Open sdruskat opened 2 years ago
The full validation with cffconvert
is quite expensive and slows down reading the dataset a lot: this should probably be optional.
Leaving open as this needs more work, e.g. --strict
flag to trigger cffconvert workflow, and fixing things in the workflow (ideally work with cffconvert's developers to make this usable as library).
This uses cffconvert to validate CFF files.