Fairly Simple (I think) programming project

[Glyph]

I need a simple program to read in a flat file, and write out a flat file based on some stuff. The input is data maps of programmed survey questions. The output is file that defines some locations and codes. That is probably a confusing description -- see the example below.


Below is what a single question in the input file looks like. They are all enclosed in {}
The key pieces of information we need are the data location. (in this case 906.2) and the response codes. (look at the bold stuff)

{ L4: #12.92 906.2
L4.
What is your current marital status? Are you ...

(READ LIST. RECORD ONE RESPONSE.)

!FLD, , 1
() 01 Single
() 02 Married
() 03 Living with significant other
() 04 Divorced/separated
() 05 Widowed
() 99 Refused }


So from a file that contains the above, we would need to locate a marital status question. (probably look for keyboards, like divorced, or married in response text)

Then determine the location, and use the location and the codes to define 4 standard categories. Below is what the output file would look like.

>def @marital1 [906.2#02] //@marital1 defines Married
>def @marital2 [906.2#01] //@marital2 defines Single / never married
>def @marital3 [906.2#03] //@marital3 defines Living with someone
>def @marital4 [906.2#04,05] //@marital4 defines Separated / widowed / divorced

That is the concept. There would be some other kinds of questions like age, ethnicity, income, gender, education, etc

I would provide some detailed specs as to what will qualify a question as marital, or income, etc -- and what terms exactly will count for each defined category within that type of question, etc

One of the companies I already got a quote from, suggested using an XML config file so that the rules could actually be tweaked after the fact. (e.g. if I wanted to add a qualifying keyword, etc) So that functionality would be nice.

Contact me or post here if you have any questions.