-
Notifications
You must be signed in to change notification settings - Fork 204
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Install project in editable mode when using a pyproject.toml manifest automatically #1083
Comments
With regards to the CLI. I guess coming from So I was wondering if it this might make it confusing, because you need an extra specifier with Now that I'm reading more about it I think editable installs by default is the correct way to go. Prior art:
W.r.t build backends its much of the same thing, hatch's is pretty well documented: https://hatch.pypa.io/latest/plugins/builder/wheel/#default-file-selection. However, |
This doesn't work for me with pixi [build-system]
requires = ["setuptools"]
build-backend = "setuptools.build_meta"
[project]
name = "fluid"
requires-python = ">=3.10"
dynamic = ["version"]
[tool.pixi.project]
channels = ["conda-forge"]
platforms = ["linux-64", "linux-aarch64", "osx-arm64"]
[tool.pixi.dependencies]
pytorch = ">=2.1.0, <2.2.0"
gymnasium = ">=0.28, <0.29"
loguru = ">=0.7.0, <0.8.0"
pytest = ">=8.0, <9.0"
[tool.pixi.host-dependencies]
python = ">=3.10, <3.11" If I add this at the bottom, things work as expected: [tool.pixi.pypi-dependencies]
fluid = { path = "./", editable = true } How can I tell it doesn't work? |
Hi @garymm you are correct in the PR that closes this issue it was decided against an automatic install. Mainly because you can have multiple environments where you might not want an editable in all cases. 'pixi unit' does add it automatically to the .toml. If you find this behavior undesirable, feel free to make an issue where we can weigh the pro's and cons :) |
Problem description
With #1044 merged, I propose to automatically add current project as a dependency in editable mode when using a pyproject.toml manifest.
I do not see any reason not to, but maybe I am missing something?
The text was updated successfully, but these errors were encountered: