Hi everyone,
I wish to extend rpclib with the possibility of generating a schema
with an "any" tag:
<any namespace="##other" processContents="lax" maxOccurs="unbounded"/>
I also want to document the process, so others will have a clue about
it.
What I think I have to do:
1. add a new class to the list of primitives
2. change the code that produces XML from a Python class
3. change the code that generates a Python object from XML
4. extend the tests, such that they include the newly created type
My questions are:
a. are there any other steps that I missed?
b. which class should I inherit from?
- AnyXml and AnyDict seem reasonable candidates, their names are
very attractive. However, I did not find any examples of how
they are used
- if my guess is correct, then this will be a simple matter of
creating a derived class and setting __namespace__ to '##other'.
This sounds too good to be true.
c. what are the functions that have to be overridden in the new class?
d. how to tell the parser about "processContents", "namespace" and the
behaviour specified by its different values?
- I think it can rely on validate_native, but my understanding is
that I must also define the behaviour of validate_native for this
type - I don't know where to do that
e. is there any documentation about this? Some kind of howto?
f. which other questions was I supposed to ask? :-)
Looking forward to your feedback.
_______________________________________________
Soap mailing list
[email protected]
http://mail.python.org/mailman/listinfo/soap
