Contributing

Dev setup¶

git clone https://github.com/RLE-Assessment/rle-python
cd rle-python
pip install -e .            # editable install with core dependencies

Add extras as needed for map/visualization work, e.g. pip install -e ".[viz]".

Running tests¶

The test suite (marked unit) is fast and needs no Earth Engine credentials. pytest is provided by the dev dependency group:

pip install --group dev     # or: pip install pytest
pytest

Building these docs¶

The documentation in docs/ is written for MyST and contains executable {code-cell} blocks that run against the package and the bundled docs/data/null_island.geojson sample. Install the CLI and a Jupyter kernel, then preview locally:

npm install -g mystmd
pip install jupyter-server ipykernel
python -m ipykernel install --user --name python3

cd docs
myst start                    # live preview at http://localhost:3000
myst build --html --execute   # execute cells + build static HTML in _build/html/

myst build --html --execute is what CI runs (see .github/workflows/deploy-docs.yml); it fails the build if any code cell errors, which keeps the examples honest.

Layout¶

docs/
├── myst.yml             # MyST project config + TOC
├── index.md             # landing page
├── installation.md
├── quickstart.md
├── ecosystems.md
├── assessment.md
├── ecosystem-codes.md
├── cli-and-backends.md
├── api.md
├── concepts.md
├── contributing.md
└── data/
    └── null_island.geojson   # sample dataset used by the executable examples

Style¶

• Keep code samples runnable against the current main branch. Prefer executable {code-cell} blocks for light, local operations; use static python blocks for anything that needs the cloud, Earth Engine, viz widgets, or writes files.

• Use MyST cross-references ([](page.md#anchor)) rather than raw URLs for in-repo links — they are checked at build time.