Searches on vocabularies should tolerate at least case changes

[MattBlissett]

Previous searches would have used all-caps enum values: https://api.gbif-uat.org/v1/occurrence/search?sex=MALE

It seems unnecessarily strict to not allow this value to continue to work.

[marcos-lg]

This can be fixed with this PR in pipelines but it requires reindexing the datasets gbif/pipelines#1103

[MortenHofft]

Instead of just lowercase normalizer would it be worth using a different normalizer like camelCase or such?
so that TYPE_STATUS also match the new vocab value TypeStatus

[marcos-lg]

Instead of just lowercase normalizer would it be worth using a different normalizer like camelCase or such? so that TYPE_STATUS also match the new vocab value TypeStatus

Concept names don't allow underscores. And the lowercase normalizer fixes all the searches with different cases, camelCase included.

[MortenHofft]

Thanks @marcos-lg
I wasn't very clear. I justed wanted to highlight that not all old values can be fixed this way. E.g. typeStatus:TYPE_GENUS wont work since the difference isn't just in upper and lowercase letters. It is the presence of an underscore as well.

That might be fine. I guess it depends on how much we want queries that use old enums values to still work. 95% will probably be caught with your suggestion. Which is probably fine.