In ci.schemas.member.MemberDetails, disabilities, hobbies, and qualifications are currently implemented as a simple dict str -> str. The fields are implemented in Compass in such a way that only one entry of each subtype (e.g. A Levels, Degree (eg BA, BSC), Doctorate) is allowd, enabling using a dict in Python.
This would also feasibly enable using a full pydantic model, but I somewhat question the usecase -- it would be ~60 lines for the three models, increasing the size of schemas.members by 12%. Willing to consider it if someone has a use, but for now will leave this issue open as a marker.
In
ci.schemas.member.MemberDetails, disabilities, hobbies, and qualifications are currently implemented as a simple dict str -> str. The fields are implemented in Compass in such a way that only one entry of each subtype (e.g.A Levels,Degree (eg BA, BSC),Doctorate) is allowd, enabling using a dict in Python.This would also feasibly enable using a full pydantic model, but I somewhat question the usecase -- it would be ~60 lines for the three models, increasing the size of
schemas.membersby 12%. Willing to consider it if someone has a use, but for now will leave this issue open as a marker.