Re: Sending exception messages from a CPP back to the caller

Thanks to all who responded.
Since this specific task is a utility being used by programmers and
performs a single function, I opted to just monitor for all messages and
send back a CFP0001 (command ended in error) to *PRV as a *INFO message .
That way the programmer knows to look for low level exception messages.



On Wed, Nov 6, 2013 at 1:38 PM, Alan Campin <alan0307d@xxxxxxxxx> wrote:

I used to build custom message id with substitution variables and tons of
second level help but finally just started using CPF9898. Just a waste of
time because no one ever looked at any of it.

Don't know if you are interested but if you download any of my projects
from www.think400.dk/downloads.htm you will find a service program called
XVERRH. This is my standard error handler.

If you find an error, just issue a ERRH_Throw and you send an exception
message back and the message is logged and can be retrieved at a higher
level. Does all the work of error handling.


On Wed, Nov 6, 2013 at 11:27 AM, CRPence <CRPbottle@xxxxxxxxx> wrote:

On 06-Nov 08:14 -0800, Jeff Young wrote:

I has been ages since I have had to do this.
Do I just send the message to *PRV or *EXT?

The CPP is /transferred to/ vs called from the Command Analyzer
feature, so the invoker of the command is the /previous/ caller in the
stack. The External message queue is a special location rather than the
effective stack [i.e. the call message queue], so avoid using that
except when specifically that target is desired; I believe there is even
a restriction for sending *ESCAPE messages there, I do not recall.

FWiW: If another program were left in the stack between the calling
program [e.g. a CLP invoking the command] and the CPP, then the message
would be sent previous to that [hopefully static] program name; i.e. the
message is sent relative, minus one caller, to the known and consistent
caller [typically by name, per "Call stack entry identifier"] of the CPP
from a *CMD invocation.

Do I need a special message id or can I use the generic CPF9898?

That generic message provided by the OS *can be* utilized, and
hard-coded or dynamically-built message data to describe the failure,
but that is IMO a poor choice as a /design/ decision. Certainly
qualifies for the various expressions prefixed with: /Good enough for/

--
Regards, Chuck
--
This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing

list

To post a message email: MIDRANGE-L@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/mailman/listinfo/midrange-l
or email: MIDRANGE-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives
at http://archive.midrange.com/midrange-l.

--
This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing list
To post a message email: MIDRANGE-L@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/mailman/listinfo/midrange-l
or email: MIDRANGE-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives
at http://archive.midrange.com/midrange-l.