bpo-36216: Add check for characters in netloc that normalize to separ…

…ators (GH-12201)

Doc/library/urllib.parse.rst

@@ -124,6 +124,11 @@ or on combining URL components into a URL string.
    Unmatched square brackets in the :attr:`netloc` attribute will raise a
    :exc:`ValueError`.
 
+   Characters in the :attr:`netloc` attribute that decompose under NFKC
+   normalization (as used by the IDNA encoding) into any of ``/``, ``?``,
+   ``#``, ``@``, or ``:`` will raise a :exc:`ValueError`. If the URL is
+   decomposed before parsing, no error will be raised.
+
    .. versionchanged:: 3.2
       Added IPv6 URL parsing capabilities.
 
@@ -136,6 +141,10 @@ or on combining URL components into a URL string.
       Out-of-range port numbers now raise :exc:`ValueError`, instead of
       returning :const:`None`.
 
+   .. versionchanged:: 3.8
+      Characters that affect netloc parsing under NFKC normalization will
+      now raise :exc:`ValueError`.
+
 
 .. function:: parse_qs(qs, keep_blank_values=False, strict_parsing=False, encoding='utf-8', errors='replace', max_num_fields=None)
 
@@ -259,10 +268,19 @@ or on combining URL components into a URL string.
    Unmatched square brackets in the :attr:`netloc` attribute will raise a
    :exc:`ValueError`.
 
+   Characters in the :attr:`netloc` attribute that decompose under NFKC
+   normalization (as used by the IDNA encoding) into any of ``/``, ``?``,
+   ``#``, ``@``, or ``:`` will raise a :exc:`ValueError`. If the URL is
+   decomposed before parsing, no error will be raised.
+
    .. versionchanged:: 3.6
       Out-of-range port numbers now raise :exc:`ValueError`, instead of
       returning :const:`None`.
 
+   .. versionchanged:: 3.8
+      Characters that affect netloc parsing under NFKC normalization will
+      now raise :exc:`ValueError`.
+
 
 .. function:: urlunsplit(parts)
 

Lib/test/test_urlparse.py

@@ -1,3 +1,5 @@
+import sys
+import unicodedata
 import unittest
 import urllib.parse
 
@@ -994,6 +996,27 @@ def test_all(self):
                 expected.append(name)
         self.assertCountEqual(urllib.parse.__all__, expected)
 
+    def test_urlsplit_normalization(self):
+        # Certain characters should never occur in the netloc,
+        # including under normalization.
+        # Ensure that ALL of them are detected and cause an error
+        illegal_chars = '/:#?@'
+        hex_chars = {'{:04X}'.format(ord(c)) for c in illegal_chars}
+        denorm_chars = [
+            c for c in map(chr, range(128, sys.maxunicode))
+            if (hex_chars & set(unicodedata.decomposition(c).split()))
+            and c not in illegal_chars
+        ]
+        # Sanity check that we found at least one such character
+        self.assertIn('\u2100', denorm_chars)
+        self.assertIn('\uFF03', denorm_chars)
+
+        for scheme in ["http", "https", "ftp"]:
+            for c in denorm_chars:
+                url = "{}://netloc{}false.netloc/path".format(scheme, c)
+                with self.subTest(url=url, char='{:04X}'.format(ord(c))):
+                    with self.assertRaises(ValueError):
+                        urllib.parse.urlsplit(url)
 
 class Utility_Tests(unittest.TestCase):
     """Testcase to test the various utility functions in the urllib."""

Lib/urllib/parse.py

@@ -396,6 +396,21 @@ def _splitnetloc(url, start=0):
             delim = min(delim, wdelim)     # use earliest delim position
     return url[start:delim], url[delim:]   # return (domain, rest)
 
+def _checknetloc(netloc):
+    if not netloc or netloc.isascii():
+        return
+    # 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:
+        return
+    _, _, netloc = netloc.rpartition('@') # anything to the left of '@' is okay
+    for c in '/?#@:':
+        if c in netloc2:
+            raise ValueError("netloc '" + netloc2 + "' contains invalid " +
+                             "characters under NFKC normalization")
+
 def urlsplit(url, scheme='', allow_fragments=True):
     """Parse a URL into 5 components:
     <scheme>://<netloc>/<path>?<query>#<fragment>
@@ -424,6 +439,7 @@ def urlsplit(url, scheme='', allow_fragments=True):
                 url, fragment = url.split('#', 1)
             if '?' in url:
                 url, query = url.split('?', 1)
+            _checknetloc(netloc)
             v = SplitResult('http', netloc, url, query, fragment)
             _parse_cache[key] = v
             return _coerce_result(v)
@@ -447,6 +463,7 @@ def urlsplit(url, scheme='', allow_fragments=True):
         url, fragment = url.split('#', 1)
     if '?' in url:
         url, query = url.split('?', 1)
+    _checknetloc(netloc)
     v = SplitResult(scheme, netloc, url, query, fragment)
     _parse_cache[key] = v
     return _coerce_result(v)

Misc/NEWS.d/next/Security/2019-03-06-09-38-40.bpo-36216.6q1m4a.rst

@@ -0,0 +1,3 @@
+Changes urlsplit() to raise ValueError when the URL contains characters that
+decompose under IDNA encoding (NFKC-normalization) into characters that
+affect how the URL is parsed.