On Tue, Aug 30, 2005 at 02:25:37PM +0000, [EMAIL PROTECTED] wrote:
> Hamish's point seems to be that these two cases don't lead to a
> conclusion about what charset Quixote should use as a default for
> text/* since it can (and does?) specify a charset parameter.

No, it cannot.  The case we are talking about is when
HTTPResponse.charset is None.  In that case the Content-Type header
does not include a charset parameter.  _encode_chunk cannot decide
on some other charset.  First, if the user specifies 'charset' as
None then Quixote should not change it to something else.  Second,
the Content-Type header may have already been sent and therefore
changing the .charset attribute would have no effect.

Can we let this issue die already?  I'm not going to dicuss it
further unless someone can provide a patch implementing their
prposed behavior.

  Neil
_______________________________________________
Quixote-users mailing list
[email protected]
http://mail.mems-exchange.org/mailman/listinfo/quixote-users

Reply via email to