Why is this JC test not-wf?

[roddey at us.ibm.com]

Ok, so I'm running through the James Clark tests with my new parser and I
don't exactly understand why not-wf\sa\081.xml is not well formed
necessarily. Here is the text:

   <!DOCTYPE doc [
   <!ENTITY e SYSTEM "nul">
   ]>
   <doc a="&e;"></doc>

So it defines an external entity 'e', which resolves to a file named 'nul',
which is an empty file. Then it references that entity as the value of the
attribute 'a'. To me, that seems perfectly fine. If could have legally
written manually:

     <doc a=""></doc>

and I think I could legally if not validating, then why would not the
result of the test file give exactly that result? And if it does give that
result, how is that less well formed than what I could have typed myself?
There is no prohibition against empty attribute values that I know of.



xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ and on CD-ROM/ISBN 981-02-3594-1
To (un)subscribe, mailto:majordomo at ic.ac.uk the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo at ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at ic.ac.uk)