by BillyBoyo » Thu Oct 11, 2012 4:03 pm
If you have something in your JCL which you know is an output file, the system doesn't know that from the JCL. You can code a concatenation on that and submit your JCL - but it won't do anything remotely useful for you.
You can, of course, read any input (concatenated or not) and write to multiple output files, but this is not concatenation, it is just multiple files.
Correct, VSAM cannot be concatenated.
You can't mix F and V. Of course you could convert one to the other, but then you'd no longer have a mix anyway.
On the "mix blocksizes" it is a "trick" question in two parts: can you mix different FB datasets?; can you mix different VB datasets, and if so with what restriction?
You can experiment. Get two different FB datatsets and try. Make an enormous record on one VB and a very short record on another and try to process on a concatenation with the short one first (this didn't used to work, maybe it has changed, I haven't checked).
DD DUMMY, and it is in the manual, will give end-of-file when encountered in a concatenation. If you have two datasets, DD DUMMY, then two more datasets on your concatentation, then only the first two datasets will be read.
- These users thanked the author BillyBoyo for the post:
- sinmani (Thu Oct 11, 2012 4:19 pm)