Replacing a NavBackStackEntry's args doesn't cause recomposition [186392337]

Fixed

Bug

Status Update

No update yet.

Description

jo...@jossiwolf.de

created issue #1

Apr 26, 2021 01:36PM

Component used: Navigation-Compose Version used: 1.0.0-alpha10, Compose 1.0.0-beta05

A bit short on time, so just posting a simple repro here. When trying to navigate to a destination that is the current destination but with different arguments using NavOptions.launchSingleTop, the NavController will replace the current NavBackStackEntry's arguments. Consider the following example:

@Composable
fun App() {
    val navController = rememberNavController()
    NavHost(
        navController = navController,
        startDestination = "home?arg={arg}"
    ) {
        composable(route = "home?arg={arg}") { backStackEntry ->
            val arg = backStackEntry.arguments?.getString("arg") ?: "Empty"
            Column {
                Text("Argument: $arg")
                Button(
                    onClick = {
                        navController.navigate("home?arg=value2") {
                            launchSingleTop = true
                        }
                    }
                ) {
                    Text("Click to navigate!")
                }
            }
        }
    }
}

In NavController#currentBackStackEntryAsState, the State's value won't be changed because the reference stays the same and the NavBackStackEntry held by the state has already been mutated, so the NavHost won't recompose after that change. Maybe the NeverEqualPolicy could be used as a quick fix.

Let me know if you need more info, cheers!

Comments

il...@google.com <il...@google.com> Jun 8, 2021 08:02PM

Assigned to jb...@google.com.

wa...@reaktor.fi <wa...@reaktor.fi> #2Jun 28, 2021 09:21AM

Yigit, do you have time to fix it?

reemission of the same liveData is racy

Screen Recording 2021-06-28 at 12.08.31.mov

1.5 MB

Download

th...@gmail.com <th...@gmail.com> #3Aug 11, 2021 05:48AM

yea i'll take it.

ap...@google.com <ap...@google.com> #4Aug 12, 2021 12:30AM

Thanks for the detailed analysis. This may not be an issue anymore since we've started using Main.immediate there but I' not sure; I'll try to create a test case.

jb...@google.com <jb...@google.com> #5Aug 12, 2021 03:02AM

Marked as fixed.

just emitting same live data reproduces the issue.

@Test
fun raceTest() {
val subLiveData = MutableLiveData(1)
val subject = liveData(testScope.coroutineContext) {
emitSource(subLiveData)
emitSource(subLiveData) //crashes
}
subject.addObserver().apply {
testScope.advanceUntilIdle()
}
}

ro...@gmail.com <ro...@gmail.com> #6Aug 12, 2021 04:02AM

With 2.2.0-alpha04 (that use Main.immediate), the issue seems to be still there (I tested it by calling emitSource() twice, like your test case)

jb...@google.com <jb...@google.com> #7Aug 12, 2021 02:47PM

yea sorry immediate does not fix it.
I actually have a WIP fix for it:

https://android-review.googlesource.com/c/platform/frameworks/support/+/1112186

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} )
}

ro...@gmail.com <ro...@gmail.com> #8Aug 12, 2021 04:29PM

Project: platform/frameworks/support
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

https://android-review.googlesource.com/1112186

https://goto.google.com/android-sha1/af12e75e6b4110f48e44ca121466943909de8f06

ro...@gmail.com <ro...@gmail.com> #9Aug 12, 2021 04:33PM

The issue in #comment2 is also still happening. For example, in the code from #comment1, notice how tapping the button does not update the value as expected: it always stays Empty. But if you leave the app open and toggle the system dark mode, now the app will get recomposed with the new value2.

If it helps, I can share the project, but it's really just the code from #comment1 put in an "Empty Compose Activity" project.

th...@gmail.com <th...@gmail.com> #10Aug 12, 2021 04:56PM

Comment has been deleted.

Message last modified on Aug 12, 2021 04:56PM

th...@gmail.com <th...@gmail.com> #11Aug 12, 2021 04:57PM

This is still reproducible with the latest snapshot version. Since this is marked as fixed, I've created another issue with the sample code and screen record(

https://issuetracker.google.com/issues/196244667)

Also, IMO, this is a core issue and should be marked as a high priority.

jb...@google.com <jb...@google.com> #12Aug 12, 2021 09:47PM

Status: Assigned (reopened)

Yeah, sorry looks like we still haven't fixed this. Somehow our test, which is very similar to the code in #1 passes, but in practice we still aren't recomposing.

ap...@google.com <ap...@google.com> #13Aug 24, 2021 10:43PM

Project: platform/frameworks/support
Branch: androidx-main

commit 5f1bdbbffc8c06355b22fdfe72ef1cf666b0d904
Author: Jeremy Woods <jbwoods@google.com>
Date: Wed Aug 18 03:45:03 2021

Ensure changing arguments causes recompose

Now that NavBackStackEntries are immutable and we don't just change
arguements, we change entire entries, we should make sure NavigatorState
properly handles launchSingleTop.

The only thing we need to do is to update both the navController and
navigator state back stacks with the proper entry.

RelNote: "Navigation Compose now properly recomposes when changing back
stack arguments and using launchSingleTop=true."
Test: all tests pass, and tested in samples from bug
Bug: 186392337
Fixes: 196390311

Change-Id: Iebd698c9e310ad84ce65238c8a5a33db86a9f1f7

M navigation/navigation-common/api/current.txt
M navigation/navigation-common/api/public_plus_experimental_current.txt
M navigation/navigation-common/api/restricted_current.txt
M navigation/navigation-common/src/main/java/androidx/navigation/Navigator.kt
M navigation/navigation-common/src/main/java/androidx/navigation/NavigatorState.kt
M navigation/navigation-runtime/src/androidTest/java/androidx/navigation/NavControllerTest.kt
M navigation/navigation-runtime/src/main/java/androidx/navigation/NavController.kt

https://android-review.googlesource.com/1799948

jb...@google.com <jb...@google.com> #14Aug 26, 2021 11:24PM

Marked as fixed.

We were able to get a fix for this in Navigation 2.4.0-alpha08.

You can try this out by following the snapshot instructions and using Build Id #7677699.

ma...@gmail.com <ma...@gmail.com> #15Jun 1, 2022 11:21AM

This still doesn't work in 2.4.2

jb...@google.com <jb...@google.com> #16Jun 1, 2022 04:39PM

Please file a new bug with a minimal sample project that reproduces the issue.

so...@gmail.com <so...@gmail.com> #17Nov 24, 2023 02:08PM

The same issue still exists.

il...@google.com <il...@google.com> #18Nov 27, 2023 06:23PM

Re comment#17 - you'll want to file a new bug with a sample project that reproduces the issue you're still experiencing.