Bruno,
You can compile and run the test i sent using just axiom jars and a
junit jar. All you need to do in that test is replace the strings with
your string and change UTF-16 to ISO-8859-1 and run the test. If it
works, then Axis2 can send and receive strings fine. So the problem
would be in the response that we get back from your server.
-- dims
Davanum, I compiled and run theses tests here and they worked fine. I changed what you said and I added a silly test:
public void testBrazilianPortuguese() throws Exception {
runtest("Sarará óóó ííí çuçuçu ãã õõ");
}
The testcase class I edited is attached.
I believe that this proves axis2 **is able** to encode and decode using different character sets when it **knows** what character set it should use.
The problem I'm experiencing appears to be that axis2 is ignoring that my server is sending iso-8859-1 encoded bytes, so axis2 decodes them as UTF-8.
As I pointed out in my previous e-mail, although my soap server does not set the Content-type http header on its responses, it indeed sets the encoding=iso-8859-1 xml attribute on its xml declaration. Axis2 could (somehow) use this infomation to define the character encoding to read from the network.
Thank you,
bruno
CharacterEncodingTest.java
Description: Binary data
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
