Issue #13169: The maximal repetition number in a regular expression has been

increased from 65534 to 2147483647 (on 32-bit platform) or 4294967294 (on
64-bit).

4 files changed, 41 insertions, 7 deletions

diff --git a/Lib/sre_compile.py b/Lib/sre_compile.py
index f52ea01459..46eac9c070 100644
--- a/Lib/sre_compile.py
+++ b/Lib/sre_compile.py
@@ -13,6 +13,7 @@
 import _sre, sys
 import sre_parse
 from sre_constants import *
+from _sre import MAXREPEAT
 
 assert _sre.MAGIC == MAGIC, "SRE module mismatch"
 
diff --git a/Lib/sre_constants.py b/Lib/sre_constants.py
index 417670bf8d..71ccb23f62 100644
--- a/Lib/sre_constants.py
+++ b/Lib/sre_constants.py
@@ -15,10 +15,6 @@
 
 MAGIC = 20031017
 
-# max code word in this release
-
-MAXREPEAT = 65535
-
 # SRE standard exception (access as sre.error)
 # should this really be here?
 
diff --git a/Lib/sre_parse.py b/Lib/sre_parse.py
index 19dd4fc4bc..045a5ebfef 100644
--- a/Lib/sre_parse.py
+++ b/Lib/sre_parse.py
@@ -15,6 +15,7 @@
 import sys
 
 from sre_constants import *
+from _sre import MAXREPEAT
 
 SPECIAL_CHARS = ".\\[{()*+?^$|"
 REPEAT_CHARS = "*+?{"
@@ -505,10 +506,14 @@ def _parse(source, state):
                     continue
                 if lo:
                     min = int(lo)
+                    if min >= MAXREPEAT:
+                        raise OverflowError("the repetition number is too large")
                 if hi:
                     max = int(hi)
-                if max < min:
-                    raise error("bad repeat interval")
+                    if max >= MAXREPEAT:
+                        raise OverflowError("the repetition number is too large")
+                    if max < min:
+                        raise error("bad repeat interval")
             else:
                 raise error("not supported")
             # figure out which item to repeat
diff --git a/Lib/test/test_re.py b/Lib/test/test_re.py
index 6b047e48db..f7e76dcfba 100644
--- a/Lib/test/test_re.py
+++ b/Lib/test/test_re.py
@@ -1,4 +1,5 @@
-from test.support import verbose, run_unittest, gc_collect, bigmemtest, _2G
+from test.support import verbose, run_unittest, gc_collect, bigmemtest, _2G, \
+        cpython_only
 import io
 import re
 from re import Scanner
@@ -883,6 +884,37 @@ class ReTests(unittest.TestCase):
         self.assertEqual(n, size + 1)
 
 
+    def test_repeat_minmax_overflow(self):
+        # Issue #13169
+        string = "x" * 100000
+        self.assertEqual(re.match(r".{65535}", string).span(), (0, 65535))
+        self.assertEqual(re.match(r".{,65535}", string).span(), (0, 65535))
+        self.assertEqual(re.match(r".{65535,}?", string).span(), (0, 65535))
+        self.assertEqual(re.match(r".{65536}", string).span(), (0, 65536))
+        self.assertEqual(re.match(r".{,65536}", string).span(), (0, 65536))
+        self.assertEqual(re.match(r".{65536,}?", string).span(), (0, 65536))
+        # 2**128 should be big enough to overflow both SRE_CODE and Py_ssize_t.
+        self.assertRaises(OverflowError, re.compile, r".{%d}" % 2**128)
+        self.assertRaises(OverflowError, re.compile, r".{,%d}" % 2**128)
+        self.assertRaises(OverflowError, re.compile, r".{%d,}?" % 2**128)
+        self.assertRaises(OverflowError, re.compile, r".{%d,%d}" % (2**129, 2**128))
+
+    @cpython_only
+    def test_repeat_minmax_overflow_maxrepeat(self):
+        try:
+            from _sre import MAXREPEAT
+        except ImportError:
+            self.skipTest('requires _sre.MAXREPEAT constant')
+        string = "x" * 100000
+        self.assertIsNone(re.match(r".{%d}" % (MAXREPEAT - 1), string))
+        self.assertEqual(re.match(r".{,%d}" % (MAXREPEAT - 1), string).span(),
+                         (0, 100000))
+        self.assertIsNone(re.match(r".{%d,}?" % (MAXREPEAT - 1), string))
+        self.assertRaises(OverflowError, re.compile, r".{%d}" % MAXREPEAT)
+        self.assertRaises(OverflowError, re.compile, r".{,%d}" % MAXREPEAT)
+        self.assertRaises(OverflowError, re.compile, r".{%d,}?" % MAXREPEAT)
+
+
 def run_re_tests():
     from test.re_tests import tests, SUCCEED, FAIL, SYNTAX_ERROR
     if verbose: