Extract and load addressbase/mastermap using id, gml, parallel

etl/0_extract_addressbase.sh

@@ -12,10 +12,12 @@ data_dir=$1
 # Unzip to GML
 #
 
-# find $data_dir -name '*.zip' -print0 | xargs -0 -P 4 -n 1 unzip
+find $data_dir -type f -name '*.zip' -printf "%f\n" | \
+parallel \
+unzip -u $data_dir/{} -d $data_dir
 
 #
-# Extract (subset) to CSV
+# Extract to CSV
 #
 # Relevant fields:
 # WKT
@@ -26,15 +28,17 @@ data_dir=$1
 # logicalStatus: 1 (one) is approved (otherwise historical, provisional)
 #
 
-# find $data_dir -type f -name '*.gml' -printf "%f\n"  | \
-# parallel \
-# ogr2ogr -f CSV \
-#     -select crossReference,source,uprn,parentUPRN,logicalStatus \
-#     {}.csv {} BasicLandPropertyUnit \
-#     -lco GEOMETRY=AS_WKT
-
-rm $data_dir/*.gml
-
-find $data_dir -type f -name '*.gml.csv' -printf "$data_dir%f\n"  | \
+find $data_dir -type f -name '*.gml' -printf "%f\n"  | \
 parallel \
-python filter_addressbase_csv.py {}
+ogr2ogr -f CSV \
+    -select crossReference,source,uprn,parentUPRN,logicalStatus \
+    $data_dir/{}.csv $data_dir/{} BasicLandPropertyUnit \
+    -lco GEOMETRY=AS_WKT
+
+#
+# Filter, grouping by TOID
+#
+
+find $data_dir -type f -name '*.gml.csv' -printf "%f\n"  | \
+parallel \
+python filter_addressbase_csv.py $data_dir/{}

etl/0_extract_mastermap.sh

@@ -1,5 +1,13 @@
 #!/usr/bin/env bash
 
+#
+# Extract MasterMap
+#
+
+: ${1?"Usage: $0 ./path/to/mastermap/dir"}
+
+data_dir=$1
+
 #
 # Extract buildings from *.gz to CSV
 #
@@ -9,12 +17,14 @@
 # Use `fid` as source ID, aka TOID.
 #
 
-: ${1?"Usage: $0 ./path/to/data/dir"}
-
-data_dir=$1
-
 find $data_dir -type f -name '*.gz' -printf "%f\n" | \
 parallel \
+gunzip $data_dir/{} -k -S gml
+
+rename 's/$/.gml/' $data_dir/*[^gzt]
+
+find $data_dir -type f -name '*.gml' -printf "%f\n" | \
+parallel \
 ogr2ogr \
     -select fid,descriptiveGroup \
     -f CSV $data_dir/{}.csv \
@@ -22,9 +32,4 @@ ogr2ogr \
     TopographicArea \
     -lco GEOMETRY=AS_WKT
 
-# then filter
-# -where "\"descriptiveGroup='(1:Building)'\"" \
-# OR toid in addressbase_toids
-
-# finally load
-# -t_srs "EPSG:3857" \
+rm $data_dir/*.gfs

etl/0_filter_for_loading.sh

@@ -0,0 +1,40 @@
+#!/usr/bin/env bash
+
+#
+# Filter and transform for loading
+#
+: ${1?"Usage: $0 ./path/to/addressbase/dir ./path/to/mastermap/dir"}
+: ${2?"Usage: $0 ./path/to/addressbase/dir ./path/to/mastermap/dir"}
+
+addressbase_dir=$1
+mastermap_dir=$2
+
+#
+# Check which TOIDs are matched against UPRNs
+#
+python check_ab_mm_match.py $addressbase_dir $mastermap_dir
+
+#
+# Filter
+# - WHERE descriptiveGroup = '(1:Building)'
+# - OR toid in addressbase_toids
+#
+python filter_mastermap.py $addressbase_dir $mastermap_dir
+
+#
+# Transform to 3857 (web mercator)
+#
+find $mastermap_dir -type f -name '*.filtered.csv' -printf "%f\n" | \
+parallel \
+ogr2ogr \
+    -f CSV $mastermap_dir/{}.3857.csv \
+    -select fid \
+    -s_srs "EPSG:27700" \
+    -t_srs "EPSG:3857" \
+    $mastermap_dir/{} \
+    -lco GEOMETRY=AS_WKT
+
+#
+# Update to EWKT (with SRID indicator for loading to Postgres)
+#
+sed -i 's/^"POLYGON/"SRID=3857;POLYGON/' $mastermap_dir/*.3857.csv

