Skip to contents

Grouped enrichments join on a bare accepted-name string (see enrich_by_group()), which collapses distinct homonymous concepts that happen to share a name. An authorship column is the only signal available to tell them apart; this picks the first alias present, in order of how likely it is to be the taxon's own authorship rather than something else. taxifydb reads the same column at build time, so that the row it keys under a name is the one this guard will accept.

Usage

enrichment_authorship_col(schema_names)

Arguments

schema_names

Character vector of column names in the enrichment .vtr.

Value

The matching column name, or NULL if none of the aliases is present.