Inter-Parameter Dependencies in Real-World Web APIs: The IDEA Dataset

Abstract

Context: Web services often impose constraints that restrict the way in which two or more input parameters can be combined to form valid calls to the service, so-called inter-parameter dependencies. Current API design languages like the OpenAPI Specification (OAS) provide no support for their formal description, making it hardly possible to automatically discover and interact with services without human intervention. Researchers and practitioners are openly requesting support for modelling and validating inter-parameter dependencies in web APIs, but this is not possible unless we share a deep understanding of how these dependencies emerge in practice. Objective: We aim to provide evidence on how inter-parameter dependencies are used in real-world web APIs. This evidence will hopefully serve as a basis for future proposals for modelling and analysing inter-parameter dependencies and will open a new range of research possibilities in areas related to service-oriented computing. Method: The documentation of 2,557 operations from 40 real-world web APIs was reviewed and carefully analysed, and 633 inter-parameter dependencies were found and classified into seven different types. Results: A machine-readable dataset was generated. This dataset helps understand the dimension and recurrence of inter-parameter dependencies in web APIs, as well as their taxonomy.