In issue #1 we explored in the summer of 2024 the benefits of Arrow and GeoArrow for IO and processing performance. In the end, we concluded that:
Immediately leveraging Arrow, through the very latest versions of GDAL/OGR and pyogrio (released while we were working on #1), provided clear benefits.
GeoArrow-Python was not sufficiently mature nor integrated with GeoPandas to use in our workflows, yet.
Since then, it appears that the GeoArrow team has focused their attention on their Rust implementation, https://github.com/geoarrow/geoarrow-rs, which includes their geoarrow.rust Python bindings. Version 0.3.0, released on September. 9, 2024, added substantially to features, and the active development since then suggests that more are to come (including pyogrio/GDAL integration).
In issue #1 we explored in the summer of 2024 the benefits of Arrow and GeoArrow for IO and processing performance. In the end, we concluded that:
Since then, it appears that the GeoArrow team has focused their attention on their Rust implementation, https://github.com/geoarrow/geoarrow-rs, which includes their
geoarrow.rustPython bindings. Version 0.3.0, released on September. 9, 2024, added substantially to features, and the active development since then suggests that more are to come (including pyogrio/GDAL integration).Let's keep an eye on progress and try it out.