Fix url parsing for S3 (#235)

piskvorky · Sep 24, 2018 · efb74ce · efb74ce
1 parent a971d06
commit efb74ce
Show file tree

Hide file tree

Showing 2 changed files with 11 additions and 2 deletions.
diff --git a/smart_open/smart_open_lib.py b/smart_open/smart_open_lib.py
@@ -84,7 +84,7 @@
 
 
 Uri = collections.namedtuple(
-    'Uri', 
+    'Uri',
     (
         'scheme',
         'uri_path',
@@ -488,7 +488,7 @@ def _parse_uri_s3x(parsed_uri):
     try:
         uri = parsed_uri.netloc + parsed_uri.path
         # Separate authentication from URI if exist
-        if ':' in uri.split('@')[0]:
+        if ':' in uri.split('@')[0] and '@' in uri:
             auth, uri = uri.split('@', 1)
             access_id, access_secret = auth.split(':')
         else:

diff --git a/smart_open/tests/test_smart_open.py b/smart_open/tests/test_smart_open.py
@@ -128,6 +128,15 @@ def test_uri_from_issue_223_works(self):
         self.assertEqual(parsed_uri.access_id, "")
         self.assertEqual(parsed_uri.access_secret, "")
 
+    def test_s3_uri_with_colon_in_key_name(self):
+        """ Correctly parse the s3 url if there is a colon in the key or dir """
+        parsed_uri = smart_open_lib._parse_uri("s3://mybucket/mydir/my:key")
+        self.assertEqual(parsed_uri.scheme, "s3")
+        self.assertEqual(parsed_uri.bucket_id, "mybucket")
+        self.assertEqual(parsed_uri.key_id, "mydir/my:key")
+        self.assertEqual(parsed_uri.access_id, None)
+        self.assertEqual(parsed_uri.access_secret, None)
+
 
 class SmartOpenHttpTest(unittest.TestCase):
     """