FileResponse has wrong Content-Type header

[wosc]

>>> import pyramid.response
>>> response = pyramid.response.FileResponse('/dev/null', content_type='image/jpeg')
>>> response.headers['Content-Type']
'image/jpeg; charset=UTF-8'

I guess the reason is that FileResponse calls its super __init__ first, which applies the default charset, and since the webob.Response.content_type property preserves parameters if the new value does not contain any, the charset doesn't go away when FileResponse sets the proper content type later on.

[wosc]

Hurm. This is actually a bug in webob, it does not apply only to FileResponses, but to normal Responses, too.
I've opened a bug there (Pylons/webob#130), closing here.

[dobesv]

See here for a workaround / solution: #1251