The current Excel/CSV bulk import template has a column where a user can supply ref_id values, but those ref_id values are not imported into newly created archival object records.
This issue existed in the earlier version of the Harvard Excel plugin and is documented here: https://github.com/harvard-library/aspace-import-excel/issues/62
I'd love to see this feature implemented in the new integrated bulk import tool for a couple of reasons:
1. The ability to supply ref_ids in the bulk import could assist with roundtripping scenarios where it's useful to preserve existing ref_ids after making batch changes via some kind of export/import workflow. This is especially true if ref_ids are used as a linking mechanism to some external system/repository.
2. Bulk import of ref_ids via spreadsheet are also useful when migrating from another system to ASpace where there is a need to retain another system's database IDs to preserve some pre-existing relationship to an external system or to other migrated records in ASpace. For example, the AT to ASpace migration tool has an option to preserve AT-assigned ref_id values in ASpace.
For what it's worth, I added one row of code (below) to the old Harvard plugin code to make this work, so I'm hoping that maybe this is a simple change?
ao.ref_id = @row_hash['ref_id'] if @row_hash['ref_id']
Could this line of code be added here in the new integrated importer in v2.8?: https://github.com/archivesspace/archivesspace/blob/v2.8.0/backend/app/lib/bulk_import/bulk_importer.rb#L257
As we were finishing up some importer-related improvements for the 2.8.1 release, we added some capability for this. The Ref ID column is now hidden in the AO Excel template and removed from the CSV, but can be unhidden/added for those that want to create their own AOs. As before, the importer will not create rows where a REF ID duplicates one in ArchivesSpace. We'll be adding this to the documentation.
Thanks so much for squeezing this into 2.8.1