Should I play or should I go? A mining study on the reproducibility of open-source video games

About

This repository contains all the materials and source code used in the paper entitled "Should I play or should I go? A mining study on the reproducibility of open-source video games" submitted to the 9th International ICSE Workshop on Games and Software Engineering  (GAS 2025).

Data Overview

The repository is structured as follows:

stats: This folder contains all the CSV files describing statistics used to answer RQ1, e.g., code metrics, GitHub stats

selected.zip: This folder contains the selected games and the corresponding source code used to answer

games_info: This folder contains the information related to games stored on itch.io

games_retrived: This folder contains the CSV files that store metadata of the games retrieved from itch.io

reproducibility_table.csv: This CSV file contains a detailed version of Table III presented in the paper

manual_labeling.csv: This CSV file contains the results of the manual labeling phase

Requirements

To replicate all the results presented in the paper, you need to install the following libraries:

Selenium

PyGithub

Pandas and Matplotlib

Source code

The following files contain the needed scripts to mining the data, compute stats, and map itch.io games to Github repos:

itch_scraping_parallel.py: It contains scripts exploit threats to speed up the itch.io mining

itch_scraping.py: It contains main scripts to collect itch.io data

postprocessing.ipynb: It contains utils for cleaning data

scraping utils: It contains support functions to the itch.io miner

data_utils.py: It contains utility functions for data analysis and GitHub crawling

main.py:  It can be used to run all the data analysis and the GitHub mapping