Note: The document accompanying this article is available for download here.

I am writing a trigger receiver program using RPG with embedded SQL that will go through the before and after record images and create an audit record in an audit file for each field that was changed on an update operation.

I am developing a “shell” trigger program that I’d like to be able to modify slightly, and use the shell as a basis to write these audit records for each file with a trigger on it.

Instead of writing multiple (IF this field changed THEN DO…) for

It seems that every time I see Copy from Import File (CPYFRMIMPF) mentioned in a Web forum, the question concerns CSV files. CPYFRMIMPF also handles files of fixed-length fields. Such files have certain advantages over CSV files, and there’s more to using them than the IBM documentation tells you.

Don’t get me wrong. CSV files are great. I use them often. But consider that fixed-length data has its advantages. Two that come to mind are:

1. Fixed-length character fields may contain any character without the nuisance of special treatment. In CSV files, character fields require special handling of the characters used

Thanks for the article on managing IBM i spooled files. We often find it useful to examine user job logs to see exactly where a problem is. But lately, we’ve been getting too many spooled files in QEZJOBLOG. We’re getting 17,000 QZSHSH and QP0ZSPWP job logs each day from website jobs. Any ideas on how to make these useless job logs go away?

–Dale

Preventing A Job Log From Being Generated

My first thought is to see where these jobs are coming from. If you can find the CL program, job scheduler, or submit jobs statement that launches these