Re: Creating a trigger program

[PaulMmn <PaulMmn@xxxxxxxxxxxxx>]

I'm coming (fashionably) late to the party to add my 2 cents...

First-- when you retrieve a journal entry for the "File of Interest" 
you have the 'raw' entry-- a bunch of journal fields (time-date-user 
who did it-etc) plus the fields from the original file.  But as a 
single field.

I think others have shown the data structure with the journal fields 
and a space for 'your fields go here.'

You create a file to match that data structure-- journal fields + the 
fields in the original file, then CPYF from the retrieved journal 
entries to this file FMTOPT(*NOCHK).  This is about the =ONLY= time 
I'll allow a *NOCHK!  It just runcates anything in the journal record 
you don't need.

As you've noticed, this is primarily a batch process.

However, IBM has another command, RCVJRNE, which allows you to tell 
IBM's journal processor to hand you a copy of every journal for a 
list of files as they're being written to the journal receiver!  This 
allows near real-time processing of the journal entries.  No more 
scanning all of the journal receivers for the whole day as a batch 
process!!

See http://systeminetwork.com and search for CBX902 for the first 
part of the sample code; the rest is mentioned on the page.

There's a link to the file with the source code.

--Paul E Musselman
PaulMmn@xxxxxxxxxxxxxxxxxxxx