| Age | Commit message (Collapse) | Author |
|---|
|
Move files to legacy package which are necessary to build legacy MediaProvider. This change isolates legacy functionality so that modifications to the upstream files won't impact the legacy MediaProvider build. Additionally, remove the playlist and dao package since these are not required
Flag: EXEMPT, code refactor
Test: m MediaProviderLegacy
Bug: 395901424
Change-Id: I652a7e397e1de65a10cb742b6f18552b6b3e79dd
|
|
Original change: https://android-review.googlesource.com/c/platform/packages/providers/MediaProvider/+/3247658
Change-Id: I7384958652e4d040889017e82d86d0604751a683
Signed-off-by: Automerger Merge Worker <android-build-automerger-merge-worker@system.gserviceaccount.com>
|
|
modules should specify the submodule of java_sdk_library that the module
actually depends on
Test: CI
Bug: 358613520
Change-Id: Ie3ad72f004a5de8fd917de17864b8ea6cf1561fe
|
|
Change-Id: I9cba95b4ee9843abe567f4af78dceff9b7947390
Test: n/a
Bug: 341190735
|
|
LegacyDatabaseHelper code affects code coverage numbers for MP module.
Test: n/a
Bug: 284258052
Change-Id: Ie28208ee10936da11eec8ac2db2fe0e50e08f04d
|
|
LegacyMediaProvider doesn't need this receiver. Test that tests
LegacyMediaProvider doesn't have any receivers is failing because of
this receiver. Removing it from the manifest explicitly.
Bug: 278130163
Test: atest android.provider.cts.media.MediaStoreTest#testLegacy
Change-Id: If5c96a6afdc9f5ea73eab07767739593b1fae130
|
|
As part of Stable URIs, we need to get an instance of MediaProvider to
interact with FuseDaemon to perform leveldb operations. Since we were
using the same DatabaseHelper class for MP and Legacy MP, we cannot MP
as local content provider in the same class.
Test: n/a
Bug: 253026009
Change-Id: I71cbbcc7c838eed5b333e0a485890b431c2b3f5b
|
|
With work profile, Media Provider process did not have the required
permission(media_rw) to set xattr on /data/media/<user-id>.
In this CL, we are making a change to set xattr on /data/media/0 for all
user profiles by adding user id suffix to xattr keys. In case of user id
reuse, it should work fine as next row id would not have been used
previously.
Also for devices with adoptable storage support, opting for adoptable
storage will not delete /data/media/0 directory.
Test: Manually on multiple user profiles and on adoptable storage
Bug: 216743720
Change-Id: I710e95742b41e19822ed7f80a5930b860611b19d
|
|
Entire project is 1p.
Test: m nothing
Change-Id: I89bfb50edff9c91cf680e883ec850c794cd510b4
|
|
- Add the @ExportedSince Build.VERSION_CODES.S on is_recording
- If the column is added @ExportedSince by X version, if the os
version is less than X, the query result won't include the column.
Test: atest MediaProviderTest
Test: atest DatabaseHelperTest
Bug: 202040814
Change-Id: Ib019feb0ef715ac1b828ae5c39140bcfb80cf51e
|
|
Use stand-alone modules-utils library.
Files import has changed accordingly.
local filegroup has been updated to java_library.
Bug: 198418216
Tag: #refactor
Test: atest MediaProviderTests MediaProviderClientTests
Ignore-AOSP-First: Merge conflict resolution
Change-Id: Ie77277cd990216ba6cf6b83885574223e0fca590
|
|
b043718839 am: 26a8485635 am: aaadd8fcca am: 5a77bd88ac am: 15e936c816
Original change: https://android-review.googlesource.com/c/platform/packages/providers/MediaProvider/+/1813477
Change-Id: I77b7557ed9ba9fa8b1df6807cd39a3e96f4a8ebd
|
|
b043718839
Original change: https://android-review.googlesource.com/c/platform/packages/providers/MediaProvider/+/1813477
Change-Id: Iba1f6c23d1fa931079a98250fbe0f4ae48bee5b8
|
|
Change-Id: I7a59c9ce791943f13885b553640e56335131085e
Fixes: 197891819
Test: android.provider.cts.media.MediaStoreTest
|
|
Currently, the DatabaseHelper assumes MediaProvider can only have 2
dbs, so if a db is NOT internal, it is external. For the photo picker,
we will introduce a separate db, picker.db which breaks this
assumption.
In preparation to introduce the picker.db, we refactor the code with
explicit methods to check if db is internal or external. In future cl,
we'll add an explicit check for picker db.
