>"PARM='&WOMBAT_1'" vs "PARM='&WOMBAT._1'"

FWIW, it likely won't simply be the presence of "_"  in the symbol name 
that is the indication of "long value" in part because JES symbols, 
introduced in z/OS 2.1, allow "underscore" as a normal part of the name. 
Underscores within names would likely become a "normal" allowed part of 
symbol names.

Gil's case is (basically) right on target. This is a case that we already 
would intend to point out in the migration information for the release and 
made clear to the ISV's (and provide a simple exec to look through a 
PDS(E) for instances).

There are some other somewhat related factors, such as: we would reject an 
attempt to create a system symbol named "&sym_more" if there is a symbol 
named "&sym".
This restriction makes it possible to say that if you see &WOMBAT_1 then
- if symbol &WOMBAT_1 exists, substitute it for &WOMBAT_1
- otherwise, if symbol &WOMBAT exists, substitute that for &WOMBAT and 
append _1.
since it is known that both cannot be system symbols and thus you cannot 
be substituting with the wrong symbol.

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