FLAMA
José Á. Galindo, José-Miguel Horcas, Alexander Felferning, David Fernández-Amorós, David Benavides
Abstract
Nowadays, feature models are the de facto standard when representing commonalities and variability, with modern examples spanning up to 7000 features. Manual analysis of such models is challenging and error-prone due to sheer size. To help in this task, automated analysis of feature models (AAFM) has emerged over the past three decades. However, the diversity of these tools and their supported languages presents a significant challenge that motivated the MOD-EVAR community to initiate a project for a new tool that supports the UVL language. Despite the rise of machine learning and data science, along with robust Python-based libraries, most AAFM tools have been implemented in Java, creating a collaboration gap. This paper introduces Flama, an innovative framework that automates the analysis of variability models. It focuses on UVL model analysis and aims for easy integration and extensibility to bridge this gap and foster better community and cross-community collaboration.