need fast parser for comma/space delimited numbers

[Les Schaffer]

Mike Miller said:

> It checks that the number of fields in each non-comment line is
> constant.  I think that part would be fairly simple to remove.

sorry, i probably wasnt clear.

what i meant was that i have to read in a few hundred files at a time,
each file has equal length rows, but the number of columns varies
__from file to file__.

les