Before sending sites for BTM deletion, we directly check for host
canonicalization. This CL also cleans up a UMA variant that never
receives traffic.
There is still follow-up work to diagnose and prevent non-canonical
hosts from entering the BTM database in the first place. That will be
handled in a future CL.
Bug: 393088777
Change-Id: I4d2d19e60d16d38fdcddf93e3bd5d4c19b4b425a
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6255112
Reviewed-by: Giovanni Ortuno Urquidi <ortuno@chromium.org>
Reviewed-by: Ryan Tarpine <rtarpine@chromium.org>
Commit-Queue: Andrew Liu <liu@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1419931}
Convert browsing history handler to use mojom instead of older webUI
message handlers. Update unit tests to use new handler. Fix dangling
ptrs. Also update browser tests to call handler. Remove unused methods
from test browser proxy.
Bug: 386401009
Change-Id: Ibcb7de1e25d7aaffbbe12069f65fecab42d30846
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6091897
Reviewed-by: John Lee <johntlee@chromium.org>
Commit-Queue: Marlon Facey <mfacey@chromium.org>
Reviewed-by: Ken Buchanan <kenrb@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1419930}
This flag was added some time ago as we were considering
changing how geo location permissions are granted.
This is adding a JNI hop every time this is checked for
a hardcoded value so cleaning that up and removing the
dead code.
Bug: 40375780
Change-Id: I58002ea5d9aa82cea5da3e8029ab912466fa537e
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6264444
Reviewed-by: Richard (Torne) Coles <torne@chromium.org>
Auto-Submit: Rupert Wiser <bewise@chromium.org>
Commit-Queue: Richard (Torne) Coles <torne@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1419927}
Fix Sanitizer to properly recurse into template contents. Also update
tests (& test support files) to take template contents into account.
Bug: 356601280
Change-Id: Idb81d0f01afe7dcc1ce5d13d719b0515d1ef4752
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6234275
Reviewed-by: Yifan Luo <lyf@chromium.org>
Commit-Queue: Daniel Vogelheim <vogelheim@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1419925}
If remote theme is an extension, upon signin the theme extension is
installed. However, if the user signs out while the extension was being
installed, this is not cleared upon - thus leaking the theme extension
after the signout.
This CL fixes that by removing the extension from the installation queue
upon signout. This CL also adds some general tests for the remote theme
extension cases.
Bug: 373573152
Change-Id: Ife787279d25b353abac5f653a725f95f10181017
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6250123
Reviewed-by: Paul Adedeji <pauladedeji@google.com>
Commit-Queue: Ankush Singh <ankushkush@google.com>
Cr-Commit-Position: refs/heads/main@{#1419922}
because the restore path can do things that aren't expected, like
having a group in the restore flow multiple times, or trying to
connect an already connected group, we can check when calling
ConnectToLocalTabGroup
Fixed: 390421559
Change-Id: If69b5299b972ff945780765e63fbf34d031d5627
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6259017
Reviewed-by: Darryl James <dljames@chromium.org>
Commit-Queue: David Pennington <dpenning@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1419918}
Boolean implicit cast from pointers are now rewritten thanks to:
https://chromium-review.googlesource.com/c/chromium/src/+/6226064
the rewritten `base::span` are appended a `.size()` to help with boolean
conversion.
However, when the pointer is coming from another an external function,
it is also matched by `size_node` and this case wasn't properly matched
by `getSourceRange`.
The fix involves handling such a case + a regression test.
Note that external function are never rewritten. They probably shouldn't
be matched in `size_node`.
Bug: None
Change-Id: I324f5f89ff6b664626bb2ace7a78b83837eda9ff
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6258052
Reviewed-by: Stephen Nusko <nuskos@chromium.org>
Commit-Queue: Arthur Sonzogni <arthursonzogni@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1419917}
This fixes the initial focus logic for the autofill credit card editor
page to only set initial focus to the card number field if TalkBack is
off. This matches the logic for the address editor, see bug for details.
Confirmed the fix manually.
Bug: 382297841
Change-Id: I368b1723a20b86564a6b0fd7cd4f9264e8f203ed
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6253188
Reviewed-by: Rouslan Solomakhin <rouslan@chromium.org>
Commit-Queue: Nick Navarro <npnavarro@chromium.org>
Reviewed-by: Slobodan Pejic <slobodan@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1419914}
Move select match for opening of custom AutocompleteMatch to
OmniboxAutocompleteController. Following CLs will move the remaining
logic of OnSelectedMatchForOpening.
Bug: 390409252
Change-Id: I7b7bda35e7afa4bbc303273d83b91b52783b4440
Skip-Clang-Tidy-Checks: google-default-arguments the method is marked virtual for mocking
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6257471
Reviewed-by: Stepan Khapugin <stkhapugin@chromium.org>
Commit-Queue: Christian Xu <christianxu@chromium.org>
Auto-Submit: Christian Xu <christianxu@chromium.org>
Reviewed-by: Ameur Hosni <ameurhosni@google.com>
Cr-Commit-Position: refs/heads/main@{#1419913}
This requires giving AuctionDownloader a new constructor that takes
parameters that AuctionURLLoaderFactoryProxy normally populates when
AuctionDownloader is used for a worklet process-initiated fetch, for use
by TrustedSignalsFetcher.
This is a step towards wiring up TrustedSignalsFetcher-driven fetches to
devtools.
