如何使用BufferReader和PrintWriter组合来输入文本

Question

I am going to participate in USACO later this year, and I am most likely going to use Java. However, I have not covered File I/O thoroughly. USACO insists that we use this BufferReader, PrintWriter, and StringTokenizer combination in order to parse inputted text. Here is the code that they showed:

import java.io.*;
import java.util.*;

class test {
  public static void main (String [] args) throws IOException {
    // Use BufferedReader rather than RandomAccessFile; it's much faster
    BufferedReader f = new BufferedReader(new FileReader("test.in"));
                                              // input file name goes above
    PrintWriter out = new PrintWriter(new BufferedWriter(new FileWriter("test.out")));
    // Use StringTokenizer vs. readLine/split -- lots faster
    StringTokenizer st = new StringTokenizer(f.readLine());
                      // Get line, break into tokens
    int i1 = Integer.parseInt(st.nextToken());    // first integer
    int i2 = Integer.parseInt(st.nextToken());    // second integer
    out.println(i1+i2);                           // output result
    out.close();                                  // close the output file
    System.exit(0);                               // don't omit this!
  }
}

However, when I literally copy/pasted the code into Netbeans, it didn't run. It gave a FileNotFoundException at the line "BufferedReader f = new BufferedReader..." I suppose it's because of the "test.out", but I am not sure what to put in it. How do I fix this?

Additionally, USACO said that this was the most efficient way to parse inputted text. Is this true? As in, is this really the most efficient way to retrieve inputted text. I am familiar with other ways like with the Scanner class, but USACO insists that using BufferedReader, PrintWriter, etc. is the best way to do it.

Answer 1

StringTokenizer is more efficient than Scanner and split . Both Scanner and split use regex to tokenize their input. StringTokenizer does not use regex, and therefore doesn't run into the overhead of using it.

test.in is a file in the working directory of your project:

YourProject
    src
        test.java
    bin 
        test.class
    test.in

Answer 2

I did USACO back in the day with pure Java. (I did really well.)

You almost always want to use Scanner -- it's pretty much perfect for programming competitions, and I/O usually isn't going to be your bottleneck. StringTokenizer is a bit more efficient, but you should only really worry about that when your program is going to be linear-time anyway.