Objects for lexical analysis

  • Authors:
  • Bernd Kühl;Axel-Tobias Schreiner

  • Affiliations:
  • University of Osnabrück, Germany;Rochester Institute of Technology

  • Venue:
  • ACM SIGPLAN Notices
  • Year:
  • 2002

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper presents a new idea for lexical analysis: lolo (language-oriented lexer objects) is strictly based on the object orientation paradigm. We introduce the idea behind the system, describe the implementation, and compare it to the conventional approach using lex[1] or flex[2].lolo[3] extracts symbols from a sequence of input characters belonging to the ASCII or Unicode sets. lolo scanners can be extended without access to the source code: symbol recognizers can be derived by inheritance and an executing scanner can be reconfigured for different contexts. Recognizer actions are represented by objects which may be replaced at any time. Recognizers need not be based on finite state automata; therefore, lolo can recognize symbols that systems like lex cannot recognize directly.