X3D File Parser Is Way Too Slow

I've recently tried to write a file parser for the .x3d file type as it's one of the few 3d model types I can find written in easy to understand (and interpret) English. While it technically does work (or what I have so far), the issue is that it takes far too long (17 minutes) to parse just a part of the file (the vertices of the model).This is the code for the object itself:

import java.io.*;
import com.sun.j3d.utils.geometry.*;
import javax.media.j3d.*;
import javax.vecmath.*;
public class ModelLoader {
BranchGroup object = new BranchGroup();
BufferedReader reader;
String line = "start";

[code]....

Anyway, here are some observations I've made about the file:

-The line in the file itself that contains all the vertices is all one giant string according to the file.
-There are apparently 18,000 vertices in all (according to the command at line 18).
-Unless Netbeans automatically terminates infinite loops after a certain amount of time, it is most definitely NOT an infinite loop as I have seen the program terminate on its own (after 17 minutes, though).

I had a theory that maybe organizing each vertex into its own line and then having the program switch to the next line when it's done reading that vertex might make the program run faster, but I'm not sure why it would, so I thought I'd come here in case that theory turned out to be a dead end.