A Constraint-Satisfaction Parser for Context-Free Grammars

TL;DR

This paper introduces Fence, a bottom-up chart parser capable of handling ambiguous, constraint-based, context-free grammars, facilitating model-based language specification by supporting constraints, ambiguity, and efficiency.

Contribution

The paper presents Fence, a novel parser that efficiently manages ambiguity and constraints in context-free grammars, enabling practical model-based language design.

Findings

Fence supports ambiguous grammars effectively.

Fence enforces custom constraints during parsing.

Fence demonstrates efficiency in parsing complex grammars.

Abstract

Traditional language processing tools constrain language designers to specific kinds of grammars. In contrast, model-based language specification decouples language design from language processing. As a consequence, model-based language specification tools need general parsers able to parse unrestricted context-free grammars. As languages specified following this approach may be ambiguous, parsers must deal with ambiguities. Model-based language specification also allows the definition of associativity, precedence, and custom constraints. Therefore parsers generated by model-driven language specification tools need to enforce constraints. In this paper, we propose Fence, an efficient bottom-up chart parser with lexical and syntactic ambiguity support that allows the specification of constraints and, therefore, enables the use of model-based language specification in practice.