How to organize a python project with poetry for a command line script

Question

When doing a poetry init I got the following structure:

/packagename
  __init__.py
  packagename.py
 /packagename
  /tests
   __init__.py
   test_packagename.py
pyproject.toml

which is fine for a package but I do not see how to make it fit a command line script. When I have a script like script.py with the following code structure:

In file script.py:

#!/usr/bin/python3

def main():
  print("Ok")

if __name__ == '__main__':
  main()

It is not intended to be used as a python module however, it may have dependencies and tests that are interesting to be handle with poetry.

In some example it is shown that with the following poetry syntax:

[tool.poetry.scripts]
cli_script = 'script.py:main'

then one can call the script with:

poetry run cli_script

I am looking for some guideline on howto organize properly my poetry project for such usage.

I have looked for option for poetry init (like poetry init --script) for instance. But it seems that kind of use case was not covered in the new/init poetry options.

Answer 1

By "poetry init" I guess you mean poetry new. But also then your structure looks a bit weird. I would suggest the following structure:

packagename
├── packagename
│   ├── __init__.py
│   └── cli.py
├── tests
│   ├── __init__.py
│   └── test_packagename.py
└── pyproject.toml

The pyproject.toml looks like this:

[tool.poetry]
name = "packagename"
version = "0.1.0"
description = ""
authors = ["finswimmer <[email protected]>"]

[tool.poetry.scripts]
cli_script = "packagename.cli:main"

[tool.poetry.dependencies]
python = "^3.6"

[tool.poetry.dev-dependencies]
pytest = "^5.2"

[build-system]
requires = ["poetry>=1.0"]
build-backend = "poetry.masonry.api"

You cli.py as in your example:

#!/usr/bin/python3

def main():
  print("Ok")

if __name__ == '__main__':
  main()

After a poetry install you can run poetry run cli_script.