basic check of null-excepting

ADBond · Jan 2, 2025 · 57bf701 · 57bf701
1 parent dd89373
commit 57bf701
Show file tree

Hide file tree

Showing 2 changed files with 31 additions and 7 deletions.
diff --git a/tests/conftest.py b/tests/conftest.py
@@ -184,6 +184,13 @@ def input_nodes_with_lat_longs():
     n_rows = 1_000
     lats = np.random.uniform(low=lat_low, high=lat_high, size=n_rows)
     longs = np.random.uniform(low=long_low, high=long_high, size=n_rows)
+
+    # introduce a few nulls to check they do not interfere
+    lats[10] = None
+    lats[45] = None
+    longs[45] = None
+    longs[30] = None
+
     # also include some names so we have a second comparison
     names = np.random.choice(
         _NAMES,

diff --git a/tests/test_cll_ch.py b/tests/test_cll_ch.py
@@ -7,21 +7,38 @@
 from splinkclickhouse.column_expression import ColumnExpression
 
 
-def test_distance_in_km_level(api_info, input_nodes_with_lat_longs):
+@mark.parametrize("not_null", [True, False])
+def test_distance_in_km_level(api_info, input_nodes_with_lat_longs, not_null):
     db_api = api_info["db_api_factory"]()
 
+    null_level_list = (
+        []
+        if not_null
+        else [cll.Or(cll.NullLevel("latitude"), cll.NullLevel("longitude"))]
+    )
+
     settings = SettingsCreator(
         link_type="dedupe_only",
         comparisons=[
             cl.ExactMatch("name"),
             cl.CustomComparison(
                 comparison_levels=[
-                    cll.Or(cll.NullLevel("latitude"), cll.NullLevel("longitude")),
-                    cll_ch.DistanceInKMLevel("latitude", "longitude", 10),
-                    cll_ch.DistanceInKMLevel("latitude", "longitude", 50),
-                    cll_ch.DistanceInKMLevel("latitude", "longitude", 100),
-                    cll_ch.DistanceInKMLevel("latitude", "longitude", 200),
-                    cll_ch.DistanceInKMLevel("latitude", "longitude", 500),
+                    *null_level_list,
+                    cll_ch.DistanceInKMLevel(
+                        "latitude", "longitude", 10, not_null=not_null
+                    ),
+                    cll_ch.DistanceInKMLevel(
+                        "latitude", "longitude", 50, not_null=not_null
+                    ),
+                    cll_ch.DistanceInKMLevel(
+                        "latitude", "longitude", 100, not_null=not_null
+                    ),
+                    cll_ch.DistanceInKMLevel(
+                        "latitude", "longitude", 200, not_null=not_null
+                    ),
+                    cll_ch.DistanceInKMLevel(
+                        "latitude", "longitude", 500, not_null=not_null
+                    ),
                     cll.ElseLevel(),
                 ],
                 output_column_name="latlong",