Merge pull request #3217 from thocevar/valuehash

[FIX] Variable: Prevent hashing of Values of DiscreteVariable.

Author: astaric

Orange/data/variable.py

@@ -207,10 +207,15 @@ def __contains__(self, other):
         raise TypeError("invalid operation on Value()")
 
     def __hash__(self):
+        if self.variable.is_discrete:
+            # It is not possible to hash the id and the domain value to the same number as required by __eq__.
+            # hash(1) == hash(Value(DiscreteVariable("var", ["red", "green", "blue"]), 1)) == hash("green")
+            # User should hash directly ids or domain values instead.
+            raise TypeError("unhashable type - cannot hash values of discrete variables!")
         if self._value is None:
             return super().__hash__()
         else:
-            return hash((super().__hash__(), self._value))
+            return hash(self._value)
 
     @property
     def value(self):

Orange/tests/test_value.py

@@ -5,7 +5,8 @@
 import unittest
 import numpy as np
 
-from Orange.data import Table, Domain, DiscreteVariable
+from Orange.data import Table, Domain, Value,\
+    DiscreteVariable, ContinuousVariable, StringVariable, TimeVariable
 
 
 class TestValue(unittest.TestCase):
@@ -50,3 +51,16 @@ def test_compare_string(self):
         zoo2 = zoo[1]['name']  # antelope
         self.assertTrue(zoo1 < zoo2)
         self.assertTrue(zoo1 >= "aardvark")
+
+    def test_hash(self):
+        v = 1234.5
+        val = Value(ContinuousVariable("var"), v)
+        self.assertTrue(val == v and hash(val) == hash(v))
+        v = "test"
+        val = Value(StringVariable("var"), v)
+        self.assertTrue(val == v and hash(val) == hash(v))
+        v = 1234.5
+        val = Value(TimeVariable("var"), v)
+        self.assertTrue(val == v and hash(val) == hash(v))
+        val = Value(DiscreteVariable("var", ["red", "green", "blue"]), 1)
+        self.assertRaises(TypeError, hash, val)