StringTokenizer的麻烦

Question

I'm getting the following error message and I can't seem to figure out the problem. Would really appreciate any help. The error message reads as:-

BaseStaInstance.java:68: cannot find symbol

symbol : constructor StringTokenizer(java.lang.Object,java.lang.String)

location: class java.util.StringTokenizer st = new StringTokenizer(buf,",");

                                  ^

Here, BaseStaInstance is my main public class.

The class that implements this StringTokenizer is as follows:-

class ServerConnect extends Thread {

Socket skt;
int iProcessId, iInProcessId;
int iOwnTimeStamp, iInTimeStamp;
ServerConnect scnt = null;

ObjectOutputStream myOutput;
ObjectInputStream myInput;

ServerConnect(){}
ServerConnect(Socket connection, int iProcessNo) {
    this.skt = connection;
    this.iProcessId = iProcessNo;
}

public void run() {
    try {

        //initialize the object "scnt" using the parameterized constructor
        ServerConnect scnt = new ServerConnect(skt, iProcessId);
        myInput = new ObjectInputStream(skt.getInputStream());

        while(true) {
            try{
                iOwnTimeStamp = Global.iTimeStamp;

                Object buf = myInput.readObject();

                //if we got input, print it out and write a message back to the remote client...
                if(buf != null){
                    scnt.replyChoice(buf);
                }
            }catch(ClassNotFoundException e) {
                e.printStackTrace();
            }
        }
    } catch(IOException e) {
        e.printStackTrace();
    }
}

void replyChoice(Object buf){       
    try{
        String sDeferReply = "";
        myOutput = new ObjectOutputStream(skt.getOutputStream());

        //the place where the basestation reads the request from the other basestation
        System.out.println("Server read:[ "+buf+" ]");

        //extract out the process id and the timestamp from the incoming request
        buf = buf.toString();

        ***StringTokenizer st = new StringTokenizer(buf,",");***

        //skip the word request
        st.nextToken();
        iInProcessId = Integer.parseInt(st.nextToken());
        iInTimeStamp = Integer.parseInt(st.nextToken());

        //check request is made
        //there is a possibility of entering the else loop only on the very first iteration
        //the control flows into the if loop even if one request has been made
        if(iOwnTimeStamp != 0){
            //if the incoming request has a larger timestamp (logical clock value, process id) than the current process, we defer the reply
            if(iOwnTimeStamp < iInTimeStamp || iProcessId < iInProcessId){
                sDeferReply="iInTimeStamp"+","+"iInProcessId";
                Global.v.addElement(new String(sDeferReply));
            }
            //incoming request has a smaller timestamp than the basestation request itself
            else{
                myOutput.writeObject("Reply");
                myOutput.flush();
            }
        }
        //if the current process is in the critical section then we defer replies
        else if(Global.iCriticalSection==1){
            sDeferReply="iInTimeStamp"+","+"iInProcessId";
            Global.v.addElement(new String(sDeferReply));
        }
        //start of execution of the thread, there is a possibility that the basestation hasn't issued a request
        else{
            myOutput.writeObject("Reply");
            myOutput.flush();   
        }
    }catch(IOException e){
        e.printStackTrace();
    }
}

}

The part that implements the StringTokenizer function has *** surrounding it.

Thanks in advance to anyone who might be able to help me out.

Answer 1

Try

StringTokenizer st = new StringTokenizer((String) buf,",");

The reason why you're getting that error is because buf , while referring to a String at that point, is still of type Object .

---

As an additional tip, you really should make the effort to try to understand the error message given by the compiler. Look at the following:

cannot find symbol constructor StringTokenizer(java.lang.Object,java.lang.String)
location: class java.util.StringTokenizer st = new StringTokenizer(buf,",");

Compiler error messages don't always make sense, but this is as good as it gets . It tells you that:

• It found the right type, java.util.StringTokenizer , so it's not an import or name obscuring problem, etc.
• It's telling you that a specific method with the given signature can not be found. Indeed, a quick check with the API confirms that StringTokenizer does NOT have a constructor that takes a (java.lang.Object, java.lang.String) .
• It's telling you exactly the line of code in your program that tries to invoke this non-existent method. And indeed , the type of your first argument is a java.lang.Object , and the type of your second argument is a java.lang.String !!!

That was how I was able to quickly pinpoint the problem in the source code and suggest a quick fix.

Being able to process error messages given by the compiler is an essential skill that you must develop, so I hope this proves to be an educational experience for you.