Joined: 23 Nov 2006 Posts: 19270 Location: Inside the Matrix
Once you construct a valid (unix and windows style names are not valid) dataset name, how will it be specified in the input DD statement?
If the input file name might change from run to run, i'd suggest you put the process in a PROCedure and specify the variable part of the name in a symbolic parameter. The same symbolic parameter could be used in the output dataset name and the value would be automatically used both places (or even in subsequent steps of the jobstream).
To learn about valid dataset names and symbolic parameters, i'd suggest you look in the JCL manuals available via the "IBM Manuals" link at the top of the page. If you read something that is not clear to you post it here as well as your question about what is confusing.