Fix param* decoding

Author: mnot

lib/response_analyse.py

@@ -280,7 +280,7 @@ def _parse_params(self, name, instr):
                     self.setMessage(name, rs.PARAM_STAR_QUOTED, param=k)
                     v = self.unquoteString(v)
                 try:
-                    enc, lang, esc_v = param_dict["filename*"].split("'", 3)
+                    enc, lang, esc_v = v.split("'", 3)
                 except ValueError:
                     self.setMessage(name, rs.PARAM_STAR_ERROR, param=k)
                     continue
@@ -293,8 +293,7 @@ def _parse_params(self, name, instr):
                         enc=enc
                     )
                 # TODO: catch unquoting errors, range of chars, charset
-                decoded_v = urllib.unquote(esc_v)
-                param_dict[k.lower()] = decoded_v
+                param_dict[k.lower()] = urllib.unquote(esc_v).decode('utf-8')
             else:
                 param_dict[k.lower()] = self._unquoteString(v)
         return param_dict
@@ -708,8 +707,6 @@ def status205(self):        # Reset Content
     def status206(self):        # Partial Content
         if not "range" in nbhttp.header_dict(self.red.req_hdrs).keys():
             self.setMessage('', rs.PARTIAL_NOT_REQUESTED)
-        if not self.red.parsed_hdrs.has_key('content-range'):
-            print self.red.parsed_hdrs.keys()
             self.setMessage('header-location', rs.PARTIAL_WITHOUT_RANGE)
     def status207(self):        # Multi-Status
         pass

test/test_response_analyse.py

@@ -1,4 +1,6 @@
 #!/usr/bin/env python
+# coding=UTF-8
+
 
 import sys
 import unittest
@@ -29,35 +31,49 @@ def parseHeader(self, name, values):
         return getattr(self.parser, name_token)(name, values)
 
     def test_content_disposition(self):
+        i = 0
         for (hdrs, expected_val, expected_msgs) in [
-            # quoted-string
+            # 0: quoted-string
             (['attachment; filename="foo.txt"'], 
              ('attachment', {'filename': 'foo.txt'}),
              []
             ),
-            # token
+            # 1: token
             (['attachment; filename=foo.txt'], 
              ('attachment', {'filename': 'foo.txt'}),
              []
             ),
-            # inline
+            # 2: inline
             (['inline; filename=foo.txt'], 
              ('inline', {'filename': 'foo.txt'}),
              []
             ),
-            # token
+            # 3: token
             (['attachment; filename=foo.txt, inline; filename=bar.txt'], 
              ('inline', {'filename': 'bar.txt'}),
              [rs.SINGLE_HEADER_REPEAT]
             ),
+            # 4: filename*
+            (["attachment; filename=foo.txt; filename*=UTF-8''a%cc%88.txt"],
+             ('attachment', {
+                'filename': 'foo.txt', 
+                'filename*': u'a\u0308.txt'
+             }),
+             []
+            ),
         ]:
+            self.red.__init__()
             val = self.parseHeader('Content-Disposition', hdrs)
-            self.assertEqual(expected_val, val)
+            self.assertEqual(expected_val, val, 
+                "[%s] %s != %s" % (i, str(expected_val), str(val)))
             diff = set(
                 [n.__name__ for n in expected_msgs]).symmetric_difference(
                 set(self.red.msg_classes)
             )
-            self.assertEqual(len(diff), 0, diff)
+            self.assertEqual(len(diff), 0, 
+                "[%s] Mismatched messages: %s" % (i, diff)
+            )
+            i += 1
 
 
 if __name__ == "__main__":