ASCII & EBCDIC Characters in the Index Attribute Name In ACIF    

Question:

I am currently using ACIF to generate an AFPDS file, on the mainframe, using an ASCII input data file. I have defined my Index attribute names (Account number, Date, etc.) in my ACIF parameters, yet when I view the results, the Index name tags are unreadable. Can you help clear up the confusion over the use of ASCII and EBCDIC in this case?

Answer:

When defining the Index attribute name parameters, it is critical to know both the language of the input data and the platform where the processing occurs.  The following paragraphs describe the correct usage.

If you are processing in a zOS, VM or VSE environment and the input data file is in EBCDIC, then you can define your attribute names using EBCDIC characters.

If you are processing in a zOS, VM or VSE environment and the input data file is in ASCII, then you must define your attribute names using hexadecimal values.

If you are processing in a Windows or AIX environment, and your input data file is in ASCII, you can define your attribute names using ASCII characters.

If you are processing in a Windows or AIX environment and the input data file is in EBCDIC, then you must define your attribute names using hexadecimal values.

For more information check out the IBM ACIF reference materials available at:

http://www.ibm.com/Search/?q=S544-5285&v=11&lang=en&cc=zz&Search.x=37&Search.y=11 .