Vectorbasemapar

Vectorbasemapar is an open‑source library designed to simplify the creation and manipulation of vector‑based parametric models in computer graphics and computational design. Developed initially in 2019 by a group of researchers at the University of Zurich, the project aims to provide a lightweight, modular framework that can be integrated into larger pipelines such as CAD systems, game engines, and scientific visualization tools. The core of the library revolves around two concepts: a generic vector base type that supports arbitrary dimensionality and precision, and a parameter set system that allows users to encode geometric construction rules as composable functions. By decoupling vector representation from parameter logic, Vectorbasemapar enables rapid experimentation with shape grammars, modular assembly, and procedural generation. The library supports multiple back‑ends, including native C++ for performance‑critical tasks and a Python wrapper for rapid prototyping. It has been adopted by several research groups working on topology optimization, generative architecture, and machine learning for geometry. The codebase is written entirely in C++17, uses the Eigen linear algebra library, and is distributed under the MIT license, encouraging both academic and commercial use. Release candidates are released annually, and the project maintains a comprehensive test suite that covers geometric consistency, numerical stability, and integration with popular mesh formats such as OBJ and STL. While the project remains niche, its design principles and modular architecture have influenced newer libraries in the parametric modeling domain.