On 2017-05-05, at 10:05, Jesse 1 Robinson wrote:

> IBM's reluctance to provide system symbol support in *batch* is well 
> motivated. From the beginning of system symbols, they were honored only in 
> STC and TSU. In batch they were ignored. The problem with batch is that a 
> number of key factors affecting symbols are unpredictable. So when and how 
> would a symbol be set?
> 
> At submit time?
> At conversion time?
> At NJE transmission time?
> At execution time?
>  
Make it an option; give the programmer control.

>     ...
> What if a system symbol happens to clash with an existing JCL or PROC 
> variable?
>  
And here, we have the horrible, irreparable design error in JCL
syntax.  Reference to an undefined symbol should have *always*
been a syntax error.

Otherwise, JCL or PROC variables should dominate system symbols.

There should be a construct to restrict scope of symbols defined
in JCLLIB members so they don't clash with symbols in open code.

I hate JCL!

-- gil

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

Reply via email to