What ancillary df duplicates_taxa should contain?

[damianooldoni]

In old version of verify_taxa the df output duplicates_taxa contained taxa from different checklists which with same scientific name.
In new verify_taxa(), I propose to use duplicates_taxa to collect all taxa which point to same bb_key-bb_acceptedKey.

Example. The taxa here below would be in duplicates_taxa:

taxonKey	scientificName	bb_key	bb_acceptedKey
1	A	10	15
2	B	10	15
3	C	10	15

Drawback of this implementation: taxa without match to GBIF Backbone would be not included. So the following two taxa which are very likely the same species, would be not returned in duplicates_taxa:

taxonKey	scientificName	bb_key	bb_acceptedKey
4	E	NA	NA
5	E	NA	NA

I can live with that 😄 @peterdesmet : you too? It is actually a question about the meaning we want to give to the expression "duplicates taxa". As we use now a key-triplets (taxonkey - bb_key - bb_acceptedKey) to identify unique taxa instead of names, I think using bb_key - bb_acceptedKey sounds to be the best option.

[peterdesmet]

As discussed, just using keys (and not names) for this df is fine. I would simply call the df “duplicates”

[damianooldoni]

Allright.

[damianooldoni]

Done.