Reproduce “A New Functional-Logic Compiler for Curry: Sprite” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Sergio Antoy, Andy Jost

Sprite introduces a native code compiler for the functional-logic language Curry, designed to be the first operationally complete implementation. The compiler achieves its goal of finding all possible results while demonstrating a 4.4x speedup for functional-logic programs compared to a leading existing compiler.

Key insights

  • Sprite successfully demonstrates that operational completeness for functional-logic programming is achievable without prohibitive performance costs, validating the Fair Scheme as a practical compilation strategy.
  • Direct native code compilation of functional-logic features, rather than simulation, yields substantial performance improvements for non-deterministic programs.
  • The rigorous implementation of pull-tabbing, work queues, and consistency management in Sprite provides a blueprint for building robust compilers for hybrid declarative languages.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.