Re: [Mono-list] implicit, explicit, and why does C# have these?

[David La Motta]

Thanks for the explanation.  I can see how the implicit operator can be useful in the example you describe; still, I think it wasn't necessary for C# to expose them to us.  I.e., let us deal with the explicit casts and spare the confusion they may cause.  In other words, an implicit cast from a Pear object to a Truck object can seem quite odd, assuming their inheritance tree has nothing in common. // David Jonathan Pryor wrote: It's good to keep this in mind: C#'s "builtin" types (int, long, etc.) are actually aliases for managed types (System.Int32, System.Int64, etc.). These managed types are (for Mono, at least) written in C#. You expect the following code to work: int n = 42; long l = n; short s = (short) n; Which means that the following code also works: System.Int32 n = 42; System.Int64 l = n; System.Int16 s = (System.Int16) n; Which means that the managed implementation of those managed types needs *some way* to represent to the compiler/runtime that some coercions are "safe" (can be done implicitly), while others are "unsafe" (can be done explicitly). C#'s implicit/explicit operators are how this is declared and defined. As for the differences between implicit & explicit, you understand the difference already. Implicit == doesn't need a cast; explicit == requires a cast. - Jon On Thu, 2003-10-16 at 16:52, David La Motta wrote: So I was reading my "C# for Java Developers" book and I came across the implicit and explicit operators. Java doesn't have these but a friend suggests C++ does. Being that I am not a C++ developer I really can't comment much on it, except to say that it looks like C# just decided to copy functionality from C++, just because it is a "cute" feature of the language. I also can't quite put my finger on the difference between implicit and explicit, so if anybody has any insight on these, please, do share. It seems to me that explicit is used when you want to force your API clients to use a cast when dealing with different types. So lets say that I have: public static implicit operator Foo(Bar bar) {...} and I also have a method called public Foo morph(Foo foo) {...} If I was ever to use my morph method with a Bar, I could issue the call like: Bar bee = new Bee(); Foo faa = morph(bee); And the compiler would be happy. If I was to change implicit for explicit in the operator's declaration, the way of calling the method would be: Foo faa = morph((Foo) bee); // with explicit cast Is this it, or is there more to it than this? I also am aware that this isn't really a "mono" question per se, but I thought some of you would be willing to shed some light on the topic... :-) Thanks! // David _______________________________________________ Mono-list maillist - [EMAIL PROTECTED] http://lists.ximian.com/mailman/listinfo/mono-list