- new args at InterceptNavigationDelegate.ShouldIgnoreNavigation: should_run_async, result_callback;
- return type changed from bool to void
Related Chromium change:
https://source.chromium.org/chromium/chromium/src/+/8c2b10e85ff860959aaf6bf86f48bd03fc998ccb
Reland "Fix Async Navigation Intercept using the wrong URL for fast redirects"
Original change's description:
Fix Async Navigation Intercept using the wrong URL for fast redirects
Prior to this change, if we received a redirect before the async
shouldIgnoreNavigation call was run (which does happen in practice) we
would still run the pending shouldIgnoreNavigation call but the
NavigationHandle would already be updated to the redirect URL and so
the result would potentially be wrong. It's also complicted to reason
about multiple checks in flight at once.
This change ensures that before processing the redirect, we finish
processing the previous step in the navigation (either the initial
navigation or a previous redirect).
Also, I moved the async task to Java to make use of the cached ExternalNavigationParams as the NavigationHandle is mutable and could be modified between when the task is posted and when it's run.
We'll still get the vast majority of the benefit from the async path
as it's extremely rare that we get a redirect before finishing the
check, so this should be performance neutral.
No new Kill Switch is needed for this change, as we can re-use the
kAsyncCheck Kill Switch to disable async nav altogether if we really
need to.
Bug: 381535042
Change-Id: I674aca65a4f2e5ca31670cf52929e8f87daf0528
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6194176
This function used to take a pointer to an observer, however it has been
recently simplified to take a callback. This is of no consequence for
the use in brave, where a `nullptr` was being passed in.
Chromium change:
https://chromium.googlesource.com/chromium/src/+/a00f28906327914733c18715a878e1b9ba274bec
commit a00f28906327914733c18715a878e1b9ba274bec
Author: James Lee <ljjlee@google.com>
Date: Thu Jan 23 01:55:15 2025 -0800
Use callback rather than observer interface in BookmarkHtmlWriter
This simplifies the error-handling branches and reduces repetition.
Bug: 390643468
This argument is getting passed along in the existing overrides. For the
existing tests the default behaviour has been preserved as true.
Chromium change:
https://chromium.googlesource.com/chromium/src/+/bed1bc72123f67fc50f94c18c71f745207d7cc19
commit bed1bc72123f67fc50f94c18c71f745207d7cc19
Author: sbingler <bingler@chromium.org>
Date: Thu Jan 23 09:00:48 2025 -0800
Propagate is_top_level_nav to ShouldUpgradeToSSL
Expose is_top_level_nav through ShouldUpgradeToSSL to its callers.
As part of this CL preconnect HSTS upgrades will be disabled whenever
kHstsTopLevelNavigationsOnly is active. This is a tentative decision
and may be adjusted in future once we have more metrics data regarding
preconnect's HSTS usage.
The metric's collection is also modified to only occur when the feature
is disabled. Since it's never been enabled yet this is effectively a
no-op on metric collection and doesn't affect existing data.
Bug: 40725781
This reverts commit 7f843663e4d5b0f4230a967739fd290e60b95fc2.
Reason: the flag is removed completely at cr134
Related Chromium change:
https://source.chromium.org/chromium/chromium/src/+/a49153686134a72ff189024392df6aad09e71fb0
[Quick Delete] Clean up kQuickDeleteAndroidFollowup flag
Clean up Quick Delete phase 2 feature checks now that the experiment is
fully launched. This includes removing the obsolete tabbed layout and
basic tab from CBD on Android.
OBSOLETE_HISTOGRAM[History.ClearBrowsingData.UserDeletedFromTab]=No longer logged
Bug: 40255099
Change-Id: I340afa9692f17ad3b472f501522a9e6ff8fb061c
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6162502
This field is gone however it was not being used in brave anyway, and it
was just being referred for instantiation/copies.
Chromium change:
https://chromium.googlesource.com/chromium/src/+/e7adca16d31ff65896b85708e5541aa20760b1c9
commit e7adca16d31ff65896b85708e5541aa20760b1c9
Author: Justin Donnelly <jdonnelly@chromium.org>
Date: Tue Jan 21 14:50:02 2025 -0800
[search_engines] Remove side_search_param and side_image_search_param.
These values are unused after the code cleanup tracked in the bug. After this lands, I will make an upstream change remove the values from prepopulated_engines.json.
