Reputation: 1 Apprentice
- Active Members
- Active Posts:
- 122 (0.15 per day)
- 06-September 11
- Profile Views:
- Last Active:
- Dec 09 2013 11:19 AM
- Who Cares
- OS Preference:
- Favorite Browser:
- Favorite Processor:
- Favorite Gaming Platform:
- Your Car:
- Who Cares
- Dream Kudos:
Posts I've Made
Posted 8 Dec 2013The haeder file can't be corrupted, otherwise you wouldn't be able to open it in the old application. Why don't you write your program, create a virtual disk and install dos on it (Free dos 1.1 you still can download). Your father puts all the new customers on your program and still can access the old database on a modern computer. Just a thought.
He want to keep his old data...stubborn man. Guess I'll just get the data that is easy to read, put it in a new Db and let him add the BTW (VAT) number by himself. Will save him work not to put every customer in it again and will save me work not figuring out how the number is encrypted.
Posted 6 Dec 2013
I already searched for solutions, but the header file is corrupt if I try to open them in a DB manager so can't do anything with them.
There's no export of data? Nothing to a CSV, text, or what not?
I would write something into the old ms-dos application to just dump it to csv. That's assuming that they wrote the old app. If not, what is the application called? Is there something open source? Freeware? Renegade developer that posted the source code?
The person writing the application has died so there isn't any source code anymore. Also I'm not an expert in old coding. Also, it was a program designed for my dad, so don't think there is anything. Name of application was Pacapim.
Could it be that the number is an int/uint?
The number is an int as far as I know. An example is 453541511. Does it matter what it is?
Posted 5 Dec 2013I have figured some stuff out by converting the text to bytes. It turns out that each field has a fixed length in bytes. Spare space in those fields are filled up with 'spaces'. The BTW (VAT) number of 9 characters is encrypted in a rather special way. Let say the number 413544355. (I read this from the application). This is translated to 5 bytes which are 24 125 49 69 157. Those 5 bytes are saying nothing at first but, there is some connection. I gonna line out some numbers here:
- 0 0 0 0 129
- 0 0 0 0 130
- 0 0 0 64 130
- 0 0 0 0 131
- 0 0 0 32 131
- 0 0 0 64 131
- 0 0 0 96 131
This goes on for a while but when the 3rd (middle) byte is added to the calculation it makes no sense at all anymore. What you see here basically is, the 5th byte starts at 129 for the number 1. At 2 it is 130. At 4 it's 131. So every power of 2 causes this 5th byte to be raised. From the looks of it, the 4th byte is a scale between 0 en 128. If 1 number is between the 2 powers of 2, then 128/2 = 64. When there are 3, 128/4 = 32 and so on. When you go above 256 the rule changes entirely leaving me in the dark how to figure out this. Maybe someone has seen this before?
- 0 0 0 0 129
Posted 3 Dec 2013Also, let me re-ask my question, is there an export function to this existing program? There's no reason to try and hack through some file structure when you can get it in a more succinct format.
Both the .FIC and .NDX files contain clear and scrambled up data. In the .FIC file, the addres and company name are clear, but the VAT number (in Belgium BTW) is encoded in the file.
You are not making sense. In your post above you showed clear data with the ".FIC" file and garbled with the ".NDX". Is that not the case?
There is no way to export the data out of the existing program. So my only chance on converting the database is working with these files.
Posted 2 Dec 2013There's no export of data? Nothing to a CSV, text, or what not?
I mean - if your FIC is the actual data it appears to be tab delimited.. so it shouldn't be that hard to do an import into an actual DB (well.. once you set up the tables).
It isn't that hard no, but the problem and confusing thing is that part of the file is readable (first columns) and a column that isn't readable (see example for the .FIC file). I'm wondering if there was a way to get that data extracted/translated to characters that make sense like a big number or words, not random characters or maybe even invisible characters.
- Member Title:
- D.I.C Head
- Age Unknown
- Birthday Unknown
KFredje hasn't added any friends yet.