This is the official repository of the "Technical Debt Dataset" [1].

There are two versions of the dataset:

• release 1.01 (32 projects, all branches analyzed, not recommended if you need to use SonarQube data (we found some issues with the analysis timeline)

• release 2.0 (31 projects, only master branch analyzed).

• download

• submit new issues and corrections

• What is it
• How to use it
• How to cite the Technical Debt Dataset
• How to contribute
• Empirical studies based on the Technical Debt Dataset
• References

Technical Debt Dataset is a curated dataset containing measurement data from four tools executed on all commits to enable researchers to work on a common set of data and thus compare their results.

The dataset was built by extracting the projects' data and analyzing all the commit using several tools. To get the data, the projects' GitHub repositories were cloned, commit information was collected from the git log using PyDriller, refactorings were classified using Refactoring Miner, and issue information was obtained by extracting issues from the Jira issue tracker. After that, code quality was inspected using two tools: Technical Debt items were analyzed with SonarQube, and code smells [2] and anti-patterns [3] with Ptidej. In addition, the fault-inducing and -fixing commits were identified by applying our implementation of the SZZ algorithm [4] for version 1, and our fork of SZZUnleashed for version 2.

The dataset is provided as an SQLite database with a .db extension. The file can be opened with a SQLite browser such as SQLiteStudio. After opening, the data can be queried using SQL by selecting "Tools" and "Open SQL Editor". Data can also be exported as a CSV file by selecting "Tools" and "Export" and then exporting a whole table or an SQL query.

SQL enables to easily query the data from the dataset.

Here is a simple example:

Get the number of "Bug" and "Vulnerability" SonarQube issues in all the branches of a project.

SELECT  SONAR_ISSUES.PROJECT_ID,  count(SONAR_ISSUES.PROJECT_ID)  AS  numberOfBugIssues,  numberOfVulnerabilityIssues
FROM SONAR_ISSUES
JOIN
(SELECT  SONAR_ISSUES.PROJECT_ID,  count(SONAR_ISSUES.PROJECT_ID)  AS  numberOfVulnerabilityIssues
FROM SONAR_ISSUES
WHERE  TYPE='VULNERABILITY'
GROUP BY SONAR_ISSUES.PROJECT_ID) AS V ON SONAR_ISSUES.PROJECT_ID = V.PROJECT_ID
WHERE TYPE='BUG'
GROUP BY SONAR_ISSUES.PROJECT_ID

It is important to note that the dataset (V1) has been created analyzing all the commits of each project, without considering branches. Moreover, the Technical Debt items were exported as reported by the tools. We did not excluded duplications nor filtered any technical debt issues, since we aimed at providing to the community exactly what is provided by the tools.

In V2, instead, we analyzed only the master branch. If you are interested to filter issues, considering only unique Technical Debt Items, filter commits in the branch you are interested to analyze (Table GIT_COMMITS, Attribute 'BRANCHES'). Considering all the commits would result in duplicated items.

Please, cite as "The Technical Debt Dataset, Version XX [1]", where XX is the appropriate version.

[1] Valentina Lenarduzzi, Nyyti Saarimäki, Davide Taibi. The Technical Debt Dataset. Proceedings for the 15th Conference on Predictive Models and Data Analytics in Software Engineering. Brazil. 2019. Download the paper

@INPROCEEDINGS{Lenarduzzi2019,
  author = {Lenarduzzi, Valentina and Saarim{\"a}ki, Nyyti and Taibi, Davide},
  title = {The Technical Debt Dataset},
  booktitle={15th Conference on Predictive Models and Data Analytics in Software Engineering}, 
  year={2019}, 
  month={January},
  }

Submit New Projects If you have analyzed a project with SonarQube and you are interested to share your data in our dataset, please send us an email ( davide [dot] taibi [ at ] tuni [ dot ] fi )

To integrate your analysis, please, report the following information:

• sonarqube_version
• project_name
• development_language
• github_url
• analyzed branch
• jira url

We will run the SZZ tool and refactoring miner on your repository and integrate your data in a new release of the dataset.

Automate the analysi pipeline We are also looking for contributors to automate the analysis pipeline. If you are interested to contribute, send us a message.

Submit Issues and correct errors Please submit new issues directly in Github

The Technical Debt Dataset has been developed only for research purposes. It includes the historical analysis of each public repository, including commit messages, timestamps, author names, and email addresses. Information from GitHub is stored in accordance with GitHub Terms of Service (GHTS), which explicitly allow extracting and redistributing public information for research purposes (GitHub Terms of Service Accessed: May 2019).

The Technical Debt Dataset is licensed under a Creative Commons Attribution-NonCommercial- ShareAlike 4.0 International license.

The Technical Debt Dataset has been used in different works:

• Nyyti Saarimäki, Valentina Lenarduzzi, and Davide Taibi. 2019. On the diffuseness of code technical debt in open source projects of the Apache Ecosystem. International Conference on Technical Debt (TechDebt 2019) 2019.

• Valentina Lenarduzzi, Antonio Martini, Davide Taibi, and Damian Andrew Tamburri. 2019. Towards Surgically-Precise Technical Debt Estimation: Early Results and Research Roadmap. In2019 IEEE Workshop on Machine Learning Techniques for Software Quality Evaluation (MaLTeSQuE)

• Valentina Lenarduzzi, Francesco Lomio, Davide Taibi, and Heikki Huttunen. 2019.On the Fault Proneness of SonarQube Technical Debt Violations: A comparison of eight Machine Learning Techniques. arXiv:1907.00376

[1] Valentina Lenarduzzi, Nyyti Saarimäki, and Davide Taibi. 2019. The Technical Debt Dataset. In 15th Conference on Predictive Models and Data Analytics in Software Engineering, 2019.

[2] Fowler, Martin. Refactoring: improving the design of existing code. Addison-Wesley Professional, 2018.

[3] Brown, William H., et al. AntiPatterns: refactoring software, architectures, and projects in crisis. John Wiley & Sons, Inc., 1998.

[4] Śliwerski, Jacek, Thomas Zimmermann, and Andreas Zeller. "When do changes induce fixes?." ACM sigsoft software engineering notes. Vol. 30. No. 4. ACM, 2005.