Re: Difference between DISP=NEW and MOD for a PDS member?



Charles

DISP is always for dataset, not for member.

That statement is a useful principle to keep in mind but it is not exactly
relevant to my question.

The statement is wrong and I think there may actually be a significance for
your threefold initial question.

For the arguments as to why that statement is wrong, please see, for
example, my posts in answer to Don Williams and Gerhard Adam.

(And more specifically, *disposition* (DELETE, CATLG, etc., as opposed to
status NEW, MOD, etc.) is always for the dataset, not the member.)

A subtlety not to be overlooked.

Chris Mason

On Sat, 5 Mar 2011 07:26:17 -0800, Charles Mills <charlesm@xxxxxxx>
wrote:

A lot of replies. Thanks all. Let me try to address various comments in a
single response.

You mis-interpreted my intention

Sorry. I took "I'd suggest you to simply perform some tests - IMHO it's much
better than asking others about it" and "Has it occurred to anyone to just
look at the JCL Reference manual?" as a rebuke. My apologies if I was
over-paranoid.

English is foreign language for me

Your English is certainly better than my Polish and I thank you for being so
tolerant of us mono-lingual Americans.

it seems the place to be having this conversation is with IBM

I am not alleging any bug in any IBM component. I was simply asking how it
worked. I think the whole of "MVS" could use some vast improvements in
documentation but that's just IMHO and I don't think IBM would be terribly
responsive to my opening a ticket. (And please, save the flames, yes, "MVS"
is much better documented than some operating systems.)

Ah! I get it. You interpreted my "I have a bug involving the treatment of
PDS members" as meaning that I was alleging a bug in "MVS." Nay! By "I have
a bug" I meant there is a bug (apparently) in my code. That is, the product
is failing at a customer's site and I have no reason to think the bug is
other than in my code.

I've found the immediate problem. Where my code apparently really should be
testing for NEW or MOD it was testing only for MOD, and as JCL apparently
populates JFCBIND2 with NEW (x'C0') for both NEW and MOD my code was
failing
for both cases.

I asked about the differences between NEW and MOD because if I was going
to
change the code I wanted to get it right. SHOULD the code be differentiating
between NEW and MOD? (Apparently not.)

I don't find that the MVS documentation is "scattered" ...
I don't recall seeing conflicting or contradictory documentation ...
I'm not sure what you might be referring to.

Agreed, on this particular topic, all of the documentation is apparently in
one place. But one does not know that, does one? If one is trying to RTFM
about how feature X works and one finds a paragraph or a chapter devoted
to
X, how does one know whether there is additional (or conflicting!)
information elsewhere? Particularly if the keywords in question are DISP or
NEW or MEMBER which occur with such frequency that a simple search is not
useful?

DISP is always for dataset, not for member.

That statement is a useful principle to keep in mind but it is not exactly
relevant to my question. FOO(BAR),DISP=NEW behaves differently depending
on
whether or not FOO(BAR) exists. So DISP is always FOR the dataset, but its
behavior is affected by the member. (And more specifically, *disposition*
(DELETE, CATLG, etc., as opposed to status NEW, MOD, etc.) is always for
the
dataset, not the member.)

Finally, I think some of you just don't appreciate the software vendor
sales/support environment. I think many of you if you were dealing with a
vendor you would be quick to say "you are asking big bucks for your product.
Like it or not, this is how our shop does things. Fix your product or go
away!" But somehow when you put on your "advice" hats you think a vendor
should be in a position to tell a customer "what you are doing is illogical
so fix your logic and THEN buy our product." You may be "right" -- but you
wouldn't sell much software (i.e., stay in business and pay your
programmers' salaries) if you were a vendor with that approach.

Thanks again. Great resource here.

Charles

-----Original Message-----
From: IBM Mainframe Discussion List [mailto:IBM-MAIN@xxxxxxxxxxx] On
Behalf
Of R.S.
Sent: Saturday, March 05, 2011 5:10 AM
To: IBM-MAIN@xxxxxxxxxxx
Subject: Re: Difference between DISP=NEW and MOD for a PDS member?


Charles,
You mis-interpreted my intention. I suggested the tests, because it's
IMHO good advice. I think you understand it as "don't bother us with
simple questions, you should RTFM or test it!". Yes, it can be
understood that way, but *it wasn't my intention*.
Please, remember: English is foreign language for me so sometimes some
nuances or allusions can be unintentional. I answered because not to say
"I know the answer", but because I knew the answer and thought it can be

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to listserv@xxxxxxxxxxx with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
.



Relevant Pages

  • Re: Difference between DISP=NEW and MOD for a PDS member?
    ... I am not alleging any bug in any IBM component. ... I don't recall seeing conflicting or contradictory documentation ... ... NEW or MEMBER which occur with such frequency that a simple search is not ... vendor you would be quick to say "you are asking big bucks for your product. ...
    (bit.listserv.ibm-main)
  • Re: strtomember and strtoset bug with hyphen
    ... I fully agree with you that the documentation should've been much clearer in ... As I explained, this cannot be considered as bug, because "-" and other ... reserved keyword as the member name without quoting it. ... >> You should either quote the special characters, ...
    (microsoft.public.sqlserver.olap)
  • Weekly Python Patch/Bug Summary
    ... Patch / Bug Summary ... Help with Python codebase ... Update to Macintosh library documentation ... Segfaults on memory error ...
    (comp.lang.python)
  • Weekly Python Patch/Bug Summary
    ... Patch / Bug Summary ... typo in liblocale.tex ... subprocess documentation ... Registry not updated by 2.4b1 installer ...
    (comp.lang.python)
  • Re: Reading local files in Netscape 6 and Mozilla (GM#001-NS)
    ... >bugzilla entry on bugzilla.mozilla.org which is the best place for bug ... This inconsistency can make it difficult for vulnerability reporters ... to contact the vendor, and some reporters feel forced to publicly ...
    (NT-Bugtraq)