> [!META]- Inline Metadata [status:: boat] [tags:: #state/boat #note/evergreen ] [up:: [[Python Poetry MOC]], [[Python MOC]]] There is a standardized way to install Python packages from VCS repos. The examples will show a complex case that include a branch and subdirectory # pyproject.toml - `package-name = {git = "[email protected]:<User>/repo-name.git}` - `package-name = {git = "[email protected]:<User>/repo-name.git", branch = "test-branch", subdirectory = "repo_dir"}` # requirements.txt - `package-name @ git+ssh://[email protected]/<User>/repo-name.git` - `package-name @ git+ssh://[email protected]/<User>/repo-name.git@ptest-branch#subdirectory=repo_dir` # pip This is basically the same format as requirements.txt. - `pip install "package-name @ git+ssh://[email protected]/<User>/repo-name.git"` - `pip install "package-name @ git+ssh://[email protected]/<User>/repo-name.git@ptest-branch#subdirectory=repo_dir"` # Private repos To install from private repos, simply ensure your Github keys are set up on your machine and that they are linked to your Github account (and that you have access to the private accounts). To do this while building a Docker image, see TBA # Sources [pip VCS support documentation](https://pip.pypa.io/en/stable/topics/vcs-support/)