Visitar URL original
Module discovery | ty
Skip to content

Module discovery

First-party modules

First-party modules are Python files that are part of your project source code.

By default, ty searches for first-party modules in the project root. It also searches the following directories, if they exist and are not themselves packages (i.e. they contain neither an __init__.py nor an __init__.pyi file):

  • ./src
  • ./<project-name>, if ./<project-name>/<project-name> exists
  • ./python

These directories are searched before the project root, in the order listed above.

If your project uses a different layout, configure the project's environment.root in your pyproject.toml or ty.toml. For example, if your project's code is in an app/ directory:

example-pkg
├── README.md
├── pyproject.toml
└── app
    └── example_pkg
        └── __init__.py

then set environment.root in your pyproject.toml to ["./app"]:

[tool.ty.environment]
root = ["./app"]

[environment]
root = ["./app"]

Third-party modules

Third-party modules are Python packages that are not part of your project or the standard library. These are usually declared as dependencies in a pyproject.toml or requirements.txt file and installed using a package manager like uv or pip. Examples of popular third-party modules are requests, numpy and django.

ty searches for third-party modules in the configured Python environment.

Python environment

The Python environment is used for discovery of third-party modules.

For a project with no explicitly configured environment, ty searches for one in the following order:

  1. An active virtual environment, using the VIRTUAL_ENV environment variable.
  2. An active non-base Conda environment.
  3. A .venv directory in the project root.
  4. An active base Conda environment.
  5. A python3 or python interpreter on PATH.

Note

When using project management tools, such as uv or Poetry, the run command usually automatically activates the virtual environment and will be detected by ty.

The Python environment may be explicitly configured using the environment.python setting or --python flag.

When setting the environment explicitly, non-virtual environments can be provided.

PYTHONPATH

ty also respects the PYTHONPATH environment variable. Each existing directory listed in PYTHONPATH is added to the module search path, just after any extra-paths and before the environment's site-packages, mirroring the resolution order of the Python interpreter itself.

PYTHONPATH uses the same format as the shell's PATH: one or more directory paths separated by the platform's path separator (: on Unix, ; on Windows). Entries that don't exist or aren't directories are ignored.