> From: petechen@math.rutgers.edu (Peter Chen) > > Hi, > I am in the processing of making a MacPerl5 module for accessing > the Butler database. Since this module currently is only useful on Macs, > and it won't contain any C interface to Perl, I am not quite sure what's > the preferred way of setting up something like this. Should I set it up in > a way similar to oraperl, or DBD::Oracle? > > You guys will have to excuse me if I don't have the full picture > about DBD/DBI, etc. I just got back onto this list today after a full year > of working on only Macs (not that I'm complaining, but sometimes I do miss > a Unix box, *sob*). To be honest, right now I would not recommend starting the development of a new DBD driver unless the need is urgent and you feel confortable with how to implement it. I'd suggest, instead, that you define a high level interface similar to the current DBI interface (connect, prepare, execute, fetch etc) and do wharever implementation is easiest for you 'behind' that. Once the ODBC DBI is usable you could think about migrating to an ODBC based driver. By that time tools will be available to simplify the task. Tim. p.s All the above presumes the 'Butler' database uses SQL.