xml parsing using dom

[sunil]

ParserEx.java
----------------
import com.ibm.xml.parser.*;
import org.w3c.dom.*;
import java.io.*;
class ParserEx{
public ParserEx(){}
public static void main(String[] args){
try{
Document doc;
String ROOT_ELEMENT_TAG = "person";
String[] colNames = {"fname","addr","rollno","first","second"};
String URL1 ="C:/Tomcat/static/AddressBook.xml";
java.io.FileReader filereader = new FileReader(URL1);
java.io.BufferedReader reader = new BufferedReader( filereader ) ;
Parser p = new Parser(URL1);
doc = p.readStream(reader);

//display table column
//for(int i=0; i<colNames.length; i++){
//System.out.println( colNames);
//}
//need to iterate the doc to get the fields in it
int rowCount = XmlUtils.getSize( doc , ROOT_ELEMENT_TAG );
System.out.println("-----------------------------------");
System.out.println("count="+rowCount);
System.out.println("-----------------------------------");
for(int r=0; r<rowCount; r++) {
Element row = XmlUtils.getElement(doc , ROOT_ELEMENT_TAG , r );
int colCount = colNames.length;
for(int c=0; c < colCount; c++) {
System.out.println( XmlUtils.getValue( row , colNames[c] ));

}//end for c=0...

}//end for r=0...
}catch(Exception e){
System.out.println( e );
}

}
}
------------------------------------
XmlUtils.java
===============

import org.w3c.dom.*;

public class XmlUtils{

public static Element getElement( Document doc , String tagName
,
int index ){
//given an XML document and a tag
//return an Element at a given index
NodeList rows = doc.getDocumentElement().getElementsByTagName(
tagName );
return (Element)rows.item( index );
}

public static int getSize( Document doc , String tagName ){
//given an XML document and a tag name
//return the number of ocurances
NodeList rows = doc.getDocumentElement().getElementsByTagName(
tagName );
return rows.getLength();
}

public static String getValue( Element e , String tagName ){
try{
//get node lists of a tag name from a Element
NodeList elements = e.getElementsByTagName( tagName );

Node node = elements.item( 0 );
NodeList nodes = node.getChildNodes();

//find a value whose value is non-whitespace
String s;
for( int i=0; i<nodes.getLength(); i++){
s = ((Node)nodes.item( i )).getNodeValue().trim();
if(s.equals("") || s.equals("\r")) {
continue;
}
else return s;
}

}
catch(Exception ex){
System.out.println( ex );
ex.printStackTrace();
}

return null;

}

public static void printNodeTypes( NodeList rows ){
System.out.println( "\tenumerating NodeList (of Elements):");
System.out.println( "\tClass\tNT\tNV" );
//iterate a given Node list
for( int ri = 0 ; ri < rows.getLength() ; ri++){
Node n = (Node)rows.item( ri );
if( n instanceof Element) {
System.out.print( "\tElement" );
}
else System.out.print( "\tNode" );

//print out Node type and Node value
System.out.println(
"\t"+
n.getNodeType() + "\t" +
n.getNodeValue()
);
}
System.out.println();
}


}//end class


to execute this code you need ibm dom parser.

 

[Joe Kesselman]

1) Did you have a question, or are you just sharing code?

2) What part of this is relying on the IBM parser, and why? Most DOM
code should work across implementations (that's the point of using a
standard API, after all). Until recently, loading a DOM from a file and
saving it back was unfortunately not standardized, but DOM Level 3 has
introduced standard calls for those purposes and DOM users should
consider converging on those... or at least, if you're working in Java,
on the JAXP interfaces which that community agreed upon... which again
would help keep your code relatively parser-independent. (I believe the
current version of IBM's parser supports both DOM Level 3 and JAXP, but
I haven't checked recently.)