UPDATE: It seems the error is for steps lower than 4secs (=0.067mins), this is the reason <0.067min intervals (like 0.05 0.02) give error
To show the card and be able to select easy/hard etc and show the next card
As in Anki PC it allows you to set intervals with decimal spaces, which is very useful especially in hard cards
When I open a deck with settings with decimals (f.e. one that synced from pc), it shows the card ok, but when I tap an option "easy" "hard" etc which though includes decimals in scheduling f.e. 0.1min, 1.5min then it sends an error as shown in the picture saying: "Collection not opened" an error occurred while accessing your collection...
Why I know it is the decimals? If I set the intervals in integers f.e. 1min 5min 2min, then everything is ok.

happen only one time?
Every time
Refer to the support page if you are unsure where to get the "debug info".
AnkiDroid Version = 2.14.2
Android Version = 9
ACRA UUID = dc880539-d555-4065-9ea9-b0d24e7f1564
Scheduler = std2
Crash Reports Enabled = true
Enter an [x] character to confirm the points below:
PLEASE fix soon so I can study from my phone optimized. thanks!!!
Hello! 馃憢 Thanks for logging this issue. Please remember we are all volunteers here, so some patience may be required before we can get to the issue. Also remember that the fastest way to get resolution on an issue is to propose a change directly, https://github.com/ankidroid/Anki-Android/wiki/Contributing
Very interesting, thanks for reporting this. Probably just an integer expectation as you imply
Interesting - I just tried this and I cannot reproduce it from current version, can you install the "parallel" version of 2.15alpha17 in order to check ?
It will be side-by-side with your main ankidroid and...maybe work with these non-integer review settings
I can't reproduce either. But now idea why this would have been solved between the reported version and master
Well, assuming @Cashburner cannot reproduce in 2.15alpha17 either only way to know for sure would be to make a test and do a git bisect vs the code, or - something you could try @Cashburner would be to bisect vs the parallel versions of the 2.15alpha series if you can't reproduce it with 2.15alpha17 either. Just start with "half of 17" (so, 2.15alpha8) to see if it works there or not, if it crashes then try half-of-the-other-half (so between 8 and 17, maybe 2.15alpha12) to re-check, and keep splitting the difference until the alpha that fixes it, that would narrow it down and we might find it.
I don't have the time for that personally, so in the absence of someone else bisecting I will leave it as is with next version just fixing it. I imagine it is an edge case used by even fewer than the few people that actually change these settings. It should definitely work just I imagine not many will run in to it?
just saw the replies, give me a minute I am going to try all these.
(So after some testing it seems that in 2.15alpha17 the same happens.) But this is not the WOW here:
Here is something interesting:
Screenshots:
0.067>4secs OK
0.0659<4sec error


Fascinating! Okay, so likely nothing changed (actually what I was thinking, based on my knowledge of changes between 2.14.2 and 2.15alpha-current) but it was slightly different data than we were testing.
So looks like we have a data validation issue of some type plus a side issue of number formatting. Not sure how many digits to show but it should probably confirm with upstream.
Thanks a bunch for testing this, and I think it's on us now to triage again
Thanks a lot for the help! Happy Christmas guys with health to you and your family. (I know it's not a place to say this but why not).
Also want to ask, what's the timeline for fixing this issue? I can't wait to use ankidroid again :)
Indeed - it is holidays and I have a small child so I'm not working very much sorry. I can't make a promise on timeline but hopefully soon