Bug: 357910230
Chromium change:
https://chromium.googlesource.com/chromium/src/+/0a9673e574a887c8fc4ee121b1299496b7804810
commit 0a9673e574a887c8fc4ee121b1299496b7804810
Author: Chris Fredrickson <cfredric@chromium.org>
Date: Tue Jan 21 08:45:14 2025 -0800
Convert ExclusionReason to a scoped enum
This CL is mechanical; it changes `enum` to `enum class`, and then
adds the missing qualifications and static_casts.
Change-Id: Id82475659afa6ee0ad12fed4c635e6e231597c9d
This is only used in tests, so it can be a default constructed instance.
Chromium change:
https://chromium.googlesource.com/chromium/src/+/5eaeb769a8d8d68f590633148be14254e3be3181
commit 5eaeb769a8d8d68f590633148be14254e3be3181
Author: Rushan Suleymanov <rushans@google.com>
Date: Tue Jan 21 07:51:54 2025 -0800
[Sync] Add EntityData for remote entity deletions
This CL adds EntityData content when EntityChange for a deletion is
created. This data is populated for remote tombstones only (i.e.
deletions for commit-only types and removed membership in collaborations
won't have this data).
Bug: 352214654, 385133876
Chromium change:
https://chromium.googlesource.com/chromium/src/+/8f3d075ebf5ff02504a72ac34e0adcc088376881
commit 8f3d075ebf5ff02504a72ac34e0adcc088376881
Author: Taylor Bergquist <tbergquist@chromium.org>
Date: Tue Jan 21 14:40:53 2025 -0800
Simplify layout when creating a window during tab dragging.
There were several compounding complexities in the old approach:
- the window and tab bounds calculations were distributed across several
methods, with unclear responsibilities and interactions
- the two cases (dragging all tabs in a window, and dragging some tabs
out into a new window) reused some of these methods, and duplicated
(with some important differences) other parts of the logic
- Wayland doesn't support screen coordinates or programmatically
positioning windows; this required special handling whenever any window
origin math was done (as that all happens in screen space)
It turns out that a lot of this complexity can just cancel out if you
arrange things the right way.
This CL does several things:
- skips calculating the detached window origin (over and over, and
incrementally nudging the window around). The move loop will reposition
it for us, so we can just calculate the drag offset once at the end.
- greatly simplifies the dragging all tabs in a window case, which was
only complicated because it was reusing too much of the other case
- clearly defines the responsibilities of, and information flow between,
the helper methods of DragBrowserToNewTabStrip
- unifies calculations between Wayland (which doesn't support screen
coordinates or programmatically positioning windows) and other platforms
- more strictly separates the calculation of different pieces of
information (mainly window size/origin and tab sizes/origins)
Bug: 382754501
This is a renaming to nudge people to use other constructs for hints.
This change doesn't go much further than to keep the code as is, with
execption of `SidebarButtonView`.
Chromium change:
https://chromium.googlesource.com/chromium/src/+/af6f4b9c4f3b02fbcba3ffa3c7f4b65b0995aba3
commit af6f4b9c4f3b02fbcba3ffa3c7f4b65b0995aba3
Author: Javier Contreras Tenorio (from Dev Box) <javiercon@microsoft.com>
Date: Tue Jan 21 11:32:35 2025 -0800
[views-ax] Cleanup of GetTooltipText (part 2)
This CL cleans up the last bits of the refactor to make Views cache
their tooltip text.
* Rename GetCachedTooltipText and SetCachedTooltipText to just
Get/SetTooltipText.
- This involved some renaming in some Views that had a
method named the same to set a custom tooltip text. That method
on those Views was then renamed to make clearer that it was "custom".
- Some Views also had some custom logic in their SetTooltipText and so
this CL handled those by adding a PropertyChangedCallback with that logic
that would run on those Views when the tooltip property changed.
* Rename GetTooltipText(Point) to GetRenderedTooltipText(Point)
- This is the method that the tooltip controller uses to expose
the tooltip, and in most cases this and GetTooltipText() return the same
string, but there are some Views such as TableView and MenuItemView that
require a Point to calculate the tooltip text.
* This CL also includes the small refactor of a couple of Views
for which their tooltip text had not been refactored. These were small
refactors, one just needed a callback and the other just had to call
SetTooltipText.
* For the Views that had their own custom logic in a local
`SetTooltipText`, this CL moves that custom logic to an override
of a function in view that will run when the tooltip text changes.