etl/1_load_geometries.py

@@ -1,105 +0,0 @@
-"""Load geometries from GeoJSON to Postgres
-
-  - create 'geometry' record with {
-        id: <polygon-guid>,
-        doc: {source_id: <toid>},
-        geom: <geom-wkb_hex>
-    }
-  - create corresponding 'building' record with {
-        id: <building-guid>,
-        doc: {},
-        geom_id: <polygon-guid>
-    }
-"""
-import glob
-import json
-import os
-import sys
-
-import fiona
-import psycopg2
-from shapely.geometry import shape
-
-
-def main(source_dir, config_path):
-    """Load config, read files and save features to the database
-    """
-    conf = read_config(config_path)
-    dbconf = conf['database']
-    conn = psycopg2.connect(**dbconf)
-
-    source_files = glob.glob("{}/*.geojson".format(source_dir))
-
-    loaded = {}
-    for source_file in source_files:
-        with fiona.open(source_file, 'r') as source:
-            with conn.cursor() as cur:
-                for feature in source:
-                    fid = feature['properties']['fid']
-                    if fid not in loaded:
-                        save_feature(cur, feature)
-                        loaded[fid] = True
-                    else:
-                        print("Skipping", fid)
-                conn.commit()
-    conn.close()
-
-
-
-def save_feature(cur, feature):
-    """Save a feature with geometry and source id
-    """
-    cur.execute(
-        """INSERT INTO geometries
-        (
-            geometry_doc,
-            geometry_geom
-        )
-        VALUES
-        (
-            %s::jsonb,
-            ST_SetSRID(%s::geometry, %s)
-        )
-        RETURNING geometry_id
-        """, (
-            json.dumps({
-                'source_id': feature['properties']['fid']
-            }),
-            shape(feature['geometry']).wkb_hex,
-            3857
-        )
-    )
-    geom_id, = cur.fetchone()
-    cur.execute(
-        """INSERT INTO buildings
-        (
-            building_doc,
-            geometry_id
-        )
-        VALUES
-        (
-            %s::jsonb,
-            %s
-        )
-        """, (
-            json.dumps({}),
-            geom_id
-        )
-    )
-
-
-def read_config(config_path):
-    """Read a JSON config file containing database connection details
-    """
-    with open(config_path, 'r') as conf_fh:
-        conf = json.load(conf_fh)
-    return conf
-
-
-if __name__ == '__main__':
-    if len(sys.argv) != 3:
-        print("Usage: {} ./path/to/source/dir ./path/to/dbconfig.json".format(
-            os.path.basename(__file__)
-        ))
-        exit()
-    main(sys.argv[1], sys.argv[2])

etl/1_load_geometries.sh

@@ -0,0 +1,40 @@
+#!/usr/bin/env bash
+
+#
+# Load geometries from GeoJSON to Postgres
+# - assume postgres connection details are set in the environment using PGUSER, PGHOST etc.
+#
+: ${1?"Usage: $0 ./path/to/mastermap/dir"}
+
+mastermap_dir=$1
+
+#
+# Create 'geometry' record with
+#     id: <polygon-guid>,
+#     source_id: <toid>,
+#     geom: <geom>
+#
+find $mastermap_dir -type f -name '*.3857.csv' \
+-printf "COPY geometries ( geometry_geom, source_id ) FROM '$mastermap_dir/%f' WITH CSV HEADER;\n" | \
+parallel \
+psql -c {}
+
+#
+# Delete any duplicated geometries (by TOID)
+#
+psql -c "DELETE FROM geometries a USING (
+    SELECT MIN(ctid) as ctid, source_id
+    FROM geometries
+    GROUP BY source_id
+    HAVING COUNT(*) > 1
+) b
+WHERE a.source_id = b.source_id
+AND a.ctid <> b.ctid;"
+
+#
+# Create corresponding 'building' record with
+#     id: <building-guid>,
+#     doc: {},
+#     geom_id: <polygon-guid>
+#
+psql -c "INSERT INTO buildings ( geometry_id ) SELECT geometry_id from geometries;"

etl/check_ab_mm_match.py

@@ -11,17 +11,23 @@ csv.field_size_limit(sys.maxsize)
 
 def main(ab_path, mm_path):
     ab_paths = sorted(glob.glob(os.path.join(ab_path, "*.gml.csv.filtered")))
-    mm_paths = sorted(glob.glob(os.path.join(mm_path, "*.gz.csv")))
+    mm_paths = sorted(glob.glob(os.path.join(mm_path, "*.gml.csv")))
 
