diff --git a/python/pyspark/sql/column.py b/python/pyspark/sql/column.py index 1ecec5b126505..0a85da7443d3d 100644 --- a/python/pyspark/sql/column.py +++ b/python/pyspark/sql/column.py @@ -396,6 +396,11 @@ def over(self, window): jc = self._jc.over(window._jspec) return Column(jc) + def __nonzero__(self): + raise ValueError("Cannot convert column into bool: please use '&' for 'and', '|' for 'or', " + "'~' for 'not' when building DataFrame boolean expressions.") + __bool__ = __nonzero__ + def __repr__(self): return 'Column<%s>' % self._jc.toString().encode('utf8') diff --git a/python/pyspark/sql/tests.py b/python/pyspark/sql/tests.py index 13f4556943ac8..e6a434e4b2dff 100644 --- a/python/pyspark/sql/tests.py +++ b/python/pyspark/sql/tests.py @@ -164,6 +164,14 @@ def test_explode(self): self.assertEqual(result[0][0], "a") self.assertEqual(result[0][1], "b") + def test_and_in_expression(self): + self.assertEqual(4, self.df.filter((self.df.key <= 10) & (self.df.value <= "2")).count()) + self.assertRaises(ValueError, lambda: (self.df.key <= 10) and (self.df.value <= "2")) + self.assertEqual(14, self.df.filter((self.df.key <= 3) | (self.df.value < "2")).count()) + self.assertRaises(ValueError, lambda: self.df.key <= 3 or self.df.value < "2") + self.assertEqual(99, self.df.filter(~(self.df.key == 1)).count()) + self.assertRaises(ValueError, lambda: not self.df.key == 1) + def test_udf_with_callable(self): d = [Row(number=i, squared=i**2) for i in range(10)] rdd = self.sc.parallelize(d) @@ -408,7 +416,7 @@ def test_column_operators(self): self.assertTrue(isinstance((- ci - 1 - 2) % 3 * 2.5 / 3.5, Column)) rcc = (1 + ci), (1 - ci), (1 * ci), (1 / ci), (1 % ci) self.assertTrue(all(isinstance(c, Column) for c in rcc)) - cb = [ci == 5, ci != 0, ci > 3, ci < 4, ci >= 0, ci <= 7, ci and cs, ci or cs] + cb = [ci == 5, ci != 0, ci > 3, ci < 4, ci >= 0, ci <= 7] self.assertTrue(all(isinstance(c, Column) for c in cb)) cbool = (ci & ci), (ci | ci), (~ci) self.assertTrue(all(isinstance(c, Column) for c in cbool))