go/cleanup-greylist-txt
These have already been greylisted, however due to bugs/omissions in the tooling have been kept in go/greylist-txt instead of being annotated in the code.
Bug: 137350495
Test: m
Change-Id: Id6c1f5e403a0e66edb1102ee45f3bf19f244fb09
Merged-In: Id6c1f5e403a0e66edb1102ee45f3bf19f244fb09
See go/UnsupportedAppUsage for more details.
These have already been greylisted, however due to bugs/omissions in the tooling have been kept in go/greylist-txt instead of being annotated in the code.
Bug: 137350495
Test: m
Change-Id: I5aa29a49b193db47aaee4d3a756c17f48cc9f0b1
Merged-In: I5aa29a49b193db47aaee4d3a756c17f48cc9f0b1
When the protobuf header (or data payload) size are a multiple of 4,
"entry_length" (called "aligned size" in the code) would have counted bytes that
are never actually written.
Bug: 139418052
Change-Id: Ia688a82a67f3807f7feb0be03670bf2827b1d6a1
(cherry picked from commit b99e50922bf8a5eb9f034c1d232646fda122e83c)
std::set only works correctly when the < comparator is a strict weak
ordering, while UsageLocation::operator< was actually implementing !=.
Bug: 134190468
Change-Id: Icb9407e9c8451f9fcb4eb9b2cea310e3bcaf159e
Tested: aapt2_tests, and b/134190468#comment1
(cherry picked from commit 44fa342eb91b90df7998fa2808e21af75aafaf39)
Add new SystemApi methods for Telecom to call in lieu of the old @hide
methods. Also moved CallerInfo and CallerInfoAsyncQuery to Telecom
because there are no more users in Telephony.
Bug: 141576016
Test: CTS
Change-Id: I458ba6bcfc03db72c0419b0cab2f0d0adfa971d4
Merged-In: I458ba6bcfc03db72c0419b0cab2f0d0adfa971d4
* changes:
Implement ApexManager UnitTest
Refine getPackageInfo for Apex match policy
ApexManager: Post populateAllPackagesCacheIfNeeded to a background thread
Move expensive code from BroadcastReceiver to background thread
The legacy match policy could confuse, for example,
the naming of MATCH_APEX do not named with "ONLY|FACTORY|ACTIVE",
it's make sense to joint both active and factory apexes.
Policy refined:
1. MATCH_FACTORY_ONLY : Return *factory* [eg. the version on the system image] APK packages only
2. MATCH_APEX : Return *active* APK or APEX packages
3. MATCH_FACTORY_ONLY | MATCH_APEX : Return *factory* APK or APEX packages
4. NONE (0) : Return *active* APK packages only
Fixes: 142365486
Bug: 142115351
Test: adb shell pm path $APEX_PKG_NAME
Test: adb shell pm path $APK_PKG_NAME
Test: adb shell pm list packages --apex-only
Test: atest ApexManagerTest
Test: atest CtsStagedInstallHostTestCases
Exempt-From-Owner-Approval: Already approved in
https://googleplex-android-review.git.corp.google.com/c/platform/frameworks/base/+/9524811/
Change-Id: I0077cc8dcabdcf7b67374e8b059b62158fa43714
Merged-In: Ifa01278201dd23672dfc1a30f67b888df9bc1bfd
It's the first patch for refounding network selection. The new
network selection will try to compare the networks by more
factors, and will try to choose the best network after doing
the evaluation of trade-off.
Create the object that will serve to represent network quality
for more comprehensive ranking.
Bug: 143676287
Test: 1. Build pass.
2. atest FrameworksNetTests
Change-Id: I4b6071d14365aa84d06be9802516fedf527e70f7
* changes:
Clean up generatePackageInfoFromApex() API
Populate error message if apexd.markStagedSessionReady fails
Populate error message if apexd verification fails
Make ApexManager an abstract class
Don't use ApexInfo.packageName in StagingManager
In Mms test log we found StrictMode printed out DrmManagerClient was
not released, it was created in PduPersister constructor.
To fix this problem, call mDrmManagerClient.release in PduPersister.
release.
Bug: 143230740
Test: Edit and send MMS, not print this log again.
