How unidata/metpy works
Project type and purpose
• MetPy is an open-source, full-featured Python library for meteorological data: reading, calculating, and visualizing weather data. It is not a web app or microservice suite but a modular library intended for researchers and educators.
File structure
• Top‐level:
• src/metpy: core library code, organized into subpackages: calc, interpolate, io, plots, remote, units, constants, etc.
• docs: Sphinx documentation source; build scripts, templates, static assets.
• examples, tutorials: usage scripts demonstrating core functionality.
• tests: unit tests mirroring library subpackages.
• tools: CLI and code-quality scripts (flake8 plugin, nexrad spec parser).
• CI config: GitHub Actions workflows under .github; devcontainer definitions; coverage & linting settings.
• Packaging: pyproject.toml, setup.cfg, MANIFEST.in.
• staticdata: sample datasets for testing or examples.
README for additional insights
• Core functionality: reading (io), calculations (calc), interpolation, plotting (matplotlib & cartopy), remote data access (AWS).
• Dependencies: numpy, scipy, pandas, pint, xarray, matplotlib, optional pyproj.
• Philosophy: modular building blocks that “plug” into scientific Python ecosystem.