@@ -987,6 +987,12 @@ def test_urlsplit_normalization(self):

self.assertIn('\u2100', denorm_chars)

self.assertIn('\uFF03', denorm_chars)

# bpo-36742: Verify port separators are ignored when they

# existed prior to decomposition

urllib.parse.urlsplit('http://\u30d5\u309a:80')

with self.assertRaises(ValueError):

urllib.parse.urlsplit('http://\u30d5\u309a\ufe1380')

for scheme in ["http", "https", "ftp"]:

for c in denorm_chars:

url = "{}://netloc{}false.netloc/path".format(scheme, c)

@@ -333,13 +333,16 @@ def _checknetloc(netloc):

# looking for characters like \u2100 that expand to 'a/c'

# IDNA uses NFKC equivalence, so normalize for this check

import unicodedata

netloc2 = unicodedata.normalize('NFKC', netloc)

if netloc == netloc2:

n = netloc.rpartition('@')[2] # ignore anything to the left of '@'

n = n.replace(':', '') # ignore characters already included

n = n.replace('#', '') # but not the surrounding text

n = n.replace('?', '')

netloc2 = unicodedata.normalize('NFKC', n)

if n == netloc2:

return

_, _, netloc = netloc.rpartition('@') # anything to the left of '@' is okay

for c in '/?#@:':

if c in netloc2:

raise ValueError("netloc '" + netloc2 + "' contains invalid " +

raise ValueError("netloc '" + netloc + "' contains invalid " +

"characters under NFKC normalization")

def urlsplit(url, scheme='', allow_fragments=True):

@@ -0,0 +1 @@

Fixes mishandling of pre-normalization characters in urlsplit().