From: Peter Dimov (pdimov_at_[hidden])
Date: 2004-10-20 15:25:40
Stefan Seefeld wrote:
> The fact that this particular implementation I'v been using is based
> on utf-8 should be transparent (though accessible).
Then you should provide constructors/functions that take iterators to UTF-8,
UTF-16, and UTF-32. Users will invoke whatever is most convenient to them,
and you'll do the conversion internally. Unless I am misunderstanding
something.