On 2016-12-09, at 09:56, Farley, Peter x23353 wrote:
>
> For myself, I have no need to explore or use such edge cases. All my symbol
> substitutions stay within the "classic DD *" limit of 80 bytes after
> substitution (IDCAMS input to delete/define VSAM files with special
> characteristics not covered by the JCL LIKE parameter in my most common use
> cases, sometimes SORT control cards too for parameterized record
> selection/modification).
>
Yet a programmer who discovers a problem in such an edge case and
fails to report it so it can be fixed is doing his peers a disservice.
I opened the SR as JES2. I already have a related one on SDSF.
It's dismaying that such problems are not discovered in product test.
I'd expect a high priority of a test team would be to explore edge
cases that developers might have overlooked in unit test.
I have an RCF, now 15 months old requesting documentation:
Please provide an explanation of what determines the attributes
(RECFM, LRECL, BLKSIZE) merged into the DCB by OPEN. ...
The answer so far:
Hi, Paul, thanks for your note. Our Development team has this
service item in their queue. I can send you a note when we have
some updates ready. This suggested update will take some time
for Development to complete and won't appear in the V2R2 GA
document.
(I wish they could send me a preview. If they do, I'll test against
it and probably generate more SRs.)
> "Doctor! Doctor! It hurts when I do that! --- Well, don't do that!"
>
Is that in the Reference Manual anywhere? Part of the specification?
-- gil
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
