首页> 外文会议>Fundamental Approaches to Software Engineering >Consistent Adaptation and Evolution of Class Diagrams during Refinement
【24h】

Consistent Adaptation and Evolution of Class Diagrams during Refinement

机译:细化过程中类图的一致适应和演化

获取原文
获取外文期刊封面目录资料

摘要

Software models are key in separating and solving independent development concerns. However, there is still a gap on how to transition design information among these separate, but related models during development and maintenance. This paper addresses the problem on how to maintain the consistency of UML class diagrams during various levels of refinement. We present a new approach to automated consistency checking called ViewIntegra. Our approach separates consistency checking into transformation and comparison. It uses transformation to translate model elements to simplify their subsequent comparison. Transformation-based consistency checking, in the manner we use it, is new since we use transformation to bridge the gap between software models. No intermediate models or model checkers are required; developers need only be familiar with the models they design with and none other. The separation of transformation and comparison makes our approach to consistency checking more transparent. It also makes our approach useful for both propagating design changes among models and validating consistency. This gives developers added flexibility in deciding when to re-generate a model from scratch or when to resolve its inconsistencies. Although this paper emphasizes the adaptation and evaluation of class diagrams, we found our technique to be equally useful on other models. Our approach is tool supported.
机译:软件模型是分离和解决独立开发问题的关键。但是,在开发和维护期间如何在这些独立但相关的模型之间转换设计信息仍然存在差距。本文解决了在各种细化级别期间如何保持UML类图的一致性的问题。我们提出了一种称为ViewIntegra的自动一致性检查新方法。我们的方法将一致性检查分为转换和比较。它使用转换来转换模型元素,以简化它们的后续比较。由于我们使用转换来弥合软件模型之间的鸿沟,因此以我们使用的方式进行的基于转换的一致性检查是新的。不需要中间模型或模型检查器;开发人员只需要熟悉与他们一起设计的模型即可。转换和比较的分离使我们的一致性检查方法更加透明。这也使我们的方法对于在模型之间传播设计更改和验证一致性都非常有用。这使开发人员在决定何时从头开始重新生成模型或何时解决其不一致之处时具有更大的灵活性。尽管本文强调了类图的适应和评估,但我们发现我们的技术在其他模型上同样有用。我们的方法受工具支持。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号