i think that i have identified why so many implementations fail echoString
(iso-8859-1, cyr) test - it seems that underlying XML parser that you use
in SOAP::Lite is interpreting response message as UTF-8 (which is default
for XML 1.0) even though on transport level encoding was forced in charset
parameter of Content-Type ...
many implementation such as SoapRMI are answering with the same charset
that client requested (SOAP:Lite forces response to be UTF-8 it seems -so
it passes its won test...):
i could probably pass this test with SOAP:lite if i added <?xml
version="1.0" encoding="iso-8859-1"?> header but it should not be required
to do it IMHO.