When data are taken the files are transferred to the USDF and ingested into a butler registry. A lot of the general tooling developed by SQuaRE rely on ObsTAP queries of an ObsCore database table. Currently we generate the content of the ObsCore table by doing an explicit export of a butler registry using dax_obscore. This works but is an additional step that would have to work as part of the automated data ingest explicitly exporting the records for the new raws and adding them to the ObsCore table. Presumably some data processing will also happen at USDF and we would like, for example, calexp images to be available to other systems as soon as the processing completes. Do we want to have to do an ObsCore export/import every time a batch job completes?
Things would be simpler if we coulde have an ObsCore view of the butler registry that automatically updates when new files are added or deleted. This ticket is to investigate the feasibility of this and to write a tech note with options.