Bounded context translation

  • Authors:
  • Robert M. Graham

  • Affiliations:
  • Massachusetts Institute of Technology, Cambridge, Mass.

  • Venue:
  • AFIPS '64 (Spring) Proceedings of the April 21-23, 1964, spring joint computer conference
  • Year:
  • 1964

Quantified Score

Hi-index 0.00

Visualization

Abstract

All translators are syntax-directed in the sense that the translator must obviously recognize the various syntactic structures and the output of the translator is a function of the syntax of the language. The term syntax-directed is usually applied to a translator which contains a direct encoding of the syntax of the language, this direct encoding being used by the translator as data. The companion paper by Cheatham and Sattley is concerned with this type of translation. In the other class of translators the syntax is essentially buried in the coding of the translator. Most of the algebraic languages in use are precedence grammars, or close enough so that the properties of precedence grammars are useful. Using the properties of precedence grammars, bounded context translation is possible. At each step in the scan of an expression in bounded context translation the decision as to what action to take next is a function of the symbol currently under scan and of N symbols on either side (where N is fixed for the particular language).