| By Kunal Mittal | Article Rating: |
|
| October 3, 2003 12:00 AM EDT | Reads: |
13,558 |
BEA Liquid Data for WebLogic provides a unified view of data aggregated from multiple resources such as databases, XML files, Web services, EJBs, or Java 2 Connector Architecture (J2CA) adapters. At a very high level, Liquid Data can be thought of as an enterprise-wide JOIN tool that provides an XML view of data from any source. This data can be accessed using evolving technologies such as XQuery.
Liquid Data provides several different mechanisms through which Data Views in Liquid Data can be leveraged by the rest of the BEA Platform. In this article I'll introduce several ways of accessing Data Views outside of the Liquid Data for WebLogic Data View Builder itself.
Quick Introduction to Data Views
Data Views are rich data sources built using XQuery
technologies to aggregate data from discrete sources using a single
query. BEA Liquid Data for WebLogic provides a rich graphical
interface, called the Data View Builder, through which users can set
up the data sources they need to access, set up security and
authentication information, and then define various queries that they
need to execute. The Data View Builder allows users to define XML
Schemas for the individual data sources or the resulting data that
they get from the execution of the query.
Data Views can aggregate data from various sources such as relational databases (using standard SQL or stored procedures), XML, or delimited files stored in a file system; Web services; and legacy EIS. Using the Data View Builder, one or more of these sources' schemas are aggregated to form the target schema. Users can define JOIN conditions, and other data aggregation or selection logic that you are used to doing while accessing relational data using SQL. Users also have the ability to design custom functions to access data in different way.
This article assumes that you have already created some Data Views. I'll focus on the various mechanisms available to access them.
Accessing Data Views
Data Views in Liquid Data are available through various
access mechanisms to the rest of the BEA Platform. Any Java program
with the appropriate security credentials, and that can access the
Java Naming and Directory Interface (JNDI), can issue queries against
a Liquid Data server. The different types of Java programs include:
- Stand-alone Java applications
- Java servlets
- JavaServer Pages (JSPs)
- JavaBeans
- Enterprise JavaBeans (EJBs)
- Web services
- Business process management (BPM) components or workflows in BEA WebLogic Integration
- Portlets in BEA WebLogic Portal
A stateless session EJB exposes the query API, which delegates queries through a query processor to the underlying data stores. Any Java program that can access EJBs can use this EJB-based API. Stand-alone Java applications, Javabeans, and other EJBs all use this API, and collectively are called EJB clients. These programs can use the EJB remote or local interfaces directly.
EJB clients are any applications that invoke queries on the Liquid Data server using the Liquid Data EJB API. All Java clients can leverage the flexibility and data integration properties offered by XQuery to meet their data access needs. All of these clients access the EJB remote interfaces directly; therefore, they can be collectively characterized as EJB clients. The Data View builder that is part of the core Liquid Data product is a special kind of EJB client.
The four basic steps to access Data Views through this mechanism are:
JNDI is used to look up the EJBs and standard EJB calls are used to connect to a Liquid Data server. Once connected, queries can be executed anonymously or through authentication itself by passing a security context to the corresponding JNDI environment properties SECURITY_PRINCIPAL and SECURITY_CREDENTIALS.
In most cases users would want to execute queries based on certain conditions. XQuery-based parameters are used to pass search conditions to a Liquid Data server. The following code snippet is an excerpt of how to pass search parameters to a data view.
import com.bea.ldi.server.common.QueryParameters;
...
QueryParameters qp = new QueryParameters();
qp.setString("title", "WebLogic Developers Journal");
execute (queryString, [qp]) command
If the query is a stored, named query, you would use
executeStored (queryName, [qp]) command
The "qp" parameter is optional based on whether you are passing query parameters.
When the query is executed, the Liquid Data server returns an unformatted XML string. You can use the toXML() or getDocument() methods to receive an XML or DOM format with the results of the query.
Access to Data Views Through JSP
Liquid Data for WebLogic provides a set of tag libraries that
can be used by a JSP to access Data Views. The basic tags provided
are a "query" tag and a "param" tag.
The "query" tag is used to connect to a Liquid Data server and execute a query. It takes the name of the query, the connection to the server, and the username and password needed to authenticate with the server
The "param" tag is used to pass parameters for the query using name/value pairs. Listing 1 shows some sample code on how to use the "query" and "param" tags.
You can use these tag libraries to execute both ad hoc and stored queries. The resulting XML can be displayed on the JSP using XLST translations. WebLogic server comes with an XSLT JSP Tag Library that can be used for an XSL transformation. Alternatively, you could use a third-party or open source library for this functionality.
