Flutterby™! : Perl to the Rescue

Next unread comment / Catchup all unread comments User Account Info | Logout | XML/Pilot/etc versions | Long version (with comments) | Weblog archives | Site Map | | Browse Topics

Perl to the Rescue

2001-10-18 19:37:34+00 by ebradway 3 comments

I am in the process of rebuilding a smaller version of the transaction processing system that my ex-employers shut down last week. It was written in Perl on Linux with Sybase on the back-end. I've been asked to recreate a small, but significant portion of the system and for various reasons, I'm having to do it on WinNT. Initially, I'm building the system around an MDB file to highlight the fact the initial system is not intended to do significant numbers of transactions. Anybody have any hints on resources for working with MDB files in Perl? I'm assuming there is a DBD::MDB module out there. I also picked up a copy of the Perl Resource Kit for Win32. Any other resouces I might need? Also, has anybody see Effective Perl Programming?

[ related topics: Free Software Books Perl Open Source Software Engineering Work, productivity and environment ]

comments in ascending chronological order (reverse):

#Comment made: 2002-02-21 05:33:03+00 by: Dan Lyke

I didn't find a DBD::MDB module anywhere. I've only done this with Tcl on Windows, but perhaps you could use ODBC drivers and set up your MDB database as an ODBC data source?

I'm also under the impression that there's a Cygnus port of PostgreSQL, but licensing restrictions on the Cygnus[Wiki] tools might make that only appropriate for proof-of-concept.

Perhaps tied hashes are in order?

#Comment made: 2002-02-21 05:33:03+00 by: Dan Lyke

You might try DBD::AnyData or DBD::CSV if super low volume for prototyping is an option. And I think the DBD::AnyData guy plans on working towards using tied hashes.

#Comment made: 2002-02-21 05:33:03+00 by: ebradway

I'm using Win32::ODBC. It works well enough. My problem is I need to create a fairly complex set of relationships and I don't have it mapped out completely in my head. Access happens to be a truly wonderful tool for prototyping database schemas. I was just hoping to stick to a more familiar API - but Win32::ODBC lets me execute SQL against the MDB file and the DSN provides locking for multiuser purposes.

Once I have the schema where I want it, I'll probably end up porting the entire mess to DB2. Access makes moving the data a breeze and I know better than to use the Access crap that gets you into trouble - macros, spaces in names, etc.

The benefit of using DBI is that I would have been able to simple change the DBD and any hangups between a DBD::MDB and DBD::DB2 and have the code ported. Oh well.