity. Change-Id: I1c5884e6c0346b066b251a1e0a351fe5c2d70470

Chromium change: https://chromium.googlesource.com/chromium/src/+/81dee0f4f06b4015b8d9176f1adcd13f6b3bd96f commit 81dee0f4f06b4015b8d9176f1adcd13f6b3bd96f Author: Taylor Bergquist <tbergquist@chromium.org> Date: Fri Jun 7 00:31:30 2024 +0000 Clean up IsChromeRefresh2023 in c/b/ui/view/frame/ This includes removing quite a bit of dead code calculating the distance between the top of the window and the top of the tabstrip. Prior to Refresh, this was up to the frame to decide; now it's a constant owned by the tab strip. Bug: 335903961

Due to a recent change, `browser_` has been discarded and it is now accessed as part of the base class. Chromium change: https://chromium.googlesource.com/chromium/src/+/ebac87839819b73cf82eab69ebaf373fa73ab3e6 commit ebac87839819b73cf82eab69ebaf373fa73ab3e6 Author: Andy Paicu <andypaicu@chromium.org> Date: Thu Jun 6 12:39:39 2024 +0000 [PEPC] Add prompt positioning setup and gold tests Modify the existing prompt positioning feature and add a FeatureParam with the 3 different flavors we want to experiment with. Also add gold UI tests for all 3 flavors. Fixed: 336988906

`ShouldTabSearchRenderBeforeTabStrip` has an override that relies on a substitution of `IsChromeRefresh2023`, however this call is no more. This change introduces a patch, as this function should always be returning false for brave. Chromium change: https://chromium.googlesource.com/chromium/src/+/b43d6eac96efad39df9ea265fdd566325ad34deb commit b43d6eac96efad39df9ea265fdd566325ad34deb Author: Dana Fried <dfried@chromium.org> Date: Fri Jun 7 00:40:40 2024 +0000 [GM2 Cleanup] tab_search_bubble_host.cc Moves tab search button to the right in all cases except on Mac. This may be temporary as the button may be moving back. #chrome-refresh-2023-cleanup Change-Id: I35c5fb7f8d2d48b65726b7dfaa646af96842939a

Chromium change: https://chromium.googlesource.com/chromium/src/+/033c53d0fced230d7085f5e9fe5222b6797c3b32 commit 033c53d0fced230d7085f5e9fe5222b6797c3b32 Author: David Pennington <dpenning@chromium.org> Date: Sat Jun 8 06:48:42 2024 +0000 [SaveTabGroupsV2] Add a learn more link to the intro IPH. Image of Learn More link in bug. Bug: 342028422

Chromium change: https://chromium.googlesource.com/chromium/src/+/d9d3307260d658a8b38dc5bbb68a871b38ee6fb3 commit d9d3307260d658a8b38dc5bbb68a871b38ee6fb3 Author: Thomas Nguyen <tungnh@chromium.org> Date: Thu Jun 6 15:15:27 2024 +0000 [PEPC]: Fire device status changed event when request finishes Before requesting the embedded permission, we will cache the device status. Once the request is complete, we will trigger a status change event. Fixed: 343010261

Chromium change: https://source.chromium.org/chromium/chromium/src/+/d2c4b3d248d0a74dd00e9956df65604eea0ef6bc [E2E] Use BottomControlsStacker to resize bottom browser controls Introduce BrowserControlsStacker to coordinate different UI that wants to change the bottom controls height. This CL is used as transitional solution, just forwarding calls from BrowserControlsStacker to the wrapped BrowserControlsSizer, while also provide the BrowserControlsStateProvider for clients to do the observation. There are still references of BrowserControlsSizer in custom_tabs, which is not in scope of this initial refactor. Bug: 345488108

Chromium change: https://chromium-review.googlesource.com/c/chromium/src/+/5584838 commit d5f3902b427fcea22c0da4d2ee63d05cdfada54f Author: Nicolas Dossou-Gbete <dgn@chromium.org> Date: Mon Jun 10 21:40:44 2024 +0000 � Disambiguate remaining Fallback vs Default template URL getters Clarifies in DefaultSearchManager and TemplateURLPrepopulateData APIs that: - "Default" refers to the default search engine, that is used for omnibox searches - "Fallback" refers to the search engine that can be used when there isn't one selected by another mean, and picked from the prepopulated set. As part of it, we remove the optional argument that allowed to get the index of Google when calling GetPrepopulatedEngines(). Instead, we can call the preexisting (but renamed) GetPrepopulatedFallbackSearch() function to obtain it. Bug: 330544981

