single dataset pipelines automate update of published index

[tgbugs]

need a reduced index of published dataset ids that will replace the combined curation export, it should be updated by sparcron probably, needed to simplify sds viewer

[tgbugs]

The reason these get stuck is because hasUriPublished is not populated without rerunning the export without fetching once publication is complete.

[jgrethe]

Probably the easiest way to get the ground truth is to pull that list from Pennsieve

[tgbugs]

Yes, adding the discover api as a third source that we check to trigger export is the full solution.