[3.8] bpo-43882 - urllib.parse should sanitize urls containing ASCII newline and tabs. (GH-25595) (#25726)

Co-authored-by: Gregory P. Smith <greg@krypto.org>
Co-authored-by: Serhiy Storchaka <storchaka@gmail.com>
(cherry picked from commit 76cd81d60310d65d01f9d7b48a8985d8ab89c8b4)
Co-authored-by: Senthil Kumaran <senthil@uthcode.com>
Co-authored-by: Senthil Kumaran <skumaran@gatech.edu>

4 files changed, 77 insertions, 0 deletions

diff --git a/Doc/library/urllib.parse.rst b/Doc/library/urllib.parse.rst
index fcad707..a6cfc5d 100644
--- a/Doc/library/urllib.parse.rst
+++ b/Doc/library/urllib.parse.rst
@@ -312,6 +312,9 @@ or on combining URL components into a URL string.
    ``#``, ``@``, or ``:`` will raise a :exc:`ValueError`. If the URL is
    decomposed before parsing, no error will be raised.
 
+   Following the `WHATWG spec`_ that updates RFC 3986, ASCII newline
+   ``\n``, ``\r`` and tab ``\t`` characters are stripped from the URL.
+
    .. versionchanged:: 3.6
       Out-of-range port numbers now raise :exc:`ValueError`, instead of
       returning :const:`None`.
@@ -320,6 +323,10 @@ or on combining URL components into a URL string.
       Characters that affect netloc parsing under NFKC normalization will
       now raise :exc:`ValueError`.
 
+   .. versionchanged:: 3.8.10
+      ASCII newline and tab characters are stripped from the URL.
+
+.. _WHATWG spec: https://url.spec.whatwg.org/#concept-basic-url-parser
 
 .. function:: urlunsplit(parts)
 
@@ -668,6 +675,10 @@ task isn't already covered by the URL parsing functions above.
 
 .. seealso::
 
+   `WHATWG`_ -  URL Living standard
+      Working Group for the URL Standard that defines URLs, domains, IP addresses, the
+      application/x-www-form-urlencoded format, and their API.
+
    :rfc:`3986` - Uniform Resource Identifiers
       This is the current standard (STD66). Any changes to urllib.parse module
       should conform to this. Certain deviations could be observed, which are
@@ -691,3 +702,5 @@ task isn't already covered by the URL parsing functions above.
 
    :rfc:`1738` - Uniform Resource Locators (URL)
       This specifies the formal syntax and semantics of absolute URLs.
+
+.. _WHATWG: https://url.spec.whatwg.org/
diff --git a/Lib/test/test_urlparse.py b/Lib/test/test_urlparse.py
index d2ec0da..0f99130 100644
--- a/Lib/test/test_urlparse.py
+++ b/Lib/test/test_urlparse.py
@@ -612,6 +612,54 @@ class UrlParseTestCase(unittest.TestCase):
         with self.assertRaisesRegex(ValueError, "out of range"):
             p.port
 
