# HG changeset patch
# User MRAB <regex@mrabarnett.plus.com>
# Date 1356834937 0
# Node ID ffce7af849a4e6a5c3af4ccac7c7460e44ad080e
# Parent  c59ee1ff6f277d9985aaf45e3b9ed3a887c0caef
Issue #16741: `int()`, `float()`, etc think python strings are null-terminated

diff -r c59ee1ff6f27 -r ffce7af849a4 Lib/test/test_int.py
--- a/Lib/test/test_int.py	Sat Dec 29 23:41:08 2012 +0200
+++ b/Lib/test/test_int.py	Sun Dec 30 02:35:37 2012 +0000
@@ -217,6 +217,28 @@
         self.assertEqual(int('2br45qc', 35), 4294967297)
         self.assertEqual(int('1z141z5', 36), 4294967297)
 
+        # Bug #16741: `int()`, `float()`, etc think python strings are null-terminated
+        s = "1\x01"
+        try:
+            int(s)
+        except ValueError as e:
+            self.assertEqual(repr(e),
+              repr(ValueError("invalid literal for int() with base 10: {!r}".format(s))))
+
+        s = "1\x00"
+        try:
+            int(s)
+        except ValueError as e:
+            self.assertEqual(repr(e),
+              repr(ValueError("invalid literal for int() with base 10: {!r}".format(s))))
+
+        s = "1" + " " * 198 + "\u03C0"
+        try:
+            int(s)
+        except ValueError as e:
+            self.assertEqual(repr(e),
+              repr(ValueError("invalid literal for int() with base 10: {!r}".format(s))))
+
     @support.cpython_only
     def test_small_ints(self):
         # Bug #3236: Return small longs from PyLong_FromString
diff -r c59ee1ff6f27 -r ffce7af849a4 Objects/longobject.c
--- a/Objects/longobject.c	Sat Dec 29 23:41:08 2012 +0200
+++ b/Objects/longobject.c	Sun Dec 30 02:35:37 2012 +0000
@@ -1976,20 +1976,21 @@
     return long_normalize(z);
 }
 
-PyObject *
-PyLong_FromString(char *str, char **pend, int base)
+/* Parses a long from a bytestring. Leading and trailing whitespace will be
+ * ignored.
+ *
+ * If successful, a PyLong object will be returned and 'pend' will be pointing
+ * to the first unused byte unless it's NULL.
+ *
+ * If unsuccessful, NULL will be returned.
+ */
+static PyObject *
+long_from_string(char *str, char **pend, int base)
 {
     int sign = 1, error_if_nonzero = 0;
     char *start, *orig_str = str;
     PyLongObject *z = NULL;
-    PyObject *strobj;
-    Py_ssize_t slen;
-
-    if ((base != 0 && base < 2) || base > 36) {
-        PyErr_SetString(PyExc_ValueError,
-                        "int() arg 2 must be >= 2 and <= 36");
-        return NULL;
-    }
+
     while (*str != '\0' && isspace(Py_CHARMASK(*str)))
         str++;
     if (*str == '+')
@@ -2238,21 +2239,45 @@
         str++;
     if (*str != '\0')
         goto onError;
-    if (pend)
+    if (pend != NULL)
         *pend = str;
     long_normalize(z);
     return (PyObject *) maybe_small_long(z);
 
   onError:
     Py_XDECREF(z);
-    slen = strlen(orig_str) < 200 ? strlen(orig_str) : 200;
-    strobj = PyUnicode_FromStringAndSize(orig_str, slen);
-    if (strobj == NULL)
+    return NULL;
+}
+
+PyObject *
+PyLong_FromString(char *str, char **pend, int base)
+{
+    PyObject *result;
+    Py_ssize_t slen;
+    PyObject *strobj;
+
+    if ((base != 0 && base < 2) || base > 36) {
+        PyErr_SetString(PyExc_ValueError,
+                        "int() base must be >= 2 and <= 36");
         return NULL;
-    PyErr_Format(PyExc_ValueError,
-                 "invalid literal for int() with base %d: %R",
-                 base, strobj);
-    Py_DECREF(strobj);
+    }
+    result = long_from_string(str, pend, base);
+    if (result != NULL && (pend == NULL || **pend == '\0'))
+        return result;
+    slen = strlen(str);
+    if (slen > 200) {
+        slen = 200;
+        /* Don't truncate in the middle of a UTF-8 multibyte sequence. */
+        while ((str[slen] & 0xC0) == 0x80)
+            slen--;
+    }
+    strobj = PyUnicode_FromStringAndSize(str, slen);
+    if (strobj != NULL) {
+        PyErr_Format(PyExc_ValueError,
+                        "invalid literal for int() with base %d: %R",
+                        base, strobj);
+        Py_DECREF(strobj);
+    }
     return NULL;
 }
 
@@ -2274,7 +2299,13 @@
     PyObject *asciidig;
     char *buffer, *end;
     Py_ssize_t buflen;
-
+    PyObject *strobj;
+
+    if ((base != 0 && base < 2) || base > 36) {
+        PyErr_SetString(PyExc_ValueError,
+                        "int() base must be >= 2 and <= 36");
+        return NULL;
+    }
     asciidig = _PyUnicode_TransformDecimalAndSpaceToASCII(u);
     if (asciidig == NULL)
         return NULL;
@@ -2283,11 +2314,17 @@
         Py_DECREF(asciidig);
         return NULL;
     }
-    result = PyLong_FromString(buffer, &end, base);
-    if (result != NULL && end != buffer + buflen) {
-        PyErr_SetString(PyExc_ValueError,
-                        "null byte in argument for int()");
-        Py_DECREF(result);
+    result = long_from_string(buffer, &end, base);
+    if (result == NULL || end != buffer + buflen) {
+        strobj = PySequence_GetSlice(u, 0, 200);
+        if (strobj != NULL) {
+            PyErr_Format(PyExc_ValueError,
+                            "invalid literal for int() with base %d: %R",
+                            base, strobj);
+            Py_DECREF(strobj);
+        }
+        if (result != NULL)
+            Py_DECREF(result);
         result = NULL;
     }
     Py_DECREF(asciidig);