Sequential Search

Question

i need to extend the below code in order to
search for a Term for example "Computer Graphics"
rather than a single word such as "Graphics"
to run the below code, compile
than at command prompt type:
java DefinitionChecker <filename>
lets say the file contains
Computer Graphics is defined as a pictorial computer output produced on a display screen, plotter, or printer # Computer Graphics is purely delimited by science # Mobile Agents are defined as autonomous, intelligent programs that move through a network, searching for and interacting with services on the user's behalf. #
# Computer Graphicsis defined as science # Graphics is as ahmad # Mobile Agent can be defined as an agent that transports itself and its execution state through the net #
Assuming the user enters "Mobile Agent"
results should be :
From File XYZ.txt
Search results:
1) Mobile Agent can be defined as an agent that transports itself and its execution state through the net.
2) Mobile Agents are defined as autonomous, intelligent programs that move through a network, searching for and interacting with services on the user's behalf.
From File TravelPLan.txt
Search Results:
1) ..
2)..
import java.util.Vector;
import java.io.*;
public class DefinitionChecker{
String keyword;
String[] token1 = new String[]{" is "," are "," was "," be "," can be "};
String[] token2 = new String[]{" defined "," described "," delimited "};
String[] token3 = new String[]{" as "," by "};
String sentences[];
public static void main(String[] args){
if (args.length < 1){
System.out.println("USAGE : DefintionChecker keyword file");
DefinitionChecker dc = new DefinitionChecker(args);
public DefinitionChecker(String[] args){
try{
sentences = getArrayFromFile(new File(args[1]));
}catch(IOException ioe){
System.out.println(ioe);
keyword = args[0].toLowerCase();
for (int i = 0; i < sentences.length; i++){
int pos = 0;
//System.out.println("Checking : " + sentences);
if (sentences[i].toLowerCase().indexOf(keyword) > -1 && containsToken(token2,sentences[i].toLowerCase(),0) > -1){
if ((pos = containsToken(token1,sentences[i].toLowerCase(),pos)) != -1){
if ((pos = containsToken(token2,sentences[i].toLowerCase(),pos)) != -1){
if ((pos = containsToken(token3,sentences[i].toLowerCase(),pos)) != -1){
System.out.println("\t\tMATCH : " + sentences[i]);
}else{
//System.out.println("\t\t" + sentences[i] + " has no token3.");
}else{
//System.out.println("\t\t" + sentences[i] + " has no token2.");
}else{
//System.out.println("\t\t" + sentences[i] + " has no token1.");
}else{
//System.out.println("\t\t" + sentences[i] + " has no keyword or has no token2.");
private int containsToken(String[] tokens, String s, int pos){
int tPosition = -1;
for (int i = 0; i < tokens.length && tPosition == -1; i++){
tPosition = s.indexOf(tokens[i],pos);
return tPosition;
private String[] getArrayFromFile(File f) throws IOException{
FileReader reader = new FileReader(f);
Vector sentences = new Vector();
char[] cbuf = new char[1];
String delimiter = "#";
String sentence = "";
String c = "";
while (reader.read(cbuf) != -1){
c = new String(cbuf);
if (c.equals(delimiter)){
sentences.add(sentence);
sentence = "";
}else{
sentence += c;
String[] sentenceArray = new String[sentences.size()];
sentenceArray = (String[])sentences.toArray(sentenceArray);
return sentenceArray;

Answer 1

Rather than using the exhaustive code currently commented out, can't you just use the keywords AS the token in your tokenizer? I am assuming the keywords need to be matched "in sequence". (So "Computer Graphics" will not match against "Graphics Computer")
The hash (#) seems to be the "end of sentence" delimiter. So, create a StringTokenizer using the keyword (args[0]) as the delimiter.
I am assuming this is for a Uni assignment or something, so I won't worry about performance issues.
Read the whole file into a String (not generally good practice, but for the sake of a Uni assigment who cares!). Use your tokenizer to parse the string.
The tokenizer will delimit the string on the basis of the keyword. For each result the tokenizer gives you, truncate the result at the point of the hash (#). This should give you a string starting with your keyword, and ending prior to the hash.
EG:
import java.util.*;
String keyword = args[0];
String fileContents; // Put the contents of your file into this
StringTokenizer st = new StringTokenizer(fileContents, keyword);
List results = new LinkedList();
String result = null;
while(st.hasMoreTokens) {
// Get the actual result
result = st.nextToken();
// truncate the result at the hash
result = result.substring(0, result.indexOf("#");
//Add the original keyword because the tokenizer will have removed it
result = keyword + result;
// Add to the list of results
results.add(result);
No guarantees on the "compilability" of this, but you should get the idea.
Now.. how about them Duke Dollars? :)