RDFa JavaScript Library

Hi all,

So I fixed a couple of edge-case bugs in my RDFa JavaScript library
(good test cases!), and am now passing 85/98. The 13 failed test cases
break down as follows:


TC 11: likely SPARQL bug (UNION)

TC 42: likely test harness bug (negative result)

TC 86: likely test harness bug (negative result)

TC 92: likely SPARQL bug (UNION)

TC 94: likely SPARQL bug (UNION)

TC 99: likely SPARQL bug (\n vs. actual newline)

TC 100: same canonical XML, it actually works.

TC 101: same canonical XML, it actually works.

TC 102: same canonical XML, it actually works.

TC 103: same canonical XML, it actually works.

TC 107: test harness bug (negative result)

TC 108: likely SPARQL bug (white space)

TC 112: likely SPARQL bug (white space)


In other words, as far as I can tell, the parser is doing the right
thing in all test cases. However, I'm not super concerned about getting
an "official" score that's higher than 85/98, I just wanted to be sure
there was no core problem and the test harness wasn't giving me a higher
score than I deserved :)

I've frozen the library:
http://www.w3.org/2006/07/SWD/RDFa/impl/js/20080817/

You can re-create the EARL report here:
http://torrez.us/code/rdfa-js/tests/

by using the following RDFa test runner:
http://www.w3.org/2006/07/SWD/RDFa/impl/js/20080817/rdfa-testrunner.js

The parser itself is at:
http://www.w3.org/2006/07/SWD/RDFa/impl/js/20080817/rdfa.js


And of course I'm attaching the EARL report if you don't want to do it
yourself. It's worth noting that this applies in *Firefox only* for now.
Other browsers should be close behind, but I'll test them another time.

-Ben