This page is a compilation of formal public feedback received so far. See Feedback for further information on this issue, how to discuss it, and how to provide feedback.
Date/Time: Fri Jan 17 07:55:53 CST 2014
Name: Anne van Kesteren
Report Type: Error Report
Opt Subject: Feedback on UTS46
I want to reference UTS46 from the URL Standard: http://url.spec.whatwg.org/ A question I had was whether the algorithms could be made more explicit. You define ToASCII, but you do not clearly stipulate what the input is to ToASCII within that section. You do not say it takes a UseSTD3ASCIIRules flag, or a Transitional Processing or Nontransitional Processing marker. Additionally you do not say what it returns in the end. The same feedback applies to ToUnicode. The text additionally assumes DNS lookup will be done, while this algorithm could be used in just a URL parsing context. There won't be any DNS queries at that level. Which brings me to my next point, which is that currently DNS length restrictions are enforced here, while for the purposes of URL parsing we might not want to have those enforced (they are not by browsers) as the names might be used in a non-DNS context as well. I also wonder whether you have given any thought to give advice as to how to write domain names outside the context of the validity algorithms. E.g. defining what sequence of code points would constitute valid input vs what sequence of code points has an effect. E.g. "A.com" could be considered "invalid" because it maps to "a.com".
Feedback above this line was considered at the February 2014 UTC meeting.
Date/Time: Mon May 5 12:47:07 CDT 2014
Name: Markus Scherer
Report Type: Error Report
Opt Subject: draft UTS #46 dangling sentence
http://www.unicode.org/draft/reports/tr46/tr46.html#ToUnicode Processing step 1 ends with Note that unlike ToASCII of [RFC3490], this always signals whether or not there was an In the current 6.3 version it says Note that unlike RFC3490 ToASCII, this always signals whether or not there was an error. It looks like the part about errors was to be moved to the end of step 2. Maybe that whole last sentence of step 1 should go? I suggest carefully reviewing the diffs.