New user (Juno) onboarding does not always return back to original orientation
Categories
(Fenix :: Onboarding, defect, P5)
Tracking
(firefox112 unaffected, firefox113 wontfix, firefox114 wontfix, firefox115 wontfix)
| Tracking | Status | |
|---|---|---|
| firefox112 | --- | unaffected |
| firefox113 | --- | wontfix |
| firefox114 | --- | wontfix |
| firefox115 | --- | wontfix |
People
(Reporter: jonalmeida, Unassigned)
References
(Regression)
Details
(Keywords: regression)
Attachments
(1 file)
Landscape mode tablet is stuck in portrait mode after Juno onboarding is complete.
Steps to reproduce
- Open a fresh install of Fenix on a tablet.
- Observe the Juno onboarding page orientation.
- Finish the onboarding.
- Observe the app's orientation.
Expected behaviour
- At step 2, the onboarding changes to portrait mode. Also see bug 1826944.
- At step 4, the onboarding changes back to the default (landscape).
Actual behaviour
- At step 2, the onboarding changes to portrait mode.
- At step 4, the onboarding stays in portrait mode.
Device information
- Firefox version: Nightly 112 (latest).
- Android device model: MalataMediatekArgon2
- Android OS version: Android 9
Any additional information?
|
Reporter | |
Comment 1•2 years ago
|
||
It looks like orientation change was intentional and is being reset here in onDestroy which is not guaranteed to be sent.
Maybe we can to change orientation in onStop - will that suffice for this feature's use case?
|
|
Reporter | |
Comment 2•2 years ago
|
||
Added a screenshot of the bug I saw with an embedded tablet.
|
||
Comment 3•2 years ago
|
||
I'm unable to reproduce this. Tested on Pixel 7, Samsung Galaxy A50 and Nexus 9 tablet emulator.
I've moved the calls to onStart and onStop as a preemptive measure, and they do seem the better place for set/getRequestedOrientation. It behaves as intended in my testing, but it worked fine for me, even when the calls were in onCreate and onDestroy. Could you please help me test that? This PR adds that.
|
||
Comment 4•2 years ago
|
||
Set release status flags based on info from the regressing bug 1826387
|
||
Updated•2 years ago
|
|
|
||
Comment 5•2 years ago
|
||
The severity field is not set for this bug.
:skhan, could you have a look please?
For more information, please visit auto_nag documentation.
|
||
Comment 6•2 years ago
|
||
Jonathan, could uyou help Rahul? (comment #3) Thanks
|
|
Reporter | |
Comment 7•2 years ago
|
||
I tested out Rahul's patch offline and was still seeing the bug reproduced (see discussion here).
I'm reducing the severity and priority of this bug because:
- The device is a niche one and not reproducible on a tablet emulator.
- I do not have easy access to the device in question.
- Tablet/landscape is still a small population of our userbase.
- The bug is of minor significance.
Marking as wontfix for 115 as well.
Description
•