Is it possible to efficiently get all the polygons inside a polygon from a shapefile without using PostGIS?

Question

I have a huge shapefile (over 400MB) of an entire city that contains many polygons of the residential area of this city. I can check it on QGIS as the following:

This shapefile contains a huge number of polygons, and I'd like cut a small shapefile from it in a circular or squared area around a point. So, let's say I give a point to a system, I want to return all the polygons around this area inside an offset (just so I can send only this segmentation of the data to the frontend of my application).

I know I could convert this shapefile to GeoJSON and find what are the polygons that exist inside the area that I have created (using turf.js for example), but that would be a brute force solution and it wouldn't be efficient.

On the other hand, I know I could install PostgreSQL and PostGIS and implement solutions like this one.

My doubt is that if I can achieve the goal of finding all polygons inside a polygon efficiently using only the command line, Python or JavaScript without implementing a spatial index for that, working directly with the shapefile. Is it possible?

Answer 1

You can achieve this using ogr2ogr with

# SHP from Natural Earth Data
ogr2ogr out.shp ne_110m_admin_0_countries.shp -dialect SQLite -sql "SELECT * FROM ne_110m_admin_0_countries WHERE ST_intersects(ST_Buffer(ST_GeomFromText('POINT (1 43)', 4326), 10), ne_110m_admin_0_countries.geometry);"

To sort out execution with index, you may want to add after ogr2ogr the option --debug ON

PS: the buffer units are degrees here because using EPSG 4326 units. The points coordinates are longitude 1 and latitude 43. You may need to reproject if you want to use meters or directly use a layer that use a projection based on meters.