Would it be interesting for observers to be able to pick scheduler and sequence job files from an URL instead of a local file?
This would allow a KStars installation to work out a dynamic job list prepared by an external server, possibly another KStars publishing files on a web server.
I'm planning to make the scheduler periodically update the schedule of jobs when it is not started, so that the list of jobs is always up-to-date in terms of startup time and score. Picking from an URL happened to me as a nice extension to this. But I'm only imagining a feature there that would need additional use cases, nothing practical for now.
I keep my .esq and .esl files on dropbox. Periodically updating the scheduler with the file content is a good idea I think. Would this mean I could alter a schedule which is in progress?
Derek
@dokeeffe multiple steps are needed. First, self-evaluation must be done by the job object, compared to the current algorithm where the scheduler itself does the evaluation. Second, evaluation must be made asynchronous, because having the UI stutter periodically is a no-go. At that point we'll have a periodic update when the scheduler is stopped. Now we could extend that to the running scheduler, but we would have to take care of executing jobs. But that is indeed feasible, IF users really see an interest (automated setups probably?).