Full Project Lifecycle with Poetry

Intermediate12 min

Manage Python project dependencies, virtual environments, versioning, and publishing with Poetry's declarative pyproject.toml workflow.

Prerequisites

-Python 3.8+ installed
-pipx or uv installed (recommended for Poetry install)

Steps

Install Poetry

Install Poetry in an isolated environment using pipx. This prevents Poetry's own dependencies from conflicting with your projects.

$ pipx install poetry

Alternatively use: curl -sSL https://install.python-poetry.org | python3 -

Do not install Poetry with pip into a project virtualenv. It should be installed globally via pipx or the official installer.

Create a new project

Scaffolds a new project with pyproject.toml, a package directory, and a tests directory. The generated structure follows Python packaging best practices.

$ poetry new myproject && cd myproject

Add production dependencies

Adds packages to the [tool.poetry.dependencies] section and installs them. The poetry.lock file is updated automatically.

$ poetry add requests flask sqlalchemy

Add development dependencies

Adds packages to the dev dependency group. These are excluded when building distribution packages.

$ poetry add --group dev pytest ruff mypy coverage

You can create custom groups like --group docs for documentation tools.

Run commands inside the environment

Execute any command within the project's virtual environment without manually activating it.

$ poetry run pytest tests/

Build and publish the package

Bump the version, build source and wheel distributions, and preview a publish to PyPI. Remove --dry-run when ready to publish for real.

$ poetry version patch && poetry build && poetry publish --dry-run

You need a PyPI account and API token configured via 'poetry config pypi-token.pypi <token>' before you can publish.

Full Script

full-script.sh

#!/bin/bash
# Full Poetry project workflow
set -e

PROJECT_NAME=${1:-myproject}

# Install Poetry if not present
if ! command -v poetry &> /dev/null; then
  pipx install poetry
fi

poetry new "$PROJECT_NAME"
cd "$PROJECT_NAME"

poetry add requests flask sqlalchemy
poetry add --group dev pytest ruff mypy coverage

poetry run pytest tests/
poetry build

echo "Project '$PROJECT_NAME' built with Poetry."
ls -la dist/

FAQ

Discussion

Loading comments...

#!/bin/bash # Full Poetry project workflow set -e PROJECT_NAME=${1:-myproject} # Install Poetry if not present if ! command -v poetry &> /dev/null; then pipx install poetry fi poetry new "$PROJECT_NAME" cd "$PROJECT_NAME" poetry add requests flask sqlalchemy poetry add --group dev pytest ruff mypy coverage poetry run pytest tests/ poetry build echo "Project '$PROJECT_NAME' built with Poetry." ls -la dist/