[lug] (OT) C++ ?, pointers to member functions

[Scott A. Herod]

"D. Stimits" wrote:
> 
> "Scott A. Herod" wrote:
> > To tie this with another C++ thread from a couple of weeks ago.  I guess
> > that I would not expect objects built with different "packing" to be
> > used together successfully.  ( I probably should check this with a
> > simple test. )  Wouldn't changing how the compiler segmented memory
> > for the layout of objects make vtables incompatable?
> 
> My own comment on this is that I wasn't expecting you to be mixing
> compilers, or to directly manipulate vtables. But in the case of
> anything declared virtual, it is the common characteristic of relative
> offset that gives the ability to work with virtual replacements. Am I
> incorrect when I assume you are using member method pointers under a
> single compiler and platform (rather than mixing)?
> 
> D. Stimits, stimits at idcomm.com
> 

No, I am.  I was just curious.  This mainly relates to the discussion
about COM that I mentioned in a previous post.  My little bit of reading
suggests that it relies heavily on the consistency of the vtable layout.
That bothers me and I am trying to rationalize my discomfort.

Scott