Add __len__ method to ExistType (#1199)

Co-authored-by: Arthur Varkentin <[email protected]>
Co-authored-by: Forest Gregg <[email protected]>

Author: 3 people

dedupe/variables/exists.py

@@ -30,6 +30,9 @@ def comparator(self, field_1: Any, field_2: Any) -> list[int]:
         else:
             return self.cat_comparator(0, 0)
 
+    def __len__(self) -> int:
+        return len(self.higher_vars)
+
     # This flag tells fieldDistances in dedupe.core to pass
     # missing values (None) into the comparator
     comparator.missing = True  # type: ignore

tests/test_exists.py

@@ -11,3 +11,10 @@ def test_comparator(self):
         assert numpy.array_equal(var.comparator(None, None), [0, 0])
         assert numpy.array_equal(var.comparator(1, 1), [1, 0])
         assert numpy.array_equal(var.comparator(1, 0), [0, 1])
+
+    def test_len_higher_vars(self):
+        # The len > 1 is neccessary for the correct processing in datamodel.py
+        var = ExistsType("foo")
+        assert len(var) > 1
+        assert len(var.higher_vars) > 1
+        assert len(var) == len(var.higher_vars)