AUTOMATIC-GENERATION OF VISUAL PROGRAMMING ENVIRONMENTS

The authors present the VLCC (Visual Language Compiler-Compiler) graph ical system for the automatic generation of visual programming environ ments. The VLCC system supports implementation of a visual language by helping the language designer define the graphical objects, syntax, a nd semantics of that language. The generation process produces an inte grated environment with an editor and interpreter for the defined visu al language. To capture the widest range of visual languages, the VLCC system can be configured for a specific class of languages (for examp le, diagrammatic, iconic, and so on). Characterization of these differ ent language classes depends on the structure of their graphical objec ts and on the way they can be composed. For example; an icon-oriented visual language is based on primitive icons, characterized by the coor dinate pair of a representative point (such as the centroid), and rela tions to spatially arrange those icons and form iconic sentences. The VLCC system uses an underlying grammar formalism called the positional grammar model. Positional grammars provide a powerful formalism to de scribe and efficiently parse a wide variety of visual languages. They directly extend from the context-free string grammars where spatial re lations in addition to string concatenation are allowed. This similari ty means many interesting results on LR parsing can be easily extended to the positional; grammars. In particular, a (Yet Another Compiler-C ompiler) YACC-like tool has been implemented and embedded in the VLCC system. This tool enables automatic parser generation of visual langua ges described by such positional grammars.