Addison Phillips wrote: > http://www.w3.org/TR/charmod-resid/#C059 That document is apparently rather old, it still mentions RFC 2396, and it has a normative (sic!) reference to [I-D IRI]. <shudder /> It also references XML 1.0 chapter 4.2.2 claiming that I can find provisions for IRIs there. Checking that in XML 1.0 4th ed. it's about the <SystemLiteral>, and the presented algorithm boils down to "use UTF-8 and percent-encode". That's no fair summary of RFC 3987, it's not good enough to get a valid STD 66 URI from a given valid IRI, it doesn't mention the IDNA issue. And for crying out loud, there's also no RFC 3987 reference in XML 1.0 4th ed. published only months ago. <sigh /> Frank (deciding to use pure US-ASCII XML for yet another decade)Received on Friday, 20 April 2007 23:20:36 GMT
This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 2 June 2009 19:17:13 GMT