UK Mater List updates not propagated to verification rules and other species lists

[JimBacon]

If there is a name change for a taxon in the UK Master List, resulting in a new preferred TVK (external_key), then auto-verification runs in to problems.

For records made against the UK Master List

• All the verification rules work by matching a TVK in the rule to the external_key of the record being checked. The rules are not updated with the UK Master List so a rule that previously applied to a taxon is no longer matched. It will therefore not be flagged if the record is outside range or phenology limits.
• Because of the above, cache_occurrences_functional.data_cleaner_result will always be true for the new record.
• The identification difficulty rule will also not find a match for the updated taxon so cache_occurrences_functional.identification_difficulty will be null for the new record.
• Because of the above, auto-acceptance depends upon the setting of auto_accept_occurrences_with_null_id_difficulty.

For records made against a different species list Other species lists are not updated with the changes made to the UK Master List so a taxon with a name change in the master list will still have the old TVK as external_key in other lists.

• Occurrences created having the old TVK will not find a match in cache_taxon_paths which only holds paths for the preferred TVKs of the master list.
• Because of the above, cache_occurrences_functional.taxon_path will be null for the new record.
• Because of the above, auto-acceptance, where limited by taxon restrictions which work by checking the taxon_path, will stop accepting such records.

[JimBacon]

Hi @johnvanbreda, do you have any thoughts, for or against,

1. Updating the TVK of verification rules on changes to preferred TVKs in the master list,
2. Updating the external_key of taxa in other lists when a preferred TVK changes in the master list?

The aspiration would be for verification rules and auto acceptance to continue functioning after changes to the master list.

[johnvanbreda]

@JimBacon I agree with these suggestions and I do something similar for associated taxa in the Pantheon project. In the longer term, the organism_key provides a more robust way of defining a single taxonomic concept that doesn't change quite as frequently as the accepted name TVK. This option wasn't available when these tables were designed but would be the way to go now.