基于目标模型的横切关注点识别及语义连接点定义方法研究

横切关注点的识别是面向方面需求工程中的一个关键问题，而面向方面软件的演化需要更加健壮的语义连接点模型。现有的一些典型方法存在横切关注点识别方法复杂、识别精度有待提高、以及连接点模型难以支持大粒度基础程序的演化等不足。本项目拟重点围绕以下两个问题开展研究：1)基于目标模型的横切关注点识别方法;2)基于目标的语义连接点模型及其实现方法。研究目标模型的元模型及横切关注点的识别规则，在元模型指导下使用XML描述目标模型，通过XQuery语言按指定的识别规则查找横切关注点，以达到方法实用、识别精度较高的目的。研究不依赖于基础程序结构的基于目标的语义连接点模型，通过程序转换把语义连接点定义转换成AspectJ语言的连接点定义，以实现支持大粒度基础程序演化的目的。项目的研究成果将为横切关注点的识别、构建可适应面向方面软件提供理论和工具支撑。

Identification of crosscutting concern is the key problem of Aspect-Oriented Requirement Engineering. Meanwhile, the evolutionary of Aspect-Oriented Software needs more robust semantic join point model. The existing typical methods have many drawbacks. Identification method of crosscutting concern is complex and the identification accuracy should be improved, moreover, the join point model is difficult to support the evolutionary of large granularity base program. The project would carry out research mainly on the following two questions: the identification of crosscutting concern based on the goal model and the semantic join point model and its implementation based on the goal model. The project would research on the meta-model of goal model and the identification rules of crosscutting concerns. Then the goal model is described by XML with the guidance of the meta-model. Last crosscutting concerns are identified by XQuery language with the identification rules. The method is practical and has well identification accuracy. The project would also research on semantic join point model based on goal and independent from the structure of base program. Through program transformation, the definition of semantic join point would be transformed to the definition of join point of AspectJ language in order to support the evolutionary of large granularity base program. The research result of project would provide theory and tools support for the identification of crosscutting concern and the construction of adaptive Aspect-Oriented Software.