fix greenspace and add more schools

pipeline/download/os_greenspace.py

@@ -1,8 +1,13 @@
-"""Download OS Open Greenspace and extract site centroids.
+"""Download OS Open Greenspace and extract access points.
 
-Downloads the OS Open Greenspace dataset as ESRI Shapefile, computes
+Downloads the OS Open Greenspace dataset as ESRI Shapefile and extracts
-WGS84 centroids for each greenspace site polygon, and outputs a parquet
+access point locations (park entrances). Each access point is tagged with
-with lat/lng/category columns compatible with the POI proximity pipeline.
+its parent site's function type (e.g. Public Park Or Garden). Sites without
+access points fall back to polygon centroids.
+
+Using access points rather than polygon centroids gives much more accurate
+distance calculations — a property next to Hyde Park won't show 400m just
+because the centroid is in the middle of the park.
 
 Source: https://osdatahub.os.uk/downloads/open/OpenGreenspace
 License: Open Government Licence v3.0
@@ -25,6 +30,115 @@ URL = "https://api.os.uk/downloads/v1/products/OpenGreenspace/downloads?area=GB&
 _to_wgs84 = Transformer.from_crs("EPSG:27700", "EPSG:4326", always_xy=True)
 
 
+def _find_field(field_names: list[str], *needles: str) -> int | None:
+    """Find the index of the first field whose lowercased name contains any needle."""
+    for i, name in enumerate(field_names):
+        lower = name.lower()
+        for needle in needles:
+            if needle in lower:
+                return i
+    return None
+
+
+def _read_site_functions(shp_path: Path) -> dict[str, str]:
+    """Build a mapping from site ID → function type from the GreenspaceSite shapefile."""
+    reader = shp.Reader(str(shp_path), encoding="latin-1")
+    field_names = [f[0] for f in reader.fields[1:]]
+
+    id_idx = _find_field(field_names, "id")
+    func_idx = _find_field(field_names, "funct")
+    if id_idx is None or func_idx is None:
+        raise ValueError(f"Missing id/function fields. Available: {field_names}")
+
+    site_funcs = {}
+    for rec in reader.iterRecords():
+        site_funcs[rec[id_idx]] = rec[func_idx]
+
+    print(f"  Loaded {len(site_funcs):,} site function mappings")
+    return site_funcs
+
+
+def _read_access_points(
+    shp_path: Path, site_funcs: dict[str, str]
+) -> tuple[list[float], list[float], list[str]]:
+    """Read access points, tagging each with its parent site's function."""
+    reader = shp.Reader(str(shp_path), encoding="latin-1")
+    field_names = [f[0] for f in reader.fields[1:]]
+
+    # The access point shapefile has a reference field linking to the parent site
+    ref_idx = _find_field(field_names, "refto", "ref_to", "greensp")
+    if ref_idx is None:
+        raise ValueError(
+            f"No site reference field found in access points. Available: {field_names}"
+        )
+
+    lats: list[float] = []
+    lngs: list[float] = []
+    categories: list[str] = []
+    skipped = 0
+
+    for sr in reader.shapeRecords():
+        site_id = sr.record[ref_idx]
+        func = site_funcs.get(site_id)
+        if func is None:
+            skipped += 1
+            continue
+
+        try:
+            geom = to_shapely(sr.shape.__geo_interface__)
+            if geom.is_empty:
+                continue
+            lng, lat = _to_wgs84.transform(geom.x, geom.y)
+        except Exception:
+            continue
+
+        lats.append(lat)
+        lngs.append(lng)
+        categories.append(func)
+
+    if skipped:
+        print(f"  Skipped {skipped:,} access points with unknown site ID")
+
+    return lats, lngs, categories
+
+
+def _read_site_centroids(
+    shp_path: Path, site_funcs: dict[str, str], covered_ids: set[str]
+) -> tuple[list[float], list[float], list[str]]:
+    """Read polygon centroids for sites that have no access points (fallback)."""
+    reader = shp.Reader(str(shp_path), encoding="latin-1")
+    field_names = [f[0] for f in reader.fields[1:]]
+    id_idx = _find_field(field_names, "id")
+    func_idx = _find_field(field_names, "funct")
+    if id_idx is None or func_idx is None:
+        return [], [], []
+
+    lats: list[float] = []
+    lngs: list[float] = []
+    categories: list[str] = []
+
+    for sr in reader.shapeRecords():
+        site_id = sr.record[id_idx]
+        if site_id in covered_ids:
+            continue
+
+        func = sr.record[func_idx]
+        try:
+            geom = to_shapely(sr.shape.__geo_interface__)
+            if geom.is_empty or not geom.is_valid:
+                continue
+            centroid = geom.centroid
+            lng, lat = _to_wgs84.transform(centroid.x, centroid.y)
+        except Exception:
+            continue
+
+        lats.append(lat)
+        lngs.append(lng)
+        categories.append(func)
+
+    return lats, lngs, categories
+
+
 def download_greenspace(output: Path) -> None:
     output.parent.mkdir(parents=True, exist_ok=True)
 
