If modifiactions on the ISPF side are allowed, it would be trivial to provide a command that solitied the file name without upper casing it, then invoked IND$FILE or implemented the same protocol.
There is no RSF command; you read inbound structured fields with a Read Modified command. There is an open source TN3270 called x3270; there may be others. HFS and zFS are file systems. The Unix files that you access exist within a file system, and look the same whether the file system is HFS (deprecated)or zFS. Unlwaa you are an admin you don't deal directly with zFS files. And, yes, if you want to handle both legacy datasets and Unix files then you must provide syntax to distinguish them. One possibility would be to treat anything not beginning with period, slash or tilde as a legacy dataset. -- Shmuel (Seymour J.) Metz http://mason.gmu.edu/~smetz3 ________________________________________ From: IBM Mainframe Discussion List [IBM-MAIN@LISTSERV.UA.EDU] on behalf of Charles Mills [charl...@mcn.org] Sent: Sunday, March 28, 2021 10:33 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: IND$FILE and zFS? More on this subject. Stop me if I am boring you. <g> I withdraw what I said. I have no way of knowing what IBM's attitude to an RFE would be. Does IBM ever (anymore) accept RFE's to TSO? Legally I think IND$FILE is a part of PC/3270 but technologically it is a TSO/E command. Adding zFS support to IND$FILE would be only part of the battle. You recall of course that ISPF uppercases TSO command operands in most cases, so the enhanced IND$FILE would only be usable with most zFS files from the true TSO command line and a few places in ISPF like option 6 and Edit. Also, I think some (many? most? all?) 3270 emulators may restrict, structure or massage the mainframe file name in some way, so zFS files might not be accessible with IND$FILE from every 3270 emulator. If you think this is a valuable idea an IBM enhancement is not the only route. IND$FILE is not magic. One could write one's own, either as a "for profit" enterprise, or as an open source or CBT tape project. The protocol is readily reverse-engineered: every 3270 emulator ("real" or "TN") author has done it. The protocol is half-documented IIRC. The historic 3270 datastream document I believe documents the 3270 commands Read and Write Structured Fields (RSF and WSF). IND$FILE simply or mostly wraps the bytes of the file in an RSF or WSF command. Is there an open source TN3270? That would be an easy source for the IND$FILE protocol. Even easier (albeit less efficient at runtime) would be a TSO command that front-ended or wrapped IND$FILE. The command would simply pass a QSAM VB dataset name to IND$FILE, and copy it to or from the user-specified zFS file. It could be written in Rexx, which would make it a pretty simple project. I think most 3270 emulators let you specify your own name for IND$FILE. One small design issue would be recognizing zFS filenames. There is no "absolute" way to recognize a zFS filename: SYS1.MACLIB is a valid zFS filename. I'm not saying it is unsolvable; just that it is an issue that would need to be solved. You might prefix zFS filenames with zFS: or you might require that every zFS file specification contain at least one slash somewhere. Recognizing lowercase characters in the filename is not sufficient; not if users have previously been permitted to specify dataset names in lowercase. -----Original Message----- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of Paul Gilmartin Sent: Saturday, March 27, 2021 2:59 PM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: IND$FILE and zFS? On Sat, 27 Mar 2021 14:45:37 -0700, Charles Mills wrote: >Waste of bits. IND$File is the b@stard child of an orphan. It's AMODE/RMODE >24, if that gives you any clue as to its currency. > >IMHO > Yes, but ... The use of IND$FILE remains so prevalent (partly because of absurd site security rules), and the use of zFS has become so prevalent, that the union should be sanctified. I can hardly imagine that transferring a Classic data set is performed except by ALLOCATing it, and a general technique for allocation is hinted at in: https://www.ibm.com/support/knowledgecenter/SSLTBW_2.4.0/com.ibm.zos.v2r4.bpxa500/tso.htm The match should be made. ---------------------------------------------------------------------- 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 ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
