I'm in the process of writing some large simulations that need > 4GB of
memory which require access to among other things data in a Pervasive
SQL v8 database.

I'm running things on a X86_64 box running 64 bit linux. Pervasive has
ODBC drivers that supposedly work from 32 bit machines but I can't get
them to work even from a 32 bit linux box. Is there some way in which
I can access these databases from an X86_64 box. Contacting Pervasive
seems to be a non-starter. I called them on their sales line and asked
them if they have a product I can buy that would solve my problem and
they said they could only answer if I bought a support contract, for a
pre-sales question!!?

Is there any open-source ODBC (or otherwise) library etc that I can
make work on an X86_64 box running linux. I'd be happy to port a 32
bit driver over to the 64 bit machine and give someone back a working
64 bit driver if they wanted it. Alternatively, is there a product
that runs on Windows that can replicate a Pervasive database into a
MySQL (or Postgres or some such) database because I have ODBC (and
native drivers) for almost all the mainstream commercial and
open-source databases up and running.

Is there some specification for what the formats for Pervasive/Btrieve
data is? I have DDF files all set up.

Thanks

Ashwin Kapur
akapur at nyc dot rr dot com

On the 32 bit Linux box type:

su - psql

You should now be able to access Pervasive ODBC, which is in fact
iODBC. It's all in the user's profile. If you copy the exports to the
local profile you should be able to run it as any user. Note that for
some functions though the user still needs to be a member of the pvsw
group.

Support for the 64 box is minimal. Currently you'll have a hard time
trying to find software other than a few OS's for it. Since running
16bit apps on a 32bit OS has also proven to be troublesome we decided
to skip the 64bit OS for now ourselves. Our 64 box is running a 32bit
SuSE and sofar we're quite happy with it.

Replication? You mean ETL. Check out Pervasive integration or
Ecosystem products:
http://www.pervasive.com/products/index_integration.asp

Not sure on the formats question. There's no metadata in the Btrieve
files. If you want to write your own tool to browse the Btrieve files
you should at least need to retrieve the pagelength indicator from the
Btrieve header. You cannot do this on a live database though since the
Pervasive engine wants exclusive access. Your best option in this case
would be to use the Butil utility to export the data to unformatted
files which are close to C-ISAM files. For each record you will find a
textual length indicator followed by a comma and an ascii dump of the
record to which you need to apply the record structure.


Gordon


On Sun, 22 May 2005 03:42:44 GMT, Ashwin Kapur <akapur (AT) nyc (DOT) rr.com>
wrote:


Gordon Bos
Q-RY Solutions
+31-(0)15-2564035

http://www.q-ry.nl/