[XML-SIG] WSDL Test Suite?

[Mark Bucciarelli]

On Thursday 27 February 2003 11:53 am, M.-A. Lemburg wrote:
> Mark Bucciarelli wrote:
> > Looks like I've got soapy parsing WSDL using only native Python 2
> > libraries.
[...]
> Did you have to make any changes to soapy to get it up and
> running ?

A few minor ones.  

- element.getAttributeNS('', 'name') had to be replaced by 
getAttribute('name').  Did the Python API change?

- namespaces had to be added

- use xml.dom.minidom instead of a PyExpat.Reader()

- a couple for loops needed to be changed.  soap.py had stuff like this: 

	for attrns, attrkey in node.attributes.keys(): 

that I couldn't get to work.  Did the python API change?

When I test it a bit more, I'll post the updated version and try to contact 
Adam Elman, the author.

I'd like to better understand how to deal with all the various namespace 
URI's; for example, it looks like xsd has had at least three different URI's:

www.w3.org/1999/XMLSchema
www.w3.org/2000/10/XMLSchema
www.w3.org/2001/XMLSchema

There are similar multiples for the soap, enc, xsi prefixes.   From my 
reading, I would assume that a properly formed WSDL doc (and SOAP doc) will 
specify what URI to associate with each prefix.  If that's the case, I'm not 
sure why the URI's need to be hardcoded in soap.py (and schema.py for that 
matter).  I'll have to read through the code more carefully.

Mark