BUG: core/pickle: use only non-interned strings as buffer in unpickled arrays (#1708)

Using strings as mutable buffers is "evil" (since strings are
immutable), but it should "work" for non-interned strings.  On Python 3,
bytes are never interned, and so should also work.

2 files changed, 20 insertions, 0 deletions

diff --git a/numpy/core/src/multiarray/methods.c b/numpy/core/src/multiarray/methods.c
index e53177519..a4c92b2a1 100644
--- a/numpy/core/src/multiarray/methods.c
+++ b/numpy/core/src/multiarray/methods.c
@@ -1481,7 +1481,13 @@ array_setstate(PyArrayObject *self, PyObject *args)
     if (!PyDataType_FLAGCHK(typecode, NPY_LIST_PICKLE)) {
         int swap=!PyArray_ISNOTSWAPPED(self);
         self->data = datastr;
+#ifndef NPY_PY3K
+        /* Check that the string is not interned */
+        if (!_IsAligned(self) || swap || PyString_CHECK_INTERNED(rawdata)) {
+#else
+        /* Bytes are never interned */
         if (!_IsAligned(self) || swap) {
+#endif
             intp num = PyArray_NBYTES(self);
             self->data = PyDataMem_NEW(num);
             if (self->data == NULL) {
diff --git a/numpy/core/tests/test_regression.py b/numpy/core/tests/test_regression.py
index bcb34d04e..867df6c9d 100644
--- a/numpy/core/tests/test_regression.py
+++ b/numpy/core/tests/test_regression.py
@@ -1580,5 +1580,19 @@ class TestRegression(TestCase):
         y = np.array(x.flat)
         assert_equal(x, [[1,3],[2,4]])
 
+    def test_pickle_string_overwrite(self):
+        import re
+
+        data = np.array([1], dtype='b')
+        blob = pickle.dumps(data, protocol=1)
+        data = pickle.loads(blob)
+
+        # Check that loads does not clobber interned strings
+        s = re.sub("a(.)", "\x01\\1", "a_")
+        assert_equal(s[0], "\x01")
+        data[0] = 0xbb
+        s = re.sub("a(.)", "\x01\\1", "a_")
+        assert_equal(s[0], "\x01")
+
 if __name__ == "__main__":
     run_module_suite()