Introduction

These are just notes taken as newbies (starting with Oliver) learn their way around ecat and the Checklist Bank (clb).

Schema Details

checklist

A checklist is a list of taxon concepts (e.g. Family "Felidae", Species "Panthera onca") organized into a hierarchy by Rank.

name_usage

dataset

name_string

This table holds verbatim scientific name strings as given by data providers. An attempt is made to parse that string into its name parts, which are then written to the parsed_name table. The name parser also assigns a NameType (enum from ecat-common) to the name_string record depending on the outcome of the parse. The enum's ordinal value is written to the type field:

• 0 sciname, // a scientific name which is not well formed
• 1 wellformed, // a well formed scientific name according to present nomenclatural rules. This is either the canonical or canonical with authorship
• 2 doubtful, // doubtful whether this is a scientific name at all
• 3 blacklisted, // surely not a scientific name
• 4 virus, // a virus name
• 5 hybrid, // a hybrid formula (not a hybrid name)
• 6 informal; // a scientific name with some informal addition like "cf."

In cases where the name parse fails there will be no corresponding entry in the parsed_name table. Name parsing can fail for a number of reasons, but primary among them are bad original scientific names (e.g. "Unknown Unknown") and viruses (which don't have genera).

The canonical_name_fk field points to the name_string record that holds the canonical name for the species in question. In cases where the name parser and lexical grouper determines a scientific_name entry to be the canonical name of a given species, it will point to itself in the canonical_name_fk field. Example (in this case the canonical name is held by record id 121672):

The fields asserted_by_fk and homonym are deprecated and can safely be ignored.

parsed_name

This table holds the parsed name parts of the scientific name held in a corresponding row in the name_string table (see above). There is at most one record in parsed_name for each entry in name_string. If name parsing fails for some reason there will be no entry in parsed_name for the bad scientific_name entry in name_string.

term

term_gbif_portal_rank