变位词程序的实现

这篇文章是 读厚《编程珠玑》系列博客 的第 2 篇,主要的内容是《编程珠玑》第二章最后提出的变位词程序的实现。

问题简述

问题来源于《编程珠玑》第二章中最后提出的变位词程序的实现。其中的变位词的概念,在第二章开篇的 C 问题中得到了阐释。

C. 给定一个英语词典,找出其中所有变位词的集合。例如,『pots』,『stop』,『tops』互为变位词,因为每一个单词都可以通过改变其他单词中字母的顺序来得到。

我们所实现的变位词程序需要做的就是:将字典读入,然后找出所有的变位词集合

实现思路

本文中的思路就按照文章中介绍的进行实现:

将程序分为三个部分:

  • sign:读入字典,对单词进行『sign』(计算标识符)。计算标识符的过程就是对单词中的字母进行排序。例如:上述三个单词『pots』,『stop』,『tops』的标识符是相同的,都是『opst』
  • sort:将所有具有相同标识符的单词放到一起(这里我们直接使用系统的 sort 程序)
  • squash:将具有相同标识符的单词放到一行打印出来

最后,通过管道将三个程序连接起来:

1
sign < words.txt | sort | squash > gramlist.txt

最终gramlist.txt中就是我们需要的结果。

代码实现

sign 程序

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
//
// sign.c
// somecode
//
// Created by 罗远航 on 16/12/2016.
// Copyright © 2016 罗远航. All rights reserved.
//

#include <stdio.h>
#include <stdlib.h>
#include <string.h>

#define WORDMAX 100

int charcomp(const void *x, const void *y){
return *(char *)x - *(char *)y;
}

int main(){
char word[WORDMAX], sig[WORDMAX];
while(scanf("%s",word) != EOF){
strcpy(sig, word);
qsort(sig, strlen(sig), sizeof(char), charcomp);
printf("%s %s\n",sig, word);
}
}

squash 程序

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
//
// squash.c
// somecode
//
// Created by 罗远航 on 16/12/2016.
// Copyright © 2016 罗远航. All rights reserved.
//

#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#define WORDMAX 100

int main()
{
char word[WORDMAX], sig[WORDMAX], oldsig[WORDMAX];
int linenum = 0;
strcpy(oldsig, "");
while (scanf("%s %s", sig, word) != EOF)
{
if (strcmp(oldsig, sig) != 0 && linenum > 0)
printf("\n");
strcpy(oldsig, sig);
linenum++;
printf("%s ", word);
}
printf("\n");
return 0;
}

运行结果

运行程序之后,我找到了几个比较长的变位词:

1
2
3
4
algorithm's logarithm's
anatomicopathological pathologicoanatomical
paradisaically paradisiacally
……

注:单词表我是在这里下载到的,大概有35万英文单词。


本文的版权归作者 罗远航 所有,采用 Attribution-NonCommercial 3.0 License。任何人可以进行转载、分享,但不可在未经允许的情况下用于商业用途;转载请注明出处。感谢配合!