Have you played with this type of database?

[normeus]

I have to transfer data from a database format called "neoaccess database technology" from 1995 which was used on a product called quickImage. The database format is "object based" so using my hex editor changing a bite on one field at a time an comparing results is getting me somewhere at a snails pace. I figured it wouldnt hurt to ask if any one had done something similar and might have code to share ( C, C++, VB, Perl, PHP, Objective-C, or anything at this point  ) where I could at least figure out why these record pointers are so random.

like offset 13F:  2 bites represent the number of records which can be a max of 0xFFFF

the first record pointer is located at 0x55BA

Thank you if you have any answers for me or even for any comments you might have.

[TaoPhoenix]

Sorry all I can add is a "useless" "moral support" answer but if anyone can, this is the crowd to do it!

Mouser, does any of your new Mewlo experience apply here?

[Shades]

I would first find out which company made the database software and search for a converter that may exist.

If such a converter doesn't exist, try to get the database software itself and run it in a virtual machine with the correct OS for the database server. Most database servers that run on Windows can be accessed through ODBC. Using that you should be able to retrieve data from the old beast and put it again in a database that is supported for a while, such as  Oracle and Microsoft (maybe IBM?).

All these companies have free versions of their database software, which are normally limited in storage space and some functionality. However, the limits of these databases is a multitude higher than anyone could dream of in 1995.

The open source databases PostGreSQL will also be here for a while still.

Some digging in the dusty archive resulted in this link, which describes a company Neologic that might be the maker of the database format you are sifting through.

My suggestion is long in the tooth as well, but getting data from old obscure data formats is a dread...at best! 

[Stoic Joker]

If such a converter doesn't exist, try to get the database software itself and run it in a virtual machine with the correct OS for the database server. Most database servers that run on Windows can be accessed through ODBC. Using that you should be able to retrieve data from the old beast and put it again in a database that is supported for a while, such as  Oracle and Microsoft (maybe IBM?).

-Shades (October 18, 2013, 07:39 PM)

We think alike here.

I did a quick google search and found these:
http://stason.org/TU...ct.html#.UmJ0cxvD_cs (mentions 1995 and gives name of companies site)

http://www.neo4j.org/download (Not sure if this is the current version of the product, but it does have a free demo)

http://web.archive.o...://www.neologic.com/ (web.archive.org takes us back to the companies site from 1996)

Note: While the (ftp) download links don't appear to be working anymore ... They do at least give the downloads filename. So if you do a bit of spelunking it could be possible to find the down in someone else's archive.

[Renegade]

^^ Good finds there! I'd checked quickly, but only turned up a dead site.

It seems like they have community versions there, which are usually free. I've never heard of a non-free community version.