summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
authorAntoine Pitrou <solipsis@pitrou.net>2012-11-20 21:30:42 (GMT)
committerAntoine Pitrou <solipsis@pitrou.net>2012-11-20 21:30:42 (GMT)
commit39bdad813a2f07bd8c373e2c88b76467ef12832c (patch)
treef7908fef135347773f5eed6ca29a49918e466ff8
parent4a1fdcf07d4ee37ec003cd29be6b17e91d5db012 (diff)
downloadcpython-39bdad813a2f07bd8c373e2c88b76467ef12832c.zip
cpython-39bdad813a2f07bd8c373e2c88b76467ef12832c.tar.gz
cpython-39bdad813a2f07bd8c373e2c88b76467ef12832c.tar.bz2
Issue #1160: Fix compiling large regular expressions on UCS2 builds.
Patch by Serhiy Storchaka.
-rw-r--r--Lib/test/test_re.py6
-rw-r--r--Misc/NEWS3
-rw-r--r--Modules/_sre.c11
-rw-r--r--Modules/sre.h6
4 files changed, 17 insertions, 9 deletions
diff --git a/Lib/test/test_re.py b/Lib/test/test_re.py
index 5162d4e..39972d5 100644
--- a/Lib/test/test_re.py
+++ b/Lib/test/test_re.py
@@ -419,6 +419,12 @@ class ReTests(unittest.TestCase):
self.assertEqual(re.match("([\u2222\u2223])",
"\u2222", re.UNICODE).group(1), "\u2222")
+ def test_big_codesize(self):
+ # Issue #1160
+ r = re.compile('|'.join(('%d'%x for x in range(10000))))
+ self.assertIsNotNone(r.match('1000'))
+ self.assertIsNotNone(r.match('9999'))
+
def test_anyall(self):
self.assertEqual(re.match("a.b", "a\nb", re.DOTALL).group(0),
"a\nb")
diff --git a/Misc/NEWS b/Misc/NEWS
index c8580e3..7aa34d9 100644
--- a/Misc/NEWS
+++ b/Misc/NEWS
@@ -167,6 +167,9 @@ Core and Builtins
Library
-------
+- Issue #1160: Fix compiling large regular expressions on UCS2 builds.
+ Patch by Serhiy Storchaka.
+
- Issue #14313: zipfile now raises NotImplementedError when the compression
type is unknown.
diff --git a/Modules/_sre.c b/Modules/_sre.c
index 472b5a3..9600a08 100644
--- a/Modules/_sre.c
+++ b/Modules/_sre.c
@@ -2695,6 +2695,13 @@ _compile(PyObject* self_, PyObject* args)
for (i = 0; i < n; i++) {
PyObject *o = PyList_GET_ITEM(code, i);
unsigned long value = PyLong_AsUnsignedLong(o);
+ if (value == (unsigned long)-1 && PyErr_Occurred()) {
+ if (PyErr_ExceptionMatches(PyExc_OverflowError)) {
+ PyErr_SetString(PyExc_OverflowError,
+ "regular expression code size limit exceeded");
+ }
+ break;
+ }
self->code[i] = (SRE_CODE) value;
if ((unsigned long) self->code[i] != value) {
PyErr_SetString(PyExc_OverflowError,
@@ -3065,10 +3072,8 @@ _validate_inner(SRE_CODE *code, SRE_CODE *end, Py_ssize_t groups)
GET_ARG; max = arg;
if (min > max)
FAIL;
-#ifdef Py_UNICODE_WIDE
if (max > 65535)
FAIL;
-#endif
if (!_validate_inner(code, code+skip-4, groups))
FAIL;
code += skip-4;
@@ -3086,10 +3091,8 @@ _validate_inner(SRE_CODE *code, SRE_CODE *end, Py_ssize_t groups)
GET_ARG; max = arg;
if (min > max)
FAIL;
-#ifdef Py_UNICODE_WIDE
if (max > 65535)
FAIL;
-#endif
if (!_validate_inner(code, code+skip-3, groups))
FAIL;
code += skip-3;
diff --git a/Modules/sre.h b/Modules/sre.h
index aec9b54..d389b46 100644
--- a/Modules/sre.h
+++ b/Modules/sre.h
@@ -14,12 +14,8 @@
#include "sre_constants.h"
/* size of a code word (must be unsigned short or larger, and
- large enough to hold a Py_UNICODE character) */
-#ifdef Py_UNICODE_WIDE
+ large enough to hold a UCS4 character) */
#define SRE_CODE Py_UCS4
-#else
-#define SRE_CODE unsigned short
-#endif
typedef struct {
PyObject_VAR_HEAD