We have such discussions at the site where I'm currently working, too.
What I miss most of the time:
there are all sorts of source code generators, most of the time little
home grown tools (written in REXX), for example: enter a name
of a DB2 table, and then the tool gets the column definitions from the
DB2 catalog and builds Cursor Selects, structure definitions, Insert
statements etc. in, say, PL/1 syntax, and inserts them directly
in the source code at the desired place; often confirming with
site-specific coding standards. They work with ISPF on the
mainframe, in the "classical" TSO environment, of course.
Or another example: I wrote similar tools in C that create source code
to help with XML validating (binary representations of XML schema
definitions, to be included in PL/1 and C source code and to be used
together with a special - very fast - XML parser).
What about those tools? Will RD/z be able to adopt them? They are
mission critical for some of my customers.
When I asked such questions in the past in the RD/z evaluating team sessions
(another customer), I got answers like: "this is not needed any more"
(which is not true, it speedens up my daily work) and "you didn't understand
what RD/z is for" ... in the end, I was excluded from the RD/z
evaluation team,
because I asked such questions :-(
but: the RD/z project at this particular site did fail dramatically,
much money
was lost.
Now, two years later, another customer, the same story - I was chosen for
RD/z evaluation again, because I have experience with the product :-)
not so much fun ...
Kind regards
Bernd
Am 29.08.2016 um 19:29 schrieb Jesse 1 Robinson:
This should maybe be part of a different thread. Here's a tiny snippet of ISPF
edit in batch. I would not call it 'interactive', but it's context sensitive,
so results depend on the data being edited. There is no separator character in
play. These lines are all in a Rexx.
ADDRESS ISPEXEC
"ISREDIT MACRO"
"VGET (cpldsn cplvolp cplvola)"
ADDRESS ISREDIT
"(memname) = MEMBER" /* current member name */
"C ALL 'DEST=TSAF' 'DEST=HOLD3' "
"C ALL 1 '// ' '//*' " /* null cards to comments */
"X ALL '//' 1" /* exclude JCL and comment lines */
"F FIRST NX 'DEFINEDS' " /* find first or only define stmt */
"F NEXT NX 'DEFINEDS' " /* find next define stmt, if any */
IF RC = 0 THEN DO /* multiple define stmts found */
"DEL .ZCSR .ZLAST" /* delete extraneous lines */
END
.
.
J.O.Skip Robinson
Southern California Edison Company
Electric Dragon Team Paddler
SHARE MVS Program Co-Manager
323-715-0595 Mobile
626-302-7535 Office
[email protected]
-----Original Message-----
From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf
Of Paul Gilmartin
Sent: Monday, August 29, 2016 9:56 AM
To: [email protected]
Subject: (External):Re: RD/z
On 2016-08-29, at 10:35, Jesse 1 Robinson wrote:
I'm not sure what's being asked for here. ISPF in batch is a well-documented
process. I have batch ISPF jobs that run daily. You don't use separator
characters. You write a Rexx that issues ISPF commands in sequence. You can
edit or use virtually any ISPF function--such as table management--that's
available interactively.
But batch isn't interactive. I perceived the OP as wanting to use interactive
(not batch) ISPF as a plug-in editor for RD/z.
Does "don't use separator characters" mean that no character is sacrificed as a
separator; than one can freely code such as:
address ISREDIT 'find a;b'
and not having the ";" (or any other character) being taken as a separator?
-- gil
----------------------------------------------------------------------
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
