D16058: Move "Remember Job Progress" to Scheduler option pane.
Eric Dejouhanet
noreply at phabricator.kde.org
Tue Oct 9 11:44:54 BST 2018
TallFurryMan added a comment.
In D16058#339684 <https://phabricator.kde.org/D16058#339684>, @mutlaqja wrote:
> Ok currently there is no way in the capture module to "Forget job progress" hence it is taken from options. So there should be a way for the user to know "Ok, I don't want to resume, I want to start from scratch now". Maybe there are other use cases where the user _always_ want to start to scratch so this has to be taken into consideration.
The end-user can click the reset button to forget all progresses. This serves the use case "start from scratch now". We don't have the use case "start this particular sequence job from the beginning", and we fixed the guiding deviation resetting the count in agreement.
There was a feature to reset job progress in the Scheduler too by double-clicking a single job, but it was weird in terms of UI and I removed it when all jobs were rescheduling to attempt again processing.
In D16058#339705 <https://phabricator.kde.org/D16058#339705>, @wreissenberger wrote:
> On the other hand, `Capture` should keep its state as good as possible even when it fails. Currently, when the `Scheduler` restarts `Capture`, it starts always with the first `SequenceJob` entry and does not continue where it failed. But this is not issue of this differential, it should be addressed later.
Yes, Scheduler is processing the sequence bluntly from scratch whenever it starts, because it starts by loading a sequence. We could imagine looking at the sequence that is requested, and if it's the same, not reload and keep the current settings. That would bring back the need for the d-bus entry point "clearSequence" though to really clear the sequence when needed, such as when repeating a job.
I should comment on exactly what happens when "Remember Job Progress" is enabled: Scheduler orders the capture frames it sees stored with no regard for what sequence job captured those, and determines the state (completed or not) of each sequence job before execution. But let's not get carried away yet :)
REPOSITORY
R321 KStars
REVISION DETAIL
https://phabricator.kde.org/D16058
To: TallFurryMan, mutlaqja, wreissenberger
Cc: kde-edu, narvaez, apol
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/kde-edu/attachments/20181009/06c2c690/attachment-0001.html>
More information about the kde-edu
mailing list