DFO Data Commons Platform

ONE Data's fork of Data Commons. It runs as a custom Data Commons instance backed by Spanner, and serves the knowledge graph that the DFO's APIs and front ends read from.

New here? Read CONTRIBUTING-DFO.md before opening a branch. It covers the branch model, how upstream syncs work, and how to release.

Layout

• custom_dc/ is ONE Data's DFO code. custom_dc/dfo/prototypes is the frontend, and custom_dc/api_server holds a legacy copy of the Search API that stays until its removal PR lands. The Search API and Datasets API are built in apis. Upstream never touches these paths.
• Everything else tracks upstream datacommonsorg/website, with the graph server (mixer) and ingestion (import) as submodules.

Quickstart

just setup       # one-time: env file, Docker auth, submodules, git config
just dev         # hot-reload dev loop (DC services in Docker, frontend native)
just run-local   # run the full container locally

just on its own lists every target.

Deploying

just deploy                    # build and push a dfo-dev-<sha> image
just DEPLOY_ENV=prod deploy    # build and push a dfo-prod-<sha> image

deploy prints the command to run from the sibling infra repo, which applies it to Cloud Run. See CONTRIBUTING-DFO.md for the full release flow.

Relationship to upstream

This is a long-lived fork. We take changes from datacommonsorg/website by merging upstream/master into dev; we do not send changes back. This README replaces upstream's, and a merge=ours rule in .gitattributes keeps it ours across syncs. Register the rule once with just setup, or directly with git config merge.ours.driver true.

For the base Data Commons platform, see the upstream docs and developer guide.

License

TBD