CSV reading with non-comma separators about primap2

[mikapfl]

Read csv using pandas function and convert in a second thep

Might actually be worth, the API surface of pd.read_csv is huge and next thing we probably want to read from database or something, which pandas can do, but we don't have a function. So, I'd say: just use pandas' reading functions, then convert_long_dataframe_if.

[JGuetschow]

I do that all the time. At some point we might remove the csv reading functions as they can basically never be used

[mikapfl]

Yeah, I think reading the interchange format with a specific function has value, but in general so-called CSVs are super diverse. Probably it was not realistic to expect them to be sufficiently well-formed to read them with one function. /-: