On Fri, 2006-03-24 at 07:26 -0800, Guido van Rossum wrote: > On 3/24/06, Barry Warsaw <[EMAIL PROTECTED]> wrote: > > On Thu, 2006-03-23 at 17:06 -0800, Guido van Rossum wrote: > > > > > The pattern with the 'empty' flag is only needed when due to API > > > constraints you have only got an iterator. > > > > Which can happen quite often actually. Perhaps making the original > > object available as an attribute of the iterator can help in those > > situations though. > > It can't work, at least not in general. How do you do this if the > iterator is a generator? Or an infinite sequence? Or a filter? It > can't be made part of the iterator protocol. You can design your own > extension of the iterator protocol, but then it wouldn't accept > arbitrary iterators any more.
Yes, absolutely true. I wasn't really proposing a change to the generic iterator protocol, just suggestion something "one" could do if "one" needed that functionality (although an agreed upon convention would make it somewhat more general). -Barry
signature.asc
Description: This is a digitally signed message part
_______________________________________________ Python-3000 mailing list Python-3000@python.org http://mail.python.org/mailman/listinfo/python-3000 Unsubscribe: http://mail.python.org/mailman/options/python-3000/archive%40mail-archive.com
