Re: [Pnet-developers] Re: Possible misinterpretation of 'protected internal'

[Rich Baumann]

On Tue, 2004-09-07 at 10:47, Carl-Adam Brengesjo wrote:
> This reveals that it's a bug in csc rather than in cscc.
> csc seems to interprent "protected internal" to allow access to the 
> member in the same assembly as the derived class also.
> X is in the same assembly as B, and foo() is marked `internal', therefor 
> allow access to foo() from B.
> Look on the testcase provided, X.SomeMethod() is allowed by csc, but not 
> X.SomeMethod2(). This seems to me to be a clearly incorrect behaviour of 
> csc as X does not belong to the same assembly as A.

X is a nested class of B, so it should be able to access all private and
protected members of B. If X were a separate class, external to B, then
a compile error would be appropriate for the attempted use of the
protected method.


Rich