REXXEXCI is my program. I pass some parameters to the program, and it issues an EXEC-CICS LINK to jump across EXCI into the CICS region to run a non-terminal transaction. I pass the APPLID of the CICS region, and a program name to run, and a transaction code to run the program under etc... I don't have any issue writing COBOL code to manage the EXCI aspects. I can code to the LINK interface or I can code to the CALL interface.
The area where I am a bit thin is in the REXX arena. I guess I will just work with the ADDRESS LINKMVS format and pass the individual parameters. I thought I would be able to do something cute by putting all the parms into a PLIST and then calling my COBOL program. I was able to put all the parms into a PLIST and then call an assembler routine, so I thought How much more work could it be to go directly to COBOL? Anyway again, I appreciate everyone's patience and assistance here. On Mon, Mar 22, 2021 at 3:25 PM Paul Gilmartin < 0000000433f07816-dmarc-requ...@listserv.ua.edu> wrote: > On Mon, 22 Mar 2021 14:53:02 -0400, Cameron Conacher wrote: > > > >when you say that my COBOL program must be coded Rexx-aware, do you mean I > >need to work with IRXEXCOM to tease the parameters out from REXX? > > > Perhaps that code exists in REXXEXCI. > > >I appreciate everyone's input. > >I don't know if the "appropriate" path forward is to use ADDRESS LINK or > >ADDRESS LINKMVS or if I should continue to try to make the CALL work. > >I don't want to waste anyone's time (other than my own). > > > >I will keep playing around to see if I can figure this out. > > > Oh my! > > o CALL invokes a Rexx subroutine (which might be coded in Assembler > with Rexx-like linkage.) > o 'CALL' (quoted) issues the TSO (or other command environment) > to invoke a program object with a JCL-like PARM. (Or do something > else.) > o address 'LINKMVS' invokes a program object with one (or more > JCL-like PARMs. (JCL allows at most one.) > o address 'LINKPGM' is close to the CALL macro interface. > o address 'LINK' is sui generis. I suspect it's a residue of > CMS conventions. > > TSO-REXX LISTSERV might correct some of my misconceptions. > > Is a description of REXXEXCI readily available? It seems to be > a CICS thing. > > -- gil > > ---------------------------------------------------------------------- > 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
