Implementor experience: two comments on tests and documentation

Hi all,

As a side effect of a current project, and inspired by the lack of a  
pure-java implementation, I just wrote an RDFa parser java-rdfa. [1]
I reckon it's just about the worst coded, although it passes the  
(accepted) tests 100% (but see below).

Syntax and processing doc [2] and tests [3] have been invaluable, and  
have been clear enough for me to bumble through. I thought I'd mention  
two issues that tripped me up, however:

1. [2] 5.4 (Sequence)

Step 6 begins: "If in any of the previous steps a [new subject] was  
set to a non-null value..." and ends:
"Note that none of this block is executed if there is no [new subject]  
value, i.e., [new subject] remains null."

So it could be null at this point.

Step 7 begins: "If in any of the previous steps a [current object  
resource] was set to a non-null value, it is now used to generate  
triples:"

But no mention of what happens if [new subject] is null.

I think at this point [new subject] should be set to [parent subject]  
if null, but I confess I may have missed something.

2. XMLLiteral tests 101, 102, 103. [4]

XMLLiteral equivalence is hard, of course. However these three tests  
swap the attribute order of:

svg:width="200" svg:height="100"

For a streaming parser, which gets things in document order, that's a  
bit annoying.

Anyway, minor gripes. Thanks to the editors,

Damian Steer

[1] <http://github.com/shellac/java-rdfa/tree/master>
[2] <http://www.w3.org/TR/rdfa-syntax/>
[3] <http://www.w3.org/2006/07/SWD/RDFa/testsuite/xhtml1-testcases/rdfa-xhtml1-test-manifest.rdf 
 >
[4] <http://www.w3.org/2006/07/SWD/RDFa/testsuite/xhtml1-testcases/010[1,2,3].sparql 
 >

Received on Tuesday, 30 June 2009 20:46:47 UTC