> I would have hoped that unloading would have been part of the test
> suite of any particular DLL, there are going to be a number of DLLS
> which are never going to get unloaded of course and with the
> interconnections a great many will be mutually or secondarily
> addref'd. Not for the first time I feel like the number has been
> screwed to the door before the footings were in place.
Some operating systems don't allow you to unload a DLL once its been
loaded. For OpenVMS I need to hardwire any CanUnload to FALSE.
That's an #ifdef then and a simple case to test :-).
Simon
Colin.
Beware knowledge cheaply gained for in the spending of it you may pay more than its worth.
S. P. Lucy
