2014-04-02 抽取最短英文描述（阿里巴巴面试题 有bug不是最优）

题目：

Alibaba笔试题：给定一段产品的英文描述，包含M个英文单词，每个英文单词以空格分隔，无其他标点符号；再给定N个英文单词关键字，请说明思路并编程实现方法

char* extractSummary(char[] description,char[] *key words)

目标是找出此产品描述中包含N个关键字（每个关键词至少出现一次）的长度最短的子串，作为产品简介输出。（不限编程语言）20分。

思路：

首先找出这N个单词在该M个单词出现的最开始出现的单词X，和最后出现的单词Y，然后分别去掉单词X,再次判别单词是否在X,Y之间，若在，后移找到新的单词X,一次操作；对于Y，可以进行同样的操作，依次前移，得到最后的最短字串单词为止。



程序实现：

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

char *left, *right;
char character[1024];
char **keywords;
int number;

char * str_reverse_str(char *word)
{
char *p = NULL;
char *temp = NULL;
char *count = word;
char *characters = character;

while((*characters != '\0') && (p = strstr(characters, word))){
temp = p;
while(count && *count != '\0'){
p++;
count++;
}
characters = p;
count = word;
}

return temp;
}

int is_key_word(char *words)
{
int i;
for(i = 0; i < number; i++){ //obtain the first and last
if(strcmp(words, keywords[i]) == 0)
return 1;
}
return 0;
}

void obtain_first_word(char *first)
{
left = first;

char p[30]; //obtain the word
char *s = p;
*s++ = ' ';
first++;
while(first && *first != ' '){
*s++ = *first++;
}
*s++ = ' ';
*s = '\0';

if(strstr(first, p)){
obtain_first_word(first);
}else if(!is_key_word(p)){
obtain_first_word(first);
}
}

void obtain_last_word(char *last)
{
right = last;
char p[30];
char *s = p;
*s++ = ' ';
last--;
last--;
char *p2 = last;
while(p2 && *p2 != ' '){
s++;
p2--;
}
*s++ =' ';
*s = '\0';
s--;
s--;
while(last && *last != ' '){
*s-- = *last--;
}
char *temp = last;
temp++;
char c;
if(temp){
c = *temp;
*temp = '\0';
}
if(strstr(left, p)){
obtain_last_word(temp);
} else if(!is_key_word(p)){
obtain_last_word(temp);
} else {
*temp = c;
}
}

char * extract_summary()
{
int i;
char *first = NULL, *last = NULL, *p ;

for(i = 0; i < number; i++){ //obtain the first and last
if((p = strstr(character, keywords[i]))){
if(!first ||first > p)
first = p;
}

if((p = str_reverse_str(keywords[i]))){
if(!last ||last < p)
last = p;
//printf("%d\n", 101);
}
}

last++;
while(last && *last != ' '){
last++;
}
*last++ = ' ';
*last = '\0';
obtain_first_word(first);
obtain_last_word(last);
printf("\nthe minest character as follows:\n%s", ++left);
}

int main(void)
{
char *s = character; //set the the side of character to ' '
*s++ = ' ';
printf("Please input the character sequence as follows:\n");
gets(s);
while(s && *s != '\0'){
s++;
}
*s++ = ' ';
*s = '\0';

printf("Please input the number of keywords:\n");
scanf("%d", &number);
printf("Please input the keyword as follows:\n");
keywords = (char **)malloc(sizeof(char *) * number);
int i;
for(i = 0; i < number; i++){
keywords[i] = (char *)malloc(sizeof(char) * 30);
s = keywords[i]; //set the the side of character to ' '
*s++ = ' ';
scanf("%s", s);
while(s && *s != '\0'){
s++;
}
*s++ = ' ';
*s ='\0';
}

extract_summary();

for(i = 0; i < number; i++){
free(keywords[i]);
}
free(keywords);

return 0;
}