Bug: 379843989
Change-Id: Ibd6743402d1ab48d2ab60dec22e3ff8945ddd921
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6217975
Reviewed-by: Maks Orlovich <morlovich@chromium.org>
Reviewed-by: Christian Dullweber <dullweber@chromium.org>
Commit-Queue: mmenke <mmenke@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1419906}
There's evidence this animation is causing 4+% of compositor frames
on Chrome on Android, which seems excessive for this UX. This reuses the
existing feature that was added but abandoned and only update field
trial to disable scrollbar fade on Android, by making the duration small
enough so there's only one step in the animation.
Bug: 40196369
Change-Id: I82ab399f4e1406173cc0bfdd927b918b51a5d453
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6258398
Commit-Queue: Bo Liu <boliu@chromium.org>
Reviewed-by: Vasiliy Telezhnikov <vasilyt@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1419903}
This CL merges EntityDataManager::{Add,Update}EntityInstance() into
one function EntityDataManager::AddOrUpdateEntityInstance().
The old EntityDataManager::UpdateEntityInstance() had exactly this
effect: it first removes the given entity if present, and then adds
the entity to the database.
Alternatively, we could let UpdateEntityInstance() first check if
the entity already exists and fail if it doesn't. But it seems that
no existing or planned code needs the distinction between "add" and
"update" (EntityDataManager wouldn't even report an error).
So merging them seems like the right thing to do.
Bug: 389631478
Change-Id: I28d06807fd40aa96bc01fdd57fcde753a87d30f7
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6256116
Commit-Queue: Christoph Schwering <schwering@google.com>
Reviewed-by: Jan Keitel <jkeitel@google.com>
Cr-Commit-Position: refs/heads/main@{#1419902}
This CL resolves a crash bug which occurs due to incorrect calculation
as currently we calculate the collapsible placeholder before the call to
|PrepareWhitespaceAtPositionForSplit| method which actually replaces
"\n" with a nbsp and which is why at the time of removal of collapsible
placeholder, it's not there anymore so the DCHECK fails.
Now on, the collapsible placeholder calculation would be done after all
adjustments done on insertion position and just before the node
insertion starts. This makes sure that we check for collapsible
placeholder at correct |insertion_pos|.
The web_test in this CL should not crash.
Bug: 395028747
Change-Id: Ibba1b868da27a0891a5ea8132295acdfacaed641
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6248618
Reviewed-by: Sambamurthy Bandaru <sambamurthy.bandaru@microsoft.com>
Reviewed-by: Kent Tamura <tkent@chromium.org>
Reviewed-by: Siye Liu <siliu@microsoft.com>
Commit-Queue: Utkarsh Pathak <utpathak@microsoft.com>
Cr-Commit-Position: refs/heads/main@{#1419895}
There are cases where WKDownload.originatingHost.securityOrigin is not
available (e.g. if the download is initiated from a data frame).
In that case, display the download origin, after redirection.
If it is still not available, display a placeholder string to inform
the user.
Also, observe the WebState to show the URL if the page navigates.
Bug: 394997959, 395232501
Change-Id: I447e8933676157cc4c345abbb946d13d9e7aff12
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6252618
Commit-Queue: Olivier Robin <olivierrobin@chromium.org>
Reviewed-by: Quentin Pubert <qpubert@google.com>
Reviewed-by: Gauthier Ambard <gambard@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1419891}
Use ASAuthorizationWebBrowserPlatformPublicKeyCredential's providerName
to identify the provider of a passkey sourced from a third-party passkey
provider on MacOS.
Fixed: 392837090
Change-Id: I0a6747c451c813768065cd4f70d13f96a89e2541
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6213993
Reviewed-by: Mohamed Amir Yosef <mamir@chromium.org>
Commit-Queue: Mohamed Amir Yosef <mamir@chromium.org>
Auto-Submit: Nina Satragno <nsatragno@chromium.org>
Reviewed-by: Adem Derinel <derinel@google.com>
Cr-Commit-Position: refs/heads/main@{#1419889}
All the tests make the same operations and check the same things, but
using different input and checking for a different error code. This
change consolidates these tests into a single parameterized test, using
the different inputs and expected failing result as parameters.
Modified inheritance structure of already existing parametrized tests.
Bug: 374787847
Change-Id: I10d528a887739b0e2a7324388aa60deec5a721da
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6239358
Reviewed-by: Caitlin Fischer <caitlinfischer@google.com>
Reviewed-by: Luc Nguyen <lucnguyen@google.com>
Commit-Queue: Ramon Cano Aparicio <rcanoaparicio@google.com>
Cr-Commit-Position: refs/heads/main@{#1419887}
This new feature, `FledgeQueryKAnonymity`, defaults to enabled,
mimicking the current behavior. When enabled, this feature causes the
browser to query k-anonymity status at interest group join and update
time.
This feature operates independently of the
existing `FledgeConsiderKAnonymity` and `FledgeEnforceKAnonymity`
features, which together decide the KAnonymityBidMode.
It may be reasonable to disable kFledgeQueryKAnonymity on clients on
which k-anonymity is not enforced, as k-anonymity status isn't used in
those auctions.
Though it's possible to enforce k-anonymity using
kFledgeConsiderKAnonymity and kFledgeEnforceKAnonymity without also
enabling kFledgeQueryKAnonymity, this would be unusual, since it means
that no keys will be queried, so that new ads added to interest groups
via subsequent joins and updates would be considered non-k-anonymous,
and as each interest group's cache of k-anonymous keys expire beyond the
kKAnonymityExpiration of seven days, all of its ads would be considered
non-k-anonymous.
Change-Id: I7ed340c4a2d5640b134de574a9c1e4ea7880dc9b
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6259656
Reviewed-by: Russ Hamilton <behamilton@google.com>
Commit-Queue: Orr Bernstein <orrb@google.com>
Cr-Commit-Position: refs/heads/main@{#1419886}
