// the find
opengeos/geospatial
A Python package for installing commonly used packages for geospatial analysis and data visualization with only one command.
A meta-package from the geemap/leafmap author that installs a stack of commonly-paired geospatial Python libraries (geopandas, rasterio, xarray, geemap, leafmap, etc.) with a single pip/conda install. It's aimed at people setting up a geospatial Python environment for teaching, workshops, or Colab/Binder notebooks who don't want to hand-pick a dozen packages.
Solves a real pain point — geospatial Python dependency hell (GDAL/rasterio/fiona especially) is notoriously fragile, and bundling known-compatible versions saves a lot of trial and error. CI matrix covers macOS/Ubuntu/Windows plus Docker and PyPI/conda-forge publishing, so the install story is actually tested across platforms, not just claimed. Dependabot is wired in to keep pins current, and the maintainer (giswqs) is also the author of several of the bundled libraries, so compatibility isn't guesswork.
There's barely any code here — geospatial/geospatial.py is a thin wrapper, so you're really just trusting a requirements list, not adopting a library; a pinned environment.yml would do the same job without the package indirection. Pulling in the full stack means installing heavy, sometimes platform-finicky deps (GDAL-based packages) you may not need, which can balloon install time and still break on uncommon platforms despite the CI matrix. Tests (tests/test_geospatial.py) look minimal and likely just check that imports succeed, not that the pinned versions actually interoperate correctly. Useful for spinning up a notebook environment fast, but not something you'd want in a production dependency tree — it's a convenience tool, not infrastructure.