First a correction of my own typo. I was familiar with a
407 at the University of Colorado circa 1970, not "mid-1950's".
(Not quite that old, yet.) And it was standalone on the job
prep floor, not connected to any computer.
In a recent note, "Shmuel Metz (Seymour J.)" said:
> Date: Wed, 15 Mar 2006 07:56:00 -0500
>
> There's a bit in the DCB for concatenation of unlike attributes, and
> it's been there since the 1960's. There's additional support for mixed
> block sizes that came in incrementally with various releases of DFP.
> For a long time it has been the cases that are not handled that are
> the exceptions rather than the cases that are handled.
>
And I understand that nowadays at least BLKSIZE is taken from the
maximum among the catenands, not the first. But the calling
program must be notified synchronously of the crossing of the
unlike-attribute-boundary so it may adjust its behavior accordingly.
(RECFM=F must be treated differently from RECFM=V.)
I tried the jobstep:
//* Test validity of catenation in SYSTSIN.
//*
//STEP EXEC PGM=IKJEFT01
//SYSTSPRT DD SYSOUT=(,)
//SYSTSIN DD *
ALLOCATE -
// DD *
SHR -
// DD *
DUMMY
// DD *
LISTALC STATUS SYSNAMES
with no errors and expected results, even for the extreme case of
concatenation within a single "statement" (or is that "command"?)
RCF submitted.
> >Now I'll stick my neck out and suggest that maybe the manual author
> >has got a few points wrong
>
> Certainly writting "statement" when he ws referring to a record rather
> than a statement.
>
Three errors of fact in sixteen lines of text. I wonder if this
is representative of the entire publication.
-- gil
--
StorageTek
INFORMATION made POWERFUL
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
