Re: myWishList
Rohan Hart <[email protected]> Mon, 20 Jun 2005 12:26:46 +1200
| Newsgroups | gmane.comp.lang.nice.general |
|---|---|
| Message-ID | <[email protected]> |
On Sat, 18 Jun 2005, Imam Tashdid ul Alam said: > --- Rohan Hart <[email protected]> wrote: >> abstract interfaces (or kinds) are (currently?) a weaker statement >> than ordinary interfaces in that they are 'provides' vs. 'is a', >> ie. you can *only* call methods on a class implementing a kind. > erm... completely lost. I *only* call methods on a class > implementing an interface. I bet you also expect to be able to store them in a field? That's not possible with abstract interfaces because they exist only at the type level. >> In any case the current implementation is really only adding an >> infinite number of commonly recognised Platonic entities to the >> set of classes considered for dispatch, in much the same way that >> singleton classes are available for dispatch. > alright.. erm... I think I understand. It depends on whether you see something like 3 being a value or type. I'll leave it to someone like Oleg to show type arithmetic in Nice. See http://okmij.org/ftp/Computation/type-arithmetics.html and http://okmij.org/ftp/Haskell/number-parameterized-types.html for examples in C++ and Haskell. >> Closed vs. open enumerations is a slightly different topic. I >> think Java provides both. > does it? could you give me an example? Sorry, by "I think" I really meant "I've never used Java1.5 enums and this is based on vague recollections from the 'whatsnew in Java1.5' pages" Rohan ------------------------------------------------------- SF.Net email is sponsored by: Discover Easy Linux Migration Strategies from IBM. Find simple to follow Roadmaps, straightforward articles, informative Webcasts and more! Get everything you need to get up to speed, fast. http://ads.osdn.com/?ad_id=7477&alloc_id=16492&op=click