W3C home > Mailing lists > Public > public-rdf-in-xhtml-tf@w3.org > October 2009

Bug in Test Case 121 description

From: Gregg Kellogg <gregg@kellogg-assoc.com>
Date: Sun, 25 Oct 2009 20:46:56 -0400
Test case 121's description is '"[]" is a valid safe CURIE', when it should be '"[]" is not a valid safe CURIE'. The Ruby RDFa parser passes this, because it treats [] as an alias for the current subject, rather than ignoring it.

The problem came to light when I was trying to generate a triple to describe a <script> as property of the current page using the following test case:

<?xml version="1.0" encoding="UTF-8"?>
<html xmlns="http://www.w3.org/1999/xhtml" xmlns:ex="http://example.com/ex#">
   <head about="http://example.com/9999.xhtml">
      <script src="/javascripts/jrails.js" rev="ex:script" resource="[]" type="text/javascript"></script>

This test case generates a triple using my parser, but should not generate anything.

<http://example.com/9999.xhtml> <http://example.com/ex#script> <http://example.com/javascripts/jrails.js> .

That said, I think that this is a reasonable thing to want to assert and perhaps RDFa 1.1 will consider some way to work with such src attributes. To get the triple in a valid way, I need to use the following element:

      <script src="/javascripts/jrails.js" rev="ex:script" resource="http://example.com/9999.xhtml" type="text/javascript"></script>

But this is inelegant.

Suggested actions:

 1.  Change the description for TC 121 to "[]" is not a valid safe CURIE'
 2.  Add the above test case to ensure that [] does not generate a valid resource.

Gregg Kellogg
e: gregg@kellogg-assoc.com<mailto:gregg@kellogg-assoc.com>
c: (415) 686-8603
p: (415) 459-3202
f: (317) 642-2755
Received on Monday, 26 October 2009 00:47:59 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:50:33 UTC