Occurrences

Paper

../papers/03-02-04/03-02-04.html

Date of Presentation

Wednesday, 22 May

Time of Presentation

11.45

Presentation Level

Technical

Abstract

Relational databases are a mature technology that dominates the implementation of database systems. XML is a new technology that is attracting a great deal of interest and there are indications that it may dominate some areas of information system development, particularly amongst distributed systems. The role of XML within information systems and legacy information systems is still to be precisely determined however. This paper describes a generalised mapping between relational databases and XML documents that is based on the tree structure of the Document Object Model (DOM). It also describes XR2 (XML to Relational Translation), an implementation of the solution we describe XR2 automates the major functions involved in the solution proposed, which achieves a mapping from a relational database to an XML format. These are: 1. Facilitate the specification of an XML schema that maps from the metadata of a relational database to the XML output structure that a user requires. Given the XML schema produced in 1., a. XR2 will construct queries to extract the data required. b. XR2 generates output that places XML formatting appropriate to the data retrieved by the queries. We have investigated the use of two types of XML schema for the purpose of database extraction, namely Document Type Definitions (DTD) and the schema specified by the XML-Data proposal of the World Wide Web Consortium. XR2 generates both DTD and XML schema. XR2 will also extract data from a database based on a DTD or XML schema. While it is possible to achieve an extraction process using either a DTD or XML schema, it is simpler to use an XML schema, the reasons for which will be explained in subsequent sections. We describe a general solution to the problem of extracting XML formatted data from a database. The solution described is to a generalised problem, that is, only natural assumptions are made on the structure of the database and on the structure of the XML documents to be produced. A natural assumption is taken here to mean that it is possible to create queries on the database that will result in the production of the XML document. The solution we describe can be applied independently of the database platform, by conforming to standard relational database mechanisms (SQL). The current implementation of the solution operates external to the database, examining database structure and querying the database when necessary. The extraction process can be based on ad-hoc XML schema. That is, the extraction can operate on a portion of the database defined in a given XML schema. This allows the needs of any particular extraction to be met through the schema definition. The database structure need not be altered.