Re: [DotGNU]ECMA doubt .. function overloading

[Rhys Weatherley]

Gopal V wrote:

> Still waiting for Rhys to respond,

Rhys wakes up and finds a full-blown ECMA spec interpretation
war going on in his mailbox ... :-)

The relevant section is "14.5.5.1 Method invocations".

    If N is applicable with respect to A (14.4.2.1), then all
    methods declared in a base type of T are removed from the set.

That's the missing piece of the puzzle.  When the compiler sees
the applicable "Foo(float)" member in "Child", it is supposed to
erase all methods in "Parent", regardless of whether they may
be better or not.

Personally, I think the current cscc behaviour is closer to
what the programmer intended, but the spec is the spec, so I'll
have a look at fixing this after I fully wake up.

Cheers,

Rhys.