The previous runAsShell calls adopted permissions spuriously. This
change fixes runAsShell to consistently adopt the correct permission
for the respective API call.
Also calls setEthernetEnabled() no matter what the current state is at
the end of the test (as there is no harm in doing so).
Test: TH
Flag: EXEMPT mainline
Change-Id: Ib11b889148213b160567cf0a3b9ea72b5b0c851e
This is a no-op, test only refactoring.
Move ethernet state management logic from EthernetManagerTest into a
dedicated JUnit test rule, RestoreEthernetRule. This new rule is
responsible for saving the initial ethernet state, allowing tests to
enable or disable it, and restoring the original state upon test
completion. This refactoring simplifies EthernetManagerTest and
promotes code reuse.
Flag: EXEMPT EXEMPT test only refacotring
Test: atest NetworkStaticLibTests:com.android.testutils.RestoreEthernetRuleTest \
CtsNetTestCases:android.net.cts.EthernetManagerTest --update-device
Bug: 410731080
Change-Id: If3dd869d6af4e0c659386621fb5d92012aa574d1
The monolithic EthernetStateListener has been refactored and split
into two specialized test classes:
- TestInterfaceStateListener: For interface state callbacks.
- TestEthernetStateListener: For ethernet state callbacks.
Both new classes now extend TestableCallback, providing a standardized
and more robust way to manage and verify listener interactions in
unit tests. This improves testability and clarity of test code.
This is a no-op, test-only change.
Flag: EXEMPT refactor
Test: atest CtsNetTestCases:android.net.cts.EthernetManagerTest
Bug: 410731080
Change-Id: I667e3607954dcb073b2732b16bedfbc71a70bb39
This is a no-op, test-only change.
Move EthernetStateListener from EthernetManagerTest.kt to a new
standalone file in testutils.
This refactoring makes the listener reusable across other tests and
decouples it from its dependency on EthernetTestInterface.
Flag: EXEMPT refactor
Test: atest CtsNetTestCases:android.net.cts.EthernetManagerTest
Bug: 410731080
Change-Id: Iacd9236f758011da0714884040aebe3e42ef1646
Junyu Laiauthored and Chalard Jeancommited 2025-07-04 13:03:55 +09:00
This change introduces a utility class to disable
Bluetooth before testing, in case it was previously connected
as the default network.
Flag: EXEMPT test only
Test: atest NetworkStaticLibTests:com.android.testutils.AutoReleaseBluetoothRuleTest \
--update-device
Bug: 410731080
Change-Id: I95afc6174698bf73b21d95a85c02e107ee0d59f7
This makes it a bit easier to use FakeSettingsProvider and
get settings change notifications. It's still difficult because
the ContentResolver#registerContentObserver methods are final and
cannot be subclassed, but at least it removes the need for some
boilerplate code.
Bug: 397521796
Flag: TEST_ONLY
Test: tests updated in next CL
Change-Id: I43b45eca964b103ad65d0db82682aa8644fcacaf
Lorenzo Colitticommited 2025-06-28 11:06:53 +09:00
RestoreDefaultNetwork is a virtual test, causing a lot of confusion among
testers who can't reproduce or run this test.
Instead, blame the test that actually failed to restore the default network.
Before :
```
[1/2] android.net.cts.MultinetworkApiTest#testFoo: PASSED (253ms)
[2/2] android.net.cts.MultinetworkApiTest#RestoreDefaultNetwork: FAILED (54ms)
STACKTRACE:
java.lang.IllegalStateException: testFoo does not restore the default network. Default network has caps null ; expected a network with transports = WIFI
at com.android.testutils.DefaultNetworkRestoreMonitor$listener$1.testFinished(DefaultNetworkRestoreMonitor.kt:64)
```
After :
```
[1/2] android.net.cts.MultinetworkApiTest#testFoo: PASSED (353ms)
[2/2] android.net.cts.MultinetworkApiTest#testFoo: FAILED (54ms)
STACKTRACE:
java.lang.IllegalStateException: testFoo does not restore the default network. Default network has caps null ; expected a network with transports = WIFI
at com.android.testutils.DefaultNetworkRestoreMonitor$listener$1.testFinished(DefaultNetworkRestoreMonitor.kt:64)
```
Test: this
Flag: EXEMPT test-only
Bug: 413188127
Change-Id: Ia17bc9ddeedbd9303e5788ef79fb190e6e9dca1e
This will make these agents testing a lot easier.
