Abstract The recent surge in the field of generative artificial intelligence (GenAI) has the potential to bring about transformative changes across a range of sectors, including software engineering and education. As GenAI tools, such as OpenAI’s ChatGPT, are increasingly utilised in software engineering, it becomes imperative to understand the impact of these technologies on the software product. This study employs a methodological approach, comprising web scraping and data mining from LeetCode, with the objective of comparing the software quality of Python programs produced by LeetCode users with that generated by GPT-4o. In order to gain insight into these matters, this study addresses the question whether GPT-4o produces software of superior quality to that produced by humans. The findings indicate that GPT-4o does not present a considerable impediment to code quality, understandability, or runtime when generating code on a limited scale. Indeed, the generated code even exhibits significantly better values across all the three code quality dimensions in comparison to the user-written code. However, no significantly superior values were observed for the generated code in terms of memory usage in comparison to the user code, which contravened the expectations. Furthermore, it will be demonstrated that GPT-4o encountered challenges in generalising to problems that were not included in the training data set. This contribution presents a first large-scale study comparing generated code with human-written code based on LeetCode platform based on multiple measures including code quality, code understandability, time behaviour and resource utilisation. All data is publicly available for further research.

RESTful APIs based on HTTP are one of the most important ways to make data and functionality available to applications and software services. However, the quality of the API design strongly impacts API understandability and usability, and many rules have been specified for this. While we have evidence for the effectiveness of many design rules, it is still difficult for nractitioners to identify rule violations in their design. We therefore present RESTRuler, a Java-based open-source tool that uses static analysis to detect design rule violations in OpenAPI descriptions. The current prototype supports 14 rules that go beyond simple syntactic checks and partly rely on natural language processing. The modular architecture also makes it easy to implement new rules. To evaluate RESTRuler, we conducted a benchmark with over 2,300 public OpenAPI descriptions and asked 7 API experts to construct 111 complicated rule violations. For robustness, RESTRuler successfully analyzed 99% of the used real-world OpenAPI definitions, with some failing due to excessive size. For performance efficiency, the tool performed well for the majority of files and could analyze 84% in less than 23 seconds with low CPU and RAM usage. Lastly, for effectiveness, RESTRuler achieved a precision of 91% (ranging from 60% to 100% per rule) and recall of 68% (ranging from 46% to 100%). Based on these variations between rule implementations, we identified several opportunities for improvements. While RESTRuler is still a research prototype, the evaluation suggests that the tool is quite robust to errors, resource-efficient for most APIs, and shows good precision and decent recall. Practitioners can use it to improve the quality of their API design.

JavaScript (JS) is one of the most popular programming languages, and widely used for web apps, mobile apps, desktop clients, and even backend development. Due to its dynamic and flexible nature, however, JS applications often have a reputation for poor software quality. While the type-safe superset TypeScript (TS) offers features to address these prejudices, there is currently insufficient empirical evidence to broadly support the claim that TS applications exhibit better software quality than JS applications.