stacrs

A small no-dependency Python package for STAC, using Rust under the hood.

Usage

Install via pip:

python -m pip install stacrs

Or via conda:

conda install conda-forge::stacrs

Then:

import stacrs

# Search a STAC API
items = stacrs.search(
    "https://landsatlook.usgs.gov/stac-server",
    collections="landsat-c2l2-sr",
    intersects={"type": "Point", "coordinates": [-105.119, 40.173]},
    sortby="-properties.datetime",
    max_items=100,
)

# Write items to a stac-geoparquet file
stacrs.write("items.parquet", items)

# Read items from a stac-geoparquet file as an item collection
item_collection = stacrs.read("items.parquet")

# Use `search_to` for better performance if you know you'll be writing the items
# to a file
stacrs.search_to(
    "items.parquet",
    "https://landsatlook.usgs.gov/stac-server",
    collections="landsat-c2l2-sr",
    intersects={"type": "Point", "coordinates": [-105.119, 40.173]},
    sortby="-properties.datetime",
    max_items=100,
)

See the documentation for details. In particular, our example notebook demonstrates some of the more interesting features.

Comparisons

This package (intentionally) has limited functionality, as it is not intended to be a replacement for existing Python STAC packages. pystac is a mature Python library with a significantly richer API for working with STAC objects. For querying STAC APIs, pystac-client is more feature-rich than our simplistic stacrs.search.

That being said, it is hoped that stacrs will be a nice complement to the existing Python STAC ecosystem by providing a no-dependency package with unique capabilities, such as searching directly into a stac-geoparquet file.

Development

Get Rust and uv. Then:

git clone [email protected]:gadomski/stacrs.git
cd stacrs
uv sync  # This will take a little while while the Rust dependencies build
uv run pytest

See CONTRIBUTING.md for more information about contributing to this project.

License

stacrs is dual-licensed under both the MIT license and the Apache license (Version 2.0). See LICENSE-APACHE and LICENSE-MIT for details.