Speeding up Generalized PSR Parsers by Memoization Techniques

TL;DR

This paper introduces memoization techniques to significantly accelerate generalized PSR parsers for hyperedge replacement grammars, making them more practical for parsing valid graphs without manual tuning.

Contribution

It proposes a novel memoization approach to improve GPSR parser efficiency, implemented within the Grappa graph parser distiller, eliminating the need for manual tuning.

Findings

Significant speed-up by orders of magnitude on valid graphs

Memoization does not improve parsing of invalid or ambiguous graphs

Enhances practicality of GPSR parsing for certain applications

Abstract

Predictive shift-reduce (PSR) parsing for hyperedge replacement (HR) grammars is very efficient, but restricted to a subclass of unambiguous HR grammars. To overcome this restriction, we have recently extended PSR parsing to generalized PSR (GPSR) parsing along the lines of Tomita-style generalized LR parsing. Unfortunately, GPSR parsers turned out to be too inefficient without manual tuning. This paper proposes to use memoization techniques to speed up GPSR parsers without any need of manual tuning, and which has been realized within the graph parser distiller Grappa. We present running time measurements for some example languages; they show a significant speed up by some orders of magnitude when parsing valid graphs. But memoization techniques do not help when parsing invalid graphs or if all parses of an ambiguous input graph shall be determined.