v0.2.2

Patch release rolling up all changes since v0.2.0 (the prior release; 0.2.1 was never published).

Install / upgrade:

pip install --upgrade rle-python

Features

• Add Criterion B status classification from EOO/AOO (#10).
• Push ecosystem filters down to the parquet and read gs:// via pyarrow for faster, lower-memory remote reads (#9).

Fixes

• Fix gs:// cache detection in _remote_file_exists (#8).
• Raise a clear error listing valid columns when a configured column is missing.

Performance

• Reduce peak memory in AOOGridVectorLocal._compute.

Maintenance / docs

• Bump GitHub Actions to Node 24 runtimes.
• Docs: pixi environment for executing cells, remote-GeoParquet quickstart, executed-cell output styling.

Docs: https://rle-assessment.github.io/rle-python/

v0.2.0

First PyPI release of rle-python — core tools for IUCN Red List of Ecosystems (RLE) analysis with local and cloud data access.

Install:

pip install rle-python

• RLE data model (Ecosystems, AOOGrid, EOO) and assessment logic (Extent of Occurrence, Area of Occupancy grids, ecosystem codes).
• Local and cloud-file data access (gs://, s3:// via fsspec); optional viz, gcs, aws extras.
• rle CLI and a pluggable backend registry (Earth Engine support lives in the separate rle-python-gee).

Docs: https://rle-assessment.github.io/rle-python/