Fencing the Lamb: A Chart Parser for ModelCC

Abstract

Traditional grammar-driven language specification techniques constrain language designers to specific kinds of grammars. In contrast, model-based language specification techniques decouple language design from language processing. They allow the occurrence of ambiguities and the declarative specification of constraints for solving them. As a result, these techniques require general parser generators, which should be able to parse context-free grammars, handle ambiguities, and enforce constraints to disambiguate them as desired by the language designer. In this paper, we describe Fence, a bottom-up chart parser with lexical and syntactic ambiguity support. Fence accepts lexical analysis graphs outputted by the Lamb (Lexical AMBiguity) lexer and efficiently resolves ambiguities by means of the declarative specification of constraints. Both Lamb and Fence are part of the ModelCC model-based parser generator. © Springer-Verlag Berlin Heidelberg 2013.