Research Article

FIRE/J—optimizing regular expression searches with generative programming

  1. Vassilios Karakoidas*,
  2. Diomidis Spinellis

Article first published online: 23 JUL 2007

DOI: 10.1002/spe.841

Issue

Software: Practice and Experience

Software: Practice and Experience

Volume 38, Issue 6, pages 557–573, May 2008

Additional Information

How to Cite

Karakoidas, V. and Spinellis, D. (2008), FIRE/J—optimizing regular expression searches with generative programming. Software: Practice and Experience, 38: 557–573. doi: 10.1002/spe.841

Author Information

  1. Department of Management Science and Technology, Athens University of Economics and Business, Patission Ave 74 GR-10434, Athens, Greece

*Department of Management Science and Technology, Athens University of Economics and Business, Patission Ave 74 GR-10434, Athens, Greece

Publication History

  1. Issue published online: 4 APR 2008
  2. Article first published online: 23 JUL 2007
  3. Manuscript Accepted: 31 MAY 2007
  4. Manuscript Revised: 24 MAY 2007
  5. Manuscript Received: 16 DEC 2005

SEARCH

SEARCH BY CITATION

ARTICLE TOOLS

Get PDF (175K)

Keywords:

  • regular expressions;
  • just-in time (JIT);
  • Java;
  • Java virtual machine (JVM);
  • automata;
  • domain specific languages (DSL);
  • generative programming

Abstract

Regular expressions are a powerful tool for analyzing and manipulating text. Their theoretical background lies within automata theory and formal languages. The FIRE/J (fast implementation of regular expressions for Java) regular expression library is designed to provide maximum execution speed while remaining portable across different machine architectures. To achieve that, FIRE/J transforms each regular expression into a tailor-made class file, which is compiled directly to Java virtual machine (JVM) bytecodes. The library is compatible with the POSIX standard. Copyright © 2007 John Wiley & Sons, Ltd.

Get PDF (175K)