Merge pull request #13907 from stestagg/master

BUG: handle weird bytestrings in dtype() (#13902)

3 files changed, 43 insertions, 2 deletions

diff --git a/numpy/core/src/multiarray/arraytypes.c.src b/numpy/core/src/multiarray/arraytypes.c.src
index 3b986ed04..c586a0b1d 100644
--- a/numpy/core/src/multiarray/arraytypes.c.src
+++ b/numpy/core/src/multiarray/arraytypes.c.src
@@ -4414,7 +4414,17 @@ PyArray_DescrFromType(int type)
 {
     PyArray_Descr *ret = NULL;
 
-    if (type < NPY_NTYPES) {
+    if (type < 0) {
+        /*
+         * It's not valid for type to be less than 0.
+         * If that happens, then no other branch of
+         * this if/else chain should be followed.
+         * This is effectively a no-op that ensures
+         * the default error is raised.
+         */
+        ret = NULL;
+    }
+    else if (type < NPY_NTYPES) {
         ret = _builtin_descrs[type];
     }
     else if (type == NPY_NOTYPE) {
diff --git a/numpy/core/src/multiarray/descriptor.c b/numpy/core/src/multiarray/descriptor.c
index 82d1ef4c9..ff85c3fcb 100644
--- a/numpy/core/src/multiarray/descriptor.c
+++ b/numpy/core/src/multiarray/descriptor.c
@@ -1546,7 +1546,8 @@ PyArray_DescrConverter(PyObject *obj, PyArray_Descr **at)
 
         /* A typecode like 'd' */
         if (len == 1) {
-            check_num = type[0];
+            /* Python byte string characters are unsigned */
+            check_num = (unsigned char) type[0];
         }
         /* A kind + size like 'f8' */
         else {
diff --git a/numpy/core/tests/test_dtype.py b/numpy/core/tests/test_dtype.py
index bc5c23caf..f60eab696 100644
--- a/numpy/core/tests/test_dtype.py
+++ b/numpy/core/tests/test_dtype.py
@@ -88,6 +88,36 @@ class TestBuiltin(object):
             assert_raises(TypeError, np.dtype, 'q8')
             assert_raises(TypeError, np.dtype, 'Q8')
 
+    @pytest.mark.parametrize(
+        'value',
+        ['m8', 'M8', 'datetime64', 'timedelta64',
+         'i4, (2,3)f8, f4', 'a3, 3u8, (3,4)a10',
+         '>f', '<f', '=f', '|f',
+        ])
+    def test_dtype_bytes_str_equivalence(self, value):
+        bytes_value = value.encode('ascii')
+        from_bytes = np.dtype(bytes_value)
+        from_str = np.dtype(value)
+        assert_dtype_equal(from_bytes, from_str)
+
+    def test_dtype_from_bytes(self):
+        # Empty bytes object
+        assert_raises(TypeError, np.dtype, b'')
+        # Byte order indicator, but no type
+        assert_raises(TypeError, np.dtype, b'|')
+
+        # Single character with ordinal < NPY_NTYPES returns
+        # type by index into _builtin_descrs
+        assert_dtype_equal(np.dtype(bytes([0])), np.dtype('bool'))
+        assert_dtype_equal(np.dtype(bytes([17])), np.dtype(object))
+
+        # Single character where value is a valid type code
+        assert_dtype_equal(np.dtype(b'f'), np.dtype('float32'))
+
+        # Bytes with non-ascii values raise errors
+        assert_raises(TypeError, np.dtype, b'\xff')
+        assert_raises(TypeError, np.dtype, b's\xff')
+
     def test_bad_param(self):
         # Can't give a size that's too small
         assert_raises(ValueError, np.dtype,