Support optional columns with default values during validation

[megin1989]

We are using frictionless-py for validating CSV files against a Table Schema (Data Package ).

We recently introduced new boolean-style flag columns (Yes / No) in our schema, for example:

• FLAG_PART_2
• FLAG_TYPE_A
• FLAG_TYPE_B

These columns represent feature or attribute flags and are optional for data providers during a transition period.

Requirement

• Validation should still pass if these columns are absent
• If absent, they should be treated as a default value (e.g. "No" / false) for downstream processing
• If present, values should be validated against allowed values (Yes, No)

Could you please advise whether this scenario is currently supported in frictionless-py, or if there is a recommended approach for handling optional columns with default values?

[gabrielbdornas]

Hello @megin1989, reading the documentation, I think you could do it using the fieldsMatch property at the Table Schema level.

I created this reprex to test it, but unfortunately, the frictionless validate datapackage.json runs with the Missing label error we're trying to avoid using the aforementioned property. We have to investigate it to see whether this error is related to something wrong in my datapackage.json or in the lib.