数据结构实践三：代码实现

文本文件单词的检索与计数

代码实现

文本检索类

import java.io.BufferedReader;
import java.io.FileReader;
import java.lang.management.BufferPoolMXBean;
import java.util.Scanner;

/**
 * 文本检索类
 */
public class Search {

    //待检索文本
    String txt;
    //目标文本
    String word;
    
    Scanner scan = new Scanner(System.in);

    /**
     * 读取文本文件内容
     */
    public void readFile() {
        try {
            StringBuffer stringBuffer = new StringBuffer();
            BufferedReader bufferedReader = new BufferedReader(new FileReader("D:\\IDEA\\WordSearch\\text.txt"));
            String t = null;
            while ((t = bufferedReader.readLine()) != null) {
                stringBuffer.append(t.trim());
            }
            txt = stringBuffer.toString();
            System.out.println("文本文件内容：\n" + txt);
        } catch (Exception e) {
            System.out.println(e.getMessage());
        }
    }

    /**
     * 检索目标文本
     */
    public void search() {
        int num;
        readFile();
        System.out.println("请输入要检索的单词：");
        word = scan.next();
        num=KMP(txt, word);
        if(num==0){
            System.out.println("未找到单词");
        }
        else{
            System.out.println("一共找到"+num+"个");
        }
    }

    /**
     * 获取next数组
     * @param T
     * @return
     */
    public static int[] next(String T) {
        char[] t = T.toCharArray();
        int[] next = new int[t.length];
        next[0] = -1;
        int j = 0;
        int k = -1;
        while (j < t.length - 1) {
            if (k == -1 || t[j] == t[k]) {
                j++;
                k++;
                if (t[j] == t[k]) { // 当两个字符相等时要跳过
                    next[j] = next[k];
                } else {
                    next[j] = k;
                }
            } else {
                k = next[k];
            }
        }
        return next;
    }

    /**
     * KMP算法检索
     * @param txt
     * @param word
     * @return
     */
    public int KMP(String txt, String word) {
        int num = 0;
        int i = 0;
        while (i < txt.length()) {
            char[] S = txt.toCharArray();
            char[] T = word.toCharArray();
            int j = 0;
            int[] next = next(word);
            while (i < S.length && j < T.length) {
                if (j == -1 || S[i] == T[j]) {
                    i++;
                    j++;
                } else {
                    j = next[j];
                }
            }
            if (j == T.length) {
                findIndex(i-j, word);
                num++;
            }
        }
        return num;
    }

    /**
     * 获取检索位置
     * @param t
     * @param word
     */
    public void findIndex(int t,String word){
        t++;
        int s=t+word.length()-1;
        System.out.println("文本发现位置："+t+"~"+s);
    }

    /**
     * 主函数
     * @param args
     */
    public static void main(String[] args) {
        new Search().search();
    }
}

运行截图