[boost] Re: Interest in bidirectional map?
by David B. Held other posts by this author
Oct 23 2002 8:25AM messages near this date
[boost] Re: Interest in bidirectional map?
|
[boost] Re: Interest in bidirectional map?
Dirk Gerrits wrote:
> [...]
> I think I'd prefer the other notation because it allows you to use
> compile-time constants as indices.
So would I, but I'm afraid I can't turn lead to gold. ;)
> Perhaps you could offer both a preferred syntax and a compatible
> syntax in the spirit of Boost.Function and Boost.Signals?
If both syntaxes were realizable, I would certainly consider it. As it
stands, someone considerably more clever than I would have to make my
original proposal work. :( It's not just VC6 that wouldn't like it.
I'm afraid there might not be any legal way to do what I suggested in
C++. In particular, foo<n> can only name a type, not an object. And as
far as I know, you can't call the subscript operator on a type. I guess
that's what happens when you talk first and think later. ;)
Dave
_______________________________________________
Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
Thread:
Dirk Gerrits
David B. Held
Dirk Gerrits
=?iso-8859-1?Q?Joaqu=EDn=20M=AA=20L=F3pez=20Mu=F1oz?=
David B. Held
David B. Held
David B. Held
David B. Held
Jeremy B. Maitin-Shepard
|