loading spinner

[AlperK61]

closes #208

[a-miscellaneous]

generating differences is sadly not cancelable, job.cancel does not work.
parentJob lauch generation and then parentjob.cancel also does not work

[akriese]

What do we do with this then? @a-miscellaneous @AlperK61

[a-miscellaneous]

What do we do with this then? @a-miscellaneous @AlperK61

next sprint, conflicts left

[akriese]

We have to make the difference generation cooperative for cancelling. Then, job.cancel() will work. I'll look into that now: https://kotlinlang.org/docs/cancellation-and-timeouts.html#making-computation-code-cancellable

[akriese]

After doing my research, it is pretty clear, that we can only cancel the job by injecting some kind of suspend function into the Algorithm.run() or some check if the job is still wanted. That would mean needing to refactor lib2 quite a bit so that it is cancel-aware... Unfortunately, the builtin Threat.yield() does not work as kotlinx.coroutines.yield and thus, lib2 must include kotlinx somehow.
Or... We redefine the algorithm to run in steps, like a ML training. Then, we could check if the job was cancelled between each step.
Some links I checked:
https://stackoverflow.com/questions/69097765/how-to-stop-or-cancel-a-kotlin-coroutine-stop-currently-running-code-block-inst
https://stackoverflow.com/questions/74065990/properly-cancelling-kotlin-coroutine-job

[akriese]

Rebased onto main and added singleton to cancel running algorithm instance.

[zino212]

LGTM and seems to work nicely. I only request restyling the AlertDialog to match our new style

issues solved