RE: RNF3826 - A subfield cannot be of type Object

Perhaps you are correct in this case; I'm not sure where the java
application is maintained, and I lack the time to scan through the history
find out, which has been trimmed off.

If you're not the creator/maintainer of the application (let's say, for
example, you need to similarly manage sets of worksheets in an HSSF
spreadsheet), this discussion is still valid and pertinent. In fact, I have
several "spreadsheet programs" where I have to manage sets of information,
and this inability to link related data has been an issue. We had the same
issue - at a different client - with a java application that did address
standardization. For comparing results of that standardization with results
of other standardizations (in a duplicate-purge process) this was quite
unwieldy, and could have been simplified via DS.

Dennis Lovelady
http://www.linkedin.com/in/dennislovelady
--
There are very few personal problems that can't be solved by a suitable
application of high explosives.

Could this not also be handled by creating a java array of connection
objects, and then saving each instance of a connection object in that
java array, as they are created?

In that way, you would not have to resort to trying to "fool" the
compiler, and introduce code that could potentially "break" at some
future release.

In other words, perhaps you are creating this array at the wrong
architectural level, e.g. as an RPG IV array, versus a Java array
object.

Those are my thoughts on this situation.

What do you all think?

All the best,

Mark S. Waterbury