Saving additional info zone helper

cea/datamanagement/zone_helper.py

@@ -72,12 +72,6 @@ def assign_attributes(shapefile, buildings_height, buildings_floors, buildings_h
               'if we do not find data in OSM for a particular building, we estimate it based on the number of floors, '
               'multiplied by a pre-defined floor-to-floor height (set to 3m by default)')
 
-        # Make sure relevant OSM parameters (if available) are passed as floats, not strings
-        OSM_COLUMNS = ['building:min_level', 'min_height', 'building:levels', 'height']
-        selected_columns = list(set(list_of_columns).intersection(set(OSM_COLUMNS)))
-        shapefile[selected_columns] = shapefile[selected_columns] \
-            .fillna(1).apply(lambda x: pd.to_numeric(x, errors='coerce'))
-
         # Make sure relevant OSM parameters (if available) are passed as floats, not strings
         OSM_COLUMNS = ['building:min_level', 'min_height', 'building:levels', 'height']
         shapefile[[c for c in OSM_COLUMNS if c in list_of_columns]] = \
@@ -191,6 +185,46 @@ def assign_attributes(shapefile, buildings_height, buildings_floors, buildings_h
 
     return shapefile
 
+def assign_attributes_additional(shapefile):
+    """
+    This script fills the zone.shp file with additional information from OSM,
+    including house number, street name, postcode, if HDB (for Singapore), city, country
+    """
+    # TODO: include different terms used by OSM in different countries
+    #  As of 19 Sept 2023, this function works fine in Switzerland and Singapore
+
+    # local variables
+    no_buildings = shapefile.shape[0]
+    list_of_columns = shapefile.columns
+
+    # Check which attributes OSM has (sometimes it does not have any) and create the column if missing
+    if 'addr:city' not in list_of_columns:
+        shapefile['addr:city'] = [''] * no_buildings
+    if 'addr:country' not in list_of_columns:
+        shapefile['addr:country'] = [''] * no_buildings
+    if 'addr:postcode' not in list_of_columns:
+        shapefile['addr:postcode'] = [''] * no_buildings
+    if 'addr:street' not in list_of_columns:
+        shapefile['addr:street'] = [''] * no_buildings
+    if 'addr:housename' not in list_of_columns:
+        shapefile['addr:housename'] = [''] * no_buildings
+    if 'addr:housenumber' not in list_of_columns:
+        shapefile['addr:country'] = [''] * no_buildings
+    if 'residential' not in list_of_columns:
+        shapefile['residential'] = [''] * no_buildings     #not HDB
+
+    # Assign the cea-formatted columns with attributes
+    shapefile['house_no'] = shapefile['addr:housenumber']
+    shapefile['street'] = shapefile['addr:street']
+    shapefile['postcode'] = shapefile['addr:postcode']
+    shapefile['house_name'] = shapefile['addr:housename']
+    shapefile['resi_type'] = shapefile['residential']
+    shapefile['city'] = shapefile['addr:city']
+    shapefile['country'] = shapefile['addr:country']
+
+    return shapefile
+
+
 
 def fix_overlapping_geoms(buildings, zone):
     """
@@ -482,6 +516,10 @@ def polygon_to_zone(buildings_floors, buildings_floors_below_ground, buildings_h
     shapefile = assign_attributes(shapefile, buildings_height, buildings_floors,
                                  buildings_height_below_ground, buildings_floors_below_ground, key="B")
 
+    # adding additional information from OSM
+    # (e.g. house number, street number, postcode, if HDB for Singapore buildings)
+    shapefile = assign_attributes_additional(shapefile)
+
     # fix geometries of buildings with overlapping polygons
     if fix_overlapping is True:
         print("Fixing overlapping geometries.")
@@ -498,7 +536,7 @@ def polygon_to_zone(buildings_floors, buildings_floors_below_ground, buildings_h
     # clean up attributes
     cleaned_shapefile = shapefile[
         ["Name", "height_ag", "floors_ag", "height_bg", "floors_bg", "description", "category", "geometry",
-         "REFERENCE"]]
+         "REFERENCE", 'house_no', 'street', 'postcode', 'house_name', 'resi_type', 'city', 'country']]
     cleaned_shapefile = cleaned_shapefile.to_crs(get_projected_coordinate_system(float(lat), float(lon)))
 
     # save shapefile to zone.shp

cea/schemas.yml

@@ -17933,6 +17933,48 @@ get_zone_geometry:
         unit: 'NA'
         values: alphanumeric
         nullable: true
+      house_no:
+        description: House number in the address(if any)
+        type: string
+        unit: 'NA'
+        values: alphanumeric
+        nullable: true
+      street:
+        description: Street name in the address (if any)
+        type: string
+        unit: 'NA'
+        values: alphanumeric
+        nullable: true
+      postcode:
+        description: Postal code in the address (if any)
+        type: string
+        unit: 'NA'
+        values: alphanumeric
+        nullable: true
+      house_name:
+        description: House number (if any)
+        type: string
+        unit: 'NA'
+        values: alphanumeric
+        nullable: true
+      resi_type:
+        description: Residential type, e.g. HDB for Singapore (if any)
+        type: string
+        unit: 'NA'
+        values: alphanumeric
+        nullable: true
+      city:
+        description: City name in the address (if any)
+        type: string
+        unit: 'NA'
+        values: alphanumeric
+        nullable: true
+      country:
+        description: Country 2-digit code in the address (if any)
+        type: string
+        unit: 'NA'
+        values: alphanumeric
+        nullable: true
   used_by:
   - archetypes_mapper
   - decentralized

cea/utilities/standardize_coordinates.py

@@ -15,7 +15,7 @@
 
 def shapefile_to_WSG_and_UTM(shapefile_path):
 
-    # read the CPG and replace whateever is there with ISO, and save
+    # read the CPG and replace whatever is there with ISO, and save
     ensure_cpg_file(shapefile_path)
 
     # get coordinate system and project to WSG 84
@@ -65,4 +65,4 @@ def get_lat_lon_projected_shapefile(data):
     data = data.to_crs(get_geographic_coordinate_system())
     lon = data.geometry[0].centroid.coords.xy[0][0]
     lat = data.geometry[0].centroid.coords.xy[1][0]
-    return lat, lon
+    return lat, lon