In this evaluation, we aim to investigate the effectiveness and efficiency of our technique for detecting errors in XPath expression processors. Specifically, we focus on answering the following questions:
1. Is XPress effective in identifying new XPath-related errors in consolidated XML processors?
2. Does the query generation strategy improve error detection efficiency compared to random generation approaches?
3. How does differential testing compare to the most advanced methods available?
4. What type of XPath-related errors could go unnoticed?
To answer these questions, we conducted tests on six widely used XPath processors: BaseX, exist-DB, Saxon-HE, PostgreSQL, libXML2, and a commercial database management system. We selected these processors due to their recognition in the community and their implementation of the latest XPath standards.
In the study, we configured our tool, XPress, to generate XML documents with between 1 and 50 nodes, and produce 200 XPath expressions per document. The queries had a controlled depth and complexity distribution to reflect real-world scenarios. We used default configurations on each XML processor and conducted the tests in a high-performance environment with a 64-core AMD EPYC processor and 512GB of memory, ensuring reliable results.
Q2BSTUDIO, a company specialized in development and technology services, focuses on innovating with advanced technologies to optimize data processing. Our dedication to technical excellence and efficient solutions allows us to analyze systems such as the XML processors evaluated in this study, validating their performance and reliability in real-world environments. With our expertise, we support the development of advanced tools to improve security and efficiency in structured data manipulation.
