clb_usage(), clb_usage_search(), clb_usage_info() and clb_synonyms() now return all name usage fields instead of a small subset: origin, the dataset, sector and verbatim keys, identifier, reference_ids, link, remarks, the scrutinizer and temporal range of taxa, and accepted_id / accepted_name for synonyms. accepted_id falls back to parent_id when the API omits the accepted usage, as for clb_synonyms(). clb_usage_search() also returns sector_dataset_key, sector_publisher_key, secondary_source_keys and secondary_source_groups.
clb_distribution(), col_distribution() and clb_usage_info()$distributions gain country_code, the ISO country of ISO gazetteer areas (BE for BE-VLG), and reference_citation, the citation of the distribution’s reference. clb_distribution() looks up each distinct reference once; pass citations = FALSE to skip that (#6).
Ids containing reserved characters such as / or : are now URL-encoded in request paths. GRIIS checklists use GBIF species URLs as taxon ids, which made clb_distribution() and the other taxon functions fail with HTTP 404 (#8).
First release of the ChecklistBank-based client. It is unrelated to the rOpenSci rcol package (versions 0.1.0 to 0.2.0, archived on CRAN in 2022), which had a different API.
col_*() shortcut functions (col_match(), col_usage(), col_tree(), …) that always target the latest extended COL release without a dataset argument. The release is resolved once to its integer key and pinned for the session via col_key(); col_refresh() re-pins to a newer release.
clb_usage_info() / col_usage_info() return the full usage information document (usage, classification, synonymy, vernacular names, distributions, references, …) in a single request.