Predictive back gesture leads to fragments not being initialized/resumed correctly [297379023]

Fixed

Bug

Status Update

No update yet.

Description

pa...@gmail.com

created issue #1

Aug 24, 2023 06:09PM

Component used: Fragment

Version used: 1.7.0-alpha03

Devices/Android versions reproduced on: Pixel 6 with Android 14

The short summary of the issue is that once the predictive back gesture is completed (or started and canceled only), the fragments seem to be stacked on top of each other, with the top one displaying its contents without being interactive and the bottom one handling the touch input, which leads to crashes and unexpected behavior.

I've attached a minimal sample project to reproduce and describe the issues:

Navigate through the app and its two fragments without using the back gesture: everything works as expected.
On the second fragment, use the back gesture to navigate back to the first fragment. Then tap on the overflow menu: the app crashes unexpectedly with an IndexOutOfBoundsException in FragmentManager.
Relaunch the app and on the second fragment, start the back gesture but cancel it. Then tap on the overflow menu of the second fragment: it doesn't react anymore. Move slightly left with the finger and tap again to open the menu: this time it opens and the toast says "First fragment", which means the touch input is handled by the first fragment and contents are displayed by the second one. Now pause the app by moving it to the background, then resume it: the contents are fully gone and the overflow menu of the first fragment can still be opened by tapping on the top right corner (but slightly left of it because the overflow menu of the second fragment seems to cover the first). Using the back gesture will bring you back to the first fragment, which is then again acting as described in step 2.
Repeat step 3 without pausing the app, then tap on the disabled(!) "Go to second fragment" button of the second fragment: the app crashes with an IllegalArgumentException because the touch input is again handled by the first fragment and it's (enabled!) top button, so the executed navigation transaction (first to second fragment with the second currently displayed) is obviously illegal.

Hope you guys can fix that for the next alpha, thanks for your amazing work! :)

test-predictive.zip

115 KB

Download

Comments

il...@google.com <il...@google.com> Aug 24, 2023 06:28PM

Assigned to jb...@google.com.

ap...@google.com <ap...@google.com> #2Aug 31, 2023 02:04AM

Yigit, do you have time to fix it?

reemission of the same liveData is racy

jb...@google.com <jb...@google.com> #3Sep 5, 2023 03:02PM

Marked as fixed.

yea i'll take it.

pr...@google.com <pr...@google.com> #4Sep 6, 2023 07:56PM

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.

pa...@gmail.com <pa...@gmail.com> #5Sep 7, 2023 04:16PM

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()
}
}

an...@gmail.com <an...@gmail.com> #6Jun 20, 2024 08:49PM

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)

ho...@myrealtrip.com <ho...@myrealtrip.com> #7Oct 15, 2024 03:10AM

Comment has been deleted.

Message last modified on Oct 15, 2024 03:11AM

ha...@gmail.com <ha...@gmail.com> #8Oct 15, 2024 03:12AM

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