Skip to content

Python framework for rapid development of Starcraft 2 AI bots

License

Notifications You must be signed in to change notification settings

DavT007/sharpy-sc2

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

sharpy-sc2

Sharpy is a Python framework for rapid development of Starcraft 2 AI bots.

Sharpy is built on top of python-sc2 and it is the framework used by Sharpened Edge bot.

Sharpy was created to make it as easy as possible to create practice dummies for testing against Sharpened Edge. Sharpy contains everything needed to run all the practise dummies, as well as means to create more.

Sharpy is a work in progress. The folder structure and classes are subject to change.

Build statuses

Master branch

Develop branch

Requirements

  1. Python 3.7/3.8/3.9 64-bit
    • This requirement comes from sc2-pathlib as the pathlib is built for 64-bit python 3.7/3.8/3.9
  2. Windows, Linux or MacOS
    • This requirement comes from sc2-pathlib
    • macOS (and possibly others as well) may be supported after sc2-pathlib is built for it.

Ladder Dummy Bots

To build dummy bots for ladder, run ladder_zip.py. Bots will appear as individual zip files in publish folder.

Getting started

Read the getting started guide in wiki.

Contributing

To contribute to sharpy-sc2 source code, please create a pull request.

We also appreciate well written issues, comments and improvements to project wiki.

Pull Request Process

  • Keep pull requests small and atomic. Change only one logical thing at a time.
  • All Github actions checks for the PR must pass before it will be reviewed.
    • Make sure that the source code is formatted according to rules (see below)
    • Make sure that the source code passes linting
    • Make sure that all tests pass

Developing sharpy-sc2

Using Virtual Environment

Windows

Virtual Environments (venv) can be used to isolate this project's Python dependencies from other projects.

You can create a virtual environment for this project with

venv-create.bat

And activate it with

venv-activate.bat

Venv needs to be activated for every new console window, so it may be helpful to create an alias such as

doskey sharpy=cd C:\Dev\sharpy-sc2 $T venv-activate.bat

More information about virtual environments can be found from the documentation.

Other operating systems

You may replicate the commands used by the above bat scripts to work on your own operating system.

Installing Depedencies

To install all dependencies required to run the bots, use

pip install -r requirements.txt

To install all development dependencies, use

pip install -r requirements.dev.txt

Pre-commit hooks

To install git pre-commit hooks that will run black and flake8 (see below), use

pre-commit install

If you ever want to uninstall the hooks, use

pre-commit uninstall

or simply uninstall the .git\hooks\pre-commit file.

Code Formatting

sharpy-sc2 uses Black for automatic Python source code formatting.

to format code automatically, run

> py -m black .
All done! ✨ � ✨
272 files left unchanged.

Black can also be integrated to your favorite editor. See Editor Integration section in Black readme.

Linting

sharpy-sc2 uses flake8 for source code linting.

To run flake8 linting, use

> py -m flake8
0

Running Tests

Tests are written using pytest framework.

To run all tests, use

pytest

To run tests from a single test file, use pytest path-to-file eg.

pytest sharpy\knowledges\knowledge_test.py

pytest follows standard test discovery rules and will run all tests in the current directory and its subdirectories.

For new tests, make sure that...

  1. file name follows the naming pattern of *_test.py
  2. the test class name starts with Test*
  3. all test methods start with test_*.

About

Python framework for rapid development of Starcraft 2 AI bots

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%