Change-Id: If3318c8e77d6272b9d83af57d910951055356d99
For HIDL, vibrator may change the value of supportsAmplitudeControl
based on context. Since the interface is being redone in AIDL, we're
avoiding this changing value and providing capabilities for amplitudes
we expect to be available in different states.
Bug: 141828236
Test: boot
Change-Id: If1e38cffa3fd93f2a658c476637db2dd7ce9f91d
Vibrator just tries to reconnect when commands fail, but because of the
HIDL->AIDL shim's handling of errors, we were dropping binder-level
errors, and the client code wasn't properly handling them. Now, we
convert dead object and other errors appropriately so that the client
code is given the errors it is expecting.
Bug: 141828236
Test: kill aidl vibrator HAL repeatedly, and ensure that it reconnects
and starts handling vibrator commands.
Change-Id: I9719ec030f514905f1c5e57a1bdf455956d88e76
Update ImsCallProfile to indicate the verstat for incoming calls.
Test: Run new GTS tests.
Bug: 135929421
Merged-In: I712a42836382e8929e40b887fd01c450a4096bab
Change-Id: I712a42836382e8929e40b887fd01c450a4096bab
populateAllPackagesCacheIfNeeded is rather expensive method and
shouldn't be run in broadcast handler thread (it was my mistake to
suggest this in ag/9204784).
It's fine to use background thread to execute
populateAllPackagesCacheIfNeeded because components that need
to match APEX are not doing it on user critical path.
Bug: 138460874
Test: adb shell pm list packages --apex-only --show-versioncode
Change-Id: I738e932d94e86e897cee2b82f658d0e89328c6d3
Merged-In: Ib17fadb978f112c26c113e3bc42b9edaeeefca8a
This changes the API for generatePackageInfoFromApex in PackageParser
in response to review suggestions made in ag/6701090.
The method is renamed to generatePackageInfo with a PackageParser.Package
parameter (as well as an ApexInfo and flags). The implementation is combined
with the main generatePackageInfo method. Callers are changed to perform
the parsing themselves, including the collection of certificates if
necessary.
Note: the method signature and implementation had already changed since
ag/6701090, so the suggestions from that CL may not apply directly.
Bug: 129261524
Test: atest PackageParserTest
Test: atest CtsStagedInstallHostTestCases
Change-Id: Iee213e025583c2201ff9d65c447e4278b925a2f0
Merged-In: I495539679812110f89d5e3d93f2622ba2dab36c0
(cherry picked from commit c2c7a22333b828cd3a630fe9efb1b18b0e6af289)
And a two implementations:
* ApexManagerImpl - in case device supports updatable APEX.
* ApexManagerNoOp - in case device does not.
This is a more future-proof way of handling both new flow that supports
APEX and legacy one that does not.
Test: atest CtsStagedInstallHostTestCases
Bug: 131697251
Exempt-From-Owner-Approval: already approved in
https://googleplex-android-review.git.corp.google.com/c/platform/frameworks/base/+/8483219/
Change-Id: Ieeeb2c7f52ff63ebf9c7e913e95e7cf62d8c7441
Merged-In: If27fcdb95d4ec60834bd5334b68f53d4235a5d61
(cherry picked from commit 278af8f43c72a67dd9c471b159e5e184fb3e5f0d)
* ApexInfo.packageName has a misleading name since it represents name
of apex module (e.g. com.android.tzdata), not the packageName from
AndroidManifest.xml of an apex (e.g. com.google.android.tzdata). Later
should be used in PackageManager-world while former is apexd-internal
name.
* Restructured code a little bit to make the pre-reboot verification
flow easier follow.
Test: atest CtsStagedInstallHostTestCases
Test: adb shell pull /system/apex/com.android.resolv.apex /tmp/resolv.apex &&
adb install /tmp/resolv.apex &&
adb reboot &&
<verify staged session was applied>
Bug: 132324953
Change-Id: Ifd88b5e5df84ba7a6aebd70dced7b9ab48ed0159
Merged-In: Iedfe87f6e70a2e1d2792487ac7b2891693ceb72a
(cherry picked from commit e0dbc98f954be1661676436b18885bc2e7d29119)
The com.android.okhttp and com.android.org.conscrypt packages are no
longer present in the input source so do not need hiding.
Bug: 142113521
Test: m checkbuild
Change-Id: I303b054ec114450b5fce3482eddf3cc3749d3814
