I'd answer the question myself by doing timings, but I have not found out
how to profile in J, by time or otherwise. So if anyone is inclined to
feed
this man by teaching him how to fish, that would actually be the best
answer. :-)
AHS
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
Arved,
you have a moot question, I think. The wise fisherman, when faced with CSV
in the J language would proceed with
load 'csv'
date =. readcsv 'myfile.csv'
I am a newbie to J myself, but I would point out your strategy might be a
little naive.. The double quotes might contain escaped double quotes (which
would have to be treated as data format errors if encountered outside of the
quoted field) ...there are a few catches The provided package as is has
sorted out all of the picayune details of CSV parsing. I have written more
than a few CSV readers myself in various Von Neumann languages and there is
nothing particularly interesting to be learned reinventing that wheel.
- michael
--
- michael dykman
- [EMAIL PROTECTED]
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
