openplanetdata-airflow

Shared Airflow infrastructure for OpenPlanetData workflows. Provides custom operators, shared constants, and static geospatial data for processing OpenStreetMap data pipelines.

Installation

pip install openplanetdata-airflow

Operators

GolOperator

Runs gol commands inside the OpenPlanetData Docker container, with optional output file redirection.

from openplanetdata.airflow.operators.gol import GolOperator

query = GolOperator(
    task_id="query",
    args=["query", "planet.gol", "a[boundary=administrative]"],
    output_file="/data/output.geojson",
)

Ogr2OgrOperator

Runs ogr2ogr inside a GDAL Docker container. Supports Airflow template rendering and dynamic task mapping with expand().

from openplanetdata.airflow.operators.ogr2ogr import Ogr2OgrOperator

convert = Ogr2OgrOperator(
    task_id="convert",
    args=["-f", "GPKG", "output.gpkg", "input.geojson"],
)

Shared defaults

The openplanetdata.airflow.defaults module provides shared constants:

Static data

• openplanetdata.airflow.data.continents — list of 7 continents with names and slugs
• openplanetdata.airflow.data.countries — 250 countries and territories with ISO 3166-1 alpha-2 codes and coastline flags

Docker image

The docker/ directory contains a multi-stage Dockerfile that builds an Airflow worker image with geospatial tools pre-installed:

• aria2 for fast downloads
• GDAL/OGR for format conversions
• GOL (Geodesk) for OSM boundary queries
• osmcoastline for coastline extraction
• pyosmium for OSM data updates

License

MIT