This is needed in many cases over adding a property callback because
these Views used the value of the tooltip text BEFORE it changes,
which was not possible to get via the property callback.
*`Label` had a local `SetTooltipText` that set a local custom
tooltip variable. To accommodate this along with the new method
of the same name in View, we renamed the Label version
to SetCustomTooltip text, which also fits the behavior better.
* More information can be found on the linked bug, but TLDR
is that Views now cache their tooltip text. This CL then renames
and cleans up a bit to finish up the refactor.
`View::GetTooltipText()` returns this cached tooltip for all Views.
`View::GetRenderedTooltipText(point)` exists so Views that require
of a point to compute their tooltip can override this method to do so.
The base version of this method will simply return the cached member.
`GetRenderedTooltipText` is the method that is used by the tooltip
manager.
This CL is part of the ViewsAX project:
https://docs.google.com/document/d/1Ku7HOyDsiZem1yaV6ccZ-tz3lO2XR2NEcm8HjR6d-VY/edit#heading=h.ke1u3utej413
Bug: 325137417, 378724151
The list of tests in `chrome/android/BUILD.gn` has been moved under
`chrome/android/javatests/BUILD.gn`. This affects an exclusion in place
for removing one of these tests.
Chromium change:
https://chromium.googlesource.com/chromium/src/+/b4c1dcd2f21521a4f3ab5fedca31b29d52653eb1
commit b4c1dcd2f21521a4f3ab5fedca31b29d52653eb1
Author: Martin Kong <martinkong@google.com>
Date: Tue Jan 21 10:02:28 2025 -0800
Split chrome_unit_test_java into multiple targets
This CL splits the chrome_unit_test_java target into multiple targets
based on the package, so that a change to a single test file will not
cause all test files to be recompiled.
Bug: 40286190
int getCurrentTabId() replaced with @Nullable Tab getCurrentTab()
Related Chromium change:
https://source.chromium.org/chromium/chromium/src/+/052f94ec2a329c0bef43a949cd84fe058a32710f
[Hub] Fix Tab Pane interaction with Bottom Toolbar for non-NTP pages
Improve interaction between the tab switcher and the bottom toolbar by
animating the shrinking/expanding tab to the top of the screen
subtracting the toolbar space if bottom toolbar is in use.
The bottom toolbar area still has a somewhat unpolished transition
because the area where the bottom toolbar is remains blank during the
animation rather than something nicer like a crossfade. For now this is
still significantly better than the previous state. However, there is
probably room to improve this with help from UX. Unfortunately, the fact
Hub is layered over the rest of the UI and we are using LayoutManager to
handle the transition makes this somewhat more difficult than it might
otherwise seem.
See demo on bug.
Fixed: 385704998
Change-Id: Id22ec322a89126e319c9a5f5a5bae575eac206f3
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6180002
Chromium now doesn't contain `Basic`/`Advanced`.
Class ClearBrowsingDataFragment now represents what used to be ClearBrowsingDataFragmentAdvanced.
Related Chromium change:
https://source.chromium.org/chromium/chromium/src/+/a49153686134a72ff189024392df6aad09e71fb0
[Quick Delete] Clean up kQuickDeleteAndroidFollowup flag
Clean up Quick Delete phase 2 feature checks now that the experiment is
fully launched. This includes removing the obsolete tabbed layout and
basic tab from CBD on Android.
OBSOLETE_HISTOGRAM[History.ClearBrowsingData.UserDeletedFromTab]=No longer logged
Bug: 40255099
Change-Id: I340afa9692f17ad3b472f501522a9e6ff8fb061c
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6162502
The naming now better reflects the size of the sample.
Chromium change:
https://chromium.googlesource.com/chromium/src/+/4a39d12c4eec40ffa9a531598ea645edc1d25650
commit 4a39d12c4eec40ffa9a531598ea645edc1d25650
Author: Ramon Cano Aparicio <rcanoaparicio@google.com>
Date: Mon Jan 20 05:00:17 2025 -0800
Remove HistogramBase::Sample type.
Rename HistogramBase::Sample to Sample32.
Add size to type name so the size is known without needing to lookup
the typedef.
Bug: 40899968
Methods GetSizeDIPs()/GetSizeDevicePx() instead of GetSize()
Related Chromium change
https://source.chromium.org/chromium/chromium/src/+/4f2bc05e8d30de72c30d6dd34e55b50b4464661b
Reland "Avoid pixel rounding errors by using device pixels directly"
This reverts commit 2754d87ac5c787d866b53ef438aadd6ea5ef5619.
