Generate random number of centroid point for each input geometry #GPKG #sqlite #ogr2ogr #python

README.md

Generate Random Number of Centroid Points For Each Geometry

Execute the following to generate a GeoPackage with multiple centroids per input geometry:

mkdir -p data
wget https://service.pdok.nl/cbs/pc4/atom/v1_0/downloads/cbs_pc4_2020.gpkg.zip -O data/cbs_pc4_2020.gpkg.zip
unzip data/cbs_pc4_2020.gpkg.zip -d data/
ogr2ogr -f GPKG data/temp.gpkg data/cbs_pc4_2020.gpkg -sql "select ST_Centroid(geom) as geom,postcode from cbs_pc4_2020" -nln cbs_pc4_2020
./generate_random_nr_centroids.py data/temp.gpkg data/output.gpkg cbs_pc4_2020 geom postcode

generate_random_nr_centroids.py

#!/usr/bin/env python3
import sqlite3
import random
from shutil import copyfile
import argparse
import pathlib


def main(input_gpkg, output_gpkg, table_name, geom_name, attributes):
    copyfile(input_gpkg, output_gpkg)
    src_con = sqlite3.connect(input_gpkg)
    cur = src_con.cursor()
    dest_con = sqlite3.connect(output_gpkg)
    dest_cur = dest_con.cursor()
    dest_con.enable_load_extension(True)
    attributes_string = ",".join(attributes)
    dest_con.execute("SELECT load_extension('mod_spatialite')")
    count = 0
    for row in cur.execute(f"SELECT hex({geom_name}),{attributes_string} FROM {table_name}"):
        random_number = random.randint(1, 10)
        for i in range(random_number):
            attributes_value_string = ""
            for i in range(len(attributes)):
                attributes_value_string += f"{row[i+1]},"
            attributes_value_string = attributes_value_string[:-1]
            insert_statement = f"INSERT INTO {table_name} ({geom_name},{attributes_string}) VALUES (X'{row[0]}',{attributes_value_string})"
            dest_cur.execute(insert_statement)
            count += 1
    dest_con.commit()
    dest_con.close()
    print(f"total number of duplicated features: {count}")


if __name__ == "__main__":
    parser = argparse.ArgumentParser()
    parser.add_argument('input_gpkg', type=pathlib.Path)
    parser.add_argument('output_gpkg', type=pathlib.Path)
    parser.add_argument('table_name')
    parser.add_argument('geom_name')
    parser.add_argument('attributes', action='store')
    args = parser.parse_args()
    input_gpkg = args.input_gpkg
    output_gpkg = args.output_gpkg
    table_name = args.table_name
    geom_name = args.geom_name
    attributes = args.attributes.split(',')
    main(input_gpkg, output_gpkg, table_name, geom_name, attributes)