I had presumed, silly me, from
>I can't find any documentation for SAVE that includes a 
>second or third operand!
that SAVE actually did not document the second and third operands. But it 
does. So at least that removes the example from the category of using 
something undocumented.

The behavior is "the length is the number of characters between the first 
quote and the last quote" and "take the characters between the first quote 
and last quote, and put that within CL8'...' for each multiple of 8 and an 
appropriate thing for whatever remains." And yes that means that if your 
quotes happen to be in "unfortunate positions" things won't work. That's 
why it works for some SYSPARMs and fails for others.

I understand why you don't like that for quotes, I don't like it either. 
But safety dictates that we leave it unchanged.

However the thing that seems to be getting lost here is that the DB2 
sample is simply incorrect and should not have the extra quotes. They 
don't want an extra quote in the generated code any more than you do. Tell 
them to fix it. And don't accept a simple "no" for an answer. It would 
probably be nice to accept a FIN.  And if they don't choose to change the 
sample, then they must document the restrictions on the lengths of 
SYSPARMs that "work" (such documentation would be pretty silly and take 
longer to figure out than changing the sample). I suspect they'll be 
pretty happy to fix the sample in a future release.

Sure, we could provide a new and improved SAVEX macro (or more simply a 
new option on the existing SAVE macro). But I would guess that in practice 
there is no one who has a need for such a construct. And no one who would 
use it. Certainly DB2 doesn't need it. And almost no one even knows of the 
existence of the option. So submit a requirement that makes a valid 
business case and I'll be glad to give you a 4th operand on SAVE (or a 
keyword) that does things the way you would expect. But don't ask us to do 
work to create something that no one will use.

It appears that no one vocal enough to ask has needed the function in the 
lifetime of the SAVE macro. Needs can certainly change. But have they, in 
this case?

Peter Relson
z/OS Core Technology Design

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to