Taskipy

1.14.1 · active · verified Sun Apr 12

Taskipy is a task runner for Python projects that streamlines development workflows by defining and executing automation tasks. It uses a `pyproject.toml` file for configuration, offering an `npm run-script`-inspired interface for common operations like linting, testing, and building. The current version is 1.14.1, with a consistent release cadence addressing bug fixes and minor features.

Warnings

breaking The `psutil` dependency version constraint was updated in Taskipy 1.14.1 to allow versions greater than 6. Older Taskipy versions (prior to 1.14.1) might have had tighter constraints (e.g., `<7`), potentially leading to conflicts if `psutil` was updated independently to version 7 or higher.
Fix: Upgrade Taskipy to 1.14.1 or newer (`pip install --upgrade taskipy`) to ensure compatibility with `psutil` versions >6. Confirm your `pyproject.toml` or `requirements.txt` allows a compatible `psutil` range.
gotcha Command-line arguments passed to a task are *not* automatically propagated to `pre_` and `post_` hooks. These hooks run without the arguments meant for the main task.
Fix: If `pre_` or `post_` hooks require arguments, define them as separate, explicit tasks and call them with arguments directly, or use Taskipy's variable system if the arguments are static and can be templated.
gotcha By default, tasks run from the current working directory where `task` is invoked, not necessarily from the project root where `pyproject.toml` resides. This can lead to unexpected behavior if tasks depend on relative paths.
Fix: Explicitly set the `cwd` (current working directory) for individual tasks or globally in `[tool.taskipy.settings]` to ensure tasks always run relative to a specific path, typically the project root. Example: `[tool.taskipy.tasks.mytask] cmd = "my_command", cwd = "."` or `[tool.taskipy.settings] cwd = "."`.
gotcha Defining multi-line commands directly within `pyproject.toml` can be awkward due to TOML's syntax favoring single-line string values.
Fix: For complex, multi-line commands, consider moving them into a separate shell script (e.g., `.taskipy/scripts/my_complex_script.sh`) and then call that script from your `pyproject.toml` task: `complex_task = "./.taskipy/scripts/my_complex_script.sh"`.
gotcha Taskipy itself needs to be installed first. It doesn't solve the initial 'bootstrapping problem' of how to run the very first setup command if `make` or `just` are being replaced.
Fix: Communicate the installation steps for Taskipy clearly (e.g., `pip install taskipy` or `poetry add --dev taskipy`) as part of your project's setup guide before instructing users to run `task` commands.

Install

pip install taskipy Pip
poetry add --dev taskipy Poetry (as dev dependency)
conda install -c conda-forge taskipy Conda

Imports

main
```
from taskipy.cli import main
```
Taskipy is primarily a CLI tool. This is the entry point for programmatic execution of the CLI.
TaskRunner
```
from taskipy.task_runner import TaskRunner
```
The core class for executing tasks programmatically, though direct use is less common for end-users than CLI interaction.

Quickstart

Define tasks in your `pyproject.toml` under the `[tool.taskipy.tasks]` section. Each task is a key-value pair where the key is the task name and the value is the shell command. Tasks can also be defined as tables to include a `help` message or specify a `cwd`. Run tasks using the `task <task_name>` command in your terminal.

# pyproject.toml
[tool.taskipy.tasks]
hello = "echo Hello, Taskipy!"
greet = { cmd = "python -c 'import os; print(f\"Hello, {os.environ.get(\"USER\", \"World\")}\")'", help = "Greets the current user" }

# In your terminal, from project root:
# task -l
# task hello
# task greet

view raw JSON →