On Tue, Nov 20, 2012 at 3:38 AM, Boris Zbarsky <[email protected]> wrote: > So I don't think we should require this behavior.
Thanks, I went with requiring decoding per UTF-8 for now causing sequences such as "%80" to turn into U+FFFD. I think that simplification over just decoding valid utf-8 sequences is justified because once you run the Punycode algorithm the original byte sequence meaning is lost anyway. -- http://annevankesteren.nl/
