An alias entry in a PDS directory doesn't point to the base name, it points to the actual member. And, yes, I know about load modules, but what the link editor/binder puts in the user halfwords doesn't count.
-- Shmuel (Seymour J.) Metz http://mason.gmu.edu/~smetz3 ________________________________________ From: IBM Mainframe Discussion List <IBM-MAIN@listserv.ua.edu> on behalf of John McKown <john.archie.mck...@gmail.com> Sent: Thursday, May 3, 2018 2:16 PM To: IBM-MAIN@listserv.ua.edu Subject: Re: [EXTERNAL] Re: IEFA107I when pointing to dataset alias On Thu, May 3, 2018 at 1:06 PM, Paul Gilmartin < 0000000433f07816-dmarc-requ...@listserv.ua.edu> wrote: > On Thu, 3 May 2018 15:22:22 +0000, Seymour J Metz wrote: > > >That was my point; if you need it at IPL time and it's cataloged in a > user catalog, you need an explicit volume serial. > > > I'm (slowly) coming to grasp that. At IPL time, the CAS is not > initialized and > user catalogs can not be searched. So data sets needed during IPL must > either > be catalogued in the master cat or accessed by explicit volume serial. > > But in the case that impacted me many years ago, I wanted both the alias > and > the related DSN in different user cats. I didn't need either during IPL > (not my > job) and in user cats they couldn't be accessed during IPL. It still > disconcerts me > that after CAS initialization a user cat can't be searched for the alias > and the HLQ > of that alias could not identify a possibly different user cat to search > for the > related DSN. > > (Ih another case I would have found it useful to have an alias of an > alias. That, > also, should be supported.) > > --gil > > This is disconcerting to me too. But I can envision what might be happening. The logic to me would be something like: 1) Find catalog in which A.B.C exists according to the standard search order 2) Read entry for A.B.C in that catalog. 3) If alias, find the base related-to entry in the catalog. This reminds me a bit of BPAM. Suppose you have member A, but not B, in PDS1. Now suppose you have B in PDS2 as an alias to A (in PDS2). If you ask for member B then you get A in PDS2, not the PDS1 version. Granted, I don't think that BPAM actually uses the A entry in PDS2 for anything when you reference B, but I can easily be wrong. I'm speaking conceptually. The way you & I want it to work is like a UNIX symlink (which can traverse to another filesystem), but it is working more like a hard link (which cannot span filesystems). -- We all have skeletons in our closet. Mine are so old, they have osteoporosis. Maranatha! <>< John McKown ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
