Re: Quotes being stripped by "--csv"

[Manuel Collado]

El 19/11/23 a las 9:37, arnold@skeeve.com escribió:
> Hi.
>
> I understand what you're saying. I don't have an answer at this point.
> I think it would be helpful for you to open an issue on the Github repo
> for Brian Kernighan's awk, as CSV handling was his idea. Maybe he can
> come up with something.
>
> In any case, opening an issue there will allow for wider discussion amongst
> AWK implementors.

To clarify things, it should be noted that there are two fundamental 
modes of handling CSV data. Either:

- Focusing on the CSV representation of fields, or
- Focusing on the real value of field data.

In the first mode $0 should be the original CSV record, and $1..$NF 
should be the fragments of $0 identified by the CSV parser. I.e. with 
their original quoting.

In the second mode $1..$NF should be the unquoted, real field data 
values. And $0 should be the concatenation of these fields delimited by 
OFS. In order to be consistent with the awk concept of records and fields.

These two modes are then incompatible. You must chose to implement one 
or the other. Of course, it is possible to support both approaches, and 
use a suitable command line flag or control variable to select the 
desired one.

And it is also possible to implement only one mode, and provide 
additional functions to convert between the real data values and their 
CSV representation.

HTH. Regards.
--
Manuel Collado - http://mcollado.z15.es