idk

pipeline/download/test_places.py

@@ -1,15 +1,22 @@
+import numpy as np
 import polars as pl
 from pyproj import Transformer
+from scipy.spatial import cKDTree
 
 from pipeline.download.places import (
     _assign_london_display_city,
+    _build_street_places,
     _display_city_from_tags,
     _is_dlr_station,
     _is_tram_station,
     _naptan_dlr_stations,
+    _normalize_street_name,
     _ofs_universities,
+    _outcode_of_postcode,
+    _pois_to_places,
     _select_university_name,
     _station_display_name,
+    _street_centroid,
 )
 
 WGS84_TO_BNG = Transformer.from_crs("EPSG:4326", "EPSG:27700", always_xy=True)
@@ -168,6 +175,73 @@ def test_ofs_universities_extracts_university_title_rows_with_postcode_coords():
     ]
 
 
+def test_street_centroid_averages_vertices():
+    assert _street_centroid([(51.0, -0.1), (53.0, -0.3)]) == (52.0, -0.2)
+    assert _street_centroid([]) is None
+
+
+def test_normalize_street_name_and_outcode():
+    assert _normalize_street_name("  High   Street ") == "high street"
+    assert _outcode_of_postcode("NW1 6XE") == "NW1"
+    assert _outcode_of_postcode("") == ""
+
+
+def test_build_street_places_groups_segments_by_name_and_outcode():
+    # Two postcodes: NW1 (north) and CR0 (south).
+    tree = cKDTree(np.array([[51.53, -0.14], [51.37, -0.10]], dtype=np.float64))
+    outcodes = ["NW1", "CR0"]
+
+    streets = [
+        {"name": "High Street", "lat": 51.531, "lon": -0.141},  # NW1
+        {"name": "High Street", "lat": 51.529, "lon": -0.139},  # NW1 (same road, 2nd segment)
+        {"name": "High Street", "lat": 51.371, "lon": -0.101},  # CR0 (different road, same name)
+        {"name": "Baker Street", "lat": 51.5305, "lon": -0.1405},  # NW1
+    ]
+
+    places = _build_street_places(streets, tree, outcodes)
+
+    # 3 distinct streets: High Street/NW1 (2 segments merged), High Street/CR0, Baker Street/NW1.
+    assert len(places) == 3
+    assert all(place["place_type"] == "street" for place in places)
+
+    nw1_high = next(
+        place
+        for place in places
+        if place["name"] == "High Street" and place["lat"] > 51.5
+    )
+    assert nw1_high["lat"] == (51.531 + 51.529) / 2
+    assert nw1_high["lon"] == (-0.141 + -0.139) / 2
+    # The same-named CR0 road stays separate.
+    assert any(
+        place["name"] == "High Street" and place["lat"] < 51.4 for place in places
+    )
+
+
+def test_pois_to_places_keeps_high_value_named_pois_only():
+    pois = pl.DataFrame(
+        {
+            "name": ["Hyde Park", "St Thomas' Hospital", "Joe's Cafe", "Hyde Park", ""],
+            "category": [
+                "leisure/park",
+                "amenity/hospital",
+                "amenity/cafe",  # everyday amenity → excluded
+                "leisure/park",  # duplicate → excluded
+                "leisure/park",  # empty name → excluded
+            ],
+            "lat": [51.507, 51.498, 51.5, 51.507, 51.6],
+            "lng": [-0.165, -0.118, -0.1, -0.165, -0.2],
+        }
+    )
+
+    places = _pois_to_places(pois)
+
+    assert [(place["name"], place["place_type"]) for place in places] == [
+        ("Hyde Park", "park"),
+        ("St Thomas' Hospital", "hospital"),
+    ]
+    assert all(place["travel_destination"] is False for place in places)
+
+
 def test_display_city_from_tags_uses_explicit_london_context():
     assert _display_city_from_tags({"is_in": "Croydon, London, UK"}) == "London"
     assert _display_city_from_tags({"is_in": "Croydon, Cambridgeshire, UK"}) is None