RE: key attribute missing in generated dtd

"Jeroen van Rotterdam" <[email protected]>
Newsgroups gmane.comp.web.dom.test-suites.general
Message-ID <[email protected]>
Thanks for the response,


> The listing of DocumentLS and DOMBuilder as potential values for 
> interface would suggest that you generated the DTD from a draft where 
> both interfaces had an "abort" method (which isn't the case 
> in the last 
> public draft).

That's correct I'm working with the latest (non-public) working draft. 
There are obvious reasons for that, we have an implementation of this
latest draft and
I don't want to rewrite my tests after a new publication.

> 
> Neither dom-to-dtd.xsl or dom-to-xsd.xsl is sufficiently 
> smart enough to 
> fabricate a common element definition from conflicting definitions of 
> the same method name.  It generates the element from one 
> (probably first 
> in document order in the combined spec) and assumes that any 
> other use 
> will be compatible.  When that isn't the case, you have to 
> suppress the 
> automatic generation and provide a hand-crafted element.  The 
> exclusion 
> of 'load' on line 240 of dom-to-dtd.xsl is an instance of 
> this effect. 
> (Related is a fixup in the dom?-combine.xsl that make the 
> argument names 
> of Element.getElementsByTagName and Document.getElementsByTagName 
> identical).
> 
> You might review whether both "abort" methods are needed or whether 
> their arguments could be synchronized.

Both are needed with different arguments

> 
> If you explicitly specify the interface attribute, the 
> test-to-xxx.xsl 
> transforms should not have a problem.

Ok, I will have a look at it.


jeroen

=====
X-Hive Corporation
Jeroen van Rotterdam, CEO
e-mail: [email protected]
phone: +31 10 2818080
http://www.x-hive.com
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.