-----BEGIN PGP SIGNED MESSAGE-----

On Sunday 12 January 2003 04:09, Alan Gutierrez wrote:

> They state that overloading by return type is not supported by the
> standard. The discussions also imply that although not allowed, it is
> certianly possible for many compilers, since many compilers put the return
> type in the mangled name.

Wasn't the problem a covariant return-type (i.e. returning a subtype of the 
original type in a reimplemented method) and not overloading by return-type?

Cheers
Carsten Pfeiffer
-----BEGIN PGP SIGNATURE-----

iQEVAwUBPiF6l6WgYMJuwmZtAQHk2ggAtaj/hs5NbiQtvXtJkmCOGd/i++qwaraU
AJ8qGwjyl6SdX9/3zT5J21AriiV4gj1GedxjKJj/IjGD0Idhx0o6XSrHor+6NTTH
/lVXJVMjLLlEQHlhU0KOE659sdLoE84GZX1ljtMTb4VgzbAznwo3/JXQqMOuy6nK
hM/4Ivyg1EpcW3GfIv+GBOUhH07TD5N6i3+J8mEhIpngXRYWc98Mmhomw1EFf72E
vwgweSxTNoqyQyvYlY1JaWANnwhJfyck53p9xRIBtLXZnbcgM2i9oBsF6ZhaXf2r
G7wvDS9rz8Su6Mn8btBFbwLQqWaVvnzZZ1N2Ve9+wNNOQLiN5eTnjg==
=/vvr
-----END PGP SIGNATURE-----