Chromium change: https://chromium-review.googlesource.com/c/chromium/src/+/5584838 Author: Nicolas Dossou-Gbete <dgn@chromium.org> Date: Mon Jun 10 21:40:44 2024 +0000 � Disambiguate remaining Fallback vs Default template URL getters Clarifies in DefaultSearchManager and TemplateURLPrepopulateData APIs that: - "Default" refers to the default search engine, that is used for omnibox searches - "Fallback" refers to the search engine that can be used when there isn't one selected by another mean, and picked from the prepopulated set. As part of it, we remove the optional argument that allowed to get the index of Google when calling GetPrepopulatedEngines(). Instead, we can call the preexisting (but renamed) GetPrepopulatedFallbackSearch() function to obtain it. Bug: 330544981

Chromium change: https://chromium.googlesource.com/chromium/src/+/390e42d97bf3145f0b88a617fef72c778fc9d2ef commit 390e42d97bf3145f0b88a617fef72c778fc9d2ef Author: Abigail Klein <abigailbklein@google.com> Date: Mon Jun 10 20:12:24 2024 +0000 Remove features::kReadAnything as it has been rolled out for a year. Bug: 1266555

On a refactoring upstram relating to sidepanel, this target has been tagged to not be used on android. This should have no functional effect. Chromium change: https://chromium.googlesource.com/chromium/src/+/80ddd807ff1937d6632020d30f8931fac3e9ba3a commit 80ddd807ff1937d6632020d30f8931fac3e9ba3a Author: Erik Chen <erikchen@chromium.org> Date: Mon Jun 10 20:50:02 2024 +0000 Modularize side panel This CL is a refactor with no intended behavior change. This CL moves side panel C++ code into a standalone module. This CL removes //chrome/browser/companion from android, which was only included by accident. Change-Id: I37dad5bc447e995d975511aced99f8f05a30eec1 Bug: 345499793

Now that ChromeRefresh2023 is active (and per the comment in the chromium_src override), we no longer need this override. Chromium change: https://source.chromium.org/chromium/chromium/src/+/81dee0f4f06b4015b8d9176f1adcd13f6b3bd96f commit 81dee0f4f06b4015b8d9176f1adcd13f6b3bd96f Author: Taylor Bergquist <tbergquist@chromium.org> Date: Fri Jun 7 00:31:30 2024 +0000 Clean up IsChromeRefresh2023 in c/b/ui/view/frame/ This includes removing quite a bit of dead code calculating the distance between the top of the window and the top of the tabstrip. Prior to Refresh, this was up to the frame to decide; now it's a constant owned by the tab strip. Bug: 335903961

This was causing violations since parts of the side panel are now located under the views hierarchy. Chromium change: https://source.chromium.org/chromium/chromium/src/+/c6bc563b617bc804404b47c6ddc3d140f5d381a4 commit c6bc563b617bc804404b47c6ddc3d140f5d381a4 Author: Erik Chen <erikchen@chromium.org> Date: Tue May 14 21:22:35 2024 +0000 Move non-overlay lens out of chrome/browser/ui/lens. This CL has no intended behavior change. There was previously a DEPS rule that chrome/browser could not depend on chrome/browser/ui/views. This DEPS rule has been removed. Previously, the header files and implementation for non-overlay lens were separated for no reason other than this DEPS rule. This CL moves the header files into chrome/browser/ui/views/lens. This CL ensures that chrome/browser/ui/lens is used only for the overlay lens feature, and that all overlay lens C++ code is in this directory. Bug: 339497734 Chromium change: https://source.chromium.org/chromium/chromium/src/+/598410c64176215b9ad3f0a3f9d90ae218a3d13a commit 598410c64176215b9ad3f0a3f9d90ae218a3d13a Author: Erik Chen <erikchen@chromium.org> Date: Thu Apr 11 21:05:51 2024 +0000 Add //chrome/browser design principle documentation. Change-Id: Iec5038a44e81dfe5fbdbe6ddbf40c55c5a3007eb

…latedEngines Chromium change: https://source.chromium.org/chromium/chromium/src/+/d5f3902b427fcea22c0da4d2ee63d05cdfada54f commit d5f3902b427fcea22c0da4d2ee63d05cdfada54f Author: Nicolas Dossou-Gbete <dgn@chromium.org> Date: Mon Jun 10 21:40:44 2024 +0000 🧇 Disambiguate remaining Fallback vs Default template URL getters Clarifies in DefaultSearchManager and TemplateURLPrepopulateData APIs that: - "Default" refers to the default search engine, that is used for omnibox searches - "Fallback" refers to the search engine that can be used when there isn't one selected by another mean, and picked from the prepopulated set. As part of it, we remove the optional argument that allowed to get the index of Google when calling GetPrepopulatedEngines(). Instead, we can call the preexisting (but renamed) GetPrepopulatedFallbackSearch() function to obtain it. Bug: 330544981

