aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
-rw-r--r--lib-python/3/test/test_urlparse.py18
-rw-r--r--lib-python/3/urllib/parse.py2
2 files changed, 19 insertions, 1 deletions
diff --git a/lib-python/3/test/test_urlparse.py b/lib-python/3/test/test_urlparse.py
index 574da5bd69..6528ea2841 100644
--- a/lib-python/3/test/test_urlparse.py
+++ b/lib-python/3/test/test_urlparse.py
@@ -724,6 +724,24 @@ class UrlParseTestCase(unittest.TestCase):
with self.assertRaises(ValueError):
p.port
+ def test_attributes_bad_scheme(self):
+ """Check handling of invalid schemes."""
+ for bytes in (False, True):
+ for parse in (urllib.parse.urlsplit, urllib.parse.urlparse):
+ for scheme in (".", "+", "-", "0", "http&", "६http"):
+ with self.subTest(bytes=bytes, parse=parse, scheme=scheme):
+ url = scheme + "://www.example.net"
+ if bytes:
+ if url.isascii():
+ url = url.encode("ascii")
+ else:
+ continue
+ p = parse(url)
+ if bytes:
+ self.assertEqual(p.scheme, b"")
+ else:
+ self.assertEqual(p.scheme, "")
+
def test_attributes_without_netloc(self):
# This example is straight from RFC 3261. It looks like it
# should allow the username, hostname, and port to be filled
diff --git a/lib-python/3/urllib/parse.py b/lib-python/3/urllib/parse.py
index 5b7193f67c..495ac42d90 100644
--- a/lib-python/3/urllib/parse.py
+++ b/lib-python/3/urllib/parse.py
@@ -482,7 +482,7 @@ def urlsplit(url, scheme='', allow_fragments=True):
clear_cache()
netloc = query = fragment = ''
i = url.find(':')
- if i > 0:
+ if i > 0 and url[0].isascii() and url[0].isalpha():
for c in url[:i]:
if c not in scheme_chars:
break