fully-qualified domain names containing differently-encoded labels ?

=JeffH Jeff.Hodges at KingsMountain.com
Mon Oct 25 23:51:38 CEST 2010


 > On Mon, Oct 25, 2010 at 01:23:45PM -0700, =JeffH wrote:
 >
 >> But in any case, it would seem that any specification of conversion
 >> operations on a multi-label domain name, that takes into account
 >> IDNA{2003,2008}, needs to properly allow for and accomodate
 >> label-by-label type diversity, yes?
 >
 > Given that neither IDNA2008 nor IDNA2003 had anything to say about
 > domain names, but only about label names, I should hope so.  You
 > should treat them only by label, and not by domain name.

Understood. However, in various protocols, e.g. HTTP, one has to handle (e.g. 
compare) strings matching RFC3986's "reg-name" production, and thus must 
decompose said strings into labels, do the IDNA magic as necessary, and then 
re-construct as appropriate and compare (for example).

 > In general
 > in the DNS, this is true anyway.  You have no way of knowing whether
 > there are actually 8-bit-clean labels inside a domain.  They're
 > perfectly legal under STD13, but they violate the hostname rules.  If
 > you're not naming a host for hostname-compatible purposes, there is
 > nothing wrong with creating the label "punctuationmistake's" inside
 > your domain "example.com".
 >
 > Moreover, note that your use of "type" there seems a little suspicious
 > to me.  If these were really different types in any important way, I
 > would have expected rather strong arguments in favour of changing the
 > xn-- prefix.

wrt the latter, fwiw, note that RFC5890 uses the terms "label type" and "label 
form" essentially synonymously; and the former is explicitly used to denote 
label types..

    For IDNA-aware systems, the valid label types are: A-labels, U-labels,
    and NR-LDH labels.


thanks,

=JeffH

















More information about the Idna-update mailing list