Abstract
Positional grammars are formalism for the definition and implementation of visual languages. They have already been used in the past as part of the VLCC system (Visual Language Compiler-Compiler) for the definition and the implementation of visual environments for editing and compiling flowcharts, chemical structures, combinatorial networks, electric circuits, etc. In this paper, we introduce the eXtended Positional Grammars (XPG, for short) that enhance the descriptive power of Positional Grammars. We also present a more powerful LR-based methodology for parsing visual languages described by XPGs. The result is the possibility of describing and compiling a much wider class of visual languages yet keeping most of LR parsing efficiency.