fix https://github.com/Pylons/webob/issues/18 (non-ascii form POST, then parse, then serialize)

3 files changed, 18 insertions, 5 deletions

diff --git a/tests/test_request_nose.py b/tests/test_request_nose.py
index cb0665b..6f0026c 100644
--- a/tests/test_request_nose.py
+++ b/tests/test_request_nose.py
@@ -180,3 +180,12 @@ def test_json_body_alternate_charset():
 def test_json_body_GET_request():
     request = Request.blank('/')
     assert_raises(ValueError, getattr, request, 'json_body')
+
+def test_non_ascii_body_params():
+    body = 'test=%D1%82%D0%B5%D1%81%D1%82'
+    req = Request.blank('/', POST=body)
+    # acessing params parses request body
+    req.params
+    # accessing body again makes the POST dict serialize again
+    # make sure it can handle the non-ascii characters in the query
+    eq(req.body, body)
diff --git a/webob/compat.py b/webob/compat.py
index 7cda0c4..6a87516 100644
--- a/webob/compat.py
+++ b/webob/compat.py
@@ -52,10 +52,11 @@ if PY3: # pragma: no cover
     from urllib import parse
     urlparse = parse
     from urllib.parse import quote as url_quote
-    from urllib.parse import urlencode as url_encode
+    from urllib.parse import urlencode as url_encode, quote_plus
     from urllib.request import urlopen as url_open
 else:
     import urlparse
+    from urllib import quote_plus
     from urllib import quote as url_quote
     from urllib import unquote as url_unquote
     from urllib import urlencode as url_encode
diff --git a/webob/request.py b/webob/request.py
index 27c3480..8e65506 100644
--- a/webob/request.py
+++ b/webob/request.py
@@ -33,6 +33,7 @@ from webob.compat import (
     url_encode,
     url_quote,
     url_unquote,
+    quote_plus,
     urlparse,
     )
 
@@ -373,7 +374,7 @@ class BaseRequest(object):
                     port = '80'
         else:
             port = e['SERVER_PORT']
-        return port    
+        return port
 
     @property
     def host_url(self):
@@ -1454,9 +1455,11 @@ class FakeCGIBody(io.RawIOBase):
         if self.file is None:
             if self.content_type.startswith(
                 'application/x-www-form-urlencoded'):
-                # TODO: check if bytes_ is necessary
-                data = bytes_(url_encode(self.vars), 'utf8')
-                self.file = io.BytesIO(data)
+                data = '&'.join(
+                    '%s=%s' % (quote_plus(bytes_(k, 'utf8')), quote_plus(bytes_(v, 'utf8')))
+                    for k,v in self.vars.items()
+                )
+                self.file = io.BytesIO(bytes_(data))
             elif self.content_type.startswith('multipart/form-data'):
                 self.file = _encode_multipart(
                     self.vars.items(),