How will you know for certain that you are not giving out the confidential information? By not giving out the data you need converted, right? I see this constraint as a potential stumbling block, but it is unavoidable. At some point, however, you must run code I write to create your csv. Ensure this computer is not on the internet so I can't send patient data back to myself, ok?
You need a program that reads the .dat files and produces .csv, reliably, even if some of the data contains various errors and/or omissions.
Fake data is what I need. BE CAREFUL when you are making this data for me you don't overwrite, delete, or otherwise damage your patient data.
I need .dat files only. gnt are compiled cobol and idx are index+sometimes data to speed searches of large data sets.
I need the files associated with a db with a single record, with every field correctly filled out and a csv that matches what my script should produce for you from that.
Then you have to do it again with multiple records, (7, ok?). Again fill out all the fields. No csv required for this set, just the .dat file(s).
I'll make a script that converts .dat to .csv and you run it on a COPY of your patient data. It might choke on weird data, I'll try to make it as resilient as possible and have informative error messages, but we may have to go back and forth a bit to get through all the records without problems.
Please, whether you accept my bid or not, don't trust anyone's code on a networked machine.