Improving DarwinCore checklist importer usability if there's no root taxon without a parentNameUsageID

[gdower]

Feature or enhancement

When using the DarwinCore Checklist importer, it could improve usability if the user interface detects and warns that there is no root taxon with an empty parentNameUsageID. Otherwise, new users might get stuck when starting out with importing a new dataset because it's not obvious why none of the records are Ready to import. It would also help to add an indicator that warns that the page needs to be refreshed after importing the parent taxon by clicking the row import button, or better yet, dynamically updating the rows to change the children records' statuses to Ready after the user pushes the import button on the parent record's row and it imports successfully. Just adding a reload page indicator or even reloading the page for the user after they click a button could help.

Location

Task - Darwin Core Archive (DwC-A) import

Screenshot, napkin sketch of interface, or conceptual description

Example with nothing being importable:

The interface at least gives a hint that the parentNameUsageID can't be found, but the root taxon also might not be the first record (like the above example). It might be clearer to put a message that the parentNameUsageID for at least one taxon needs to be removed. Should the import fail to even enter Ready state if there is no taxon with an empty parentNameUsageID? Or maybe it's better to just let the user remove the parentNameUsageID from the root taxon in Ready state?

When you do remove the parentNameUsageID for the top-most root taxon and try to re-import errored records, the root taxon record still fails to go into Ready state, so maybe it does make sense to not even allow the ImportDataset to enter the Ready state unless there is at least 1 row without an empty parentNameUsageID? This is what the interface looks like after trying to fix the root taxon issue and trying to re-import errored records:

At that point, I just started over with a new import because it doesn't seem to be possible to fix the root taxon in the table interface.

Your role

No response

[debpaul]

• @gdower, with help from @mjy and testing I got this to work ... let's chat
• we all have some ideas for how to add functionality (I'll summarize more fully later) that would benefit all of us, e. g.
  • use an existing ID for parents when the name exists in TW already to hang children off of
  • use an existing TW Source ID to stub the "citation by source" for a given name
  • using native Plazi output we can then imagine being able to import other fields they parse (e. g. of some ideas: Source page numbers, Zenodo identifiers, Plazi treatment record IDs)

(FYI for @LordFlashmeow )