Research And Implementation Of MOF-based 2-Level Modeling Tool

Human ability to discern and understand complex question is limited. In order to narrow down the scope of research, people always focus on one aspect of the studied things. Through multiple studying we can understand the outline of things. The abstraction and simplification of something that exists in reality is called model. The abstract objects are either concrete things or one part of some system. The abstraction of something that exists in reality and the description of abstracted thing are two main factors of model. We can define accurately model as follows: A model is a description of (part of) a system written in a well-defined language. A well-defined language is a language with well-defined form (syntax), and meaning (semantics), which is suitable for automated interpretation by a computer. A good model gets rid of secondary and irrelevant properties and focuses on those properties that can reflect system's nature. In using traditional modeling tools, modelers just make use of given modeling elements to model. If modelers need extend modeling elements, they only wait until the designer of the modeling tool goes to update it. The modeling element is classified in the general and the special. The general modeling elements are applicable in most domains and the specific modeling elements are only applicable in given domain. Commonly, we facing a complicated system, it is difficult for us to descript the system. We need a new modeling mechanism. Firstly, according to question domain we create our own domain modeling language, and then we can begin to model domain model by means of this domain modeling language. This is the idea of 2-level modeling. Model Driven Architecture (MDA) and Meta Object Facility (MOF) which are defined by Object Management Group (OMG) support the 2-level modeling mechanism in theory studying and implementation. The central theme of the MOF approach to metadata management is extensibility. The aim is to provide a framework that supports any kind of metadata, and that allows new kinds to be added as required. In order to achieve this, the MOF has layered metadata architecture. The key feature of the MOF metadata architecture is the meta-metamodeling layer that ties together the metamodels and models. The MOF layer architecture provides the theoretical layer concept for 2-level modeling mechanism. We can depict 2-level modeling architecture with MOF layer. According to the layer of 2-level modeling architecture, the central theme of 2-level modeling is defining a meta-metamodel, namely a metamodeling language. A language is defined by three main things: structure, editor, and semantics. Its structure defines its abstract syntax, what concepts are supported and how they can be arranged. Its editor defines its concrete syntax, how it should be rendered and edited. Its semantics defines its behavior, how it should be interpreted and how it should be transformed into executable code. Of course, language also has other aspects, such as constraints and type systems. The MOF model that is in third layer of MOF architecture entirely defines the structure of modeling language. The description ability of MOF is powerful. The MOF model provides a set of modeling components, including class, association, aggregation, conference, datatype, package, constraint and consistency. These modeling components are enough to define the abstract syntax of any kind of language. As for semantics, it is defined by the framework that uses models. Different framework perhaps provides different semantics for same model. Ifmodels are used for transformation, the semantics should be defined by transformation tools. The process of 2-level modeling is iterative. This process is different from ordinary programming language. In most cases, modelers will work with two levels of modeling, the metamodeling level and modeling level. They define the metamodeling languages in the meta level, and model in the modeling level. In 2-level modeling, each node in modeling level has a type which is a link to another node the meta level. The node in the modeling level is said to be an instance of the type. The meta level type node defines what relationships its instances can have and also what properties the will have. There are two kinds of models in MDA: Platform-Independent Model (PIM) and Platform-Specific Model (PSM). In 2-level modeling, the final models by modeling are PIMs. According to different use context, those PIMs can be transformed into other PSMs. OMG defines XMI (XML Metadata Interchange) that provides a mechanism of model transformation. XMI supports the map from MOF to XML. XML is popular in information interchanging so that XMI can be soon accepted. Many modeling tools have supported XMI, including Rational Rose. Furthermore, OMG is defining QVT(Query View Transformation) specification. This project aims to provide a standard for model transformation tools. Once the specification is defined, the transformation and management of models could be unified. In MDA, RAS (Reusable Asset Specification) provides a conference standard for model reuse. This Specification provides a set of guidelines and recommendations about the structure, content, and descriptions of reusable software assets. Recognizing that there are different categories of reusable software assets, the specification identifies some categories, or rather types or profiles, and provides general guidelines on these profiles. RAS addresses the...