It might be possible to have a program (Cobol, pl/1, etc.) or script (REXX, clist, Unix Perl, etc.) generate the JCL and create the wild card function that way.
Since I do not know the full expectation of the process from the OP, it would help to just see a sample of what is being requested. If the request is to have jcl like the following, then there is no current process in z/OS JCL to do that //DD1 DD DISP=SHR,DSN=*.MYDSN.LIST Collect all datasets that end in MYDSN.LIST //DD1 DD DISP=SHR,DSN=SYS*.**.CNTL Collect all datasets that begin with SYS and end in CNTL I would think a REXX process that uses ISPF functions to emulate the option 3.4 that would allow the collection of Datasets with Masks and then build the JCL and place it somewhere OPCA could get it, might work. Perhaps OPCA has a coding interface (I know that Workload Manager - ESP from CA does) that would allow the JCL to be created. Then all the OP has to do is create the file and OPCA would be able to create the JCL needed. Since OPCA is part of IBM Tivoli, the OP could open an SR with IBM for assistance. Lizette > -----Original Message----- > From: IBM Mainframe Discussion List [mailto:[email protected]] On > Behalf Of John McKown > Sent: Saturday, July 26, 2014 7:06 AM > To: [email protected] > Subject: Re: to use wildcard characters in JCl > > I agree with Gil, but wlll also mention mention that the BPXBATCH he was > talking > about runs a UNIX program, not normal z/OS batch programs. > And can only generate UNIX file names, not legacy data set names (DSN=). > > If you're thinking of what the Windows cmd prompt, or a UNIX shell can do, > where > you can "wild card" to match multiple files in a given directory (or folder > in Windows- > speak), then you simply can't do that with z/OS JCL. This "wild carding" is > sometimes called "file globbing" > in UNIX and z/OS JCL does not have _anything_ like that for JCL in the DSN= > of a > DD, or in a PARM=, or anywhere else. The closest for z/OS batch might be the > LISTCAT command in the IDCAMS program, but that doesn't help you directly. And > that "wild carding" is done by the IDCAMS program internally. There is not a > z/OS > system API call to do it for you. At least not directly. And, again, not in > JCL. JCL is > _primitive_! > > The "best" z/OS can do, and it's none too good, is what is called GDG-ALL > processing. A pretty technical explanation is given here: > http://www-01.ibm.com/support/docview.wss?uid=isg1II08285 > A nice tutorial on GDG data sets is here: > http://www.simotime.com/gdgone01.htm > > I include the above only for completeness, just in case you are not familiar > with > GDGs. > > On Sat, Jul 26, 2014 at 12:58 AM, Rajesh Janakiraman > <[email protected]> wrote: > > Dear Mates, > > > > If you have any idea about using wildcard characters in JCL, kindly let me > > know. > > > > I need to get an extract file for every one hour and I need to add the job > > in OPCA, > so that by using wildcard characters for every one hour the extract may get > genrated automatically. > > > > Please let me kow the detailed explanation i.e. wildcard characters that is > > found in > JCL to describe a dataset. > > > > Thanks in advance. > > > -- > There is nothing more pleasant than traveling and meeting new people! > Genghis Khan > > Maranatha! <>< > John McKown > ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
