Re: Extracting data from Oracle 2.1

From: Daniel S. G. Dickerson <dickersn_at_pacbell.net>
Date: 1996/09/08
Message-ID: <32332ECE.AE1_at_pacbell.net>#1/1


> >We have a large application running on Oracle 2.1 on a MAC.
>
> Why? Oracle 2.1? You're kidding, right? If the application was truly
> important to your business, why didn't you follow any upgrade path?
> What do you consider "large"? 100M or 100G?

This application was written several (~4-5?) years to run on machines donated by Apple to a local museum. There is no money to upgrade the machines, and besides, the purpose of this Oracle installation is to generate code a tables that run on other machines in the network. Since the company that wrote the original code is no longer associated with the project, the chances of porting the code to a new processor is remote. Also, to allow for maintenace and whole system backups, all 40 of the machines in the network are Macintosh Quadras. This enables the staff to swap out any "dead" machines and swap in any working one.  

> >We need to extract the data to some other format. We would prefer to
> >export to another database application like Access, Foxpro, or Paradox,
> >but can also use Excel running on a PC. If all else fails, we could
> >use an ASCII format on a UNIX box. We are writing an application that
> >needs this data, but can not depend upon a particular database engine
> >or server.
>
> Well, pick a database and go for it. Why can't the application
> "depend" on a particular database engine or server?

We don't really need a Server, just some way to extract the data. We have
been unable to find an existing database engine that can read these files,
that is why we posted the original message.

> >We performed a complete export (using EXP) of the database on the MAC
> >and have stored the export files on a Syquest cartridge that we can
> >read using MacOpener on the PC.
>
> Provided you have a tool similar to SQL*Plus, just spool the output of
> a select statement to a text file (with delimiters, i.e.; ","). Run
> the data into your new system with their import utility or roll your
> own.
>
> >
> >Does anybody have any tools that we can use to translate this data?
>
> See above.

Our experience with Oracle is VERY limited, again that is why we posted the message. The size of the database and the number of table is one of the reasons we are hoping to find and export/import utility

> >There is a second VERY important reason for translating this data:
> >Our version of the database is no longer supported by Oracle, and
> >they do not seem to have a new version that will run on our Quadra.
>
> So replace the Quadra.

It is not possible to replace the Quadra, see above.

> >In addition, our install disks seem to be defective, so if this system
> >goes belly up, we are really sunk. Does anybody know how to get new
> >install disks?
>
> Why not just back up your machine? You do have a current back up,
> don't you?

Yes we do have a backup, but we would still like to be able to reinstall the system if needed. If we have to put Oracle on an existing system, we do not want to risk corrupting the preexisting system by restoring a backup on top of it. Oracle also seems to store some files in folders other than its own folder. We tried "installing" it by copying the Oracle
folder contents to a new machine, but something was missing.

> > Since Oracle will not supports us, we need to move to a
> >system that will.
>
> Oracle, I'm sure, will be happy to support you on a fairly current
> release of the database. If you plan to be multi-user, Oracle would
> seem to be the way to go.

Since we can't move machines, we are stuck with Oracle 2.1 on a Quadra. Oracle does not seem to have any products that will run on Apple machines other that the PowerMac. Received on Sun Sep 08 1996 - 00:00:00 CEST

Original text of this message