Reason for revert: fixed TabsTest#testNewTabSetsContentViewSize, AndroidScrollIntegrationTest#testPageDown
Diff from original land to fix tests:
https://chromium-review.googlesource.com/c/chromium/src/+/6169396/4..5
Original change's description:
> Revert "Avoid pixel rounding errors by using device pixels directly"
>
> This reverts commit a4951af7d99bc472b9577f85cc2002b9e213f4af.
>
> Reason for revert: Potential culprit for failures on Android. See b/389752076.
>
> Original change's description:
> > Avoid pixel rounding errors by using device pixels directly
> >
> > Previously we were (on Android) dividing device pixels by devicePixelRatio and rounding off to integers to get a size in DIPs, then when sending the size to Blink, it was multiplied again by devicePixelRatio and rounded to integers again. This can lead to rounding errors, which are avoided by not dividing in the first place.
> >
> > This fixes an issue on Android, because ViewAndroid receives its
> > sizing in device pixels. On other platforms the sizing is received in
> > DIPs; to the extent this bug is on those platforms other OS APIs reporting device pixels could be integrated in the future.
> >
> > The PR includes one more test, but there are many tests that exercise
> > the various parts of code changed in this CL already (discovered because earlier patchsets failed them before the code was fixed).
> >
> > Bug: 40234130
> >
> > Change-Id: Ie5bc209b5c4e2c88bf024c32022a562eeaf05cf0
> > Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6142555
> > Reviewed-by: Tom Sepez <tsepez@chromium.org>
> > Reviewed-by: Dmitry Gozman <dgozman@chromium.org>
> > Reviewed-by: Nico Weber <thakis@chromium.org>
> > Commit-Queue: Chris Harrelson <chrishtr@chromium.org>
> > Reviewed-by: Bo Liu <boliu@chromium.org>
> > Cr-Commit-Position: refs/heads/main@{#1405652}
>
> Bug: 40234130
> Change-Id: I29227280eb3e99e4cfbcfbb7364f90bcd74978f7
> No-Presubmit: true
> No-Tree-Checks: true
> No-Try: true
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6173616
> Reviewed-by: Yue Zhang <yuezhanggg@chromium.org>
> Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
> Owners-Override: Yue Zhang <yuezhanggg@chromium.org>
> Commit-Queue: Yue Zhang <yuezhanggg@chromium.org>
> Cr-Commit-Position: refs/heads/main@{#1405760}
Bug: 40234130, 389752076
Change-Id: Ibaa09fb6ba469c9c22ae26819d2ed59c224a8f9c
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6169396
from void to std::unique_ptr<PermissionDialogDelegate>
Related Chromium change
https://source.chromium.org/chromium/chromium/src/+/eea12451eed010fb897e3171b9c4cb7979c082ff
Reland "[PEPC]Add new JNI call to update screen variant"
This is a reland of commit aeb64aa747c14aa5bb461e0bc0f5c4a83764b9f7
We took out the assertion from the original change because the
PermissionDialogDelegate Java class won't be null right after the
tab's gone. It needs to wait for a signal from C++. Removing the check
might even help us close the dialog in some corner cases.
Original change's description:
> [PEPC]Add new JNI call to update screen variant
>
> The CL moves the owner of native permission dialog delegate to C++
> native side instead of Java. This way, we can decide to destroy modal
> dialog or update screen variant from C++ after accepting a prompt.
>
> Also, this CL is introducing the first shape of the API to update custom
> view based on the current screen variant.
>
> Bug: 388407662
> Change-Id: I62239454c625c34259f6a495a6a53136ff1f26f1
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6167529
> Commit-Queue: Thomas Nguyen <tungnh@chromium.org>
> Reviewed-by: Andy Paicu <andypaicu@chromium.org>
> Cr-Commit-Position: refs/heads/main@{#1406569}
Bug: 388407662
Change-Id: I82717d973ae65ef7cc6f0a197210929bc2bd11e8
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6178268
Reviewed-by: Andy Paicu <andypaicu@chromium.org>
This file is now moving under:
//third_party/search_engines_data/resources/definitions/
As a result the patching for the list of search engines has to be
updated to be applied on the third party path.
