An Automatic Approach to Generate XML Schemas from Relational Models

Data storage in enterprise information systems is usually performed in relational databases, and XML is a common standard for exchanging data within and among organizations. Automating databases update from data in XML format remains a challenge. In this paper, we propose a methodology to efficiently generate, from a relational model, an XML schema which can then be used by an XML middleware platform that performs the mediation management between the source and target XML representations. Our method is composed of two main steps, with the objective of both avoiding any data redundancy by performing a high level of nesting, and preserving the integrity constraints. The first step consists in an algorithm that stratifies all the relations of the database model in different levels depending on the existing functional dependencies and on the foreign keys degrees. The second step consists in an algorithm based on a set of generic XML Schema fragments templates that automatically creates an XML schema according to the classification performed by the previous step. Our methodology has been successfully tested in the eHealth domain between an XML representation of an open format ISO standard communications protocol embedding biosignals and related metadata and a European relational reference model including these data.