Clarification: @optional means full, empty, or partially full?
Another trivial clarification, I'm afraid 🙁
I am using @optional currently in a csvs to mean a column may hold one of
- any of the allowed values defined for that column (in all cells)
- any of the allowed values defined for that column (in some of the cells with the rest empty)
- no values (totally empty column, just a header and n blank cells)
Is that the intended meaning? It seems to be from the way the validator works?
It's not totally clear to me from the doc if a partially empty column is OK:
http://digital-preservation.github.io/csv-schema/csv-schema-1.1.html#column-definitions-examples
Is there a different directive for this?
Testing is always row-by-row, so yes a partially empty column would be fine - @optional is effectively equivalent to adding "or empty" to your column test.
I see if I can think of some clearer wording.
Thanks for clarifying. So "column" is shorthand for "the cell in this column in this row". As I surmised.