On Mon, 29 May 2023 19:10:55 -0500, Charles Mills wrote: > >>Support for PDSE as a member* of a GDG was added in z/OS 2.1 or 2, iirc. >>*(not to be confused with members of a PDS[E] > >And also not to be confused with PDSE 2 member generations, right? > >> maybe they could have >> tried harder not to be confusing > >Agreed. > IBM couldn't have made it more confusing. The proof is that they did not do so.
>>note, while GDGs usually consist of "generations" of the same dataset, >>there's no rule that says they must be. > >Interesting. Did not know that. > Explain; even provide two examples; one of a GDG consisting of "generations" of the same data[ ]set, and one which does not. Can't an Assembler programmer using STOW create PDS members with names beginning with '+', '-', '0', 'π', ... just about anything? I don't believe it's proper for higher layers such as JCL to introduce syntactic restrictions harsher than those of core layers. The only member name prohibited is 8X'FF', indicaring the end of a PDS directory. How do LM services accessible, e.g. through REXX support member "generations"? -- gil ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
