Skip to content

CONTRIBUTING.md

Latest commit

 

History

History
44 lines (35 loc) · 1.75 KB

CONTRIBUTING.md

File metadata and controls

44 lines (35 loc) · 1.75 KB
Raw

Contributing

T1D Engine is a personal project first, an OSS project second. PRs and issues are welcome, but scope is conservative — open an issue before building anything large.

Dev setup

Requires Python 3.12+ and uv.

uv sync                 # install dependencies
uv run pytest           # default suite (legacy tests deselected)
uv run pytest -m legacy # opt-in: quarantined v1 detection tests

There is no real-data requirement: tests run against anonymized fixtures in test_data/ and in-memory storage.

Rules that PRs are reviewed against

  • core/ stays storage-agnostic. It may import stdlib, pandas, numpy, pydantic, typing only. Backend SDKs (psycopg2, parquet I/O) are allowed only in core/storage/parquet.py, core/storage/supabase.py, and core/storage/_postgres_converters.py. Shells inject a Storage implementation; core never picks a backend.
  • No hardcoded thresholds or personal parameters. Everything tunable lives in config/user_config.yaml and is read through detection.config.get_config().
  • Real-time detection uses trailing windows only — no future BG context.
  • Do not extend or import detection/legacy/ from production code.
  • Bump ingestion.pipeline_version.PIPELINE_VERSION (with a changelog entry) when a builder/enricher changes output schema or timestamp semantics.
  • Never commit real patient data or a real .env. data/ and .env* (except .env.example) are gitignored for a reason.
  • Substantive changes add a dated docs/updates/YYYY-MM-DD-*.md entry — the append-only audit trail.

Commit style

Conventional commits (feat(scope): …, fix: …, docs: …). Keep commits focused; the test suite must pass on every commit.