You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Sometimes people simply forget, or programs enter in a particular state, where a completion callback is not called. This can cause unexpected problems, that unfortunately should happen. We should do something to avoid this situation.
My current thoughts are we could detect when we are idling and we are still waiting on a task to complete, if so, then we can output a warning message that lets you know which task we are still waiting for. This should be able to be turned off via a configuration option.
It seems we can use process.memoryUsage() perhaps. However, that is for the entire app, not just the particular scope.
So if the app has something happening in the background somewhere else, then this check will fail. So we can't use it.
We could implement a timeout check on the Task class, however, however, that would also fire if sub-tasks are also taking a long time.
So perhaps the detection can happen in the Task class, but the reporting happens in the TaskGroup class perhaps.
Or maybe, detection and reporting happens in the Task class, but TaskGroup turns it off for TaskGroup Tasks.
The issue then, is that the idle detection will only work best if all the task groups are attached together using the v3 nested abilities. However, as that makes a lot of sense, I'm happy encouraging that as the way forward.
Sometimes people simply forget, or programs enter in a particular state, where a completion callback is not called. This can cause unexpected problems, that unfortunately should happen. We should do something to avoid this situation.
My current thoughts are we could detect when we are idling and we are still waiting on a task to complete, if so, then we can output a warning message that lets you know which task we are still waiting for. This should be able to be turned off via a configuration option.
Would help realise problems such as #6
The text was updated successfully, but these errors were encountered: