To reject or abend for incorrectly using same dataset for input and output

Thanks for submitting the request. The DFSORT documentation clearly discusses the considerations for using the same data set for input and output. It is up to the user to determine if this is a desirable strategy or not, and implement it or
avoid it accordingly. For example, in cases where the user has a backup of the input data set or creates it in a previous step, using the same data set for input and output may be a desirable strategy.
Issuing a warning message and setting a non-zero return code after the fact if the same data set was used for input and output would be of little value if the data set was already corrupted by the sort application.
Requiring an explicit request by a parameter in order to allow the same data set to be used for input and output would change the way DFSORT currently works and cause jobs to fail which did not fail before. This would cause problems for many DFSORT users.
Since DFSORT can have multiple input data sets (concatenated input data sets for sort and copy, or multiple input data sets for merge) and multiple output data sets (SORTOUT, OUTFIL), cross-checking all of the input and output data set names would require significant code and resources.

Given these considerations, we respectfully decline this RFE request. However, we encourage you to submit new requests in the future, and we will be happy to review them.