Home:Learn XQuery:White Papers:Case for XQuery:Simplifies Data Integration

XQuery Simplifies Data Integration

XQuery is designed to support multiple XML information sources as input. An XQuery program’s primary functions are to select, filter, transform, join, and aggregate data across multiple data sources. The trick is that these data sources must be represented to the XQuery program as XML. Fortunately, there are also several products on the market, including Stylus Studio, that provide visual tools for building adapters that transform non-XML data sources (flat files, EDI, relational databases, and others) into XML. The net result is that a developer using current technology can build an XQuery program to join or aggregate data from diverse data sources, and produce XML as output. That XQuery program can then be deployed as a Web service which can be imported into another XQuery program, one that creates a composite view by combining this data source with other data sources in its integration scheme. This is just one example of the value of reusable, standards-based XQuery code.

XQuery's inherent data integration capability makes it a powerful tool for the modern application developer. Take Service-Oriented Architecture (SOA) applications, for example. Data integration in the emerging SOA world means dealing with data from multiple sources (relational databases, XML files, legacy applications, and Web services, to name a few). XML is the perfect language for uniformly expressing all of this data, and XQuery is the easiest and most powerful way to process it.

To best appreciate the problem XML and XQuery solve, consider how much time developer's today spend dealing with dynamic requirements for inter- and intra- enterprise information flows that must be integrated, current, and correct. A prime example of this can be seen in supply-chain management applications and the many other applications that integrate data from various sources in order to present unified customer and product information.

Building this data integration logic can be a costly, complex, and time-consuming process – some analysts believe that up to 70% of the effort on a typical systems integration project is devoted to ‘hand-coding’ data-level integration logic. What’s more, this code is very sensitive to any kind of change in the environment or even in the intended application usage. The net result is that developers often end up writing throw-away code. And spend 70% of their time doing it.

XML solves some of this problem by providing a lingua franca for data integration. To this end, XML Schemas exist for almost every industry sector imaginable to facilitate data exchange within organizations, as well as between customers, partners, distributors, and suppliers.

Even with XML, many developers are using hand-coded programming approaches that incorporate Java, DOM, XPath, XSLT and other methods, all in an effort to query and manipulate XML data. Low level approaches like DOM are difficult to write and maintain because the query expressions, the aggregation, and the transformation logic to be evaluated (the what) are so tightly bound to the underlying query processing strategy (the how) that even small changes in application requirements can require substantial re-coding efforts.

XQuery greatly simplifies XML querying and transformation by virtue of its simple and concise syntax. In addition, the developer using XQuery works with all data as an XML abstraction and can expect the underlying XQuery implementation to deal with accessing the physical data sources appropriately.