1530143 - Crash in [@ java.lang.NullPointerException: at android.app.assist.AssistStructure$WindowNode.<init>(AssistStructure.java)]

Status: RESOLVED INCOMPLETE

(Firefox for Android Graveyard :: General, defect, P3)

Description

[Julian Seward [:jseward]]

This bug is for crash report bp-c143b049-cb4d-4e6b-a707-9923b0190221.

Seen in the Android nightly for 20190221093143.

Java stack trace:

java.lang.NullPointerException
	at android.app.assist.AssistStructure$WindowNode.<init>(AssistStructure.java:484)
	at android.app.assist.AssistStructure.<init>(AssistStructure.java:1908)
	at android.app.ActivityThread.handleRequestAssistContextExtras(ActivityThread.java:3574)
	at android.app.ActivityThread$H.handleMessage(ActivityThread.java:2238)
	at android.os.Handler.dispatchMessage(Handler.java:108)
	at android.os.Looper.loop(Looper.java:166)
	at android.app.ActivityThread.main(ActivityThread.java:7523)
	at java.lang.reflect.Method.invoke(Native Method)
	at com.android.internal.os.Zygote$MethodAndArgsCaller.run(Zygote.java:245)
	at com.android.internal.os.ZygoteInit.main(ZygoteInit.java:921)

Comment 1

[James Willcox (:snorp) ([e-mail]) (he/him)]

Eitan, this is crashing in Android, but maybe due to something we gave it?

Comment 2

[Eitan Isaacson [:eeejay]]

I'm not sure. Looks like it is crashing because the RootViewImpl's view is null[1]. Possibly this message is handled after the view is detached from the window?

Three quarters of the crashes don't have a11y enabled so this isn't exclusively an a11y issue. Although 25% with a11y enabled is unusually high, perhaps there is correlation with password managers that flip on accessibility and also use the autofill api.

https://android.googlesource.com/platform/frameworks/base.git/+/refs/tags/android-8.0.0_r34/core/java/android/app/assist/AssistStructure.java#484

Comment 3

[David Bolter [:davidb] (NeedInfo me for attention)]

Extremely low volume.

Comment 4

[Chris Peterson [:cpeterson]]

Eitan, is this a11y code path relevant for GeckoView in Fenix? Or is this a Fennec-specific code path?

Comment 5

[Eitan Isaacson [:eeejay]]

The crash data suggests that this is only in Fennec, although this is rare enough that it's hard to know.

Comment 6

[Chris Peterson [:cpeterson]]

(In reply to Eitan Isaacson [:eeejay] from comment #5)

The crash data suggests that this is only in Fennec, although this is rare enough that it's hard to know.

In that case, I'll move this bug to the Firefox for Android Bugzilla component. The GeckoView team is trying to figure out what to do with bugs in the "Core::Widget: Android" component when we also have separate components for Firefox for Android and GeckoView.

Comment 7

[BMO Automation]

We have completed our launch of our new Firefox on Android. The development of the new versions use GitHub for issue tracking. If the bug report still reproduces in a current version of [Firefox on Android nightly](https://play.google.com/store/apps/details?id=org.mozilla.fenix) an issue can be reported at the [Fenix GitHub project](https://github.com/mozilla-mobile/fenix/). If you want to discuss your report please use [Mozilla's chat](https://wiki.mozilla.org/Matrix#Connect_to_Matrix) server https://chat.mozilla.org and join the [#fenix](https://chat.mozilla.org/#/room/#fenix:mozilla.org) channel.