Impact of Issue Templates on Software Development Metrics: A Comparative Analysis of GitHub and Jira

<b>Replication Package</b><br>This repository contains the code and data for issue templates analysis on GitHub and Jira<br><b>Data</b><br>Data is available under src/main/java/outputs directory.<br><b>Code</b><br>- Main file for mining, processing, and analyzing of the project is src/main/java/Main.java.- Main file for creating figures and calculation p-values of the project is src/main/python/timeline_n.ipynb.<br>The following prerequisites should be installed before running the project:<br><b>Prerequisites</b><br>- Python 3.10- Poetry- Jupyter Notebooks- Node.js and prettier (Required for formatting Markdown and YAML files)<br><b>Running the project</b><br>1. Set src/main/java/Environment.java's required properties. - github_project (can be set to "accumulo", "bookkeeper", "couchdb", "hop", "netbeans", "trafficserver") - jira_project (can be set to "ACCUMULO", "BOOKKEEPER", "COUCHDB", "HOP", "NETBEANS", "TS") - token (can be taken from GitHub)<br>2. Run the main file for Mining/Processing/Analyzing where all parameters given as true.<br>3. Run Jupyter notebook for drawing figures and calculating p-values.<br><b>Results</b><br>The output of the study (results) are available under outputs directory. Figures and p values can be found under timelines, ttr_per, and p-values directories.<br><br><b>Explanation of Code</b><br>- Java Side - Environment.java - In this java folder, api urls of GitHub and Jira, token for GitHub, and output file locations are given. - Main.java - In the main class, three steps of the process (Mining, Processing, Analyzing) can be controlled with given boolean values for each step. - Mining.java - In this class, GitHub issues and Jira issues' title, key, creation date, closing date, number of comments, number of reopenings information are documented on a json files which are named as _githubIssueSummary.json and _jiraIssueSummary.json. - Processing.java - In this class, json outputs of Mining.java are taken and GitHub issues' titles are compared with Jira issues. If the similarity is high enough, those issues from GitHub and Jira are excluded. After comparison is done, remaining issues are saved on _uniqueGithubIssueSummary.json and _uniqueJiraIssueSummary.json. - Analyzing.java - In this class, json outputs of Processing.java are taken and time-to-resolution, number of comments, number of reopenings values' mean, median, variance values are calculated. Final results are printed on _analysisReport.txt. Also, each calculated value (for either time to resolution, number of comments or number of reopenings) recorded on txt files which are named as follows: _Github.txt and _Jira.txt where can be ttr (for time to resolution), noc (for number of comments), and nor (for number of reopenings). Those txt files are used to print distribution graphs on Python side.<br>- Python Side - timeline_n.ipynb - In this jupyter notebook, timeline figures, distribution figures are created and p-values are calculated for Mann-Whitney U test.