On 15-Aug-2002, John Norwood <[EMAIL PROTECTED]> wrote: > >Rotor *does* have CLS compliance checking, doesn't it? ;-) > I don't believe there's really anything a CLI implementation has to do > for CLS compliance checking.
For a CLI implementation, sure. I'm talking about the C# implementation, which is also part of Rotor, and in particular about whether it conforms to the requirements for a CLS consumer and CLS extender. Looking at the ECMA standard, I see that it defines the notion of "consumer" and "extender" only in informative text, rather than in normative text. It does say that "Compile-time enforcement of the CLS rules is strongly encouraged.", but this is not strictly speaking a requirement. > I believe the Rotor implementation of the C# compiler is the same as > the .NET Framework C# compiler in this area. In other words, the .NET Framework C# compiler is no better in this respect -- neither one implements the ECMA standard's recommendation ;-) > >I think the ECMA standard required it, last time I looked... > This is an interesting point. Sorry, I should have remembered that in the final ECMA standard it was only a recommendation, not a requirement. I was remembering how I wanted the standard to be, not how it ended up. The committee chose not to formalize the notions of CLS consumer, CLS extender, etc., I think primarily due to lack of time. -- Fergus Henderson <[EMAIL PROTECTED]> | "I have always known that the pursuit The University of Melbourne | of excellence is a lethal habit" WWW: <http://www.cs.mu.oz.au/~fjh> | -- the last words of T. S. Garp.
