Merge 4d2e10f into 508604c

goodtables/utilities/helpers.py

@@ -6,6 +6,7 @@
 
 import os
 import io
+import re
 import json
 import inspect
 import requests
@@ -202,6 +203,7 @@ def validate_handler(handler, argcount=1):
         if not len(spec[0]) == argcount:
             raise exceptions.InvalidHandlerError
 
+
 def make_valid_url(url):
     """Make sure url doesn't contain unsupported characters
 
@@ -214,8 +216,13 @@ def make_valid_url(url):
         return (glue).join(quoted)
 
     scheme, netloc, path, query, fragment = compat.urlsplit(url)
-    quoted_path = compat.quote(path.encode('utf-8'))
-    quoted_query = compat.quote_plus(query.encode('utf-8'))
-    new_url_tuple = (scheme, netloc, quoted_path, quoted_query, fragment)
+    encoded_path = url_encode_non_ascii(path)
+    encoded_query = url_encode_non_ascii(query)
+    new_url_tuple = (scheme, netloc, encoded_path, encoded_query, fragment)
     quoted_url = compat.urlunsplit(new_url_tuple)
     return quoted_url
+
+
+def url_encode_non_ascii(text):
+    # http://stackoverflow.com/questions/4389572/how-to-fetch-a-non-ascii-url-with-python-urlopen
+    return re.sub('[\x80-\xFF]', lambda c: '%%%02x' % ord(c.group(0)), text)

tests/test_utilities.py

@@ -31,3 +31,7 @@ def test_make_valid_url(self):
               'data_url=http://data.defra.gov.uk/ops/government_procurement_card/over_£500_GPC_apr_2013.csv')
         assertion = '£' not in utilities.helpers.make_valid_url(url)
         self.assertTrue(assertion)
+
+    def test_make_valid_url_dont_break_query(self):
+        url = 'http://next.openspending.org/fdp-adapter/convert?url=https%3A%2F%2Fraw.githubusercontent.com%2Fkravets-levko%2Fdata%2Fmaster%2Ftest.xlsx.csv'
+        self.assertEqual(utilities.helpers.make_valid_url(url), url)