On Fri, 11 Sep 2015 13:24:44 -0400, Steve Thompson wrote:
>>
>> If this affects ordinary TSO/E users, not just customizers, shouldn't it
>> also be mentioned in:
>> SUBMIT command
>> z/OS TSO/E Command Reference
>> SA32-0975-00
>>
>> I don't see it there. RCF time?
>>
>Chapter 34. However, P 300 is probably more accurate since
>Chapter 34 is sizeable.
>
... where I read:
>What you looking for is "Writing an Exit for the SUBMIT Command".
>Then look for "Summary of using the bytes of word 5"
>
z/OS 2.1.0>z/OS TSO/E>z/OS TSO/E Customization>Customizing TSO/E
commands>Customizing the SUBMIT command and job output processing>Writing an
exit for the SUBMIT command>Parameter descriptions>Summary of using the bytes
of word 5
... where I read:
If the current statement has // in columns 1 and 2, but the SUBMIT command
processor
does not recognize the statement as a JOB, EXEC, or DD statement, it
assumes that it is an
operator command entered into the input stream. ...
... How many decades has it been since there were only those 3 JCL
statements? That leaves out comment, CNTL, IF, THEN, ELSE, ENDIF,
INCLUDE, JCLLIB, OUTPUT, PROC, PEND, SET, XMIT, anything within the
scope of DD DATA,DLM= ... to name a few.
>After you do this, if you are familiar with C/I, could you say
>with a straight face that SUBMIT did NOT attempt to implement C/I
>functions to pass flags to the customer's exit code?
>
I think they meant to say that they have an ad-hoc parser that
sometimes misclassifies JCL statements, and they don't care to
fix it.
>Yeah, that's what TSO CT said. We do not want to get into writing
>a Converter/Interpreter. Looks to me like they already have, saw
>that they are now in beyond their ability to maintain it and are
>pushing back.
>
Yup
Did you encounter problems with JCL statements or with SYSIN? Are
you using the IBM-supplied EXIT or an installation-supplied EXIT?
(Would you care to share an example?)
>So, I am writing an ISPF edit SUB/SUBMIT macro that will use the
>INTRDR.
>
I have one. It also allocates INTRDR with the attributes of the file
being edited. I agonize over whether to respect EXCLUDE and
PRESERVE. But I won't add options. See below.
>Me thinks this is now deprecated and as such is a dead horse.
>Moving on.
>
At the USENIX Summer Conference of 1983 Rob Pike made a presentation
titled 'UNIX Style, or cat -v Considered Harmful' and together with Brian
Kernighan ...
(Only the abstract is available without registration.) Its thrust is that
utilities
should be implemented sparsely, with clearly defined objectives. For example,
SUBMIT should just submit jobs; the bells and whistles should be left to other
utilities. (But the command structure of UNIX is more hospitable to agluinating
fine-grained utilities than that of OS/360.)
-- gil
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
