improve

pipeline/transform/postcode_boundaries/test_postcode_boundaries.py

@@ -3,6 +3,8 @@
 Each test targets a specific bug or edge case identified during code review.
 """
 
+import json
+
 import numpy as np
 import polars as pl
 import pytest
@@ -11,7 +13,12 @@ from shapely.ops import unary_union
 
 from .oa_boundaries import parse_gpkg_geometry
 from .greenspace import subtract_greenspace
-from .output import _fill_holes, merge_fragments, to_wgs84_geojson
+from .output import (
+    _fill_holes,
+    merge_fragments,
+    to_wgs84_geojson,
+    write_district_geojson,
+)
 from .process_oa import _extract_polygonal, process_oa
 from .uprn import get_oa_uprns, load_uprns
 from .voronoi import _equal_split_fallback, compute_voronoi_regions
@@ -154,6 +161,7 @@ class TestWhitespacePostcodes:
                 "pcds": ["AA1 1AA", "AA1 1AB"],
                 "east1m": [500010, 500030],
                 "north1m": [180010, 180020],
+                "oa21cd": ["E00000001", "E00000001"],
                 "doterm": ["2020-01-01", None],
                 "ctry25cd": ["E92000001", "E92000001"],
             }
@@ -165,6 +173,65 @@ class TestWhitespacePostcodes:
 
         assert loaded_df["PCDS"].to_list() == ["AA1 1AB"]
 
+    def test_arcgis_filters_to_active_english_postcodes(self, tmp_path):
+        uprns = pl.DataFrame(
+            {
+                "GRIDGB1E": [500010, 500020],
+                "GRIDGB1N": [180010, 180020],
+                "PCDS": ["AA1 1AA", "CF1 1AA"],
+                "OA21CD": ["E00000001", "E00000001"],
+            }
+        )
+        uprn_path = tmp_path / "uprn.parquet"
+        uprns.write_parquet(uprn_path)
+        arcgis = pl.DataFrame(
+            {
+                "pcds": ["AA1 1AA", "CF1 1AA"],
+                "east1m": [500010, 300010],
+                "north1m": [180010, 220010],
+                "oa21cd": ["E00000001", "W00000001"],
+                "doterm": [None, None],
+                "ctry25cd": ["E92000001", "W92000004"],
+            }
+        )
+        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 1AA"]
+
+    def test_arcgis_adds_centroid_seed_for_active_postcode_without_uprn(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", "BB1 1BB"],
+                "east1m": [500010, 510000],
+                "north1m": [180010, 190000],
+                "oa21cd": ["E00000001", "E00000002"],
+                "doterm": [None, 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 set(loaded_df["PCDS"].to_list()) == {"AA1 1AA", "BB1 1BB"}
+        points, postcodes = get_oa_uprns(loaded_df, offsets, "E00000002")
+        assert postcodes == ["BB1 1BB"]
+        assert points.tolist() == [[510000.0, 190000.0]]
+
 
 # ---------------------------------------------------------------------------
 # Bug 3: Voronoi deduplication is first-seen-wins
@@ -450,7 +517,9 @@ class TestProcessOAInspireParcelAssignment:
         )
         postcodes = ["A", "B"]
 
-        fragments = process_oa(oa_geom, points, postcodes, inspire_candidates=[left, right])
+        fragments = process_oa(
+            oa_geom, points, postcodes, inspire_candidates=[left, right]
+        )
         frag_dict = dict(fragments)
 
         assert "A" in frag_dict and "B" in frag_dict
@@ -494,7 +563,9 @@ class TestProcessOAInspireParcelAssignment:
         )
         postcodes = ["A", "B"]
 
-        fragments = process_oa(oa_geom, points, postcodes, inspire_candidates=[left, right])
+        fragments = process_oa(
+            oa_geom, points, postcodes, inspire_candidates=[left, right]
+        )
         frag_dict = dict(fragments)
 
         assert "A" in frag_dict and "B" in frag_dict
@@ -539,7 +610,9 @@ class TestProcessOAInspireParcelAssignment:
         )
         postcodes = ["A", "B"]
 
-        fragments = process_oa(oa_geom, points, postcodes, inspire_candidates=[straddling])
+        fragments = process_oa(
+            oa_geom, points, postcodes, inspire_candidates=[straddling]
+        )
 
         for _, geom in fragments:
             assert geom.difference(oa_geom).area < 0.01
@@ -651,6 +724,22 @@ class TestToWgs84Geojson:
             assert lon_dp <= 6, f"Longitude {lon_s} has {lon_dp} decimal places"
             assert lat_dp <= 6, f"Latitude {lat_s} has {lat_dp} decimal places"
 
+    def test_write_district_geojson_replaces_stale_units(self, tmp_path):
+        stale_units = tmp_path / "units"
+        stale_units.mkdir()
+        (stale_units / "ZZ1.geojson").write_text(
+            json.dumps({"type": "FeatureCollection", "features": []})
+        )
+
+        file_count = write_district_geojson(
+            {"AA1 1AA": box(530000, 180000, 530100, 180100)}, tmp_path
+        )
+
+        assert file_count == 1
+        assert not (stale_units / "ZZ1.geojson").exists()
+        written = json.loads((stale_units / "AA1.geojson").read_text())
+        assert written["features"][0]["properties"]["postcodes"] == "AA1 1AA"
+
 
 # ---------------------------------------------------------------------------
 # Edge case: parse_gpkg_geometry rejects unknown envelope types