I am not positive these are Informix files, but, am fairly confident they are. They came off a third-party accounting/inventory tracking system's server (Fedora Core). Unforunately, I cannot find out the name of the application.

I have two data collections. Each collection has exactly 1 .dat file and several .idx files. I presume the .idx files are just indexes for the .dat. Using Textpad and a hex editor I am able to identify several different sections of data in the .dat file. The first section contains what appear to be field headers. In examining the binary I have found clear patterns indicating header fields, but, cannot clearly denote datatypes from the binary yet. (Maybe some more analysis will help me sort it out). The actual records consist of rather long strings (3500 characters) where each separated by a set of null characters. I was able to parse the .dat file, extract the raw data, and, begin attempting to line up the headers with the fields. This works fine for some of them, but, for others I cannot get a 100% match based on obvious cues.

I know this is a long shot, but, does anyone have experience reversing from the .dat file to match records, rows and headers?

Hi,

if these files are contained in a directory called xxx.dbs, these are either
Informix pre-SE tables, either Informix SE tables.

if these files are not contained in a xxx.dbs directory, chances are that they are Informix C-Isam files.

I can take a look on one of these files to check this. If they are, tools exist to extract data and produce a schema

I have those tools :-)

Regards,
Eric