David, Understood. I use REXX extensively for ISPF edit macros and such. I was agreeing that there were some "nice to have things" which clist did have. When I started, clist was the only way to go. Once we were introduced to REXX, I began converting clists to EXECs, and found WRITENR was a "clist only" entity. I agree with others in that the clist "PROC" could have been useful in REXX. I suspect most of us have written some sort of general-purpose parsing logic for use in REXX.
At this point, I have only a handful of clists which are still in use. Most are "1 time only" type things and are not worth converting. Bill Hitefield Dino-Software Corporation 800.480.DINO www.dino-software.com > -----Original Message----- > From: IBM Mainframe Discussion List <[email protected]> On > Behalf Of David Spiegel > Sent: Thursday, January 9, 2025 7:26 PM > To: [email protected] > Subject: Re: ISREDIT CAPS ON OFF > > Hi Bill, > I also have an Exec which calls my Writenr clist.. My point was counter the > idea that with Rexx, CLISTs are never necessary. > > Regards, > David > > On 2025-01-09 19:18, Bill Hitefield wrote: > > I wrote a dinky CLIST to issue WRITENR. I call it from REXX as needed. > > > > Bill Hitefield > > > >> -----Original Message----- > >> From: IBM Mainframe Discussion List <[email protected]> On > >> Behalf Of David Spiegel > >> Sent: Thursday, January 9, 2025 5:42 PM > >> To: [email protected] > >> Subject: Re: ISREDIT CAPS ON OFF > >> > >> Hi Tony, > >> CLISTs also have the WRITENR statement, which Rexx lacks. > >> > >> Regards,. > >> David > >> > >> On 2025-01-09 16:49, Tony Harminc wrote: > >>> On Wed, 8 Jan 2025 at 19:51, Lennie Bradshaw > >>> <[email protected]> > >>> wrote: > >>> > >>>> CLIST design is over 50 years old I think. > >>>> > >>> CLISTs as we know them arrived with OS VS/2 Release 3, so the second > >>> release of MVS around 1974 or so. Before that, on e.g MVT, SVS, and > >>> the first MVS, CLISTs had no control flow statements - they were > >>> just lists of commands with some substitution for arguments. > >>> > >>> At introduction, everyone was all excited at the shiny new CLISTs, > >>> and I remember well wondering why there was no lower case support, > >>> and if it could be easily fixed. Indeed it could, and I found and > >>> happily zapped out the upcasing and tried a CLIST with some lower > >>> case text in it. It blew up all over the place, and I discovered > >>> that pretty much all byte values beyond the upper case letters, > >>> numbers, and a few more (i.e. roughly the JCL character set at the > >>> time) was being used as internal syntactic markers for various > >>> aspects of the language. So much for > >> that quick fix. > >>> No, I don't remember the details, but I imagine much of today's > >>> CLIST processor is unchanged from the 1970s version, and the source > >>> code for that is available if anyone wants to research it further. > >>> > >>> But seriously, why does anyone use CLISTs today? There is to my > >>> knowledge just one reason: there is tight integration with TSO > >>> parsing and the CLIST header, whereas REXX requires that the command > >>> be treated as a single string, and then parsed with the (unarguably > >>> powerful) > >> REXX parsing scheme. > >>> But that doesn't allow for a syntax error to be discovered in the > >>> REXX and the user prompted using TSO facilities; the REXX has to do > >>> all the work itself. > >>> > >>> Tony H. > >>> > >>> -------------------------------------------------------------------- > >>> -- For IBM-MAIN subscribe / signoff / archive access instructions, > >>> send email to [email protected] with the message: INFO > >>> IBM-MAIN > >> --------------------------------------------------------------------- > >> - For IBM-MAIN subscribe / signoff / archive access instructions, > >> send email to [email protected] with the message: INFO > >> IBM-MAIN > > ---------------------------------------------------------------------- > > For IBM-MAIN subscribe / signoff / archive access instructions, send > > email to [email protected] with the message: INFO IBM-MAIN > > ---------------------------------------------------------------------- > For IBM-MAIN subscribe / signoff / archive access instructions, send email to > [email protected] with the message: INFO IBM-MAIN ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
