Soucy, Michael wrote on 8/1/2018 4:36 PM:
I seem to recall not too long ago someone posted a question regarding the program called "caller-id" written by Lloyd Deviney from NEWS/400. It was stated that there is an issue with this program working on V7R3. I forgot to print off the solution to the problem. I've tried searching the archives, but I can't find the answer. Is the issue related to the APIs that are being called in this program? When I Google "caller-id" I came across a question I had posted on the RPG Café. One of the replies references a link to the original article which doesn't seem to exist any more. Can someone please post a link to the solution to fixing this?
I can't tell if the issue is that you are on 7.3 and are experiencing
a problem, or
Want to proactively fix the code before you get to 7.3, or
Want a generic 'what does my call stack look like' solution.
Searching the Midrange archives for 'Lloyd Deviney' tells me that
Lloyd used the QMHSNDPM API. A likely 'problem' with this API
surfaces when you take a working example from an existing program and
move it into a sub-procedure. That alters the depth of the call
stack, so if you were looking one level back before, you need to go
two levels back now. It's not an actual problem; it's a
misunderstanding of how the call stack works, and the results of that
misunderstanding are 'results not as expected'.
The API I use for this is QWVRCSTK. A search in the archives here
shows several promising examples.
The more general internet search for the API name + 'example' turns up
some ideas, too:
Carsten Flensburg is a treasure when thinking about API examples. He
has made his API articles available at
Searching for 'Call stack' returns this:
Click the 'download
source' link to see everything including the articles.
FWIW, this interrogates the call stack, so moving it higher or lower
in the call stack will alter the results the API returns.