Author time application development

 < Day Day Up > 



The model for Web content differs from the model for a traditional document because Web content is made up of fragments that can be assembled in multiple combinations to create Web pages. Those fragments can be reusable assets that can be used in multiple locations of different pages, or local assets that exist only as part of a particular page. Those assets and the pages themselves are based on templates that specify common properties and features of all assets or pages based on a particular template. Those fragments are usually created, tested, and managed by different contributors. Therefore, a rigorous workflow system is essential to create and publish verifiable and approved Web pages. Moreover, the WCM application also has to maintain referential integrity of the links between published pages.

In the WCM application, each page consists of a hierarchical structure of nested fragments. Each fragment is represented as an XML object. The Content Manager data modeling features allow the creation of a data model that reflects the different components of a page such as reusable assets, local assets, and templates. The data model was captured in XML files and, using Java APIs, imported into Content Manager, dynamically creating the Content Manager internal representation of the model. The metadata and nonbinary fragments are stored as XML data in the Content Manager library server. Images and other binary content are stored in the Content Manager resource manager. Each fragment may have references to supporting or dependent fragments using Content Manager references or links.

The WCM application is a Web application running inside WebSphere Application Server Version 4. For programming ease, a Java object model that reflects the data model stored in Content Manager was developed. In the Web application, the methods and properties exposed by the Java objects to manipulate content were used, along with a layered approach that makes possible future enhancements and additions. The extensibility of Content Manager's access control mechanisms were leveraged to implement a sophisticated and unique access control policy required by the customer and tied with a workflow process built entirely in Content Manager that controlled each step of the process of creating, managing, and publishing the content. A tool was developed to import information about users from a corporate LDAP repository and to maintain the synchronicity of this information. Content Manager's connection pooling features were used in order to efficiently support multiple concurrent users. The XPath and XQuery features were heavily leveraged to search, access, and retrieve content from Content Manager.

As required by the customer, the new interface was developed entirely in a browser. There are two options for assembling pages from fragments for use in a browser.

  1. Assemble a page in a combination of JavaServer Pages (JSP), JavaBeans, and servlets in the application server and sending the assembled page to the browser.

  2. Send the XML fragments to the browser, then assemble and present the fragments at the browser.

The second option allows caching of small bits of data and allows the invalidation of a single fragment instead of the entire page. The disadvantage to this option is that currently, only Internet Explorer Version 6 supports this feature, making it impossible to use this approach in an Internet application. However, because the WCM application is an intranet application, it would be easy to restrict the browser to Internet Explorer.

The second option was chosen and the entire user interface was developed using JavaScript, DHTML, and XML. This browser-based user interface avoids distribution of the client software package on a client machine. Both administration and content creation are managed by a single user interface allowing users with different roles and responsibilities to reuse a common interface.

The tools used to build this WCM application were WebSphere Studio Application Developer (WSAD), VisualAge for Java along with XMLSpy and TogetherJ.



 < Day Day Up > 



High-Volume Web Sites Team - More about High-Volume Web Sites
High-Volume Web Sites Team - More about High-Volume Web Sites
ISBN: N/A
EAN: N/A
Year: 2003
Pages: 117

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net