On Tue, Mar 18, 2008 at 5:22 PM, John Cowan <[EMAIL PROTECTED]> wrote: > felix winkelmann scripsit: > > > > A real module system would solve all these problems cleanly. > > It wouldn't solve the data-punning problem. As long as the same object > can be seen one way by one module and another way by another, problems > will continue to be endemic. To fix that, we need two run-time types, > which I hereby dub restricted and unrestricted strings. >
The contents of a string a purely defined by the operations you perform on them. Let me add that I don't think this discussion leads anywhere. Unfortunately every time the word "uincode" appears, everybody feels pressed to discuss the different solutions at length (it's a particularly perverse form of the bikeshed syndrome). I admit that finding the optimal way of handling unciode is an unsolved and interesting problem, but it's much easier to talk about it than to implement it. The core *must* stay minimal (and it pains me to see the stuff that is regularly added to the extras unit - yes, that's you, kon). This is both for performance and portability reasons, and because it is simply the way it is currently implemented. There is enough low-level support to handle unicode in higher level layers (like the utf8 egg). Changing the string representation is absolutely out of the question - too much depends on it, and, let's be frank, nobody would be able to implement the required support, anyway. cheers, felix _______________________________________________ Chicken-users mailing list [email protected] http://lists.nongnu.org/mailman/listinfo/chicken-users