Test: presubmit
Bug: 190713331
Change-Id: Iab0fbe9a9047fc63a043685312a1e1fd394f9b48
|
|
am: 95cf803022 am: c8d5dbb7d3 am: 229793cb0f
Original change: https://android-review.googlesource.com/c/platform/packages/providers/MediaProvider/+/1588510
MUST ONLY BE SUBMITTED BY AUTOMERGER
Change-Id: Ide7b90d69f7d0fd266ca22003898d8d2d7de2aef
|
|
Added SPDX-license-identifier-Apache-2.0 to:
Android.bp
apex/Android.bp
apex/framework/Android.bp
apex/testing/Android.bp
errorprone/Android.bp
jni/Android.bp
legacy/Android.bp
tests/Android.bp
tests/client/Android.bp
tools/dialogs/Android.bp
Bug: 68860345
Bug: 151177513
Bug: 151953481
Test: m all
Exempt-From-Owner-Approval: janitorial work
Change-Id: Ibbbf07ceca50a345680a911a3860321ab71c83c3
|
|
Bug: 161526501
Test: atest --test-mapping packages/providers/MediaProvider
Change-Id: I844f463d7f98185a507d24c4d7970d7d1de2ec72
|
|
Previously, LegacyMediaProvider mapped all uris to "files" table. To
migrate playlist->audio map, we need to query "audio_playlists_map"
table. To be able to query "audio_playlists_table", LegacyMediaProvider
should map Playlist#Members uri to "audio_playlists_map" table.
This is a minimal change in LegacyMediaProvider to map Playlist#Members
uri to "audio_playlists_map" table. Any other uri still maps to "files"
table.
Bug: 165904660
Test: LegacyProviderMigrationTest#testLegacy_PlaylistMap
Change-Id: Ib6210c214bab32dfca2a2254e7ca4cf9ff31a6db
(cherry picked from commit b9f6864f54dade89afc7afad3a78da567f1c4f14)
(cherry picked from commit 9a5c84884f6268ba463864924965419a1772fe12)
|
|
Previously, LegacyMediaProvider mapped all uris to "files" table. To
migrate playlist->audio map, we need to query "audio_playlists_map"
table. To be able to query "audio_playlists_table", LegacyMediaProvider
should map Playlist#Members uri to "audio_playlists_map" table.
This is a minimal change in LegacyMediaProvider to map Playlist#Members
uri to "audio_playlists_map" table. Any other uri still maps to "files"
table.
Bug: 165904660
Test: LegacyProviderMigrationTest#testLegacy_PlaylistMap
Change-Id: Ib6210c214bab32dfca2a2254e7ca4cf9ff31a6db
|
|
This ensures the version code is higher than for the previous release.
This is also only for the legacy version.
Bug: 160230380
Test: version number is now the same as non-legacy
Change-Id: If63138b1e8dbe862beca3c0273a9f832d1b9d9e9
|
|
When migrating thousands of media items from a legacy provider, we
can't actually fit everything inside a single SQLite transaction,
since it'll fail with SQLITE_NOMEM.
To handle this, we need to shift the migrateFromLegacy() logic to
manage its own transactions, instead of running inside the single
transaction driven by the onUpgrade() flow. We do this by shifting
to onOpen() which has no automatic transaction.
We also add a "silent" option to the legacy provider to skip creating
thousands of media files during the newly added test; without this
we'd timeout.
Bug: 153842860
Test: atest MediaProviderClientTests:com.android.providers.media.client.LegacyProviderMigrationTest
Change-Id: Ieed9036ed7d308d8b60ce2025e8d43b418dd67a0
|
|
We recently tried fixing some obscure SQLite deadlocking in a51c3d1c,
but the cure was worse than the disease, since any threads just about
to perform a database operation before we close() the database would
encounter an "Cannot perform this operation because the connection
pool has been closed" exception.
Since we don't have control over the deep inner workings of SQLite
which result in the deadlock, the best we can do is to create a
ReadWriteLock where the "readers" are all normal database mutations
or queries, and the "writers" are schema modifications.
We're going to land this in two phases; the first phase here shifts
all database interactions to be driven by runWithTransaction() or
runWithoutTransaction(), which we'll use to acquire the upcoming
schema ReadWriteLock.
Any SQLiteDatabase instances obtained through getReadableDatabase()
or getWritableDatabase() would be unprotected by this schema lock,
so we adjust those methods to explicitly throw, to ensure that all
future changes interact with the database safely.
Bug: 152005629, 152802030
Test: atest --test-mapping packages/providers/MediaProvider
Change-Id: Ia79fb94e71352744006a7e1182dfeb112889bdf3
|
|
On create & rename, if the new path was previously deleted, saved
deleted row ID is restored.
Added custom database functions to generate row ID for insert and update
operations. Deleted row ID is restored only if insert and update are
initiated by FUSE thread.
Test: atest packages/providers/MediaProvider
Bug: 151076202
Change-Id: Ib6cc0b84f6f79988226cbd960395321facabafb6
|
|
We've seen evidence of the remote caller attempting multiple
migrations, so adjust the logic to ensure that we always have a
valid database while an active migration is taking place.
Also protect the Mainline code from a legacy provided that is
having trouble with lifecycle events.
Bug: 151960212
Test: atest --test-mapping packages/providers/MediaProvider
Change-Id: I3920bb652352ca3a08d2004fb5e3e135cbc0ce82
|
|
-- Promote generally-useful PFD API to public.
-- Add @WorkerThread annotations.
