StringTokenizer doesn't work like that.  There are kludges, but ideally,
you'd use a class that does.  For example, Jon's JlsTokenizer at
http://www.yoda.arachsys.com/java/skeetutil/.

Since StringTokenizxer is known not to be able to handle this problem, and
IMHO is really dreadful to use, why not do this instead:

String source = "dasbndc|716372|3|5|||||abcd|||";

String[] array = source.split("\\|",-1);

The "-1" argument forces processing of trailing blank fields. Note that I
included some in the data for this test.

Like this:

**************************************************

public class Test {
 
  public static void main(String[] args)
  {
     String source = "dasbndc|716372|3|5|||||abcd|||";
     
     String[] array = source.split("\\|",-1);

     for(int i = 0;i < array.length;i++) {
        System.out.println("[" + array[i] + "]");
     }
  }
}

**************************************************

Result:

[dasbndc]
[716372]
[3]
[5]
[]
[]
[]
[]
[abcd]
[]
[]
[]

Try a third parameter to the constructor - check the docs.

You need to return the delimiters as tokens in order to have
StringTokenizer not skip multiple delimiters.

But if you can use 1.4.x APIs, do as others have suggested and use
String.split().

    String tmpstr = "";
    while (tk.hasMoreTokens()) {
        tmpstr = tk.nextToken();
        if (tmpstr == null)
            tmpstr = "";
        System.out.println(tmpstr);
    }

Much better

String[] array = line.split("|");

Use the com.Ostermiller.util.StringTokenizer which has a method that
you can set to have it return empty tokens.

http://ostermiller.org/utils/StringTokenizer.html