@@ -35,77 +149,54 @@ def download_greenspace(output: Path) -> None:
         download(URL, zip_path, timeout=300)
         extract_zip(zip_path, extract_dir)
 
-        # Find the GreenspaceSite shapefile (not the AccessPoint one)
+        # Find both shapefiles
-        shp_files = list(extract_dir.rglob("*GreenspaceSite*.shp"))
+        site_shps = list(extract_dir.rglob("*GreenspaceSite*.shp"))
-        if not shp_files:
+        access_shps = list(extract_dir.rglob("*AccessPoint*.shp"))
-            shp_files = [
-                f
-                for f in extract_dir.rglob("*.shp")
-                if "AccessPoint" not in f.name
-            ]
-        if not shp_files:
-            raise FileNotFoundError(
-                "No GreenspaceSite shapefile found in download"
-            )
 
-        print(f"Reading {shp_files[0].name}...")
+        if not site_shps:
-        reader = shp.Reader(str(shp_files[0]), encoding="latin-1")
+            raise FileNotFoundError("No GreenspaceSite shapefile found")
+        if not access_shps:
+            raise FileNotFoundError("No AccessPoint shapefile found")
 
-        # Find the "function" field (greenspace type)
+        # Step 1: Build site ID → function mapping
-        field_names = [f[0] for f in reader.fields[1:]]  # skip deletion flag
+        print(f"Reading {site_shps[0].name} for function types...")
-        func_field = None
+        site_funcs = _read_site_functions(site_shps[0])
-        for name in field_names:
-            if "funct" in name.lower():
-                func_field = name
-                break
-        if func_field is None:
-            raise ValueError(
-                f"No 'function' field found. Available: {field_names}"
-            )
-        func_idx = field_names.index(func_field)
 
-        # Find a name field if available
+        # Step 2: Read access points (primary — park entrances)
-        name_idx = None
+        print(f"Reading {access_shps[0].name}...")
-        for name in field_names:
+        ap_lats, ap_lngs, ap_cats = _read_access_points(access_shps[0], site_funcs)
-            if "distname" in name.lower():
+        print(f"  {len(ap_lats):,} access points loaded")
-                name_idx = field_names.index(name)
-                break
 
-        lats = []
+        # Step 3: Fall back to centroids for sites without any access points
-        lngs = []
+        covered_ids = set()
-        categories = []
+        reader = shp.Reader(str(access_shps[0]), encoding="latin-1")
-        names = []
+        field_names = [f[0] for f in reader.fields[1:]]
+        ref_idx = _find_field(field_names, "refto", "ref_to", "greensp")
+        if ref_idx is not None:
+            for rec in reader.iterRecords():
+                covered_ids.add(rec[ref_idx])
 
