FLOC 2018: FEDERATED LOGIC CONFERENCE 2018
MaxSMT-Based Type Inference for Python 3

Authors: Mostafa Hassan, Caterina Urban, Marco Eilers and Peter Müller

Paper Information

Title:MaxSMT-Based Type Inference for Python 3
Authors:Mostafa Hassan, Caterina Urban, Marco Eilers and Peter Müller
Proceedings:CAV All Papers
Editors: Georg Weissenbacher, Hana Chockler and Igor Konnov
Keywords:Type Inference, SMT Solving, Dynamic Languages
Abstract:

ABSTRACT. We present Typpete, a sound type inferencer that automatically infers Python 3 type annotations. Typpete encodes type constraints as a MaxSMT problem and uses a system of optional constraints and specific quantifier instantiation patterns to make the constraint solving process efficient. Our experimental evaluation shows that Typpete scales to real world Python programs and outperforms state-of-the-art tools.

Pages:8
Talk:Jul 16 09:15 (Session 109A: Tools)
Paper: