使用Java从文本文件中计算字数

Question

I am trying to write a simple code that will give me the word count from a text file. The code is as follows:

import java.io.File; //to read file
import java.util.Scanner;

public class ReadTextFile {
   public static void main(String[] args) throws Exception { 
      String filename = "textfile.txt";
      File f = new File (filename);
      Scanner scan = new Scanner(f);
      int wordCnt = 1;

      while(scan.hasNextLine()) {
          String text = scan.nextLine();
          for (int i = 0; i < text.length(); i++) {
              if(text.charAt(i) == ' ' && text.charAt(i-1) != ' ') {
                  wordCnt++;
              }
          }
      }
      System.out.println("Word count is " + wordCnt);
   }

}

this code compiles but does not give the correct word count. What am I doing incorrectly?

Answer 1

Right now you are only incrementing wordCnt if the character you are on is a whitespace and the character before it is not. However this discounts several cases, such as if there is not a space, but a newline character. Consider if your file looked like:

This is a text file\n
with a bunch of\n
words. 

Your method should return ten, but since there is not space after the words file , and of it will not count them as words.

If you just want the word count you can do something along the lines of:

while(scan.hasNextLine()){
   String text = scan.nextLine();
   wordCnt+= text.split("\\s+").length;  
}

Which will split on white space(s), and return how many tokens are in the resulting Array

Answer 2

First of all remember about closing resources. Please check this out.

Since Java 8 you can count words in this way:

String regex = "\\s+"
String filename = "textfile.txt";

File f = new File (filename);

long wordCnt = 1;
try (var scanner = new Scanner (f)){
        wordCnt scanner.lines().map(str -> str.split(regex)).count();
} catch (IOException e) {
        e.printStackTrace();
}



System.out.println("Word count is " + wordCnt);