I'm writing a python package that I would like to distribute as a standalone terminal app. The structure of the project folder is the following:
energy-monitor/
config/
doc/
tests/
energymonitor/
__init__.py -> (empty)
main.py -> def main()
data/
..other packages..
project.toml
I'm using setuptools to generate a .tar.gz archive, some relevant parts of the project.toml file are:
[build-system]
requires = ["setuptools>=68.0"]
build-backend = "setuptools.build_meta"
[project]
name = "energy-monitor"
version = "0.0.1"
...
[tool.setuptools.packages.find]
where = ["energymonitor"]
[tool.setuptools.package-data]
data = ["data/*"]
[project.scripts]
energy-monitor = "energymonitor.main:main"
I generate the .tar.gz and the .whl files with the command python -m build
, then I run pipx install path/to/energy-monitor.tar.gz
. The installation is succesful, but when calling energy-monitor
from the command line I get:
Traceback (most recent call last):
File "/home/mattia/.local/bin/energy-monitor", line 5, in <module>
from energymonitor.main import main
ModuleNotFoundError: No module named 'energymonitor'
Why is this happening? I was not able to find any helpful solution online. It's the first that I build a python package so sorry if the issue is trivial.
From setuptools' docs,
I'd take it that the where
value in [tool.setuptools.packages.find]
is the search path your packages should be found inside. You gave it the package folder directly, which should probably be done like this:
[tool.setuptools]
packages = ["energymonitor"]
As a result, the energymonitor
package never actually ended up in your build, which you can easily verify by unpacking it.