D19393: Prevent rescheduling aborted jobs until all jobs are processed

[Eric Dejouhanet]

TallFurryMan added a comment.


  When Scheduler assigns a startup date to the job, it considers the altitude restriction and adds the altitude cutoff, then checks if the job has time to run.
  On the opposite, when Scheduler is executing the job, it only aborts it when the altitude restriction is crossed. The altitude cutoff is not considered.
  Because that's confusing to the end-user, I first provided the customization of that cutoff as an option. That's not extremely clearer that way, but at least a simpler test can be run by setting the cutoff to zero.
  
  As I mentioned, the case that is of interest is the situation where the job has been executing for some time, and crosses the altitude restriction (or whatever restriction or operational issue that causes it to abort occurs).
  The runtime part of Scheduler marks it aborted, and the first block of evaluateJobs bypasses its re-evaluation. The job is then removed from the sorted list before evaluating the schedule.
  Yes, it ends up just after the last job that can still be run, but there's not much more the Scheduler can do while running.
  
  So I'll be considering the JOB_ABORTED as a special case that must remain in the list in the original order.
  When only JOB_ABORTED jobs remain in the list, I'll reset them to

REPOSITORY
  R321 KStars

REVISION DETAIL
  https://phabricator.kde.org/D19393

To: TallFurryMan
Cc: wreissenberger, kde-edu, narvaez, apol
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/kde-edu/attachments/20190307/6e3060b4/attachment.html>