Is it normal that poetry show doesn't show our own package(s)?

[nbro]

I have a folder x with 3 sibling packages, which I specify in the pyproject.toml as follows

[tool.poetry]
name = "my-project"
version = "0.1.0"
description = "Nice multiple packages"
authors = ["Hi <hi@hello.bye>"]
license = "Free or not free"
readme = "README.md"

packages = [
    { include = "x", from = "." },
    { include = "y", from = "." },
    { include = "z", from = "." },
]

My project's structure looks like this

x/
    x/
    y/
    z/
    pyproject.toml

If I do poetry show, I don't see x, y or z. If I do pip list, I see my-project.

I don't know if this is the intended behaviour or not.

I also don't know if poetry supports my scenario or not. I think it depends on what I want to do. I've seen some discussions on whether poetry supports multiple pyproject.tomls.

In my case, I don't have a pyproject.toml under x, y or z, so I suppose that's why they are not installed as separate packages. It seems that include is only used to specify other packages to include in the distribution, when we do poetry build. Is this correct? Anyway, why doesn't poetry show also show my package?

I'm asking this question more because VSCode is not suggesting some code from y in x. I think this is a problem with pylance, but still I am curious to know why poetry show doesn't show my package.

[dexhunter]

Yes, this is expected. Your configuration defines one installable distribution, my-project, which contains three import packages: x, y, and z. The packages list controls which code belongs to that distribution; it doesn't create three separately installed distributions.

poetry show lists packages from the dependency lock file, so it omits the root project itself. The implementation reads the locked repository. pip list lists installed distributions, which is why you see my-project there.

Your layout is supported. In package mode, poetry install installs the root project in editable mode too; the package selection isn't limited to poetry build. See the packages setting.

For VS Code, select the interpreter printed by poetry env info --executable. Then check the imports with:

poetry run python -c "import x, y, z; print(x.__file__, y.__file__, z.__file__)"

That separates an import problem from editor discovery. Missing entries in poetry show don't establish that the imports are unavailable.