DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
<http://nagoya.apache.org/bugzilla/show_bug.cgi?id=15533>.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND
INSERTED IN THE BUG DATABASE.
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=15533
Java types not generated if opertion name matches element name
butek@us.ibm.com changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|REOPENED |RESOLVED
Resolution| |WONTFIX
------- Additional Comments From butek@us.ibm.com 2002-12-19 16:59 -------
This is working as design and I don't think any of us intend to 'fix' it. The
reason we have such an odd mode is because most (all?) .NET WSDL is generated
with this pattern. If we did not unwrap, then for all WSDL coming from .NET
we'd end up with 2 beans for every operation, which looks horrible.
Let me answer your two questions...
"i) You have different generation behaviour depending on the operation names.
This seems fragile to me. Wouldn't it be better to always do things one way and
then provide a flag or flags to to 'wrapped' or unwrapped"
No. JAX-RPC dictates that we recognized this pattern.
"ii) If I mix and match styles in the WSDL then I get compile errors, then for
the types I wan't 'wrapped' I don't get any types generated. In a simple case
the porttype class has two methods one that takes two doubles and returns a
diuble (unwrapped?) and one that takes an AddRequestType and returns an
AddResponseType (wrapped), as defined by the WSDL. However the AddResponseType
and AddRequestType Java classes aren't being generated. "
Our original intent was to work in mix-and-match mode, but we haven't been
diligent in preserving that, so it doesn't surprise me that it might not work.
HOWEVER, it is common practice to follow ONE mode throughout a WSDL file. In
fact, I believe WS-I is advocating that thou shalt not mix-and-match.
If you still think we should fix the mix-and-match failure, feel free to reopen
this bug, but don't expect any action on this soon (not from me, anyway - Tom?).
My personal position is that if you mix-and-match, you should use the
--noWrapped flag.
|