> On behalf of the MEMS Exchange, I am pleased to announce
> the release of Quixote 2.1.
> http://www.mems-exchange.org/software/quixote/

It still has this stupid function in http_response.py

    def _encode_chunk(self, chunk):
        """(chunk : str | unicode) -> str
        """
        if self.charset == 'iso-8859-1' and isinstance(chunk, str):
            return chunk # non-ASCII chars are okay
        else:
            return chunk.encode(self.charset)

so in the case of self.charset == 'utf8' and chunk is a 'str' instance,
you are calling chunk.encode ('utf8') on a str object.. which fails in
most cases you'd set self.charset to be utf8. I already proposed this
much better function some months ago:

    def _encode_chunk(self, chunk):
        """(chunk : str | unicode) -> str
        """
        if isinstance(chunk, unicode):
            return chunk.encode(self.charset)
        else:
            return chunk # non-ASCII chars are okay

..encode has real meaning on unicode objects

--
damjan | дамјан
This is my jabber ID --> damjan@bagra.net.mk <-- not my mail address!!!