Fix fractfact_opt for Python3 (#10)

* Add issue #9 test

* Fix integer division

* Fix np.array data hashing

* Assert returned value

Author: relf

pyDOE3/doe_factorial.py

@@ -406,7 +406,7 @@ def fracfact_opt(n_factors, n_erased, max_attempts=0):
     def n_comb(n, k):
         if k <= 0 or n <= 0 or k > n:
             return 0
-        return math.factorial(n) / (math.factorial(k) * math.factorial(n - k))
+        return math.factorial(n) // (math.factorial(k) * math.factorial(n - k))
 
     if n_factors > 20:
         raise ValueError("Design too big, use 20 factors or less")
@@ -503,8 +503,8 @@ def fracfact_aliasing(design):
     for combination in all_combinations:
         contrast = np.prod(design[:, combination], axis=1)
         contrast.flags.writeable = False
-        aliases[contrast.data] = aliases.get(contrast.data, [])
-        aliases[contrast.data].append(combination)
+        aliases[contrast.data.tobytes()] = aliases.get(contrast.data.tobytes(), [])
+        aliases[contrast.data.tobytes()].append(combination)
 
     aliases_list = []
     for alias in aliases.values():

tests/test_factorial.py

@@ -1,5 +1,5 @@
 import unittest
-from pyDOE3.doe_factorial import fullfact
+from pyDOE3.doe_factorial import fracfact_opt, fullfact
 from pyDOE3.doe_factorial import ff2n
 from pyDOE3.doe_factorial import fracfact
 from pyDOE3.doe_factorial import fracfact_by_res
@@ -98,3 +98,23 @@ def test_factorial6(self):
         ]
         actual = fracfact_by_res(6, 3)
         np.testing.assert_allclose(actual, expected)
+
+    def test_issue_9(self):
+        ffo_doe = fracfact_opt(4, 1)
+        self.assertEqual(ffo_doe[0], "a b c abc")
+        self.assertEqual(
+            ffo_doe[1],
+            [
+                "a = bcd",
+                "b = acd",
+                "c = abd",
+                "d = abc",
+                "ab = cd",
+                "ac = bd",
+                "ad = bc",
+                "abcd",
+            ],
+        )
+        np.testing.assert_array_equal(
+            ffo_doe[2], np.array([0.0, 0.0, 3.0, 4.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0])
+        )