Identifying Code Generation Candidates Using Software Categories

Code generators are a crucial part of the model-driven development (MDD) approach. They systematically transform abstract models to concrete executable source code. Typically, a generator developer determines the code parts that should be generated and separates them from handwritten code. Since performed manually, this task often is timeconsuming, labor-intensive, difficult to maintain and may produce more code than necessary. This paper presents an iterative approach for identifying candidates for generated code by analyzing the dependencies of categorized code parts. Dependency rules are automatically derived from a predefined software category graph and serve as basis for the categorization process. Generator developers can use this approach to systematically identify code generation candidates. The ideas and concepts of this paper were introduced at the MODELSWARD conference [1] and are extended in this contribution.