Status Update
Comments
jb...@google.com <jb...@google.com>
be...@google.com <be...@google.com> #2
reemission of the same liveData is racy
ap...@google.com <ap...@google.com> #3
il...@google.com <il...@google.com> #4
b9...@gmail.com <b9...@gmail.com> #5
@Test
fun raceTest() {
val subLiveData = MutableLiveData(1)
val subject = liveData(testScope.coroutineContext) {
emitSource(subLiveData)
emitSource(subLiveData) //crashes
}
subject.addObserver().apply {
testScope.advanceUntilIdle()
}
}
be...@google.com <be...@google.com> #6
b9...@gmail.com <b9...@gmail.com> #7
I actually have a WIP fix for it:
if your case is the one i found (emitting same LiveData multiple times, as shown in #5) you can work around it by adding a dummy transformation.
val subLiveData = MutableLiveData(1)
val subject = liveData(testScope.coroutineContext) {
emitSource(subLiveData.map {it })
emitSource(subLiveData.map {it} )
}
b9...@gmail.com <b9...@gmail.com> #8
Branch: androidx-master-dev
commit af12e75e6b4110f48e44ca121466943909de8f06
Author: Yigit Boyar <yboyar@google.com>
Date: Tue Sep 03 12:58:11 2019
Fix coroutine livedata race condition
This CL fixes a bug in liveData builder where emitting same
LiveData source twice would make it crash because the second
emission registry could possibly happen before first one is
removed as source.
We fix it by using a suspending dispose function. It does feel
a bit hacky but we cannot make DisposableHandle.dispose async
and we do not want to block there. This does not mean that there
is a problem if developer disposes it manually since our emit
functions take care of making sure it disposes (and there is
no other way to add source to the underlying MediatorLiveData)
Bug: 140249349
Test: BuildLiveDataTest#raceTest_*
Change-Id: I0b464c242a583da4669af195cf2504e2adc4de40
M lifecycle/lifecycle-livedata-ktx/api/2.2.0-alpha05.txt
M lifecycle/lifecycle-livedata-ktx/api/current.txt
M lifecycle/lifecycle-livedata-ktx/api/public_plus_experimental_2.2.0-alpha05.txt
M lifecycle/lifecycle-livedata-ktx/api/public_plus_experimental_current.txt
M lifecycle/lifecycle-livedata-ktx/api/restricted_2.2.0-alpha05.txt
M lifecycle/lifecycle-livedata-ktx/api/restricted_current.txt
M lifecycle/lifecycle-livedata-ktx/src/main/java/androidx/lifecycle/CoroutineLiveData.kt
M lifecycle/lifecycle-livedata-ktx/src/test/java/androidx/lifecycle/BuildLiveDataTest.kt
ap...@google.com <ap...@google.com> #9
Branch: androidx-master-dev
commit 524a4f074b688975041f6ddd9db9eb31a88410e8
Author: Ben Weiss <benweiss@google.com>
Date: Tue Dec 15 12:26:06 2020
Add missing onViewCreated call
DefaultProgressFragment did not call super.onViewCreated which
could lead to installation progress not completing.
Bug: 169636207
Test: N/A
RelNote: Fix stuck installation progress
Change-Id: Ib27a77bc4572fc69cea38d738255c987449d6137
M navigation/navigation-dynamic-features-fragment/src/main/java/androidx/navigation/dynamicfeatures/fragment/ui/DefaultProgressFragment.kt
b9...@gmail.com <b9...@gmail.com> #10
Why this fix doesn't included in 2.3.3?
il...@google.com <il...@google.com> #11
Re #10, yes, you're right this didn't make it into Navigation 2.3.3. Sorry about that. We'll look at doing a Navigation 2.3.4 with this fix.
b9...@gmail.com <b9...@gmail.com> #12
za...@gmail.com <za...@gmail.com> #13
I believe I'm facing another issue because of this fix:
When navigating to an activity that belongs to a dynamic module, the AbstractProgressFragment
is not removed from the back stack. The line that's supposed to do it was moved from onResume
to onViewCreated
:
if (navigated) {
findNavController().popBackStack()
return
}
And onViewCreated
is not called since we're coming back from another activity.
Moving that block back to the onResume
method should solve the problem. In the meantime (as a workaround), I guess we can do as follows:
class CustomProgressFragment : AbstractProgressFragment() {
private var navigated = false
override fun onResume() {
super.onResume()
// Workaround:
// AbstractProgressFragment is not being removed from the back stack when returning from another activity
if (navigated) findNavController().popBackStack()
}
override fun onInstalled() {
testing = true
}
...
...
}
za...@gmail.com <za...@gmail.com> #14
Update:
Regarding the workaround I mentioned, this is the correct implementation of the onInstalled
method
override fun onInstalled() {
navigated = true
}
il...@google.com <il...@google.com> #15
Re onViewCreated()
is indeed called when you return to the AbstractProgressFragment
from another fragment, which is the only thing your newly loaded graph should have as its start destination. You...don't happen to have a start destination that is an activity destination? You shouldn't be using Dynamic Navigation at all in that case (as the system will auto-download the module your activity is in upon request).
Description
Component used: androidx.navigation:navigation-dynamic-features-fragment
Version used:2.3.0 Devices/Android versions reproduced on: all devices/android versions
override fun onResume() { super.onResume() if (navigated) { findNavController().popBackStack() return } var monitor = installViewModel.installMonitor if (monitor == null) { Log.i(TAG, "onResume: monitor is null, navigating") navigate() monitor = installViewModel.installMonitor } if (monitor != null) { Log.i(TAG, "onResume: monitor is now not null, observing") monitor.status.observe(this, StateObserver(monitor)) } }
Any time onResume() is called, monitor.status observed. When dynamic feature module is installed, each StateObserver will call navigate() and destination activity/fragment will be navigated multiple time