Re: What would cause KEYLOC to be ignored?

On Mon, Aug 8, 2022 at 3:46 PM Jon Paris <jon.paris@xxxxxxxxxxxxxx> wrote:

There is really no reason to use the information because, while the file may be internally described by the program, it is externally described to the system and it know exactly where the key actually is.

Thanks. That explanation certainly fits my observations.

It seems IBM insists on maintaining the facade:

https://www.ibm.com/docs/en/i/7.4?topic=if-valid-search-arguments

I suspect it is a no-op probably left in place to document to the maintenance programmer where the key is supposed to be

Incorrect documentation is worse than no documentation. I think
especially so when the "documentation" looks like syntax rather than a
comment. I definitely disagree with this choice. Either it should work
the way it's documented, or it should be a syntax error enforced by
the compiler. I'd vote for the latter, but with IBM's commitment to
not break legacy code, I would think they had already decided on the
former. It's disappointing that they did neither of those.

John Y.