not-wf-not-sa-005 again

I am researching the reason why not-wf-not-sa-005 is classified under 
"not-wf".

I have seen the previous discussion on

  http://lists.w3.org/Archives/Public/public-xml-testsuite/2004Sep/0002.html

which refers to

  http://lists.w3.org/Archives/Public/public-xml-testsuite/2002Jun/0018.html

but the explanation given there does not seem to be accurate -- the test in 
question does not have anything to do with attributes.

That message says to refer to 

  http://www.w3.org/XML/Group/2002/02/xml10-test-suite-issues

for further info. The referenced document also happens to be linked to from

 http://www.w3.org/XML/Test/

but the document is not accessible to the general public. It requires a W3C 
username and password to access.

Perhaps in that now-inaccessible document there is an explanation of why the
test is filed under not-wf, but if not, I will restate the source of confusion 
on this one.

According to http://www.w3.org/XML/Test/xmlconf-20031210.html#not-wf, the test 
"Tests the Entity Declared VC by referring to an undefined parameter entity 
within an external entity."

This is the test:

xmlconf/xmltest/not-wf/not-sa/005.ent
=====================================
<!ELEMENT doc (#PCDATA)>
%e;

xmlconf/xmltest/not-wf/not-sa/005.xml
=====================================
<!DOCTYPE doc SYSTEM "005.ent">
<doc></doc>

It is true that the files violate the Entity Declared VC, but I do wonder why 
it is filed under "not-wf". How is this document not well-formed?

If I understand correctly (and I'm not sure that I do), a nonvalidating parser 
that elects to read external entities should not have a problem with this 
document, because standalone="no" (implicitly). The processor would just have 
to stop processing 005.ent when it encounters the reference to parameter 
entity 'e', which it will not (cannot) read. So this example doesn't violate 
the Entity Declared *WFC*. I also don't see that it violates the general 
grammar.

Comments?

Thanks,
Mike

Received on Monday, 14 March 2005 05:14:51 UTC