Tracking Service-Oriented and Web-Oriented Architecture

SOA & WOA Magazine

Subscribe to SOA & WOA Magazine: eMailAlertsEmail Alerts newslettersWeekly Newsletters
Get SOA & WOA Magazine: homepageHomepage mobileMobile rssRSS facebookFacebook twitterTwitter linkedinLinkedIn


SOA & WOA Authors: Andreas Grabner, TJ Randall, Lori MacVittie, Dynatrace Blog, Cynthia Dunlop

Related Topics: XML Magazine, SOA & WOA Magazine

XML: Article

Index XML Documents with VTD-XML

How to turn the indexing capability on in your application

Traditionally DOM or SAX-based enterprise applications have to repeat CPU-intensive XML parsing when accessing the same documents multiple times. VTD-XML 2.0 introduces a simple general-purpose XML index called VTD+XML (http://vtd-xml.sourceforge.net/persistence.html) that eliminates the need for repetitive parsing of those applications.

This article combines various examples and the latest benchmark reports to show you how to get started with this indexing. This article also discusses various scenarios and use cases where you may find VTD+XML useful.

Avoid Repetitive XML Parsing with VTD-XML
As discussed in "Simplify XML processing with VTD-XML," to date one of underlying assumptions in XML application development is that an XML document must be parsed before anything else can be done with it. In other words, the processing logic of XML applications can't start without parsing. Frequently considered a threat to database performance, XML parsing is usually many times slower than other XML operations such as XPath evaluation. When those applications perform multiple read-only access to XML data that don't change very often, wouldn't it be nice to able to eliminate the overhead of associated repetitive parsing?

With the native XML indexing feature introduced in version 2.0 of VTD-XML, you can do precisely that. VTDGen, the class encapsulating various parsing routines, now adds "readIndex(...)" and "writeIndex(...)." VTD-XML 2.0 also introduces two new exceptions: indexWriteException and indexReadException.

Let me put those new methods into action and show you how to turn on the indexing capability in your application. Consider the following XML document:

   <purchaseOrder orderDate="1999-10-21">
     <item partNum="872-AA">
       <productName>Lawnmower</productName>
       <quantity>1</quantity>
       <USPrice>148.95</USPrice>
     </item>
   </purchaseOrder>

Below is a simple pre-2.0 VTD-XML code named "printPrice.java" that prints out the content of the element "USPrice." Notice that it parses the XML file and then uses XPath to filter out the target nodes.

import com.ximpleware.*;
import com.ximpleware.xpath.*;
public class printPrice{
   public static void main(String args[]){
     VTDGen vg = new VTDGen();
     try{
       if (vg.parseFile("po.xml",true)){
         VTDNav vn = vg.getNav();
         AutoPilot ap = new AutoPilot(vn);
         ap.selectXPath("/purchaseOrder/item/USPrice/text()");
         int i=-1;
         while((i=ap.evalXPath())!=-1){
           System.out.println(" USPrice ==> "+vn.toString(i));
         }
       }
     }catch(Exception e){

     }
   }
}

A few changes are needed to add VTD-XML's new indexing capability to the Java code above. First, you need to read in the XML document, parse it, and then write out the indexed version of the same XML document. From that point onward, your application can run XPath query or processing logic directly on top of the index, saving the CPU cycles of parsing the XML document again. The following code snippets (named "genIndex.java" and "accessIndex.java" respectively) show you how to generate and access the index. Notice that, when executed sequentially, both applications produce the identical output as "printPrice.java."
The first application (genIndex.java) reads in "po.xml" and produces "po.vxl."

import com.ximpleware.*;
import com.ximpleware.xpath.*;
public class genIndex{
   public static void main(String args[]){
     VTDGen vg = new VTDGen();
     try{
       if (vg.parseFile("po.xml",true)){
         vg.writeIndex("po.vxl");
       }
     }catch(Exception e){
     }
   }
}

The second application (accessIndex.java) loads "po.vxl" and filters the document using XPath expression "/purchaseOrder/item/USPrice/text()."

import com.ximpleware.*;
import com.ximpleware.xpath.*;
public class accessIndex{
   public static void main(String args[]){
     VTDGen vg = new VTDGen();
     try{
       VTDNav vn = vg.loadIndex("po.vxl");
       AutoPilot ap = new AutoPilot(vn);
       ap.selectXPath("/purchaseOrder/item/USPrice/text()");
       int i=-1;
       while((i=ap.evalXPath())!=-1){
         System.out.println(" USPrice ==> "+vn.toString(i));
       }
     }catch(Exception e){
     }

   }
}


More Stories By Jimmy Zhang

Jimmy Zhang is a cofounder of XimpleWare, a provider of high performance XML processing solutions. He has working experience in the fields of electronic design automation and Voice over IP for a number of Silicon Valley high-tech companies. He holds both a BS and MS from the department of EECS from U.C. Berkeley.

Comments (0)

Share your thoughts on this story.

Add your comment
You must be signed in to add a comment. Sign-in | Register

In accordance with our Comment Policy, we encourage comments that are on topic, relevant and to-the-point. We will remove comments that include profanity, personal attacks, racial slurs, threats of violence, or other inappropriate material that violates our Terms and Conditions, and will block users who make repeated violations. We ask all readers to expect diversity of opinion and to treat one another with dignity and respect.