[PATCH] WebOb should leave req.body as is after parsing POST body

[wangchun]

For an empty Content-Type, req.body returns empty string (v0.9.6) or raises an error (trunk) after req.POST is called.

For Content-Type == application/x-www-form-urlencoded:

import webob
from StringIO import StringIO

body = 'HELLO, WORLD!'

environ = {
    'REQUEST_METHOD': 'POST',
    'CONTENT_TYPE': 'application/x-www-form-urlencoded',
    'CONTENT_LENGTH': str(len(body)),
    'wsgi.input': StringIO(body),
}

req = webob.Request(environ)
print req.body
print req.POST
print req.body

macbook:~/tmp$ python test.py
HELLO, WORLD!
MultiDict([('HELLO, WORLD!', '')])
HELLO%2C+WORLD%21=
macbook:~$ 

[wangchun]

also for req.body_file

[maluke]

• accessing .POST ends up normalizing the .body, so that will not be changed
• if content_type is not form-encoded, .POST does not corrupt .body in any way. if it does for you, please add a test case.