Code generation by model transformation: a case study in transformation modularity

Abstract

The realization of model-driven software development requires effective techniques for implementing code generators for domain-specific languages. This paper identifies techniques for improving separation of concerns in the implementation of generators. The core technique is code generation by model transformation, that is, the generation of a structured representation (model) of the target program instead of plain text. This approach enables the transformation of code after generation, which in turn enables the extension of the target language with features that allow better modularity in code generation rules. The technique can also be applied to 'internal code generation' for the translation of high-level extensions of a DSL to lower-level constructs within the same DSL using model-to-model transformations. This paper refines our earlier description of code generation by model transformation with an improved architecture for the composition of model-to-model normalization rules, solving the problem of combining type analysis and transformation. Instead of coarse-grained stages that alternate between normalization and type analysis, we have developed a new style of type analysis that can be integrated with normalizing transformations in a fine-grained manner. The normalization strategy has a simple extension interface and integrates non-local, context-sensitive transformation rules. We have applied the techniques in a realistic case study of domain-specific language engineering, i.e. the code generator for WebDSL, using Stratego, a high-level transformation language that integrates model-to-model, modelto-code, and code-to-code transformations.

WebDSL: A Case Study in Domain-Specific Language Engineering

“…In particular, some utilities for the generation of multiple output files were developed. Furthermore, in a refactoring of the WebDSL generator several measures were taken to increase the locality of generation rules [54]. In particular, an extension of Java has been developed to support identifier composition, partial classes, partial methods, and interface derivation.…”

“…However, the extension of WebDSL with access control and the addition of new user interface components, eventually required a number of refactorings to maintain the modularity of the generator [54].…”

“…It should be possible to provide such extensions as a plug-in to the generator, which requires an extensibility mechanism. Using the extensibility of strategy definitions in Stratego and an extension of Java to support partial classes, such extensibility is realized in a refactoring of the WebDSL generator [54]. A concern in the design of such extensions should be a proper separation between declaration of the structure of page content and visual formatting.…”

Amalgamation of domain specific languages with behaviour

“…DSLs offer concepts specifically targeted at a particular domain, which allow experts in such domains to express their problems and requirements in their own languages. On the other hand, the higher amount of knowledge embedded in these concepts allows for much more complete and specialised generation of executable solution code from DSL models [35].…”

Behaviour Protection in Modular Rule-Based System Specifications

“…On the one hand this makes it easier for domain experts to express their problems and requirements. On the other hand, the higher amount of knowledge embedded in each concept allows for much more complete generation of executable solution code from a DSML model [27] as compared to a model expressed in a general-purpose modelling language.…”