-- Add docs to explain how AUTHORITY_LEGACY works, and add explicit
START/FINISH_LEGACY_MIGRATION_CALL verbs to communicate progress.
-- Expand docs for createWriteRequest() to explain broader access
that it grants. Expand all request methods to indicate the action
is fully completed before the result is delivered.
-- Expand docs for new generation fields to explain that version
also needs to be inspected.
-- Explicitly call out why getKindSize() is deprecated, and guide
towards better alternative.
Bug: 148867182, 148403418, 148011561, 147760276
Test: none
Change-Id: Ibfae5a0b490d4b4b3f1725b72e246833f79c50e2
|
|
|
|
This change adds the forceQueryble attribute to the manifests of both
the module and legacy media providers so that consumers of the media
provider don't need to modify their manifests.
Test: atest ExternalStorageHostTest
Bug: 142386643
Change-Id: Ib6fce200e23b6bcaea001553fdef726c7e965711
|
|
Also adding a dependancy on the annotations lib to allow defining
changes.
Test: m
Test: flash device, inspect unpacked APEX and the config is present.
Bug: 142650523
Change-Id: I221d8fc945cf8eb2f1d360ef4be8b2db1fb18187
|
|
In upcoming changes we're preparing to start sending change
notifications that include the actual operation type, such as insert,
update, or delete. We could try gathering this up in MediaProvider,
but that's already proven tedious and hard to ensure that we always
gather the correct set of IDs that have changed. That would be even
more difficult since update() or delete() calls don't always have
easy access to the MEDIA_TYPE of the items being changed.
The most robust solution would be to always get direct notifications
from SQLite as media items are changed. Thankfully we can use the
recently added SQLiteDatabase.setCustomScalarFunction() along with
triggers to get precise information on exactly when and how items
change in the database.
This new approach fixes several subtle and long-standing bugs where
update() and delete() thumbnail invalidation and change notifications
may have been incorrect when called without a transaction held, or
not updating MediaDocumentsProvider state. It also lets us bypass
postProcessMetadataForFuseRename() entirely, since any database
changes performed from FUSE will now automatically result in the
right side-effects.
Bug: 144464323, 147778086
Test: atest --test-mapping packages/providers/MediaProvider
Change-Id: I261f410b47e08e407bc584581df1ed073566b1f8
|
|
Add generation numbers to the metadata for each media item. This is
useful for apps that are attempting to quickly identify exactly
which media items have changed since a previous point in
time. Generation numbers are monotonically increasing over time, and
can be safely arithmetically compared.
Detecting media changes using generation numbers is more robust than
using DATE_MODIFIED, since those values may change in unexpected
ways when apps use File.setLastModified() or when the system clock
is set incorrectly.
This is implemented by ensuring that all mutations to go through
our local SQLiteQueryBuilder, which then forces those mutations to
always happen inside an active transaction. Then each time we
create a transaction, we increment the generation counter, and
force-copy the current generation into any mutation operations.
This design also means that all mutations that should affect the
generation counter must go through DatabaseHelper to ensure that
the logic above is applied; there's mechanical refactoring across
MediaProvider to accomplish this.
Fix bug in getWritableDatabase() that could return a readable
database. Fix bug in LegacyMediaProvider to create missing
directories. Adjust logging script to enable log cleaner SQL from
SQLQueryBuilder instead of noisy raw logs.
Bug: 140248585
Test: atest --test-mapping packages/providers/MediaProvider
Change-Id: I15f05c79ed5ff445e20679dedafb1b800755458b
|
|
As requested by partners, this changes begins emitting important
events into a persistent log files on disk to aid in later debugging
investigations. This also logs non-standard ".nomedia" files that
we encounter on disk.
Also fixes bug where dump() would end up trying to obtain the
database from the main thread, which would could trigger ANRs.
Bug: 143723019, 142072796, 141506637, 137970002, 145014907
Test: atest --test-mapping packages/providers/MediaProvider
Change-Id: I5b681d9007fc344e87e53f4a9a37270f65a6b016
|
|
As a recent bug just reminded us, we can't have MediaProviderLegacy
depending on hidden APIs delivered through the APEX, since they're
unstable and can change at any time.
This requires a slight pivot of our original design strategy, where
the only shared code between the legacy and modern providers is the
DatabaseHelper logic. This requires some side-stepping to ensure
that DatabaseHelper and its dependencies safely build against
"system_current".
We also adjust DatabaseHelper to only log metrics when running as
the real production database, to avoid noise from the legacy provider
and unit tests.
These changes also helped uncover that our unit tests were
technically running against MediaProviderLegacy, which no longer
holds the bulk of the code that needs to be tested. It's tricky to
get an accurate instrumentation target because our tests live in
AOSP, but a device may have either the AOSP or Mainline version of
the module. Thus we pull everything directly into MediaProviderTests
to ensure that we can run tests against it.
Bug: 146670970
Test: atest --test-mapping packages/providers/MediaProvider
Change-Id: Ib257793f5d8a12eb83e18b0eed6284e25d7c011c
|