+    try:
         assert len(ab_paths) == len(mm_paths)
+    except AssertionError:
+        print(ab_paths)
+        print(mm_paths)
+
     zipped_paths = zip(ab_paths, mm_paths)
 
+    # parallel map over tiles
     with Pool(4) as p:
         p.starmap(check, zipped_paths)
 
 def check(ab_path, mm_path):
     tile = str(os.path.basename(ab_path)).split(".")[0]
-    print(tile)
+    output_base = os.path.dirname(ab_path)
     ab_toids = set()
     mm_toids = set()
 
@@ -35,22 +41,20 @@ def check(ab_path, mm_path):
         for line in r:
             mm_toids.add(line['fid'])
 
-    print("MasterMap", len(mm_toids))
-    print("Addressbase", len(ab_toids))
     missing = ab_toids - mm_toids
-    print("in AB but not MM", len(missing))
+    print(tile, "MasterMap:", len(mm_toids), "Addressbase:", len(ab_toids), "AB but not MM:", len(missing))
 
-    with open('missing_toids_{}.txt'.format(tile), 'w') as fh:
+    with open(os.path.join(output_base, 'missing_toids_{}.txt'.format(tile)), 'w') as fh:
         for toid in missing:
             fh.write("{}\n".format(toid))
 
-    with open('ab_toids_{}.txt'.format(tile), 'w') as fh:
+    with open(os.path.join(output_base, 'ab_toids_{}.txt'.format(tile)), 'w') as fh:
         for toid in ab_toids:
             fh.write("{}\n".format(toid))
 
 
 if __name__ == '__main__':
     if len(sys.argv) != 3:
-        print("Usage: filter_addressbase_csv.py ./path/to/addressbase/dir ./path/to/mastermap/dir")
+        print("Usage: check_ab_mm_match.py ./path/to/addressbase/dir ./path/to/mastermap/dir")
         exit(-1)
     main(sys.argv[1], sys.argv[2])

etl/filter_mastermap.py

@@ -0,0 +1,60 @@
+"""Filter MasterMap to buildings and addressbase-matches
+
+- WHERE descriptiveGroup = '(1:Building)'
+- OR toid in addressbase_toids
+"""
+import csv
+import glob
+import json
+import os
+import sys
+
+from multiprocessing import Pool
+
+csv.field_size_limit(sys.maxsize)
+
+def main(ab_path, mm_path):
+    mm_paths = sorted(glob.glob(os.path.join(mm_path, "*.gml.csv")))
+    toid_paths = sorted(glob.glob(os.path.join(ab_path, "ab_toids_*.txt")))
+
+    try:
+        assert len(mm_paths) == len(toid_paths)
+    except AssertionError:
+        print(mm_paths)
+        print(toid_paths)
+    zipped_paths = zip(mm_paths, toid_paths)
+
+    # parallel map over tiles
+    with Pool(4) as p:
+        p.starmap(filter, zipped_paths)
+
+def filter(mm_path, toid_path):
+    with open(toid_path, 'r') as fh:
+        r = csv.reader(fh)
+        toids = set(line[0] for line in r)
+
+    output_path =  "{}.filtered.csv".format(str(mm_path).replace(".gml.csv", ""))
+    alt_output_path =  "{}.filtered_not_building.csv".format(str(mm_path).replace(".gml.csv", ""))
+    output_fieldnames = ('WKT', 'fid', 'descriptiveGroup')
+    with open(mm_path, 'r') as fh:
+        r = csv.DictReader(fh)
+        with open(output_path, 'w') as output_fh:
+            w = csv.DictWriter(output_fh, fieldnames=output_fieldnames)
+            w.writeheader()
+            with open(alt_output_path, 'w') as alt_output_fh:
+                alt_w = csv.DictWriter(alt_output_fh, fieldnames=output_fieldnames)
+                alt_w.writeheader()
+                for line in r:
+                    if 'Building' in line['descriptiveGroup']:
+                        w.writerow(line)
+
+                    elif line['fid'] in toids:
+                        alt_w.writerow(line)
+
+
+
+if __name__ == '__main__':
+    if len(sys.argv) != 3:
+        print("Usage: filter_mastermap.py ./path/to/addressbase/dir ./path/to/mastermap/dir")
+        exit(-1)
+    main(sys.argv[1], sys.argv[2])