Migrate from Poetry to uv

[skearnes]

Summary

• Replace Poetry with uv for Python dependency management
• Convert pyproject.toml from [tool.poetry] layout to PEP 621 [project] + [dependency-groups]
• Regenerate lockfile (poetry.lock → uv.lock)
• Update both Dockerfiles, entrypoint.sh, and the tests / checks workflows to use uv sync / uv run
• Promote greenlet to a first-class dependency (previously injected via poetry add greenlet in CI)
• Drop Python 3.10 / 3.11 from the tests.yml matrix — they never satisfied requires-python = ">=3.12,<3.13" and were silently failing under the old Poetry config
• Pin astral-sh/setup-uv to a commit SHA (SonarCloud hotspot)

Notes

• [tool.uv] package = false — we don't build/publish ord_app as a wheel; uv sync installs only dependencies, matching the previous poetry install --no-root behavior.
• Version specifiers translated from Poetry carets (^x.y.z) to PEP 440 ranges (>=x.y.z,<next). python = "~3.12" became requires-python = ">=3.12,<3.13".
• Dockerfiles set UV_PROJECT_ENVIRONMENT=/usr/local so uvicorn/alembic land on PATH, letting entrypoint.sh and supervisord.conf call them directly (no wrapper needed).
• Dockerfile.single uses uv sync --no-dev for production; Dockerfile (dev image) syncs the full env including the dev group.

Test plan

• uv sync --frozen succeeds locally
• uv run ruff check ord_app/ passes
• docker build -f Dockerfile . succeeds; resulting image has uvicorn/alembic on PATH and import ord_app, fastapi, sqlalchemy, alembic works
• docker build -f Dockerfile.single . succeeds; resulting image has uvicorn/alembic on PATH, imports work, and pytest is correctly absent (--no-dev)
• uv run pytype -j auto passes in CI
• uv run alembic upgrade head succeeds in CI against Postgres
• uv run pytest -vv --cov=ord_app passes in CI
• Running containers serve traffic end-to-end (not exercised locally — no Postgres wired up)

🤖 Generated with Claude Code

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud