I've written CSV exports in C from scratch, no external dependencies required.
It's "Comma Separated Variables", it doesn't really need anymore specification than that.
These files have always imported into M$ and libre office suites without issue.
It's "Comma Separated Variables", it doesn't really need anymore specification than that.
These files have always imported into M$ and libre office suites without issue.
Normally, I wouldn't nitpick (hold for laughter) but this is just a perfect example of how CSV is not CSV and how it's probably impossible to support everything people call CSV, simply because of the tiny (or not so tiny) differences between formats.
- supporting "" (single)
- Supporting newlines in "", oops, now you can't getline() and instead need to getdelim()
- Supporting comments # (why is this even a thing)
- Supporting multiple "" in a field
- Escaping " with "" or \"
- length based csv, so all fields are seekable.
It's a mess, which one's your csv?
No CSV I have ever encountered has comments.
So yes, I am happy with code that works for a subset of files. There are thousands of applications which work with CSVs and they all do exactly this.
(Oh and I forgot about headers, headers should be mandatory but people are aloof about it)