Manual Context Handling

Abstract

Although attribute grammars allow us to generate context processing programs automatically, their level of automation has not yet reached that of lexical analyzer and parser generators, and much context processing programming is still done at a lower level, by writing code in a traditional language like C or C++.We will give here two non-automatic methods to collect context information from the AST; one is completely manual and the other uses some reusable software. Whether this collected information is then stored in the nodes (as with an attribute grammar), stored in compiler tables, or consumed immediately is immaterial here: since it is all handy-work, it is up to the compiler writer to decide where to put the information.