Chromium change:
https://chromium.googlesource.com/chromium/src/+/a3afc766c5b7c12798a67ccf2deca1a99a79bcb1
commit a3afc766c5b7c12798a67ccf2deca1a99a79bcb1
Author: Tomasz Wiszkowski <ender@google.com>
Date: Thu Jan 16 13:28:16 2025 -0800
Build SE definitions off of scrubbed source of truth.
This change ensures Search Engine definitions are built from data
file scrubbed from the internal source of truth, ensuring unattended
propagation of changes.
** There's no functional changes in this CL **
The CL merely uses a new, target location of the
prepopulated_engines.json file to generate derived C++ files, and
updates all included headers (and DEPS files) accordingly.
Bug: b:382104753
Related Chromium change
https://source.chromium.org/chromium/chromium/src/+/d223f2e8f4900df442947eb2303b356468625d6f
[Tab Model] Remove onTabModelSelected
Replace onTabModelSelected with observing getCurrentTabModelSupplier.
A previous CL removed most usages, but a few were leftover. This CL
completes the migration.
This change in observation style is beneficial as it avoids adding
TabModelSelectorObservers that get notified of various spammy onChange
updates and subsequently no-op all observer events except for tab model
selected which is the only action most observers cared about in the
first place.
Change-Id: I9a657404c65b8e6b182924656e264945b7ddb35e
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6165827
- bottomControlsOffsetSupplier arg at ToolbarSwipeLayout.ctor
- bottomToolbarControlsOffsetSupplier arg at TopToolbarCoordinator.initializeWithNative
Related Chromium change
https://source.chromium.org/chromium/chromium/src/+/3c074e90a12a90c9c0d128773300b209b7b29d6f
[Anchor] Make bottom controls offset for overlay observable
This allows us to drive immediate changes to the position of the overlay
instead of passively adjusting when a new frame is composed. This is
necessary to be able to set up animations to start the overlay in the
correct location.
Bug: 369134461
Change-Id: I897a2ad7943e56660255a1fd2a5b485b70d64466
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6107628
Chromium change:
https://chromium.googlesource.com/chromium/src/+/9c7ccb3ff3af0a85d29c32475921aa129962d6cf
commit 9c7ccb3ff3af0a85d29c32475921aa129962d6cf
Author: Ryan Sultanem <rsult@google.com>
Date: Wed Jan 15 12:46:54 2025 -0800
[ProfilePicker][Glic] Implement the Learn More button
The Learn More text is currently part of a bigger string. Therefore the
click even is added dynamically when the page is loaded by connecting
the part of the text through the class tag to override the click event.
On that event, the handler is notified to load the last used profile,
open or use a browser with it and load the "Sign in to Chrome" page from
the Help Center.
The page cannot be loaded directly, since the Picker uses the System
Profile which cannot be used to open a Browser.
The picker is not expected to be closed after this action.
Bug: b:385726690
This member has been changed into `BookmarkMergedSurfaceService`, and
now the bookmark model is retrieved from it.
Chromium change:
https://chromium.googlesource.com/chromium/src/+/333703279b44da5c6f246f2a3057960fb0ed04b1
commit 333703279b44da5c6f246f2a3057960fb0ed04b1
Author: Monica Basta <msalama@chromium.org>
Date: Wed Jan 15 01:35:03 2025 -0800
[B4B] Migrate BookmarkBarView to a merged surface
This CL replaces the use of `BookmarkModel` in `BookmarkBarView` with
`BookmarkMergedSurfaceService`. This completes the migration work for
the `BookmarkBarView`.
Tasks left to do:
- Adapt the observer once the new `BookmarkMergedSurfaceService`
observer is introduced.
- Add tests for the behavior when account nodes exist. This will be
added once full support for account nodes is added to
`BookmarkMergedSurface` (almost done).
Bug: 369304373
A `RegionalCapabilitiesService` instance is now required to instantiate
`SearchEngineChoiceService`.
Chromium change:
https://chromium.googlesource.com/chromium/src/+/01008cac8b5afc198e85a60761fbe766aac4f645
commit 01008cac8b5afc198e85a60761fbe766aac4f645
Author: Nicolas Dossou-Gbete <dgn@chromium.org>
Date: Wed Jan 15 05:19:52 2025 -0800
rcaps: Make RegionalCapabilitiesService a Profile-Keyed Service
Moves it from being created as an internal dependency of
SearchEngineChoiceService to being a full PKS.