chrome_test_launcher.cc no longer includes ChromeContentBrowserClient. Chromium change: https://source.chromium.org/chromium/chromium/src/+/66759752273c0966338d488e7df917f9166e1b02 commit 66759752273c0966338d488e7df917f9166e1b02 Author: Will Harris <wfh@chromium.org> Date: Thu Jun 6 15:33:13 2024 +0000 Revert "Use unique App Container ids for each instance of a browser test." This reverts commit bab0ea3f555e619055346a1df492452c08a7107e. Reason for revert: could be causing arm64 bot failures This revert will re-introduce flakes but that's been decided is better than the current situation where the additional app container profiles can cause some arm64 bots to completely fail due to an issue with the Windows API in use. Subsequent CLs will investigate other options for preventing the flakes from happening in the first place, without causing as much negative impact on arm64 bots. Note: this is not a clean revert, see PS 1..4 for diff. Original change's description: > Use unique App Container ids for each instance of a browser test. > > There is a race condition in CreateAppContainerProfile which can cause > two non-cooperating processes trying to create the identically named > app container profile to fail non-deterministically. > > This is not a problem in production because this never happens as > Chrome maintains a singleton instance to prevent multiple Chrome > browsers running at the same time pointed to the same App-Container > profile. > > However, in tests, they can run at the same time causing this to > occasionally trigger as seen in the attached bug. > > This CL fixes this by, in tests only, appending the data dir which is > unique for each test instance, to the App Container profile name. > > This (longer) name is later SHA1 hashed by the sandbox code to > generate the unique App Container profile name. > > It is hoped this CL will eliminate the current flakes on the CQ and CI > while keeping production/Chrome functionality the same as we have not > seen this issue occur in the wild. > > No functional changes in production/shipping code are intended > by this CL. > > BUG=40223285 > > Change-Id: I85f6eee2f9079f2673620fdfdbd24dfe6a0e184f > Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5468745 > Reviewed-by: John Abd-El-Malek <jam@chromium.org> > Reviewed-by: Scott Violet <sky@chromium.org> > Commit-Queue: Will Harris <wfh@chromium.org> > Cr-Commit-Position: refs/heads/main@{#1290236} Bug: 40223285, 342424178

The feature flag was enabled in https://chromium-review.googlesource.com/c/chromium/src/+/5605644 which got implemented in https://chromium-review.googlesource.com/c/chromium/src/+/5413915 with the following changes ``` if (features::IsUseMoveNotCopyInMergeTreeUpdateEnabled()) { // While experimenting with moving `details`, we have to ensure this call // order. This won't be the final structure of the code. delegate_->AccessibilityEventReceived(details); // This call steals the contents of `details` to avoid copying. SendAccessibilityEventsToManager(details); } else { SendAccessibilityEventsToManager(details); delegate_->AccessibilityEventReceived(details); } ``` When we receive the ax event and check the ax tree, the change hasn't been reflected yet due to the above enabled block.

…pgradeDecision Chromium change: https://source.chromium.org/chromium/chromium/src/+/d5d0088ff707332266eb901c9aab3db5d1762720 commit d5d0088ff707332266eb901c9aab3db5d1762720 Author: Joe Mason <joenotcharles@google.com> Date: Thu Jun 6 20:40:32 2024 +0000 Add metric for how often HSTS upgrades come from dynamic state This will be used to estimate how much work should go into optimizing the dynamic state mechanism. R=cthomp Bug: 345470315

We override the DisallowKeyedServiceFactoryRegistration method call for iOS so that early registration enforcement isn't enabled. This reverts us to our pre-existing behavior on iOS. Chromium change: https://source.chromium.org/chromium/chromium/src/+/f1e2cc21eefb9805d95bcf1833d6bccda9131a72 commit f1e2cc21eefb9805d95bcf1833d6bccda9131a72 Author: Sylvain Defresne <sdefresne@chromium.org> Date: Thu May 30 16:36:27 2024 +0000 [ios] Enforce early registration of KeyedServiceFactories Call DependencyManager::DisallowKeyedServiceFactoryRegistration() once EnsureBrowserStateKeyedServiceFactoriesBuilt() completes. It enforces that all KeyedServiceFactories are correctly registered before creation of the ChromeBrowserState. Add registration of all KeyedServiceFactories that we missing and sort alphabetically the registered factories (since the comment ask them to be kept sorted). Bug: none

