* separated values

Robin Munn rmunn at pobox.com
Tue Jan 15 16:04:47 CET 2002


On Tue, 15 Jan 2002 16:53:09 +0300, Oleg Broytmann <phd at phd.pp.ru> wrote:
>On Tue, Jan 15, 2002 at 01:42:10PM +0000, Magnus Lie Hetland wrote:
>> Basically, what I'm wondering is whether it would be possible to add
>> such a beast to the standard library. It seems like a very natural
>> format to support, doesn't it?
>
>   I don't think so.

I do. I once needed to translate a couple of .csv files to another
format, and started out writing my own parser since "it'll be simple,
right?" Discovered there are several subtleties that make it not
particularly easy, decided someone must have already invented this
wheel, and found a csv parser on-line. Not too much time wasted, but if
it had been in the standard library, I wouldn't have wasted *any* time
at all.

>> I mean, if we have ConfigParser, why
>> not csv?
>
>   May be, better way will be to remove ConfigParser? No kidding, I really
>mean it - why bload the library? Instead let us create Python CPAN and move
>modules there. I think even XML libraries should be removed from the
>standard lib.

This would seem to go against the Python philosophy that Python comes
"with batteries included". I say the more we have in the standard
library, the better -- as long as there's no duplication within the
library, of course.

-- 
Robin Munn
rmunn at pobox.com



More information about the Python-list mailing list