It depends what the WSDL says...
If the WSDL defines the type as
http://www.something.com/schema/20050101}QueueName and the application
instead sends xsd:string, then Axis is *required* to send the xsi:type
information.
If the WSDL defines the type as xsd:string, then it might be a bug.
Anne
On 6/29/05, Jim Cross <[EMAIL PROTECTED]> wrote:
>
>
>
> Workaround found:
>
> The generated proxy code contained this line (word "something" was
> substituted for the real name):
>
> elemField.setXmlType(new
> javax.xml.namespace.QName("http://www.something.com/schema/20050101";,
> "QueueName"));
>
> When it was changed to this:
>
> elemField.setXmlType(new
> javax.xml.namespace.QName("http://www.w3.org/2001/XMLSchema";,
> "string"));
>
> It works correctly and does not send the xsi:type.
>
> Is this a bug in the wsdl2java, a bug in the serialization, or a problem
> with the schema that was used to generate to the proxy?
>
>
>
> -----Original Message-----
> From: Jim Cross
> Sent: Tuesday, June 28, 2005 2:15 PM
> To: '[email protected]'
> Subject: Client insists on sending xsi type
>
> I'm assessing a server from the client side. I'm not sure what the server is
> written with. I've tried this with 1.1, 1.2, and 1.2.1.
>
> The generated proxies have this statement before the service request.
>
>
> _call.setProperty(org.apache.axis.client.Call.SEND_TYPE_ATTR,
> Boolean.FALSE);
>
> I can also confirm that
>
> MessageContext property name=sendXsiTypes value=false
>
> Nevertheless, with a particular string array, the client insists on
> generating
>
> xsi:type="xsd:string" on the element
>
> Removing the xsi:type and sending the SOAP XML via a raw post to the server
> works, so it is definite that this is problem.
>
> As far as I tell from some limited debugging, the code gets deep into its
> serialization and compares the xmlType with an actualType (not sure where or
> how this is created or what purpose it serves). If these are different, it
> seems to generate the xsi:type no matter what the property setting is.
>
> Any idea how to work around this?
>
