gh-123067: Fix quadratic complexity in parsing "-quoted cookie values with backslashes (GH-123075)

serhiy-storchaka · web-flow · commit 44e458357fca · 2024-08-17T16:30:52.000+03:00
This fixes CVE-2024-7592.
diff --git a/Lib/http/cookies.py b/Lib/http/cookies.py
@@ -184,8 +184,13 @@ def _quote(str):
         return '"' + str.translate(_Translator) + '"'
 
 
-_OctalPatt = re.compile(r"\\[0-3][0-7][0-7]")
-_QuotePatt = re.compile(r"[\\].")
+_unquote_sub = re.compile(r'\\(?:([0-3][0-7][0-7])|(.))').sub
+
+def _unquote_replace(m):
+    if m[1]:
+        return chr(int(m[1], 8))
+    else:
+        return m[2]
 
 def _unquote(str):
     # If there aren't any doublequotes,
@@ -205,30 +210,7 @@ def _unquote(str):
     #    \012 --> \n
     #    \"   --> "
     #
-    i = 0
-    n = len(str)
-    res = []
-    while 0 <= i < n:
-        o_match = _OctalPatt.search(str, i)
-        q_match = _QuotePatt.search(str, i)
-        if not o_match and not q_match:              # Neither matched
-            res.append(str[i:])
-            break
-        # else:
-        j = k = -1
-        if o_match:
-            j = o_match.start(0)
-        if q_match:
-            k = q_match.start(0)
-        if q_match and (not o_match or k < j):     # QuotePatt matched
-            res.append(str[i:k])
-            res.append(str[k+1])
-            i = k + 2
-        else:                                      # OctalPatt matched
-            res.append(str[i:j])
-            res.append(chr(int(str[j+1:j+4], 8)))
-            i = j + 4
-    return _nulljoin(res)
+    return _unquote_sub(_unquote_replace, str)
 
 # The _getdate() routine is used to set the expiration time in the cookie's HTTP
 # header.  By default, _getdate() returns the current time in the appropriate
diff --git a/Lib/test/test_http_cookies.py b/Lib/test/test_http_cookies.py
@@ -5,6 +5,7 @@
 import doctest
 from http import cookies
 import pickle
+from test import support
 
 
 class CookieTests(unittest.TestCase):
@@ -58,6 +59,43 @@ def test_basic(self):
             for k, v in sorted(case['dict'].items()):
                 self.assertEqual(C[k].value, v)
 
+    def test_unquote(self):
+        cases = [
+            (r'a="b=\""', 'b="'),
+            (r'a="b=\\"', 'b=\\'),
+            (r'a="b=\="', 'b=='),
+            (r'a="b=\n"', 'b=n'),
+            (r'a="b=\042"', 'b="'),
+            (r'a="b=\134"', 'b=\\'),
+            (r'a="b=\377"', 'b=\xff'),
+            (r'a="b=\400"', 'b=400'),
+            (r'a="b=\42"', 'b=42'),
+            (r'a="b=\\042"', 'b=\\042'),
+            (r'a="b=\\134"', 'b=\\134'),
+            (r'a="b=\\\""', 'b=\\"'),
+            (r'a="b=\\\042"', 'b=\\"'),
+            (r'a="b=\134\""', 'b=\\"'),
+            (r'a="b=\134\042"', 'b=\\"'),
+        ]
+        for encoded, decoded in cases:
+            with self.subTest(encoded):
+                C = cookies.SimpleCookie()
+                C.load(encoded)
+                self.assertEqual(C['a'].value, decoded)
+
+    @support.requires_resource('cpu')
+    def test_unquote_large(self):
+        n = 10**6
+        for encoded in r'\\', r'\134':
+            with self.subTest(encoded):
+                data = 'a="b=' + encoded*n + ';"'
+                C = cookies.SimpleCookie()
+                C.load(data)
+                value = C['a'].value
+                self.assertEqual(value[:3], 'b=\\')
+                self.assertEqual(value[-2:], '\\;')
+                self.assertEqual(len(value), n + 3)
+
     def test_load(self):
         C = cookies.SimpleCookie()
         C.load('Customer="WILE_E_COYOTE"; Version=1; Path=/acme')
diff --git a/Misc/NEWS.d/next/Library/2024-08-16-19-13-21.gh-issue-123067.Nx9O4R.rst b/Misc/NEWS.d/next/Library/2024-08-16-19-13-21.gh-issue-123067.Nx9O4R.rst
@@ -0,0 +1 @@
+Fix quadratic complexity in parsing ``"``-quoted cookie values with backslashes by :mod:`http.cookies`.

Original file line number	Diff line number	Diff line change
`@@ -0,0 +1 @@`
	`1`	+Fix quadratic complexity in parsing ``"``-quoted cookie values with backslashes by :mod:`http.cookies`.