Norman:
  You did not specify what platform you will be implementing this on.  I
think everyone assumes Windows.

A quick google search leads me to:
http://gigamegatech.com/2008/09/16/u2-and-adonet-relatively-relational/
which mentions that IBM has done some recent work with an ADO.NET driver.

dotNET things can be conveniently run from IronPython.

The fepy project has several simple .NET based SQL modules which
more-or-less follow the db API 2.0 standard and are written in pure
IronPython.

There is also the adodbapi module which could be used as a Python code
example. It fully implements the API, but I have not yet gotten around to
forking a version which is specific to ADO.NET -- the existing code uses the
COM interface to ADO. Microsoft fixed the last bug in the IronPython COM
interface yesterday, which is what I have been waiting for before starting
on another update. Some collaboration might be possible...

I have had good success talking to an AS400 DB2 database using adodbapi.
--
Vernon Cole

On Wed, Aug 26, 2009 at 2:23 PM, Paul Boddie <paul@[...].uk>  wrote:

>  On Wednesday 26 August 2009 19:20:21 Norman Bauer wrote:
>  >
>  > I am beginning to dive into developing a module connecting to an IBM
>  > UniVerse database. I'll first start by saying that this is the largest
>  > undertaking I have ever had and so far the experience has been fun,
>  > yet frustrating at times... I know wahhhh....
> 
>  Didn't IBM get round to making an ODBC connector for UniVerse or continue
>  the
>  plans to merge UniVerse with UniData (which did have such a connector, as I
>  recall)? UniData seemed very archaic when I had to do work against it, but
>  there was some kind of SQL/ODBC support, even though the documentation was
>  mediocre (again, as I recall).
> 
>  Paul
>  _______________________________________________
>  DB-SIG maillist  -  DB-SIG@[...].org
>  http://mail.python.org/mailman/listinfo/db-sig
>