Toshi, I dont think the problem is the namespace, since stub.getHeaders() returns 0 size array in axis1.1... so basically somehow, the headers are not in stub anymore in axis1.1. do you have more pointers where problem may be? thanks Nan
> -----Original Message----- > From: Toshiyuki Kimura [mailto:[EMAIL PROTECTED] > Sent: Tuesday, October 14, 2003 8:36 PM > To: [EMAIL PROTECTED] > Subject: RE: Stub.getHeader(namespace, partname) in axis1.1 > > > Nan, > > Thank you for sending the response message. > I think this is not an intentional behavior. The stub.getHeader() > has never changed between 1.1 RC2 and 1.1 final, and the code is > as follows; > > public SOAPHeaderElement getHeader(String namespace,String partName){ > for(int i=0;i<headers.size();i++) { > SOAPHeaderElement header = (SOAPHeaderElement)headers.get(i); > if(header.getNamespaceURI().equals(namespace) && > header.getName().equals(partName)) > return header; > } > return null; > } > > Please pay attention to the code - > "header.getNamespaceURI().equals(namespace)". > > If the value of header.getNamespaceURI() doesn't equal to > the namespace what you passed, you should get 'null'. > That's a namespace issue, as I guessed. > You're lucky. You have a chance to submit a patch, if you > can debug the code. Please. :-) > > -- > Toshi <[EMAIL PROTECTED]> > > -----Original Message----- > From: Nan Xiong [mailto:[EMAIL PROTECTED] > Sent: Wednesday, October 15, 2003 11:12 AM > To: [EMAIL PROTECTED] > Subject: RE: Stub.getHeader(namespace, partname) in axis1.1 > > Toshi, > > Thanks for your reply. I'm usng .NET 1.1 (2003). > the soap response from .NET is: > > <?xml version="1.0" encoding="utf-8"?> > <soap:Envelope > xmlns:soap="http://schemas.xmlsoap.org/soap/envelope/" > xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" > xmlns:xsd="http://www.w3.org/2001/XMLSchema"> > <soap:Header> > <SessionTokenHeader > xmlns="http://www.viewstar.com/webservices/2002/11"> > <Token> > VE9WXV5HS0zigKZd4oChX05hXMWSXmVswpBWWkbigJ1JSUxMTU9TUV > NSWFdFWlhiYmNlXV/CqmBlwq1kY2NoamZpaWp0XG5weXXCunx3wqnD > geKEosuGwrvCvcKzwrhsw4DCssK8wrHCv8K7w4XDiMOHw4rDhg== > </Token> > </SessionTokenHeader> > </soap:Header> > <soap:Body> > <LoginResponse > xmlns="http://www.viewstar.com/webservices/2002/11"> > <LoginResult>true</LoginResult> > </LoginResponse> > </soap:Body> > </soap:Envelope> > > I'm accessing header from Stub as following: > > stub.getHeader("http://www.viewstar.com/webservices/2002/11", > "SessionTokenHeader"); > > the above call returns null in axis1.1, but returns correctly in axis > 1.1 RC2, against the same .NET service. > > I'm able to work around it using client side handler, but I would like > to know whether this is intentionally disabled? Will the soap handler > be a perfered way in terms of compliance to jaxrpc spec. > > Thanks > Nan > > > -----Original Message----- > > From: Toshiyuki Kimura [mailto:[EMAIL PROTECTED] > > Sent: Tuesday, October 14, 2003 6:49 PM > > To: [EMAIL PROTECTED] > > Subject: Re: Stub.getHeader(namespace, partname) in axis1.1 > > > > Nan, > > > > I guess that it might be related to a change of handling > > for namespace between RC2 and 1.1 final. A couple of days > > before, I've met an interop issue - using Axis 1.1 final as > > the server and .Net(InfoPath 2003) as the client. > > > > Did you snoop the messages ? What's the version of .Net ? > > Could you post your request/response messages and its WSDL > > files for both using 1.1 RC2 and 1.1 final ? > > > > -- > > Toshi <[EMAIL PROTECTED]> > > > > -----Original Message----- > > From: Nan Xiong [mailto:[EMAIL PROTECTED] > > Sent: Wednesday, October 15, 2003 5:46 AM > > To: [EMAIL PROTECTED] > > Subject: Stub.getHeader(namespace, partname) in axis1.1 > > > > Hello, > > > > I have a .NET web services that passes session id (a string) > > in the soap header, we were able to use Stub.getHeader(...) > > and Stub.setHeader() to make axis client talking to the .NET > > services using axis1.1 RC2, however, in axis1.1 final release, > > getHeader returns null. Anybody has the same problem? > > Can someone confirm this is a bug? or intentionally disabled? > > Whats the alternatives to access the header? (client handler?) > > > > Thanks in advance > > Nan > >
