Bidirectional GBIF Registry Synchronisation

[mdoering]

Up to now we sync checklist datasets with good licenses from the GBIF registry to ChecklistBank, but not the other way around. In the future we want CLB to be the only entry for new checklist registrations and only track these with the GBIF registry. For the GBIF portal to show links to ChecklistBank we also want the native CLB integer datasetKeys in the GBIF registry.

As a start extend the GBIF registry sync with the following:

• update existing GBIF checklist entries
  • adding the CLB dataset key as an alternative identifier of new type CHECKLISTBANK
• create new checklist datasets for missing entries
  • only use
    • public datasets of origin=EXTERNAL, ignoring projects and releases for now
    • require an CC0, CCBY or CCBYNC license
  • publisherOrg = ChecklistBank
  • hostOrg = ChecklistBank
  • installationOrg = ChecklistBank
  • endpoint: the original archive in CLB, not a generated one

For this to work CLB needs write access to the registry API. For adding identifiers CLB needs access to all checklists in the registry. For adding new datasets it might be good enough to use the existing permission system that gives access to a single publishing organisation, CLB in this case.

The GBIF API needs changed to cover:

• EndpointType: ACEF & TEXT_TREE
• IdentifierType: CHECKLISTBANK

[mdoering]

@fmendezh We could start only with adding CLB dataset keys to existing gbif checklists