Tim Bunce wrote:
> Actually, though, it's crossed my mind more than once to add the
> core C code to parse a CSV record into the DBI. It should be tiny.
That would be great for users of real DBI, one less module to download.
It wouldn't do much for pure perl users though unless you want to
wrap it in an if(){...} and include a Friedl or AnyData style alternative.
> Speaking of which, I just took a look at the CSV read_fields code:
> ...
> I suspect it might run a little quicker this way (untested):
Great, thanks, I'll play with it.
> (Could probably be made faster still if read_fields was dynamically
> generated to have the current regex as a literal.)
Yep, that's been on my to-do list a while :-) I'll eventually re-write
much of AnyData based on what I learned from your code while doing DBI::PP.
--
Jeff