-        for sr in reader.shapeRecords():
+        print("Adding centroids for sites without access points...")
-            func = sr.record[func_idx]
+        fb_lats, fb_lngs, fb_cats = _read_site_centroids(
-            site_name = sr.record[name_idx] if name_idx is not None else ""
+            site_shps[0], site_funcs, covered_ids
+        )
+        print(f"  {len(fb_lats):,} centroid fallbacks added")
 
-            try:
+    lats = ap_lats + fb_lats
-                geom = to_shapely(sr.shape.__geo_interface__)
+    lngs = ap_lngs + fb_lngs
-                if geom.is_empty or not geom.is_valid:
+    categories = ap_cats + fb_cats
-                    continue
-                centroid = geom.centroid
-                lng, lat = _to_wgs84.transform(centroid.x, centroid.y)
-            except Exception:
-                continue
-
-            lats.append(lat)
-            lngs.append(lng)
-            categories.append(func)
-            names.append(site_name or "")
 
     df = pl.DataFrame(
         {
             "lat": np.array(lats, dtype=np.float64),
             "lng": np.array(lngs, dtype=np.float64),
             "category": categories,
-            "name": names,
         }
     )
 
     df.write_parquet(output)
     size_mb = output.stat().st_size / (1024 * 1024)
-    print(f"Wrote {output} ({size_mb:.1f} MB, {len(df):,} greenspace sites)")
+    print(f"Wrote {output} ({size_mb:.1f} MB, {len(df):,} points)")
 
     counts = df.group_by("category").len().sort("len", descending=True)
     for row in counts.iter_rows(named=True):
@@ -114,7 +205,7 @@ def download_greenspace(output: Path) -> None:
 
 def main() -> None:
     parser = argparse.ArgumentParser(
-        description="Download OS Open Greenspace site centroids"
+        description="Download OS Open Greenspace access points"
     )
     parser.add_argument(
         "--output", type=Path, required=True, help="Output parquet file path"

pipeline/download/tiles.py

@@ -12,7 +12,7 @@ from io import BytesIO
 from pathlib import Path
 
 PROTOMAPS_BASE = "https://build.protomaps.com"
-UK_BBOX = "-10.5,49.5,2.5,61"
+UK_BBOX = "-10.5,49,5,61"
 MAX_AGE_DAYS = 14
 
 

pipeline/transform/merge.py

@@ -59,6 +59,8 @@ _AREA_COLUMNS = [
     # Schools
     "Good+ primary schools within 5km",
     "Good+ secondary schools within 5km",
+    "Good+ primary schools within 2km",
+    "Good+ secondary schools within 2km",
     # Demographics
     "Median age",
 ]
@@ -331,6 +333,8 @@ def _build(
                 "noise_lden_db": "Noise (dB)",
                 "good_primary_5km": "Good+ primary schools within 5km",
                 "good_secondary_5km": "Good+ secondary schools within 5km",
+                "good_primary_2km": "Good+ primary schools within 2km",
+                "good_secondary_2km": "Good+ secondary schools within 2km",
                 "max_download_speed": "Max available download speed (Mbps)",
                 "serious_crime_avg_yr": "Serious crime (avg/yr)",
                 "minor_crime_avg_yr": "Minor crime (avg/yr)",

pipeline/transform/school_proximity.py

@@ -60,9 +60,14 @@ def main():
     # Load all postcodes for proximity counting
     postcodes = arcgis.rename({"lng": "lon"})
 
-    result = count_pois_per_postcode(
+    counts_5km = count_pois_per_postcode(
         postcodes, schools, radius_km=5, groups=SCHOOL_GROUPS
     )
+    counts_2km = count_pois_per_postcode(
+        postcodes, schools, radius_km=2, groups=SCHOOL_GROUPS
+    )
+
+    result = counts_5km.join(counts_2km, on="postcode")
 
     result.write_parquet(args.output)
     size_mb = args.output.stat().st_size / (1024 * 1024)