URI sets as handled by POWDER

[Joseph A Holsten <joseph@josephholsten.com>]

For those interested in LRDD, I just wanted to point to the spec for  
IRI sets in POWDER:
   http://www.w3.org/TR/powder-grouping/

IRI sets allow you to specify some acceptable values for certain parts  
of an IRI:
http://jdoe@www.example.com:1234/example1/example2?query=help#fragment
\  /   \  / \             / \  /\                / \        / \      /
  --     --   -------------   --  ----------------   --------   ------
   |      |         |          |         |              |        |
scheme iuser    ihost        port     ipath         iquery  ifragment
        info

For each rule about the IRI, a processor generates an equivalent  
regular expression, and these get chained together for a decision rule.

Some examples of how these look:
   <iriset>
     <includehosts>example.org</includehosts>
     <excluderesources>http://www.example.org/stylesheet.css http://www.example.org/jslib.js 
</excluderesources>
   </iriset>
which compiles into two matching regex:
   must match: /(([^\/\?\#]*)\@)?([^\:\/\?\#\@]+\.)?(example\.org)(: 
([0-9]+))?\//
   must not match: /^(http\:\/\/www\.example\.org\/stylesheet\.css|http 
\:\/\/www\.example\.org\/jslib\.js)$/


For those who care about the details:
* "The regular expression syntax used is defined by XML schema as  
modified by XQuery 1.0 and XPath 2.0 Functions and Operators." that's http://www.w3.org/TR/xpath-functions/
* The OWL/RDF property used for this stuff is http://www.w3.org/2007/05/powder-s#matchesregex
* Before matching agaist the regular expression, IRIs are cannocalized  
according to 2.1.4

http://josephholsten.com