Access to Data Views Through Web Services
A stored query in Liquid Data can be exposed as a Web service
through the Liquid Data administration console. The URL of the WSDL
of a generated Web service has the following pattern:
http://host:port/liquiddata/query_name/webservice?WSDL
Once you have exposed a Liquid Data query as a Web service, you would use standard Web services mechanisms to look up the WSDL and execute the Web service.
BEA Liquid Data for WebLogic 8.1 also provides a control in WebLogic Workshop that can be used to invoke this query. WebLogic Workshop can also be used to test the Liquid Data Web service-based query.
Access to Data Views Through WebLogic Integration BPM Components
BPM components in WebLogic Integration would typically use
the stateless session EJB API discussed above to access Liquid Data
queries. A business operation in WebLogic Integration would define
the JNDI name of the Query EJB to be invoked
(com.bea.ldi.server.QueryHome) and the method to invoke on the EJB.
Liquid Data helps BEA WebLogic Integration integrate data from various relational, non-relational, and Web sources. The Liquid Data views are read-only. WebLogic Integration components such as workflows are used to initiate two-way transactions from the rich set of data that is provided using the Data Views in BEA Liquid Data for WebLogic.
Access to Data Views Through Portlets
Portlets and other components in BEA WebLogic Portal can use
all the different mechanisms discussed to access Liquid Data queries.
Portlets could use Web services or the JSP tag libraries. Other
Portal components could use the EJB APIs.
Liquid Data thus becomes a rich data feed and WebLogic Portal becomes the layer that handles the presentation of this rich data.
Summary
This article has briefly introduced the several ways to
leverage Data Views in BEA Liquid Data for WebLogic outside of Liquid
Data itself. Basically, any piece of Java code that can access an EJB
or a Web service can execute queries on a Liquid Data server. Even
JSP-based applications can leverage Liquid Data queries directly,
without having to go through an EJB tier.
References
Published October 3, 2003 Reads 13,558
Copyright © 2003 SYS-CON Media, Inc. — All Rights Reserved.
Syndicated stories and blog feeds, all rights reserved by the author.
More Stories By Kunal Mittal
Kunal Mittal is a consultant for implementation and strategy for Web services and services-oriented architectures. He has co-authored and contributed to several books on Java, WebLogic and Web services.
- The Economics of Cloud Computing Analyzed
- GovIT Expo Highlights Cloud Computing
- Cloud Computing Best Practices
- Tactical Cloud Computing Panel at 1st Annual GovIT Expo
- Why SOA Needs Cloud Computing - Part 1
- The Cloud Transition: What Does It Mean For You?
- Cloud Computing Strategy
- IBM’s Mainframe Monopoly Threatened by BMC Founder’s Shop
- Economy Drives Adoption of Virtual Lab Technology
- Virtualization Expo Call for Papers Deadline December 15
- Oracle in Leader's Quadrant for Enterprise Application Servers
- Oracle Fusion Middleware Delivers World Record Single-Node Result
- The Economics of Cloud Computing Analyzed
- The Difference Between Web Hosting and Cloud Computing
- GovIT Expo Highlights Cloud Computing
- Cloud Computing Best Practices
- Tactical Cloud Computing Panel at 1st Annual GovIT Expo
- Citrix Aims To Cripple VMware’s Cloud Designs
- Product Evaluation: JBoss TCO Calculator
- Why SOA Needs Cloud Computing - Part 1
- Build Reliability into Cloud Computing for SMBs
- Perhaps SOA is More Strategy Than Architecture
- EC Wrong, Wrong, Wrong – and Sloppy to Boot: Intel
- Five Reasons to Choose a Private Cloud
- Java vs C++ "Shootout" Revisited
- Where Are RIA Technologies Headed in 2008?
- Configuring Eclipse for Remote Debugging a WebLogic Java Application
- Migrating a JBoss EJB Application to WebLogic
- XA Transactions
- The Top 250 Players in the Cloud Computing Ecosystem
- An Introduction to Abbot
- WebLogic Tutorial: "Integrating Apache Poi in WebLogic Server"
- Eclipse "Pollinate" Project to Integrate with Apache Beehive
- Failover and Recovery of Enterprise Applications - Part 1
- Cover Story: A Practical Solution to Internationalization of a J2EE Web App
- WebSphere vs WebLogic: IBM and BEA Spar Over SPEC Results
































