Generating Constraint Programs for Variability Model Reasoning: A DSL and Solver-...
summary
Claims a fully generic, solver-agnostic pipeline that decouples variability modeling languages from solvers by generating CLIF pivot semantics and then concrete solver code; prototype implements two VMLs (static and dynamic SPL with soft constraints) and targets two solver families (CSP and CLP) — demonstrated with MiniZinc, SWI Prolog and Z3, exposes a web API and exports runnable solver programs, and handles industrial-scale models (≈10,000 features). CLIF pivot semantics solver controller + bridge
AI-written summary · about
chapters (16)
- 0:00Introduction
- 1:02Software product line engineering
- 4:11Automated analysis of variability models
- 5:35Solver transformation limitations
- 6:59Decoupling models and solvers
- 8:02Cliff pivot representation
- 9:26Cliff semantic specification
- 11:15MiniZinc configuration query
- 11:56Satisfiability with iteration
- 13:21Dynamic optimization with Z3
- 14:23Architecture overview
- 17:11Prototype and supported solvers
- 18:15Live demo
- 20:25Contributions and future work
- 21:48Open source resources
- 22:09Q&A
related talks
24mBuilding Information Modeling Using Constraint Logic Programming (Extended Ab...
01h 01mCDCL SAT Solving and Applications to Optimization Problems
18mOn the Expressive Power of Languages for Static Variability
30mUnconstrained Variable Oracles for Faster Static Analyses
16mCounterexample Driven Quantifier Instantiations with Applications to Distribute...
42mASP: Theory, Practice, and Beyond
22mVerification Modulo Tested Library Contracts
21mLazy Validation and Self-Healing for Agentic Programs