Keep this service disabled on iOS, to match previous behavior. Chromium change: https://source.chromium.org/chromium/chromium/src/+/6952ce1a5bab9ac1cf2c190883bd0e152ebc5f41 commit 6952ce1a5bab9ac1cf2c190883bd0e152ebc5f41 Author: Marc Treib <treib@chromium.org> Date: Tue May 14 21:50:04 2024 +0000 Migrate GoogleGroupsUpdaterService to SyncServiceObserver Before this CL, GoogleGroupsUpdaterService exposed a public ClearSigninScopedState() and relied on users (in practice, the implementations of SyncClient::OnLocalSyncTransportDataCleared()) to call it at the right time. This CL migrates it to implement SyncServiceObserver instead, and internally clear the state at the right time (i.e. when the sync TransportState becomes DISABLED). The new behavior is not 100% identical to the old one, but it's very close and should work out to the same thing in practice. The new implementation is cleaner architecture-wise, and will (after some additional changes) allow us to get rid of SyncClient::OnLocalSyncTransportDataCleared(). Bug: 40724759, 337034860

The usage of `LocalOrSyncableBookmarkModelFactory` is supposed to be based on the `syncer::kEnableBookmarkFoldersForAccountStorage` feature flag and needs to be swapped for `BookmarkModelFactory` when the flag is enabled, which it is now by default. Also included threading fixes for bookmarks importer & exporter. Chromium change: https://source.chromium.org/chromium/chromium/src/+/947848e0c05c279bd358cd0b36ebf95738411eee commit 947848e0c05c279bd358cd0b36ebf95738411eee Author: Mikel Astiz <mastiz@chromium.org> Date: Tue May 14 14:14:01 2024 +0000 [ios] Enable flag EnableBookmarkFoldersForAccountStorage by default Ongoing experiments suggests the functionality is working as intended and the feature can be safely enabled by default. One unit-test needs updating due to an artifact of how the test itself is implemented: with the flag enabled, account bookmarks happen to be initialized, whereas the flag-disabled case doesn't have equivalent logic.

Chromium change: https://source.chromium.org/chromium/chromium/src/+/d864e62dddc14e305f0a274f6058ab082ad2b99a [SurfacePolish][CleanUp] Clean up surface polish code for MVT (Part 5) This CL represents part 5 of the cleanup for the surface polish code in the MVT, focusing on refining the code associated most visited tiles layout. The update removes the grid layout from most visited tiles and renames the most visited tiles carousel layout to simply most visited tiles layout. Bug: 331667743

Made |pinned_toolbar_actions_container_| null in toolbar. Also disabled some browser tests that failed or crashed due to this change. Also updated FeatureDefaultsTest.DisabledFeatures to check features::kToolbarPinning feature is disabled or not because |pinned_toolbar_actions_container_| is referred w/o its null checking when that feature is enabled.

We disable this service on iOS, but let's also disable it on desktop. Chromium change: https://source.chromium.org/chromium/chromium/src/+/6952ce1a5bab9ac1cf2c190883bd0e152ebc5f41 commit 6952ce1a5bab9ac1cf2c190883bd0e152ebc5f41 Author: Marc Treib <treib@chromium.org> Date: Tue May 14 21:50:04 2024 +0000 Migrate GoogleGroupsUpdaterService to SyncServiceObserver Before this CL, GoogleGroupsUpdaterService exposed a public ClearSigninScopedState() and relied on users (in practice, the implementations of SyncClient::OnLocalSyncTransportDataCleared()) to call it at the right time. This CL migrates it to implement SyncServiceObserver instead, and internally clear the state at the right time (i.e. when the sync TransportState becomes DISABLED). The new behavior is not 100% identical to the old one, but it's very close and should work out to the same thing in practice. The new implementation is cleaner architecture-wise, and will (after some additional changes) allow us to get rid of SyncClient::OnLocalSyncTransportDataCleared(). Bug: 40724759, 337034860

Replaced upstream's chrome refresh icon version that we've been using.

… for tabs Now that ChromeRefresh2023 is active, the ThemeColorPickerHandler is using a tri-color picker and foreground/background tabs and the tab strip were not taking on theme colors as expected. This change creates overrides to: - Replace SetUserColorAndBrowserColorVariant with BuildAutogeneratedThemeFromColor to correct the theme coloring for tabs and the tab strip. - Replace kGeneratedColorsInfo with kCustomizeChromeColors, allowing us to return to using the bi-color theme picker in brave://settings and the profile dialog. We call getChromeColors with extended_list=true from TypeScript in order to take advantage of this code path in the theme handler. Chromium change: https://source.chromium.org/chromium/chromium/src/+/05baea744f4111671b9e3b9baf28f8c1727f78c5 commit 05baea744f4111671b9e3b9baf28f8c1727f78c5 Author: Alison Gale <agale@chromium.org> Date: Wed May 15 17:38:31 2024 +0000 Remove references to launched kChromeRefreshSecondary2023 feature Also cleaned up some test-only references to kChromeRefresh2023 when they were adjacent. #chrome-refresh-2023-cleanup Bug: 335904039

This reverts commit f2a3b1b.

