Mailing-List: contact axis-dev-help@ws.apache.org; run by ezmlm
Precedence: bulk
Reply-To: axis-dev@ws.apache.org
Date: 29 Jul 2003 12:33:04 -0000
Message-ID: <20030729123304.29331.qmail@nagoya.betaversion.org>
From: bugzilla@apache.org
To: axis-dev@ws.apache.org
Cc: 
Subject: DO NOT REPLY [Bug 20109]  -
    WSDL2Java "soapenc" instead of "xsd" for boolean and long

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=20109>.
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=20109

WSDL2Java "soapenc" instead of "xsd" for boolean and long


------- Additional Comments From cchabanois@cognicase.fr  2003-07-29 12:33 -------
The problem comes from the getNillableQName in
org.apache.axis.wsdl.symbolTable.Utils.java.

Here is the comments for this method :
"
getNillableQName returns the QName to use if the nillable=true attribute is
used.                             
For example, in JAX-RPC:
The QName "xsd:int" maps to a java int.
However if an element with a type="xsd:int" also has the "nillable=true"
attribute, the type should be an Integer (not an int).
So in these circumstances, this routine is called with xsd:int to get a suitable
qname (soapenc:int) which maps to Integer.
"

I think too that it is not the right thing to do. (Note that the problem is
probably present too when minOccurs=0 and maxOccurs=1)

Moreover there is no way to know if an element can be nillable. In ElementDecl,
there is an attribute minOccursIs0 but no isNillable attribute. And for simple
types, type.getNode() is null so there is really no way to know if a xsd:string
is nillable or not.