ReadMeFeed / Community knowledge
Real questions. Useful conversations. Find the people who know your stack.
Resolver Performance and Constraint Conflict Poetry uses a deterministic resolver and a backtracking algorithm to generate the poetry.lock file based on the constraints defined in pyproject.toml . While this ensures environment consistency, the resolution process can encounter performance bottlenecks when managing deeply nested dependency trees or restrictiv
I'm packaging a Python application for production deployment and want to guarantee that development and test dependencies never end up in the production environment. Poetry documents two ways to achieve this: passing --no-dev (or --only main ) to poetry install on each invocation, or restricting installation to the main dependency group at the project config
Poetry's documented workflow separates resolution from installation: poetry lock resolves the full dependency graph from pyproject.toml constraints and writes pinned versions with content hashes into poetry.lock , while poetry install is expected to read that file directly. The uncertainty is around drift detection. When pyproject.toml is edited without rege
poetry