Introduced alongside , the Pipfile is a modern, superior replacement for requirements.txt . It leverages the TOML (Tom's Obvious, Minimal Language) format to provide a more robust, human-readable, and deterministic way to manage your project’s dependencies. What is a Pipfile?
Here are some best practices to keep in mind when using Pipfile:
Pipfile is primarily used with , a dependency and virtual environment manager that brings together the functionality of pip and virtualenv into a single, unified tool.
[[source]] url = "https://pypi.org/simple" verify_ssl = true name = "pypi" Pipfile
If you don't have one yet, running any install command creates it.
[[source]] url = "https://private-repo.yourcompany.com/simple" verify_ssl = true name = "private"
is a human-readable, TOML-formatted file that declares your project's dependencies. It replaces the traditional requirements.txt file with a more powerful and flexible format. Introduced alongside , the Pipfile is a modern,
This section isolates your development and testing tools. Packages listed here—such as test runners ( pytest ), code formatters ( black ), or linters ( flake8 )—will not be installed when deploying your application to a production environment. This keeps production containers small and secure. 4. [requires]
You can specify different sources for different packages, allowing you to use private repositories alongside PyPI:
Historically, Python developers relied on requirements.txt files generated via pip freeze . While simple, this approach presents significant engineering flaws: Here are some best practices to keep in
[[source]] url = "https://pypi.org" verify_ssl = true name = "pypi" [packages] requests = "*" fastapi = "==0.100.0" alembic = version = ">1.7.0", extras = ["tz"] [dev-packages] pytest = ">=7.0.0" black = "*" [requires] python_version = "3.11" Use code with caution. 1. [[source]]
This section specifies the required environment constraints, most notably the target Python version. This prevents team members from accidentally running the project on an incompatible version of Python. Specifying Package Versions
: You can define loose constraints (e.g., "any version above 2.0") in the Pipfile, while the Pipenv lock file