adding optimization to shortcirtuit get_block_data early if blocking …

…key value is missing

src/recordlinker/linkage/simple_mpi.py

@@ -34,6 +34,10 @@ def get_block_data(
         key = models.BlockingKey[key_name]
         # Get all the possible values from the data for this key
         vals = [v for v in key.to_value(data)]
+        # If there are no values for a blocking key in the pass, we can skip
+        # the query and return an empty list, this is just an optimization
+        if not vals:
+            return []
         # Create a dynamic alias for the Blocking Value table using the index
         # this is necessary since we are potentially joining the same table
         # multiple times with different conditions

tests/unit/test_simple_mpi.py

@@ -176,3 +176,9 @@ def test_block_on_multiple_names(self, session, prime_index):
         algo_config = {"blocks": [{"value": "first_name"}, {"value": "last_name"}]}
         matches = simple_mpi.get_block_data(session, data, algo_config)
         assert len(matches) == 4
+
+    def test_block_missing_keys(self, session, prime_index):
+        data = {"birthdate": "01/01/1980"}
+        algo_config = {"blocks": [{"value": "birthdate"}, {"value": "last_name"}]}
+        matches = simple_mpi.get_block_data(session, data, algo_config)
+        assert len(matches) == 0