The software system includes a builtin system for handling classifications which support for internationalization. Classifications are seeded into the system on startup. See examples in the code for how this is performed.
Each classification has the following properties:
- label what should be displayed in the UI.
- locale which is the locale for the classifications. It is specified in
accordance with RFC 5646, as
for example
en-US. - manual. Classification should not be shown by default but can be explicitly requested.
- obsolete if the classification has been deprecated. See section later on.
- parentUuid is the UUID of the classification parent if nested.
- uuid unique value that is consistent across data base migrations.
All classifications are stored in seed files, for example a file named
gender_en-US.csv containing:
uuid,label,manual,obsolete,parentUuid
b66e1c9f-1381-472c-af75-2c2bc2166124,"Male",false,false,
1c7a1cef-f7ea-4022-85dc-81bcca0b79d7,"Female",false,false,
79e18bbf-8c29-4d2e-97e3-0ddb3188fa03,"Prefer not to say",false,false,
It is possible to control which environment the different sets of classifications are seeded to.
Classifications can have parents in order to achieve nested forms.
Classifications should never be removed from the system, only deprecated. This is to maintain potential existing relationships and avoid the old UUID from being reused.