taskcond 0.1.0

pythonic task runner.

taskcond

taskcond is a Pythonic task runner that executes tasks based on dependencies and file modification times, similar to make.

Features

• Pythonic Task Definition: Define tasks as Python functions or shell commands in a TaskFile.py.
• Dependency Management: Specify dependencies between tasks to ensure they run in the correct order.
• Incremental Builds: Tasks are skipped if their output files are newer than their input files, saving execution time.
• Parallel Execution: Run independent tasks concurrently using threads or processes to speed up workflows.
• Command-Line Interface: A simple and intuitive CLI powered by click for running and listing tasks.
• Flexible Configuration: Configure default behaviors in your pyproject.toml file.
• Cycle Detection: Automatically detects and reports cyclic dependencies in your task graph.

Quick Start

1. Installation

   pip install taskcond
   

2. Create a TaskFile.py Create a file named TaskFile.py in your project root and define your tasks.

   from pathlib import Path
   from taskcond import Task, register
   
   # Define a task that creates a file
   register(
       Task(
           name="create_file",
           description="Create an output file.",
           shell_command="echo 'Hello, taskcond!' > output.txt",
           output_files=(Path("output.txt"),),
       )
   )
   
   # Define a task that depends on the first one
   register(
       Task(
           name="process_file",
           description="Process the created file.",
           shell_command="cat output.txt",
           depends=("create_file",),
           input_files=(Path("output.txt"),),
       )
   )
   

3. Run Tasks Use the taskcond CLI to run your tasks.

   # List available tasks
   taskcond list
   
   # Run a specific task and its dependencies
   taskcond run process_file
   

   The process_file task will only run if output.txt has been modified or is missing.

Configuration

You can configure taskcond by adding a [tool.taskcond] section to your pyproject.toml:

[tool.taskcond]
taskfile = "TaskFile.py"  # Path to the task definition file
force = false             # Force execution of all tasks
n_jobs = 1                # Number of parallel jobs (-1 for all cores)
use_processes = false     # Use ProcessPoolExecutor instead of ThreadPoolExecutor
visible_progressbar = true # Show a progress bar during execution

CLI options will always override the settings in pyproject.toml.