Rescheduling on success

[DrewEaster]

Hi, thanks for the great library!

We have a requirement where, rather than having fixed recurring tasks, we'd like to have one time tasks that get to decide at the end of their execution when they'd like to be rescheduled. This is actually very similar to the failure handling scenario where you get to choose, on failed execution, when to try again - only in our case we'd like to be able to do this in the success case too.

Problem I have is that, currently the SchedulerClient does not allow rescheduling of a task that's been picked and, at the point that I need to call reschedule(), it's still within the execution handling of the task (in which case the the execution is still considered picked).

I could, of course, be devious and directly mutate the row in the DB table, but it would be considerably more preferable to avoid that, especially as I'm conscious of not wanting to create any unforeseen issues by bypassing the code.

Do you have any thoughts on how this could be achieved without changes to db-scheduler itself?

[kagkarlsson]

This is currently supported using a CustomTask fre. 13. okt. 2023 kl. 16:42 skrev Andrew Easter ***@***.***>:

…

Hi, thanks for the great library! We have a requirement where, rather than having fixed recurring tasks, we'd like to have one time tasks that get to decide at the end of their execution when they'd like to be rescheduled. This is actually very similar to the failure handling scenario where you get to choose, on failed execution, when to try again - only in our case we'd like to be able to do this in the success case too. Problem I have is that, currently the SchedulerClient does not allow rescheduling of a task that's been picked and, at the point that I need to call reschedule(), it's still within the execution handling of the task (in which case the the execution is still considered picked). I could, of course, be devious and directly mutate the row in the DB table, but it would be considerably more preferable to avoid that, especially as I'm conscious of not wanting to create any unforeseen issues by bypassing the code. Do you have any thoughts on how this could be achieved without changes to db-scheduler itself? — Reply to this email directly, view it on GitHub <#429>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AABKEHEXWF2EM3YOADDNU3LX7FHNRANCNFSM6AAAAAA57IMAYQ> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[DrewEaster]

Ah, great, I can see it now. Thanks for your help!