Polars ST provides spatial operations on Polars DataFrames, Series and Expressions. Just like Shapely and Geopandas, it make use of the library GEOS, meaning that its API is mostly identical to theirs.
>>> import polars as pl
>>> import polars_st as st
>>> gdf = st.GeoDataFrame([
... "POINT (0 0)",
... "LINESTRING (0 0, 1 1, 2 2)",
... "POLYGON ((0 0, 1 0, 1 1, 0 1, 0 0))",
... ])
>>> gdf.select(st.centroid().st.to_geojson())
┌──────────────────────────────────────────┐
│ geometry │
│ --- │
│ str │
╞══════════════════════════════════════════╡
│ {"type":"Point","coordinates":[0.0,0.0]} │
│ {"type":"Point","coordinates":[1.0,1.0]} │
│ {"type":"Point","coordinates":[0.5,0.5]} │
└──────────────────────────────────────────┘Geometries are stored as EWKB in regular Polars Binary columns. EWKB is a extension to the WKB standard popularized by PostGIS, that can also store information about the CRS of each geometry as an integer code called SRID.
For every spatial operations, the WKB binary blob will be parsed into a Geometry object so the operation can be done. If the operation result is a geometry, it will then be serialized back to EWKB. Because of that round-trip, some operations might turn out to be slower than GeoPandas. In most cases however, the performance penalty of that round-trip will be marginal compared to the spatial operation, and you will fully benefit from the parallelization capabilities of Polars.
GeoPolars is an incredibly promising tool for manipulating geographic data in Polars based on GeoArrow that likely will outperform this library's performance by a long shot. It however seems to be quite a long way from being ready and feature-complete, mostly due to Polars lack of support for Arrow Extension Types and subclassing of core datatypes.
Storing geometry as EWKB and delegating core functionality to GEOS allows polars-st to be ready now, and provide additional features such as XYZM coordinates, curved geometry types and per-geometry CRS information.
I really hope Geopolars get there soon, and that maybe some of the API design explorations made here will have helped make it even more pleasant to use.
This project is not affiliated with Polars. The design language was made very close to that of Polars because I found it amazingly appealing and liked the challenge of adding geographic meaning to it, and to also hilight the fact that this project is an exclusive extension to Polars.
In order to save myself (and yourself) from the burden of GEOS version-specific features and bugs, polars-st statically link to GEOS main. This means that all documented features are guaranteed to be available. This also means that this project is LGPL licensed, the same way GEOS is.
Copyright (C) 2024 Aurèle Ferotin
This library is free software; you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation; either version 2.1 of the License, or (at your option) any later version.
This library is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with this library; if not, write to the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA