ABEND grouse 
Author Message
 ABEND grouse

On Monday, 97/05/19, "Kent A. Wendler" wrote to All about "ABEND
grouse" as follows:

KW> So far as I know, the current mainframe versions of the compiler
KW> only document how to cause an abend of the program for specific
KW> oncodes.  In other words you have to anticipate and specifically
KW> list each oncode in order to get the abend for conditional
KW> disposition processing of the datasets used by the program.
KW>
KW> Way back, in the documentation for an earlier release of the V.1
KW> compiler there were instructions for replacing the IBMBEER1 library
KW> subroutine so that *any* raising of the error condition would cause
KW> a program abend.
KW>
KW> Well, this still works with the V.2 compiler, but it's not very
KW> well documented.  It seems to me that this is sufficiently useful
KW> (how can one anticipate *every* possible oncode that might happen?)
KW> that it should once again become a well-documented feature of the
KW> current compiler versions.

You might care to address this directly to IBM via e-mail to

In the latest language standard there is a condition called ANYCOND,
and a default handler for this could generate an ABEND. The ANYCOND
condition is only raised after all earlier levels of handler have been
tried or omitted. So, it's the court of final appeal, where the death
sentence is finally confirmed.

The ability to define a default handler for this (and any other)
condition would be a boon. These handlers could simply be added to the
resident and/or transient library by the systems programmer installing
the PL/I compiler and library.

I think if enough sites ask for this facility, IBM will be amenable to
its development. [As if Peter Elderon, et al., don't have enough to do
already.]

Regards

Dave
<Team PL/I>

 * KWQ/2 1.2i * Be a {*filter*}; wear boxer shorts.
--
Please remove the '$' in the from line before reply via email.
Anti-UCE filter in operation.



Fri, 05 Nov 1999 03:00:00 GMT  
 ABEND grouse

So far as I know, the current mainframe versions of the compiler only
document how to cause an abend of the program for specific oncodes.  In
other words you have to anticipate and specifically list each oncode
in order to get the abend for conditional disposition processing of the
datasets used by the program.

Way back, in the documentation for an earlier release of the V.1 compiler
there were instructions for replacing the IBMBEER1 library subroutine
so that *any* raising of the error condition would cause a program abend.

Well, this still works with the V.2 compiler, but it's not very well
documented.  It seems to me that this is sufficiently useful (how can
one anticipate *every* possible oncode that might happen?) that it should
once again become a well-documented feature of the current compiler
versions.



Fri, 05 Nov 1999 03:00:00 GMT  
 
 [ 2 post ] 

 Relevant Pages 

1. OT - Ultimizer Utility and Re-entrant COBOL programs and Turbo abends(or hard abends) with Ultimizer on.

2. Intentional abend

3. Help trapping S113 ABEND

4. ABEND=S000 U4093 REASON=0000001C

5. Subtask abend

6. wtor abend e23

7. x37 Abend trapping in ASM

8. x37 Abend trapping in ASM

9. System abend '0E37'?

10. Demystifying LE and LE Abends

11. LOAD abends ...

12. ICHRIX02 RACF exit abend

 

 
Powered by phpBB® Forum Software