Since the IsolateSandboxedIframes feature flag was enabled, the chrome-untrusted://line-chart-display Wallet browser tests for the portfolio chart have been failing. In order to temporarily work around those failures, we've added allow-same-origin to the sandbox attribute for that chart. Entered brave/brave-browser#38771 to address this in a follow-up. Chromium change: https://source.chromium.org/chromium/chromium/src/+/b2c49e9ef800bc606041fc9356e665d11869cb30 commit b2c49e9ef800bc606041fc9356e665d11869cb30 Author: W. James MacLean <wjmaclean@chromium.org> Date: Wed May 15 19:54:12 2024 +0000 Enable IsolateSandboxedIframes by default. This CL enables the IsolateSandboxedIframes feature by default. Bug: 40082497

This avoids the following DCHECK: Check failed: instance->IsTestRenderFrameHost(). You may want to instantiate RenderViewHostTestEnabler.

Disabling this feature flag (which is enabled on iOS and Android by default) for now as it requires a Google account and possibly other backend support.

After deleting 2023UIRefresh features, upstream's all new color mixers are added and it overwrites our colors. Some of them are called from our code and others are not needed.

After deleting 2023UIRefresh, upstream's toolbar margin is increased. This causes our toolar has more long height. Back to previous value.

After deleting 2023UIRefresh, more larger text button's vertical padding is set. This made dialog button bigger. Set to previous vertical padding.

This crate has been dropped in upstrem but it is still required in brave. This change checks it out in our tree. Chromium change: https://chromium.googlesource.com/chromium/src/+/2a140ac6fc6943d7689fc41f8c4573e3d27cba7d commit 2a140ac6fc6943d7689fc41f8c4573e3d27cba7d Author: danakj <danakj@chromium.org> Date: Mon May 27 15:45:04 2024 +0000 Roll winapi-util: 0.1.6 => 0.1.8 in //third_party/rust. This change moves the dependency for windows bindings from winapi to the windows-rs crates published by Microsoft. Those crates come with .lib files that must be made available for linking, so we add them to gnrt_config.toml under the `native_libs_roots` variable. Native libs are copied into the toolchain's `./rustlib` directory and that directory is added to the library search path in order to find them there. This ensures the correct architecture library is used for host toolchains. This CL has been created semi-automatically. The expected review process and other details can be found at //tools/crates/create_update_cl.md Updated crates: * winapi-util: 0.1.6 => 0.1.8 New crates: * windows-sys@0.52.0 * windows-targets@0.52.5 * windows_aarch64_gnullvm@0.52.5 (as placeholder) * windows_aarch64_msvc@0.52.5 * windows_i686_gnu@0.52.5 (as placeholder) * windows_i686_gnullvm@0.52.5 (as placeholder) * windows_i686_msvc@0.52.5 * windows_x86_64_gnu@0.52.5 (as placeholder) * windows_x86_64_gnullvm@0.52.5 (as placeholder) * windows_x86_64_msvc@0.52.5 Removed crates: * winapi-i686-pc-windows-gnu@0.4.0 * winapi-x86_64-pc-windows-gnu@0.4.0 * winapi@0.3.9 Chromium `supply-chain/config.toml` policy requires that the following audit criteria are met (note that these are the *minimum* required criteria and `supply-chain/audits.toml` can and should record a stricter certification if possible; see also //docs/rust-unsafe.md): * winapi-util@0.1.8, windows-sys@0.52.0, windows-targets@0.52.5, windows_aarch64_gnullvm@0.52.5, windows_aarch64_msvc@0.52.5, windows_i686_gnu@0.52.5, windows_i686_gnullvm@0.52.5, windows_i686_msvc@0.52.5, windows_x86_64_gnu@0.52.5, windows_x86_64_gnullvm@0.52.5, windows_x86_64_msvc@0.52.5: does-not-implement-crypto, safe-to-run Fixed: 342194487

Crates under this path were triggering format. This mirrors what is being done in upstream.

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Upgrade from Chromium 126 to Chromium 127 #23829

Upgrade from Chromium 126 to Chromium 127 #23829

Commits on Jun 25, 2024