Issue #10182: The re module doesn't truncate indices to 32 bits anymore.

Patch by Serhiy Storchaka.

3 files changed, 24 insertions, 5 deletions

diff --git a/Lib/test/test_re.py b/Lib/test/test_re.py
index ff2c953..befe0e8 100644
--- a/Lib/test/test_re.py
+++ b/Lib/test/test_re.py
@@ -1,4 +1,5 @@
 from test.test_support import verbose, run_unittest, import_module
+from test.test_support import precisionbigmemtest, _2G
 import re
 from re import Scanner
 import sys
@@ -819,6 +820,21 @@ class ReTests(unittest.TestCase):
         # Test behaviour when not given a string or pattern as parameter
         self.assertRaises(TypeError, re.compile, 0)
 
+    # The huge memuse is because of re.sub() using a list and a join()
+    # to create the replacement result.
+    @precisionbigmemtest(size=_2G, memuse=20)
+    def test_large(self, size):
+        # Issue #10182: indices were 32-bit-truncated.
+        s = 'a' * size
+        m = re.search('$', s)
+        self.assertIsNotNone(m)
+        self.assertEqual(m.start(), size)
+        self.assertEqual(m.end(), size)
+        r, n = re.subn('', '', s)
+        self.assertEqual(r, s)
+        self.assertEqual(n, size + 1)
+
+
 def run_re_tests():
     from test.re_tests import tests, SUCCEED, FAIL, SYNTAX_ERROR
     if verbose:
diff --git a/Misc/NEWS b/Misc/NEWS
index 89f7e22..3d49637 100644
--- a/Misc/NEWS
+++ b/Misc/NEWS
@@ -154,6 +154,9 @@ Core and Builtins
 Library
 -------
 
+- Issue #10182: The re module doesn't truncate indices to 32 bits anymore.
+  Patch by Serhiy Storchaka.
+
 - Issue #16573: In 2to3, treat enumerate() like a consuming call, so superfluous
   list() calls aren't added to filter(), map(), and zip() which are directly
   passed enumerate().
diff --git a/Modules/_sre.c b/Modules/_sre.c
index ab4f269..baf5f6a 100644
--- a/Modules/_sre.c
+++ b/Modules/_sre.c
@@ -1636,7 +1636,7 @@ static PyObject*pattern_scanner(PatternObject*, PyObject*);
 static PyObject *
 sre_codesize(PyObject* self, PyObject *unused)
 {
-    return Py_BuildValue("l", sizeof(SRE_CODE));
+    return PyLong_FromSize_t(sizeof(SRE_CODE));
 }
 
 static PyObject *
@@ -2448,7 +2448,7 @@ next:
         return NULL;
 
     if (subn)
-        return Py_BuildValue("Ni", item, n);
+        return Py_BuildValue("Nn", item, n);
 
     return item;
 
@@ -3389,7 +3389,7 @@ match_start(MatchObject* self, PyObject* args)
     }
 
     /* mark is -1 if group is undefined */
-    return Py_BuildValue("i", self->mark[index*2]);
+    return PyLong_FromSsize_t(self->mark[index*2]);
 }
 
 static PyObject*
@@ -3412,7 +3412,7 @@ match_end(MatchObject* self, PyObject* args)
     }
 
     /* mark is -1 if group is undefined */
-    return Py_BuildValue("i", self->mark[index*2+1]);
+    return PyLong_FromSsize_t(self->mark[index*2+1]);
 }
 
 LOCAL(PyObject*)
@@ -3562,7 +3562,7 @@ static PyObject *
 match_lastindex_get(MatchObject *self)
 {
     if (self->lastindex >= 0)
-	return Py_BuildValue("i", self->lastindex);
+        return PyLong_FromSsize_t(self->lastindex);
     Py_INCREF(Py_None);
     return Py_None;
 }