Skip to Content.
Sympa Menu

freetds - [freetds] M$ ODBC driver work.

freetds AT lists.ibiblio.org

Subject: FreeTDS Development Group

List archive

Chronological Thread  
  • From: "Steve Murphree" <smurph AT smcomp.com>
  • To: <freetds AT lists.ibiblio.org>
  • Subject: [freetds] M$ ODBC driver work.
  • Date: Mon, 16 Jun 2003 15:46:25 -0400

Hello all,

I've tried subscribing to the list but I still don't get list related
emails. :-( So if you want to reply you will have to reply directly, as I
won't see it in the list. I alos use ASK, so yo will need to confirm that
you send me mail.

What I have done is take the ODBC part of the FreeTDS v0.61 and really added
some more functionality. Things like real native error codes, greatly
improved SQL States, and support for things like SQLPrimaryKeys,
SQLForeignKeys. Support for real connection and statement attributes plus
intial "real" descriptor support. There is now a complete IRD that has a
handle associated with it making SQLDescField and friends work. I saw that
some of this has already been done in the CVS tree. I concentrated on
interoperability with the current flavors of M$ SQL Server that the company
I work for uses; SQL Server 7 and SQL Server 2000. The SQLGetInfo function
returns information based on the server version for M$ servers only. I
don't know if the Sybase stuff was exactly the same, so I made another
directory in the freetds tree and called it msodbc. The driver, as it
stands, functions flawlessly with our ODBC apps and PHP ODBC apps although
we only use forward only "snapshots" for information presentation. I have
also written a unixODBC frontend that is part of the driver and provides a
freetds.conf less configuration. It's miles of added code.

Diffing this thing out against the current tree would be a pretty dificult
thing to do, but if anyone is interested, I will diff it against the 0.61
release and make a nice little (BIG) patch.

Also, if given CVS tree access, I would integrate it into the current tree.
A big diff splatted against the tree would be too hard to follow. I would
incrementally integrate the code bit by bit. So if the authors are
interested?








Archive powered by MHonArc 2.6.24.

Top of Page