scraping and data

pipeline/transform/postcode_boundaries/test_postcode_boundaries.py

@@ -121,6 +121,50 @@ class TestWhitespacePostcodes:
         loaded_df, _ = load_uprns(path)
         assert len(loaded_df) == 0
 
+    def test_non_english_oas_excluded(self, tmp_path):
+        df = pl.DataFrame(
+            {
+                "GRIDGB1E": [500010, 300010],
+                "GRIDGB1N": [180010, 220010],
+                "PCDS": ["AA1 1AA", "CF1 1AA"],
+                "OA21CD": ["E00000001", "W00000001"],
+            }
+        )
+        path = tmp_path / "uprn.parquet"
+        df.write_parquet(path)
+
+        loaded_df, offsets = load_uprns(path)
+
+        assert set(offsets) == {"E00000001"}
+        assert loaded_df["PCDS"].to_list() == ["AA1 1AA"]
+
+    def test_terminated_postcodes_are_remapped(self, tmp_path):
+        uprns = pl.DataFrame(
+            {
+                "GRIDGB1E": [500010],
+                "GRIDGB1N": [180010],
+                "PCDS": ["aa1 1aa"],
+                "OA21CD": ["E00000001"],
+            }
+        )
+        uprn_path = tmp_path / "uprn.parquet"
+        uprns.write_parquet(uprn_path)
+        arcgis = pl.DataFrame(
+            {
+                "pcds": ["AA1 1AA", "AA1 1AB"],
+                "east1m": [500010, 500030],
+                "north1m": [180010, 180020],
+                "doterm": ["2020-01-01", None],
+                "ctry25cd": ["E92000001", "E92000001"],
+            }
+        )
+        arcgis_path = tmp_path / "arcgis.parquet"
+        arcgis.write_parquet(arcgis_path)
+
+        loaded_df, _offsets = load_uprns(uprn_path, arcgis_path)
+
+        assert loaded_df["PCDS"].to_list() == ["AA1 1AB"]
+
 
 # ---------------------------------------------------------------------------
 # Bug 3: Voronoi deduplication is first-seen-wins