Passing a unicode object to request.write corrupts the entire response

New submission from radix <[email protected]>:

I passed a unicode object to request.write, and my HTTP response came out like this:

'H\x00\x00\x00T\x00\x00\x00T\x00\x00\x00P\x00\x00\x00/\x00\x00\x001\x00\x00\x00.\x00\x00\x001\x00\x00\x00'

and so on. (That's the string "HTTP/1.1" encoded in, apparently, UCS-4).

idnar points out that this may be because of cStringIO usage:

{{{
<idnar> >>> StringIO(u'abc').getvalue()
<idnar> 'a\x00\x00\x00b\x00\x00\x00c\x00\x00\x00'
}}}

In my opinion, either an exception should be immediately raised when unicode is passed to request.write, or the data should be encoded with the response's encoding, if one has been set.

----------
Type     : defect
Component: web
Keywords : 
Priority : normal
Nosy     : 
----------
http://twistedmatrix.com/trac/ticket/3896