Skip to content
/ tqm Public

Python-based parallel task queuing and execution framework with PySide2 GUI for efficient workflow management.

License

Notifications You must be signed in to change notification settings

sisoe24/tqm

Repository files navigation

tqm: Task Queue Manager

A simple PySide2 no-dependency framework for managing and executing tasks in parallel with a nice UI. Helps you create, queue, and track multiple tasks with dependencies and progress reporting.

demo

What is this?

TQM lets you run multiple tasks in the background while showing their progress in a UI. It's built with only PySide2 and handles all the complex stuff like threading, dependencies between tasks and UI.

Basically, a Task Queue Manager.

Features

  • Task Management

    • Run multiple tasks in parallel (multi-threaded)
    • Create task groups with parent-child relationships
    • Set dependencies so tasks wait for others to finish
    • Various retry strategies including delays, conditional logic, and custom policies
  • UI Features

    • Progress bars (both determinate and indeterminate)
    • Task filtering and search
    • Context menus for task actions
    • Detailed debug view showing task properties
  • Developer Features

    • Simple API for creating and queuing tasks
    • Callbacks for task lifecycle events (start, finish, fail, complete)
    • Predicates to control when tasks should run
    • Customizable task colors and more
    • Comprehensive typing annotations

Installation

poetry add tqm
# or pip
pip install tqm

Dependencies:

  • PySide2

Quick Start

import time
from PySide2.QtWidgets import QApplication
from tqm import TQManager

def some_task(task):
    # This will run in a separate thread
    time.sleep(2)
    task.log("Task completed!")

# Create the application
app = QApplication([])
app.setStyle('Fusion')

# Create the task manager window
task_manager = TQManager(app_name="my_app")
task_manager.show()

# Add a simple task
task_manager.add_event(
    execute=some_task,
    label="My First Task",
    comment="This is a sample task"
)

# Start executing tasks
task_manager.start_workers()

# Run the application
app.exec_()

Examples

Creating a simple task

# Add a single task that does something
task_manager.add_event(
    execute=lambda task: print("Hello, world!"),
    label="Simple Task",
    comment="Prints a greeting"
)

Creating a task group

# Create a group of related tasks
with task_manager.create_group("Download Files") as group:
    group.add_event(lambda t: t.log("Downloading file 1"), label="File 1")
    group.add_event(lambda t: t.log("Downloading file 2"), label="File 2")
    group.add_event(lambda t: t.log("Downloading file 3"), label="File 3")

Using the TaskBuilder for more control

from tqm import TaskBuilder

# Create a task with the builder pattern
task = (
    TaskBuilder("Complex Task")
    .with_event(lambda t: t.log("Task running..."))
    .with_comment("A task with callbacks")
    .with_on_start(lambda t: print("Starting task"))
    .with_on_completed(lambda t: print("Task completed"))
    .with_on_failed(lambda t: print("Task failed"))
    .with_color("lightblue")
    .with_retry_failed(max_retries=3)  # Retry up to 3 times if it fails
    .build()
)

# Add it to the task manager
task_manager.add_tasks(task)

Tasks with dependencies

# Create tasks with dependencies
task1 = TaskBuilder("Task 1").with_event(lambda t: t.log("Task 1 done")).build()
task2 = (
    TaskBuilder("Task 2")
    .with_event(lambda t: t.log("Task 2 done"))
    .with_wait_for(task1)  # This will wait for task1 to complete
    .build()
)

task_manager.add_tasks(task1, task2)

Configuration

tqm stores its settings in:

  • Windows: %LOCALAPPDATA%\tqm\[app_name]\
  • Linux/Mac: ~/.config/tqm/[app_name]/

Environment

  • TQM_CONFIG_PATH - Configuration directory path
  • TQM_SETTINGS_PATH - Settings file path
  • TQM_QSS_PATH -Qt stylesheet path
  • TQM_MONO_FONT - Monospace font family
  • TQM_DEBUG - Enables debug logging when set to '1'
  • TQM_IDLE_TIMEOUT - Timeout in milliseconds for idle system detection

Python Version Support

TQM currently supports Python 3.7+ to ensure compatibility with VFX environments where older Python versions are still common. We intentionally keep zero dependencies (except PySide2) so VFX applications can use TQM right away. However, we'll phase out support for older versions with major releases:

  • Current (0.x): Python 3.7+
  • Future (1.x): Python 3.10+ (dropping 3.7, 3.8, and 3.9 support)

Older TQM versions will remain available on the GitHub releases page, but won't receive updates once we move to newer Python versions.

Current Status

tqm is still under development but stable enough for most use cases. The core functionality has been battle-tested in production environments for over a year, though this current version is a significant rewrite with improved architecture and features. Some features and documentation might be improved in future updates.

See the GitHub Project for more information.

Theme and style

This application was designed under the OS dark theme using Qt's Fusion style. I have never tried it under light themes or other Qt styles (Windows, macOS native, Breeze, etc.) which means that I assume no responsibility for any eye damage or aesthetic trauma.

Custom Styling

The application provides a style.qss file in the default config location which users can modify to customize the appearance of common widgets:

  • TqmHeaderView
  • TqmTasksLogs
  • TqmTasksViewToolbar
  • TqmSearchBar
  • TqmStatusLabel
  • TqmTasksTreeView
  • TqmFrame
  • TqmToolButton

License

This project is licensed under MIT.

About

Python-based parallel task queuing and execution framework with PySide2 GUI for efficient workflow management.

Topics

Resources

License

Stars

Watchers

Forks

Languages