Also fix a small bug where waiting for callbacks that
are not expected would take a lot longer than it should.
Also trivial improvements to logging that I needed to stabilize
this.
Test: TH
Flag: EXEMPT test-only
Change-Id: I654d63eb226990ad1c5087cade7cf0a82d8526a1
Have TestableNetworkCallback adopt it, as a first client. It's also by
far the most complex.
Test: TH
Flag: EXEMPT test-only
Change-Id: Ie58d051b4478c901b7f0f54e89a67d22f1ee2ce2
This reverts commit a77d5ac126.
Reason for revert: Droidmonitor created revert due to b/422782308. Will be verifying through ABTD before submission.
Fix: 422782308
Change-Id: I00825e0f17a9a9e0d152ac8ebe767da3f267fbb1
Have TestableNetworkCallback adopt it, as a first client. It's also by
far the most complex.
Test: TH
Flag: EXEMPT test-only
Change-Id: Ifd636a09d0695f04b17330c4987f2b7b742508fa
This class doesn't actually serve a purpose, it only makes
TestableNetworkCallbck more complex.
Test: CtsNetTestCases ConnectivityCoverageTests
Flag: EXEMPT test-only
Change-Id: I39a598b82a3215384bdfe04980f4b280ab2c2716
The IA_PD Option is modeled as a separate "packet", because it can
contain various options itself. It turns out that with the current
packet builder architecture, building nested packets (of which DHCPv6 is
an extreme example) is actually quite difficult. We would encounter
similar problems if we ever were to try and build netlink packets.
One thing that could be done is to add another overload for
PacketBuilder#div that takes another PacketBuilder object, allowing an
syntax such as:
dhcp6_header / dhcp6_rt_sol_max / (dhcp6_ia_pd / dhcp6_ia_prefix)
Test: TH
Flag: EXEMPT mainline
Bug: 1000001
Change-Id: I35694627285759dd17d8a8bf5994aaa96e6b391d
This adds support for building the Dhcp6Pkt header. Options are added
later.
Test: TH
Flag: EXEMPT mainline
Bug: 1000001
Change-Id: I588d8fc6ba68d1132a3beb1b436d8175ac0ae577
Note that a new neighbor can still be added while processing a packet,
which may influence the result, though this shouldn't affect tests.
Test: TH
Flag: EXEMPT mainline
Bug: 1000001
Change-Id: If6860f3a52a406bbbbdd323061bfb918f1fb866c
This is a shared public utility. Its main utility is for methods
that can fail, have an error message of their own, and also want
to take an error message from a caller and want to combine them.
Example use :
fun <T> assertEquals(expected: T, actual: T, msg: String? = null) {
if (expected != null) {
failWithErrorReason(msg, "Expected $expected, got $actual")
}
}
...which allows someone to say
assertEquals(a, b, "Checking a vs b")
Test: TH
Flag: EXEMPT test-only change
Change-Id: Id3eaa8384b4177f38f13576d9b4d742c52027ae4
Update it to follow the latest conventions and add support for
eventuallyExpect.
Test: ConnectivityCoverageTest
Flag: EXEMPT test-only change
Change-Id: Iabc2fad051b874e3e18a0645361409f0427891fb
Addressing leftover comments in ag/33040619 regarding carrier privilege
acquisition in CarrierConfigRule with the following changes:
1) Refactored CarrierPrivilegeCallback to match records using a lambda.
2) Added various helper methods to handle calls with permissions.
3) Moved wait for carrier privlege to overrideConfigAndWait.
4) Fixed acquireCarrierPrivilege to append to current cert strings
instead of overwriting them.
5) Fixed dropCarrierPrivilege to remove from current cert strings
instead of cleaning up all configs.
Flag: EXEMPT test only change
Bug: 409710349
Test: atest TetheringManagerTest
Change-Id: I1165d95775e809f1aa8905af029fb13df0d3a1a8
Quang Anh Luongcommited 2025-04-25 12:50:32 +09:00
Use a common code path for getTestableNetworks, which obtains Wi-Fi and
cellular from network callbacks. This avoids mixing callbacks and
synchronous calls, as getTestableNetworks may be called after
ensureWifiConnected, which uses NetworkCallbacks.
