Release v2.0 - add Netherlands POI analysis

osm-poi-import.py

@@ -0,0 +1,136 @@
+#!/usr/bin/env python3
+"""Importeer alleen benodigde POI's uit een OSM PBF naar PostGIS.
+Veel lichter dan volledige osm2pgsql-import voor heel Nederland.
+"""
+
+import argparse
+import os
+
+import osmium
+import psycopg2
+from psycopg2.extras import execute_values
+
+DB_DSN = os.environ.get(
+    "ADRES_ANALYSE_DB_DSN",
+    "host=127.0.0.1 port=5433 dbname=adresanalyse user=adres password=adres",
+)
+
+SCHEMA_SQL = """
+CREATE EXTENSION IF NOT EXISTS postgis;
+DROP TABLE IF EXISTS osm_pois;
+CREATE TABLE osm_pois (
+  id bigserial PRIMARY KEY,
+  osm_type text NOT NULL,
+  osm_id bigint NOT NULL,
+  category text NOT NULL,
+  name text NOT NULL,
+  brand text,
+  operator text,
+  lon double precision NOT NULL,
+  lat double precision NOT NULL,
+  geom geometry(Point, 4326) NOT NULL
+);
+CREATE INDEX osm_pois_category_geom_gix ON osm_pois USING GIST (geom);
+CREATE INDEX osm_pois_category_idx ON osm_pois (category);
+CREATE INDEX osm_pois_name_idx ON osm_pois (lower(name));
+
+CREATE OR REPLACE VIEW poi_supermarkten AS
+SELECT osm_id, name, 'supermarkt'::text AS category, 'supermarket'::text AS shop, NULL::text AS amenity, brand, operator, geom
+FROM osm_pois WHERE category = 'supermarkt';
+
+CREATE OR REPLACE VIEW poi_scholen AS
+SELECT osm_id, name, 'school'::text AS category, NULL::text AS shop, 'school'::text AS amenity, brand, operator, geom
+FROM osm_pois WHERE category = 'school';
+"""
+
+
+def poi_from_tags(tags):
+    if tags.get("shop") == "supermarket":
+        return "supermarkt", tags.get("name") or tags.get("brand") or "Supermarkt"
+    if tags.get("amenity") == "school":
+        return "school", tags.get("name") or "School"
+    return None, None
+
+
+class PoiHandler(osmium.SimpleHandler):
+    def __init__(self, conn, batch_size=1000):
+        super().__init__()
+        self.conn = conn
+        self.batch_size = batch_size
+        self.rows = []
+        self.count = 0
+        self.factory = osmium.geom.WKBFactory()
+
+    def flush(self):
+        if not self.rows:
+            return
+        with self.conn.cursor() as cur:
+            execute_values(
+                cur,
+                """
+                INSERT INTO osm_pois (osm_type, osm_id, category, name, brand, operator, lon, lat, geom)
+                VALUES %s
+                """,
+                self.rows,
+            )
+        self.conn.commit()
+        self.rows.clear()
+
+    def add_row(self, osm_type, osm_id, tags, lon, lat):
+        category, name = poi_from_tags(tags)
+        if not category:
+            return
+        self.rows.append((
+            osm_type,
+            int(osm_id),
+            category,
+            name,
+            tags.get("brand"),
+            tags.get("operator"),
+            float(lon),
+            float(lat),
+            f"SRID=4326;POINT({float(lon)} {float(lat)})",
+        ))
+        self.count += 1
+        if len(self.rows) >= self.batch_size:
+            self.flush()
+
+    def node(self, n):
+        if not n.location.valid():
+            return
+        self.add_row("node", n.id, n.tags, n.location.lon, n.location.lat)
+
+    def area(self, a):
+        category, _ = poi_from_tags(a.tags)
+        if not category:
+            return
+        try:
+            wkb = self.factory.create_multipolygon(a)
+            with self.conn.cursor() as cur:
+                cur.execute("SELECT ST_X(p), ST_Y(p) FROM (SELECT ST_PointOnSurface(ST_GeomFromWKB(%s, 4326)) p) s", (psycopg2.Binary(wkb),))
+                lon, lat = cur.fetchone()
+            self.add_row("area", a.orig_id(), a.tags, lon, lat)
+        except Exception:
+            return
+
+
+def main():
+    parser = argparse.ArgumentParser()
+    parser.add_argument("pbf")
+    args = parser.parse_args()
+
+    with psycopg2.connect(DB_DSN) as conn:
+        with conn.cursor() as cur:
+            cur.execute(SCHEMA_SQL)
+        conn.commit()
+        handler = PoiHandler(conn)
+        handler.apply_file(args.pbf, locations=True)
+        handler.flush()
+        with conn.cursor() as cur:
+            cur.execute("ANALYZE osm_pois")
+        conn.commit()
+        print(f"Imported {handler.count} POIs into osm_pois")
+
+
+if __name__ == "__main__":
+    main()