Gil, IMO, that is a wonderful idea. I especially endorse the latter idea of enhancing the C runtime fopen() as a generalized way to introduce this ability to _any_ C language code which uses fopen(). Just thinking about it, I think this code be done most compatibly by introducing a new keyword parameter for the "mode". Perhaps something like: "enq=shr|old|rispf|wispf". Hopefully the first two are obvious. The "rispf" says use DISP=SHR to allocate and issue a ISPF compatible ENQ "shared" on the data set plus member (if specified in the filename parameter) whereas "wispf" says use DISP=SHR to allocate and issue an ISPF compatible write enq on the data set + member (if specified).
It might even be nice to have a C runtime configuration parameter which indicate what this default fopen() enque is. On Tue, Jun 28, 2016 at 5:27 PM, Paul Gilmartin < 0000000433f07816-dmarc-requ...@listserv.ua.edu> wrote: > Would there be much support for an RFE like this? (Would > it be a duplicate -- the idea seems obvious?) Any > suggested changes or paraphrases or clearer terminology? > > Enhance OMVS "cp" and other utilities to employ ISPF-like > member-level serialization. This would allow OMVS to update > PDS libraries that are othewise allocated SHR. For example: > > I'm editing CLIST(FOO) with z/OS ISPF Edit. From a Solaris NFS > client, I try: > > user@SunOS.5.11: cd clist > user@SunOS.5.11: date >foo > ksh: foo: cannot create [Permission denied] > > ( ... which really means "data set in use".) So I CANcel my > Edit session, then: > > user@SunOS.5.11: date >foo # Succeeds! > > NFS appears to be using ISPF serialization with member granularity. > In my experience, FTP Server does likewise. > > Now, from z/OS UNIX System services: > > user@OS/390.25.00: date | cp /dev/fd/0 "//CLIST(foo)" > cp: FSUM6259 target file "//CLIST(foo)": EDC5061I An error occurred > when attempting to define a file to the system. > > ... CLIST is allocated; member is not in use, but "cp" doesn't know any > better. > I log off TSO, then: > > user@OS/390.25.00: date | cp /dev/fd/0 "//CLIST(foo)" # Succeeds! > > It's an undue burden to need to log off in order to update with OMVS > data sets that are allocated in TSO LOGON PROC. OMVS should use > the techniques that NFS and FTP use. > > If the restriction is intrinsic to the "C" Runtime, consider this > an RFE on "C". So much the better to have a broader solution. > > Thanks, > gil > > ---------------------------------------------------------------------- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN > -- "Pessimism is a admirable quality in an engineer. Pessimistic people check their work three times, because they're sure that something won't be right. Optimistic people check once, trust in Solis-de to keep the ship safe, then blow everyone up." "I think you're mistaking the word optimistic for inept." "They've got a similar ring to my ear." >From "Star Nomad" by Lindsay Buroker: Maranatha! <>< John McKown ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