Test: atest
Bug: 253698734
Flag: TEST_ONLY
Change-Id: I8bf946b5258270554286e52b8aa3189aa2ba52e1
@JvmSerializableLambda was used in NetworkAgentTest to get around the
link error caused by JUnit enumerating through lambda methods that
referenced classes that weren't available on older platform versions.
Since this code was ported from NetworkAgentTest to CarrierConfigRule
whose lambda methods are not enumerated by JUnit, this workaround isn't
necessary anymore.
Flag: EXEMPT test only change
Bug: 409710349
Test: atest TetheringManagerTest
Change-Id: Ib876e237b239661548ae3c1d23fcb52f389f6254
Quang Anh Luongcommited 2025-04-17 17:10:44 +09:00
CarrierConfigRule cleanup does not wait for carrier privilege value
to change before continuing to the next test, which may cause test
failures if the value isn't fully reset on time.
Make sure the value has time to properly be reset by waiting for
onCarrierPrivilegesChanged before proceeding.
Flag: EXEMPT test only change.
Bug: 409710349
Test: atest TetheringManagerTest#testCarrierPrivilegedIsTetheringSupported TetheringManagerTest#testCarrierPrivilegedStopTetheringWifiWithConfigSucceeds
Change-Id: If67884514a2c1678834ab9ff67947e75e37db0a7
Quang Anh Luongcommited 2025-04-16 15:02:54 +09:00
Line breaks make tests harder to read, so this change should help
fitting everything on a single line.
Test: TH
Flag: EXEMPT mainline
Bug: 1000001
Change-Id: I286c4398a2f738ab82498b93f9f9344577befeb8
Renames srcPort to sport and dstPort to dport and generates a random
port number in the range of [1, 65536) if no argument is passed.
Test: TH
Flag: EXEMPT mainline
Bug: 1000001
Change-Id: If446cde9a7ff414f52c60f70458ff046a2c4dc33
Figuring out what packets were received is useful for debugging.
Test: TH
Flag: EXEMPT mainline
Bug: 1000001
Change-Id: I3184845b76e33da186f39a2344548c2d16715173
Bundle the version of the tethering APEX in the build branch into
CtsNetTestCases as a resource, and include it in the failure header on
test failure.
This should make it easy to identify what version of the module matches
the built test when investigating test failures.
Test: atest
Flag: TEST_ONLY
Change-Id: I15a53b3d7649ebde9706b10ee504d08aa78ee7de
Not all U devices support this. It's only supported by U QPR1+.
Bug: 404415649
Test: NetworkAgentTest
Flag: EXEMPT bugfix
Change-Id: I66d30b5f0be5628561bfeb1f90f5318954544e31
Chalard Jeanauthored and Jean Chalardcommited 2025-04-08 00:51:44 -07:00
UDP packets can be built as:
val p = ether / ipv6 / udp / data
val p2 = ether / ipv4 / udp / data
val p3 = ether / ipv4 / udp / ipv4 / udp / data
This also gets rid of Icmp6Pkt in favor of a function call that
calculates the checksum over an L4 packet.
This does add a bit of duplication: the build() function for RA and NA
packets is identical, and UdpPkt#build() is not far off. However, subtle
differences in L4 packet format (namely the checksum and length
positions) make generalizing this code difficult. This is possible by
passing the checksum position, proto byte, and adding an abstract
function that fills in the L4 length field before performing checksum
calculation. However, this complexity is likely not worth it. If the
repetition becomes obvious once more protocols (such as ICMPv4/6 echo,
etc.) are added, this decision will be revisited.
Test: TH
Flag: EXEMPT mainline
Bug: 1000001
Change-Id: Ia78792418f29a3be091ebdc9bb2c26e45da9f2b1
Patrick Rohrauthored and Maciej Żenczykowskicommited 2025-04-07 14:22:43 -07:00
Options and fragments aren't currently supported.
Test: decoded test packet on packetor.com
Flag: EXEMPT mainline
Bug: 1000001
Change-Id: Ia3a094cb1d2b093cae2f372cc84a72ad7e107105
Patrick Rohr
commited 
Junyu Lai
commited 
Chalard Jean
commited 
Lorenzo Colitti
authored and
Android (Google) Code Review
commited 
Priyanka Advani (xWF)
authored and
Maciej Żenczykowski
authored and
Quang Luong
authored and
Treehugger Robot
authored and
Remi NGUYEN VAN
commited