Bug: 388792357
This function has been moved into a base IDL file, and the patches are
for them are being addjusted accoridingly.
Chromium change:
https://chromium.googlesource.com/chromium/src/+/5f753b51a296d3e55dff42557e1e4aeb34387a2e
commit 5f753b51a296d3e55dff42557e1e4aeb34387a2e
Author: Yi Xu <yiyix@chromium.org>
Date: Tue Jan 14 21:32:11 2025 -0800
Factor out the common functions in Canvas2D and paint canvas
Both paint canvas and HTML/Offscreen 2D canvas share the same Canvas
rendering context 2D API. This CL refactors out the common functions and
puts them in a new CanvasRecordingContext2D IDL mixin, which all three
2D context interfaces implement.
Bug: 380095402
Updated our variable name to match the new nomenclature as well.
Chromium change:
https://source.chromium.org/chromium/chromium/src/+/7c6c78ad4e0ed6a0e1204264b02db8f85d34994e
commit 7c6c78ad4e0ed6a0e1204264b02db8f85d34994e
Author: Dirk Pranke <dpranke@google.com>
Date: Tue Jan 14 16:19:05 2025 -0800
Use `exec_script_allowlist` in the GN dotfile (//.gn).
GN now supports `exec_script_allowlist` as an alternative/synonym
for `exec_script_whitelist`, so we should move to it to be more
inclusive and accurate.
Bug: 389986807
Support for this preference was removed upstream a year ago, but the preference
itself was retained. It has now been removed, so we need to delete our
references to it.
Chromium change:
https://source.chromium.org/chromium/chromium/src/+/b89929785bd064a4a041705e8aa998d34797f310
commit b89929785bd064a4a041705e8aa998d34797f310
Author: Daniel Rubery <drubery@chromium.org>
Date: Fri Jan 10 11:34:00 2025 -0800
Remove one year expired prefs
It's January 2025, so prefs deprecated in January 2024 can be removed
from browser_prefs.cc.
Change-Id: I76fbb6a8b7e5ca1040d880628ba6a01ac51a801b
Chromium change:
https://source.chromium.org/chromium/chromium/src/+/5d75c527e312d3c18915498cce640b876a4ec9aa
commit 5d75c527e312d3c18915498cce640b876a4ec9aa
Author: Nihar Majmudar <niharm@google.com>
Date: Thu Jan 2 07:27:32 2025 -0800
[searchbox] Move generic methods up to SearchboxHandler.
No-op change. Right now a lot of generic searchbox methods (methods that
all searchboxes need) are implemented in RealboxHandler. We want to move
these up to the base SearchboxHandler class so other handlers can
implement handler specific methods.
Follow ups plan to move lens specific code out of the RealboxHandler.
Change-Id: Ic021780c717ce2109b1a65492f7f314295ea24e3
Bug: 386819657
This warning has been enabled for v8, and this has run into build
failures in brave in certain places.
Chromium change:
https://chromium.googlesource.com/v8/v8/+/d8634af5939710d21fd8e64b053af697dc0751bd
commit d8634af5939710d21fd8e64b053af697dc0751bd
Author: Darius Mercadier <dmercadier@chromium.org>
Date: Tue Jan 14 08:34:18 2025 +0100
Re-enable Clang -Wshadow warning
Change-Id: Ieca876e2384519ca4333857337835c9802a10f61
This class now requires a `SearchAggregatorSuggestionsService`, however
the use in brave can default to a `nullptr`.
Chromium change:
https://chromium.googlesource.com/chromium/src/+/81cd2f8a5022b1040e106c5064f703202356af53
commit 81cd2f8a5022b1040e106c5064f703202356af53
Author: Ananya <ananyaseelam@google.com>
Date: Mon Jan 13 10:46:30 2025 -0800
[search aggregator] Connect SearchAggregator to RemoteSuggestionsService
This CL adds SearchAggregatorSuggestionsService to the input parameters
of the RemoteSuggestionsService constructor. This CL also adds the
CreateSearchAggregatorSuggestionsRequest method which will later
be used in the EnterpriseSearchAggregatorProvider.
Note: The naming of SearchAggregatorSuggestionsService will
change to EnterpriseSearchAggregatorSuggestionsService in a
follow up CL. This is why the factory is named
enterprise_search_aggregator_suggestions_service_factory.
Bug: 388320853
AlexeyBarabash