A Tool for Generating Exceptional Behavior Tests With Large Language Models
This addresses a gap in software testing for developers by automating EBT generation, though it is incremental as it builds on existing LLM and testing methods.
The paper tackles the problem of developers prioritizing 'happy path' tests over exceptional behavior tests (EBTs) in software development by introducing exLong, a framework that automatically generates EBTs using a fine-tuned large language model, resulting in effective assistance for developers in creating comprehensive tests.
Exceptional behavior tests (EBTs) are crucial in software development for verifying that code correctly handles unwanted events and throws appropriate exceptions. However, prior research has shown that developers often prioritize testing "happy paths", e.g., paths without unwanted events over exceptional scenarios. We present exLong, a framework that automatically generates EBTs to address this gap. exLong leverages a large language model (LLM) fine-tuned from CodeLlama and incorporates reasoning about exception-throwing traces, conditional expressions that guard throw statements, and non-exceptional behavior tests that execute similar traces. Our demonstration video illustrates how exLong can effectively assist developers in creating comprehensive EBTs for their project (available at https://youtu.be/Jro8kMgplZk).