+    def test_urlsplit_remove_unsafe_bytes(self):
+        # Remove ASCII tabs and newlines from input, for http common case scenario.
+        url = "h\nttp://www.python\n.org\t/java\nscript:\talert('msg\r\n')/?query\n=\tsomething#frag\nment"
+        p = urllib.parse.urlsplit(url)
+        self.assertEqual(p.scheme, "http")
+        self.assertEqual(p.netloc, "www.python.org")
+        self.assertEqual(p.path, "/javascript:alert('msg')/")
+        self.assertEqual(p.query, "query=something")
+        self.assertEqual(p.fragment, "fragment")
+        self.assertEqual(p.username, None)
+        self.assertEqual(p.password, None)
+        self.assertEqual(p.hostname, "www.python.org")
+        self.assertEqual(p.port, None)
+        self.assertEqual(p.geturl(), "http://www.python.org/javascript:alert('msg')/?query=something#fragment")
+
+        # Remove ASCII tabs and newlines from input as bytes, for http common case scenario.
+        url = b"h\nttp://www.python\n.org\t/java\nscript:\talert('msg\r\n')/?query\n=\tsomething#frag\nment"
+        p = urllib.parse.urlsplit(url)
+        self.assertEqual(p.scheme, b"http")
+        self.assertEqual(p.netloc, b"www.python.org")
+        self.assertEqual(p.path, b"/javascript:alert('msg')/")
+        self.assertEqual(p.query, b"query=something")
+        self.assertEqual(p.fragment, b"fragment")
+        self.assertEqual(p.username, None)
+        self.assertEqual(p.password, None)
+        self.assertEqual(p.hostname, b"www.python.org")
+        self.assertEqual(p.port, None)
+        self.assertEqual(p.geturl(), b"http://www.python.org/javascript:alert('msg')/?query=something#fragment")
+
+        # any scheme
+        url = "x-new-scheme\t://www.python\n.org\t/java\nscript:\talert('msg\r\n')/?query\n=\tsomething#frag\nment"
+        p = urllib.parse.urlsplit(url)
+        self.assertEqual(p.geturl(), "x-new-scheme://www.python.org/javascript:alert('msg')/?query=something#fragment")
+
+        # Remove ASCII tabs and newlines from input as bytes, any scheme.
+        url = b"x-new-scheme\t://www.python\n.org\t/java\nscript:\talert('msg\r\n')/?query\n=\tsomething#frag\nment"
+        p = urllib.parse.urlsplit(url)
+        self.assertEqual(p.geturl(), b"x-new-scheme://www.python.org/javascript:alert('msg')/?query=something#fragment")
+
+        # Unsafe bytes is not returned from urlparse cache.
+        # scheme is stored after parsing, sending an scheme with unsafe bytes *will not* return an unsafe scheme
+        url = "https://www.python\n.org\t/java\nscript:\talert('msg\r\n')/?query\n=\tsomething#frag\nment"
+        scheme = "htt\nps"
+        for _ in range(2):
+            p = urllib.parse.urlsplit(url, scheme=scheme)
+            self.assertEqual(p.scheme, "https")
+            self.assertEqual(p.geturl(), "https://www.python.org/javascript:alert('msg')/?query=something#fragment")
+
     def test_attributes_bad_port(self):
         """Check handling of invalid ports."""
         for bytes in (False, True):
diff --git a/Lib/urllib/parse.py b/Lib/urllib/parse.py
index 36fd8fe..f0d9d4d 100644
--- a/Lib/urllib/parse.py
+++ b/Lib/urllib/parse.py
@@ -77,6 +77,9 @@ scheme_chars = ('abcdefghijklmnopqrstuvwxyz'
                 '0123456789'
                 '+-.')
 
+# Unsafe bytes to be removed per WHATWG spec
+_UNSAFE_URL_BYTES_TO_REMOVE = ['\t', '\r', '\n']
+
 # XXX: Consider replacing with functools.lru_cache
 MAX_CACHE_SIZE = 20
 _parse_cache = {}
@@ -414,6 +417,11 @@ def _checknetloc(netloc):
             raise ValueError("netloc '" + netloc + "' contains invalid " +
                              "characters under NFKC normalization")
 
+def _remove_unsafe_bytes_from_url(url):
+    for b in _UNSAFE_URL_BYTES_TO_REMOVE:
+        url = url.replace(b, "")
+    return url
+
 def urlsplit(url, scheme='', allow_fragments=True):
     """Parse a URL into 5 components:
     <scheme>://<netloc>/<path>?<query>#<fragment>
@@ -421,6 +429,8 @@ def urlsplit(url, scheme='', allow_fragments=True):
     Note that we don't break the components up in smaller bits
     (e.g. netloc is a single string) and we don't expand % escapes."""
     url, scheme, _coerce_result = _coerce_args(url, scheme)
+    url = _remove_unsafe_bytes_from_url(url)
+    scheme = _remove_unsafe_bytes_from_url(scheme)
     allow_fragments = bool(allow_fragments)
     key = url, scheme, allow_fragments, type(url), type(scheme)
     cached = _parse_cache.get(key, None)
diff --git a/Misc/NEWS.d/next/Security/2021-04-25-07-46-37.bpo-43882.Jpwx85.rst b/Misc/NEWS.d/next/Security/2021-04-25-07-46-37.bpo-43882.Jpwx85.rst
new file mode 100644
index 0000000..a326d07
--- /dev/null
+++ b/Misc/NEWS.d/next/Security/2021-04-25-07-46-37.bpo-43882.Jpwx85.rst
@@ -0,0 +1,6 @@
+The presence of newline or tab characters in parts of a URL could allow
+some forms of attacks.
+
+Following the controlling specification for URLs defined by WHATWG
+:func:`urllib.parse` now removes ASCII newlines and tabs from URLs,
+preventing such attacks.