Auto merge of #5967 - wimglenn:bugfix/empty_string_array, r=seberg

BUG: fix inconsistency with np.array(['']) being truthy ``` a = np.array([0]) b = np.array([None]) c = np.array(['']) d = np.array([' ']) ``` Why should we have this inconsistency: ``` >>> bool(a) False >>> bool(b) False >>> bool(c) True >>> bool(d) False ```
numpy · Jun 16, 2015 · 8c86a0a · 8c86a0a
2 parents 7e04882 + 8749a9a
commit 8c86a0a
Show file tree

Hide file tree

Showing 2 changed files with 52 additions and 2 deletions.
diff --git a/numpy/core/src/multiarray/arraytypes.c.src b/numpy/core/src/multiarray/arraytypes.c.src
@@ -2297,9 +2297,13 @@ STRING_nonzero (char *ip, PyArrayObject *ap)
     int len = PyArray_DESCR(ap)->elsize;
     int i;
     npy_bool nonz = NPY_FALSE;
+    npy_bool seen_null = NPY_FALSE;
 
     for (i = 0; i < len; i++) {
-        if (!Py_STRING_ISSPACE(*ip)) {
+        if (*ip == '\0') {
+            seen_null = NPY_TRUE;
+        }
+        else if (seen_null || !Py_STRING_ISSPACE(*ip)) {
             nonz = NPY_TRUE;
             break;
         }
@@ -2320,6 +2324,7 @@ UNICODE_nonzero (npy_ucs4 *ip, PyArrayObject *ap)
     int len = PyArray_DESCR(ap)->elsize >> 2;
     int i;
     npy_bool nonz = NPY_FALSE;
+    npy_bool seen_null = NPY_FALSE;
     char *buffer = NULL;
 
     if ((!PyArray_ISNOTSWAPPED(ap)) || (!PyArray_ISALIGNED(ap))) {
@@ -2335,7 +2340,10 @@ UNICODE_nonzero (npy_ucs4 *ip, PyArrayObject *ap)
     }
 
     for (i = 0; i < len; i++) {
-        if (!PyArray_UCS4_ISSPACE(*ip)) {
+        if (*ip == '\0') {
+            seen_null = NPY_TRUE;
+        }
+        else if (seen_null || !PyArray_UCS4_ISSPACE(*ip)) {
             nonz = NPY_TRUE;
             break;
         }

diff --git a/numpy/core/tests/test_multiarray.py b/numpy/core/tests/test_multiarray.py
@@ -5731,5 +5731,47 @@ def test_subclass_other(self):
             assert_(isinstance(f(b, a), self.Other), msg)
 
 
+class TestBytestringArrayNonzero(TestCase):
+
+    def test_empty_bstring_array_is_falsey(self):
+        self.assertFalse(np.array([''], dtype=np.str))
+
+    def test_whitespace_bstring_array_is_falsey(self):
+        a = np.array(['spam'], dtype=np.str)
+        a[0] = '  \0\0'
+        self.assertFalse(a)
+
+    def test_all_null_bstring_array_is_falsey(self):
+        a = np.array(['spam'], dtype=np.str)
+        a[0] = '\0\0\0\0'
+        self.assertFalse(a)
+
+    def test_null_inside_bstring_array_is_truthy(self):
+        a = np.array(['spam'], dtype=np.str)
+        a[0] = ' \0 \0'
+        self.assertTrue(a)
+
+
+class TestUnicodeArrayNonzero(TestCase):
+
+    def test_empty_ustring_array_is_falsey(self):
+        self.assertFalse(np.array([''], dtype=np.unicode))
+
+    def test_whitespace_ustring_array_is_falsey(self):
+        a = np.array(['eggs'], dtype=np.unicode)
+        a[0] = '  \0\0'
+        self.assertFalse(a)
+
+    def test_all_null_ustring_array_is_falsey(self):
+        a = np.array(['eggs'], dtype=np.unicode)
+        a[0] = '\0\0\0\0'
+        self.assertFalse(a)
+
+    def test_null_inside_ustring_array_is_truthy(self):
+        a = np.array(['eggs'], dtype=np.unicode)
+        a[0] = ' \0 \0'
+        self.assertTrue(a)
+
+
 if __name__ == "__main__":
     run_module_suite()