Contributing
How to contribute to crapssim
The current top priorities for the package are to improve
Do you want to help the documentation?
There’s many ways to improve the documentation for current and future users (including us!):
Write a short tutorial with some example usage of the package
Add more descriptions or type hints to internal package functions
Do you want to help supported strategies?
Craps has so many possible strategies, and it’s hard to implement them all. The ultimate goal of the package is to make building strategies easy for end users, but we also want to have commonly used and well known versions available as in the package as examples.
If you saw a strategy online or in a book, and have implemented with “crapssim”, then it most likely makes a great addition to the package. Please mention in a new discussion, file an issue, or open a pull request and we can work together to make sure it fits well.
Did you find a bug?
Please double check the bug has not already been reported in the Github issues
If your issue has not already been reported, open a new issue with as much detail to reproduce your problem as possible. The more details you provide, the easier it will be to isolate and fix the problem!
Contributing — Documentation and Examples
Conventions
This project uses the Black code style for formatting, which is easily installed or available in most IDEs
Code structure is moving towards Google Python Style Guide unless it conflicts with Black formatting
Documentation is moving towards Google Style Python Docstrings (See also here)
Please double check these conventions if submitting a PR.
Function and Type Hinting
All internal functions and classes should include:
A one-line summary docstring describing purpose and domain.
Explicit type hints for all parameters and return values.
Example:
def payout_ratio(number: int) -> float:
"""Return the true odds payout ratio for a given point number."""
As above, please use Google Style Python Docstrings (See also here)
Testing Philosophy
Tests are expected to cover both numerical and structural correctness. Each feature addition should include:
A unit test verifying direct functional behavior.
An integration or stress test demonstrating stable interaction with other modules.
Running Tests and Gauntlet
To verify correctness locally:
pytest -q
For optional stress and batch validation:
pytest -q -m stress
python tools/vxp_gauntlet.py # single run
bash -lc 'for i in $(seq 1 25); do python tools/vxp_gauntlet.py; sleep 0.2; done' # batch
Artifacts will appear under reports/vxp_gauntlet/