|
|
1b92cc7 |
diff --git a/Lib/test/test_urlparse.py b/Lib/test/test_urlparse.py
|
|
|
1b92cc7 |
index 0faf2bb..d0365ec 100644
|
|
|
1b92cc7 |
--- a/Lib/test/test_urlparse.py
|
|
|
1b92cc7 |
+++ b/Lib/test/test_urlparse.py
|
|
|
1b92cc7 |
@@ -1011,6 +1011,12 @@ class UrlParseTestCase(unittest.TestCase):
|
|
|
1b92cc7 |
self.assertIn('\u2100', denorm_chars)
|
|
|
1b92cc7 |
self.assertIn('\uFF03', denorm_chars)
|
|
|
1b92cc7 |
|
|
|
1b92cc7 |
+ # bpo-36742: Verify port separators are ignored when they
|
|
|
1b92cc7 |
+ # existed prior to decomposition
|
|
|
1b92cc7 |
+ urllib.parse.urlsplit('http://\u30d5\u309a:80')
|
|
|
1b92cc7 |
+ with self.assertRaises(ValueError):
|
|
|
1b92cc7 |
+ urllib.parse.urlsplit('http://\u30d5\u309a\ufe1380')
|
|
|
1b92cc7 |
+
|
|
|
1b92cc7 |
for scheme in ["http", "https", "ftp"]:
|
|
|
1b92cc7 |
for c in denorm_chars:
|
|
|
1b92cc7 |
url = "{}://netloc{}false.netloc/path".format(scheme, c)
|
|
|
1b92cc7 |
diff --git a/Lib/urllib/parse.py b/Lib/urllib/parse.py
|
|
|
1b92cc7 |
index 8b6c9b1..e2f7b69 100644
|
|
|
1b92cc7 |
--- a/Lib/urllib/parse.py
|
|
|
1b92cc7 |
+++ b/Lib/urllib/parse.py
|
|
|
1b92cc7 |
@@ -402,13 +402,16 @@ def _checknetloc(netloc):
|
|
|
1b92cc7 |
# looking for characters like \u2100 that expand to 'a/c'
|
|
|
1b92cc7 |
# IDNA uses NFKC equivalence, so normalize for this check
|
|
|
1b92cc7 |
import unicodedata
|
|
|
1b92cc7 |
- netloc2 = unicodedata.normalize('NFKC', netloc)
|
|
|
1b92cc7 |
- if netloc == netloc2:
|
|
|
1b92cc7 |
+ n = netloc.rpartition('@')[2] # ignore anything to the left of '@'
|
|
|
1b92cc7 |
+ n = n.replace(':', '') # ignore characters already included
|
|
|
1b92cc7 |
+ n = n.replace('#', '') # but not the surrounding text
|
|
|
1b92cc7 |
+ n = n.replace('?', '')
|
|
|
1b92cc7 |
+ netloc2 = unicodedata.normalize('NFKC', n)
|
|
|
1b92cc7 |
+ if n == netloc2:
|
|
|
1b92cc7 |
return
|
|
|
1b92cc7 |
- _, _, netloc = netloc.rpartition('@') # anything to the left of '@' is okay
|
|
|
1b92cc7 |
for c in '/?#@:':
|
|
|
1b92cc7 |
if c in netloc2:
|
|
|
1b92cc7 |
- raise ValueError("netloc '" + netloc2 + "' contains invalid " +
|
|
|
1b92cc7 |
+ raise ValueError("netloc '" + netloc + "' contains invalid " +
|
|
|
1b92cc7 |
"characters under NFKC normalization")
|
|
|
1b92cc7 |
|
|
|
1b92cc7 |
def urlsplit(url, scheme='', allow_fragments=True):
|