geopandas

Use gpd.sjoin(gdf1, gdf2, predicate='intersects') to join two GeoDataFrames based on spatial relationships like intersects, within, or contains. Both GeoDataFrames must share the same CRS, so reproject with to_crs() first if needed.

Use geopandas.read_file() with the file path, URL, or even a ZIP archive. You can filter during reading with bbox, mask, rows, or where parameters, and pass use_arrow=True with pyarrow installed for 2-4x faster I/O.

set_crs() only assigns CRS metadata without changing coordinates, and should be used only when CRS information is missing. to_crs() actually transforms coordinates to a different coordinate reference system, such as reprojecting from EPSG:4326 to EPSG:3857.

Yes, use gpd.read_postgis() with a SQLAlchemy engine to run SQL queries against PostGIS, and gdf.to_postgis() to write results back with replace, append, or fail modes. This requires psycopg2 and geoalchemy2 installed.

Area and distance calculations in a geographic CRS like EPSG:4326 return values in degrees, which are meaningless. Reproject to a projected CRS such as a UTM zone or EPSG:3857 using to_crs() before calling .area or .length.

Call gdf.explore() with folium installed to generate an interactive Leaflet map, optionally coloring by a column and adding tooltips. Save the result with m.save('map.html') for sharing in a browser.