fhir-to-bigquery

Stateless Python ETL tool for streaming FHIR Bulk Data Export ($export) payloads into BigQuery.

Implementation details live in DESIGN.md. Current repo stage: CI and packaging scaffold only.

Development

Requires Python from .python-version and uv.

uv sync --all-groups
uv run ruff format .
uv run ruff check .
uv run ty check .
uv run pytest

Docker

The image uses a multi-stage build. uv and build tooling stay in the builder stage; the runtime image contains the virtual environment and runs the installed console script from /app/.venv/bin.

Build locally:

docker build -t fhir-to-bigquery .
docker run --rm fhir-to-bigquery

CI

GitHub Actions workflows live in .github/workflows.

Workflows:

• Checks: installs deps with uv, checks formatting with ruff, lints with ruff, type-checks with ty, runs unit tests with pytest.
• Docker: on pushes and v* tags, builds image and publishes to GitHub Container Registry (ghcr.io/<owner>/<repo>).

Workflow actions use the latest tagged versions and are pinned to full commit hashes with source tag comments.

Image tags:

• Branch/commit pushes: long commit SHA tag.
• Git tag pushes matching v*: Git tag name.

Image labels include:

• org.opencontainers.image.revision: commit SHA.
• org.opencontainers.image.version: Git tag when present, else commit SHA.