While source transformation systems and languages like DMS, Stratego, ASF + SDF, Rascal and TXL provide a general, powerful base from which to attack a wide range of analysis, transformation and migration problems in the hands of an expert, new users often find it difficult to see how these tools can be applied to their particular kind of problem. The difficulty is not that these very general systems are ill-suited to the solution of the problems, it is that the paradigms for solving them using combinations of the system's language features are not obvious. In this paper we attempt to approach this difficulty for the TXL language in a non-traditional way - by introducing the paradigms of use for each kind of problem directly. Rather than simply introducing TXL's language features, we present them in context as they are required in the paradigms for solving particular classes of problems such as parsing, restructuring, optimization, static analysis and interpretation. In essence this paper presents the beginnings of a "TXL Cookbook" of recipes for effectively using TXL, and to some extent other similar tools, in a range of common source processing and analysis problems. We begin with a short introduction to TXL concepts, then dive right in to some specific problems in parsing, restructuring and static analysis.
展开▼
