Visual modeling of XML constraints based on a new extensible constraint markup language.

摘要

The eXtensible Markup Language (XML) is rapidly becoming the industry standard format for exchanging information across the networks. The validity of XML documents is very important for users and/or applications that consume these documents. The data model of an XML document is usually defined in Document Type Definitions (DTDs) or XML Schemas that can be used to validate their instances—XML documents. It is not enough to check the validity of XML documents using XML Schema or DTD because they cannot specify some non-structural constraints like dependencies and consistencies.; Research has been conducted on how to express XML constraints and how to use them to validate XML documents. However, they do not have a comprehensive classification of XML constraints. Some of them are not flexible enough to define reusable XML constraint documents. Some of them are not convenient to design XML constraint documents. Also, none of them discusses the automation of XML constraint document design. In this research, we conduct a critical review of the existing XML constraint languages, from which a comprehensive classification of XML constraints is made. We propose a new XML constraint language—eXtensible Constraint Markup Language (XCML). It provides the syntax to define XML constraints that are either assertion-based or rule-based. It is much more expressive than the existing XML constraint languages by supporting dynamic constraints, composite constraints, and more. We also propose a framework of visual modeling of XCML constraints that is highly integrated with XML data modeling at the design phase. It automates the generation of XML schema and XCML instance documents.; We have developed a prototype of the model-driven automation of XML Schema and XCML instance document generation. The generated XML schemas are used for syntactic validation. The XCML document is transformed to an XSLT stylesheet by a reusable stylesheet we developed. The transformed stylesheet is then used for semantic validation. The result of this research can be used in XML data management, data mining and data warehousing, and decision support systems for various industry domains such as healthcare information technology and E-Commerce.