Re: JCL History (was: ... PARMDD ... )
On Tue, 28 Feb 2017 09:41:43 -0800, Charles Mills wrote: >There were no APF authorized programs ... > But there was the hazard of buffer overruns even in unauthorized programs. IBM provides features to protect its code without timely extension of such features to customer-written code. Consider the decades during which IBM ignored the need for REFRPROT. REFRPROT should be mandatory, not optional. >-Original Message- >From: Tom Marchant >Sent: Tuesday, February 28, 2017 9:22 AM > >On Tue, 28 Feb 2017 10:19:49 -0600, Paul Gilmartin wrote: > >>Ah! That shows that (at least at one time) it was possible to increase >>the length of the PARM without introducing intolerable incompatibilities. > >Perhaps because the potential integrity issues were not understood at the >time. Thanks, gil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: JCL History (was: ... PARMDD ... )
On Tue, 28 Feb 2017 11:21:50 -0600, Tom Marchant wrote: > >>I'm curious because others have said (in this forum?) that earliest PROCs >>had no arguments; all modification was done by overrides. So, at that >>time symbols didn't exist in JCL, neither as PROC formal parameters nor >>in the (relatively recent) SET statment. > >That agrees with the description of Cataloged procedures in the 1967 manual. > And, apparently prior to instream PROCs, there was no PROC nor PEND statement. IIRC, PEND used to be prohibited in library PROCs (why?) Lately, it's allowed. This permits INCLUDEing a library PROC to make it an instream PROC. Is there anything done with overrides that couldn't nowadays be done as well with arguments/symbols? (But the problem of dangling commas when attributes are nullified would have to be addressed. Most easily by ignoring excess commas.) And symbols could address the deficiencies of overrides when PROC calls are nested. Referback deficiencies persist. Are arguments to outer PROCs available as symbols in nested-called PROCs without being declared as formal parameters in the inner PROC? Thanks, gil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: JCL History (was: ... PARMDD ... )
There were no APF authorized programs ... Charles -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of Tom Marchant Sent: Tuesday, February 28, 2017 9:22 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: JCL History (was: ... PARMDD ... ) On Tue, 28 Feb 2017 10:19:49 -0600, Paul Gilmartin wrote: >Ah! That shows that (at least at one time) it was possible to increase >the length of the PARM without introducing intolerable incompatibilities. Perhaps because the potential integrity issues were not understood at the time. >Where's the PROC statement described? I find it neither in the ToC nor >in the Index although there are numerous mentions of "catalogued" >proc[edures]. (PROCs aren't catalogued; procedure libraries are.) In-stream procedures were introduced in release 19. See the 1970 manual. >I'm curious because others have said (in this forum?) that earliest >PROCs had no arguments; all modification was done by overrides. So, at >that time symbols didn't exist in JCL, neither as PROC formal >parameters nor in the (relatively recent) SET statment. That agrees with the description of Cataloged procedures in the 1967 manual. -- Tom Marchant -- 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
Re: JCL History (was: ... PARMDD ... )
On Tue, 28 Feb 2017 10:19:49 -0600, Paul Gilmartin wrote: >Ah! That shows that (at least at one time) it was possible to increase >the length of the PARM without introducing intolerable incompatibilities. Perhaps because the potential integrity issues were not understood at the time. >Where's the PROC statement described? I find it neither in the ToC nor >in the Index although there are numerous mentions of "catalogued" >proc[edures]. (PROCs aren't catalogued; procedure libraries are.) In-stream procedures were introduced in release 19. See the 1970 manual. >I'm curious because others have said (in this forum?) that earliest PROCs >had no arguments; all modification was done by overrides. So, at that >time symbols didn't exist in JCL, neither as PROC formal parameters nor >in the (relatively recent) SET statment. That agrees with the description of Cataloged procedures in the 1967 manual. -- Tom Marchant -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
JCL History (was: ... PARMDD ... )
On Tue, 28 Feb 2017 07:25:32 -0600, Elardus Engelbrecht wrote: >Tom Marchant wrote: > >>Some time between 1967 and 1970. > >I think the limits of 40 or 100 characters were based on a quick way (without >using tapes or extra punch cards) to give shortish parameters to a program >using puch cards. Or so it was told to me by an oldie years ago. > My surmise, also. >Amazing how they worded same things then and now. > When was "dataset" banished in favor of "data set"? >>See page 85 of >>http://bitsavers.trailing-edge.com/pdf/ibm/360/os/R19_Jun70/GC28-6704-0_JCL_Reference_Rel_19_Jun70.pdf On Tue, 28 Feb 2017 06:33:42 -0600, Tom Marchant wrote: > >>When did it change to 100? > >Some time between 1967 and 1970. > >See page 85 of >http://bitsavers.trailing-edge.com/pdf/ibm/360/os/R19_Jun70/GC28-6704-0_JCL_Reference_Rel_19_Jun70.pdf >for OS/360, dated June, 1970 > >See also page 18 of the fifth edition of the OS/360 JCL manual, dated >March, 1967, where the limit is specified as 40 characters. > >http://bitsavers.trailing-edge.com/pdf/ibm/360/os/R01-08/C28-6539-4_OS_JCL_Mar67.pdf > Ah! That shows that (at least at one time) it was possible to increase the length of the PARM without introducing intolerable incompatibilities. Where's the PROC statement described? I find it neither in the ToC nor in the Index although there are numerous mentions of "catalogued" proc[edures]. (PROCs aren't catalogued; procedure libraries are.) I'm curious because others have said (in this forum?) that earliest PROCs had no arguments; all modification was done by overrides. So, at that time symbols didn't exist in JCL, neither as PROC formal parameters nor in the (relatively recent) SET statment. -- gil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
