Re: Converting from UTF-8 to ISO-8859-I with Axis2 and Java

[Bruno Negrao]

On 11/1/06, Davanum Srinivas <[EMAIL PROTECTED]> wrote:

Content-Type: text/html; ??? what brain-dead soap server are you going
against? :)

The problem is supposed to be the soap server developers, not the specific technology in itself...

Yes, if we can't find the charset, there is no way we will know that
it is ISO-8859-1

Davanum, now carefully revise the example of an answer from my SOAP server. I put in bold some interesting NOTE's things I've noticed:

HTTP/1.0 200 OK
Content-length: 1269
                                  <== NOTE 1*
<?xml version="1.0" encoding="iso-8859-1"?>  <== NOTE 2*
   <soap:Envelope xmlns:soap="http://schemas.xmlsoap.org/soap/envelope/">   
      <soap:Body>       
         <AgendaPesquisa>           
            <status>0</status>           
            <ListaContatosPesquisa>               
               <tipo>C</tipo>               
               <dono>lucia</dono>               
               <posicao>177</posicao>               
               <nome>Abricó Gimarães</nome>       <==NOTE 3*
               <email></email>           
            </ListaContatosPesquisa>       
         </AgendaPesquisa>   
      </soap:Body>
   </soap:Envelope>

Note 1: the Content-Type header is missing.
Note 2: But the xml definition contains "encoding=iso-8859-1". Should not this suffice for axis2 to discover the proper character encoding?
Note 3: the SOAP server is indeed sending the ISO-8859-1 strings correctly enconded. The problem  is when I decode them using axis2.

Thank you,
bruno