<2-3 sentences summarising what the project addresses and solves, the target audience, and the benefit.> xarray-spatial delivers over 150 spatial analysis algorithms for xarray DataArrays, targeting GIS professionals and researchers. It provides fast, accurate, and scalable raster operations across CPU, Dask, and GPU backends, notably without GDAL/GEOS dependencies.

How It Works

The library leverages Numba for JIT compilation, Dask for parallel/out-of-core processing, and CuPy for GPU acceleration. Functions automatically dispatch to the most efficient backend. This pure Python/Numba approach bypasses complex C/C++ dependencies like GDAL/GEOS for both raster I/O and core analysis.

Quick Start & Requirements

Installation via pip install xarray-spatial or conda install -c conda-forge xarray-spatial. Core dependencies include NumPy, Numba, SciPy, Xarray, Matplotlib, and Zstandard. Optional dependencies for GPU (CuPy, libnvcomp), Dask, and cloud storage (fsspec) require separate installation. GPU acceleration needs compatible hardware and CUDA drivers. Starter examples and data can be downloaded via CLI commands.

Highlighted Details

• Extensive Functionality: Over 150 algorithms for hydrology, surface analysis, classification, interpolation, multispectral indices, etc.
• Multi-Backend Dispatch: Seamlessly utilizes NumPy, Dask, CuPy, and Dask+CuPy for optimized performance.
• GDAL/GEOS-Free: Native GeoTIFF/COG/VRT I/O in pure Python/Numba, simplifying setup and enabling GPU-accelerated decompression.
• Cloud & VRT Support: Reads/writes from cloud storage (S3, GCS, Azure) via fsspec and handles VRT mosaics.
• Performance: Claims pixel-exact compatibility with rasterio/GDAL and offers GPU-accelerated operations.

Maintenance & Community

The project is under a feature freeze, preparing for its v1.0.0 release. Only bug fixes, performance, and documentation updates are accepted. New features will be triaged post-release. The project seeks contributors for AI-assisted workflows and mandates adherence to a specific AI review process for all PRs.

Licensing & Compatibility

The README mentions "License" but does not specify the license type, preventing assessment of commercial use compatibility.

Limitations & Caveats

A strict feature freeze is in effect until v1.0.0. The mandatory AI-assisted contribution workflow may be a barrier. Crucially, the absence of a declared license prevents assessment of commercial usability or integration restrictions.