Description

In whiteblack on blackwhite is written the utterance that has been censored by the Ministry of Truth. Its author has already disappeared along with his whole history, and now, while Big Brother is watching somebody else, you, as an ordinary official of the Minitrue, have to delete some letters from the utterance so that another utterance will appear, which has been approved of by the Ministry.
The Ministry of Truth defines a word as a nonempty sequence of English letters and an utterance as a sequence of one or more words separated with one or more spaces. There can also be spaces before the first word and after the last word of an utterance. In order to compare two utterances, one should delete all the leading and trailing spaces and replace each block of consecutive spaces with one space. If the resulting strings coincide, then the utterances are considered to be equal. When the official deletes a letter from the utterance, this letter turns into a space.

Input

The first line contains the original utterance and the second line contains the utterance that must be obtained. The length of each utterance is at most 100000 symbols. The words in both utterances are separated with exactly one space; there are no leading or trailing spaces in each line. The original and the required utterances are different.

Output

If you can't carry out your order, output “I HAVE FAILED!!!” in the only line. Otherwise, output the original utterance replacing the letters that are to be deleted with the underscore character.

题目大意:给两行字符串,可以删掉上面的字符串的某些字符,删掉的字符变成空格,删完后能否变成下面的字符串。若能则输出上面的字符串,其中删掉的字符变成下划线。

思路:KMP给下面的每一个单词匹配上面的字符串就行了,注意细节处理。区分大小写。

PS:输入没有多余空格不要误会了。

PS2:之前误解了题意所以可能代码会有点奇葩……

代码(281MS):

 #include <cstdio>
#include <cstring>
#include <iostream>
#include <algorithm>
#include <queue>
using namespace std;
typedef long long LL; const int MAXN = ; void getFail(char *P, int f[]) {
int m = strlen(P);
f[] = f[] = ;
for(int i = ; i < m; ++i) {
int j = f[i];
while(j && P[i] != P[j]) j = f[j];
f[i + ] = (P[i] == P[j] ? j + : );
}
} int KMP(char *T, char *P, int f[]) {
int n = strlen(T), m = strlen(P);
getFail(P, f);
int j = ;
for(int i = ; i < n; ++i) {
while(j && P[j] != T[i]) j = f[j];
if(P[j] == T[i]) ++j;
if(j == m) return i - m + ;
}
return -;
} char s1[MAXN], s2[MAXN], ans[MAXN];
char tmp[MAXN];
int f[MAXN], ans_cnt; bool get_next(char *&src, char *ret) {
while(*src == ' ') ++src;
if(*src == ) return false;
int i = ;
while(*src != ' ' && *src != ) ret[i++] = *src, ++src;
ret[i] = ;
return true;
} bool solve() {
int now = ;
ans_cnt = ;
char *ss = s2;
while(get_next(ss, tmp)) {
int pos = KMP(s1 + now, tmp, f);
if(pos == -) return false;
//cout<<pos<<endl;
for(int i = now; i < now + pos; ++i) {
ans[ans_cnt++] = (s1[i] == ' ' ? ' ' : '_');
}
now += pos;
for(int i = ; tmp[i]; ++i, ++now) ans[ans_cnt++] = tmp[i];
//while(s1[now] != ' ' && s1[now] != 0) ans[ans_cnt++] = '_', ++now;
if(s1[now]) {
if(s1[now] != ' ') ans[ans_cnt++] = '_';
else ans[ans_cnt++] = ' ';
++now;
} }
for(int i = now; s1[i]; ++i) {
ans[ans_cnt++] = (s1[i] == ' ' ? ' ' : '_');
}
ans[ans_cnt++] = ;
return true;
} int main() {
gets(s1);
gets(s2);
if(!solve()) puts("I HAVE FAILED!!!");
else printf("%s\n", ans);
}

URAL 1732 Ministry of Truth(KMP)的更多相关文章

  1. URAL 1732. Ministry of Truth ( KMP 多模式串匹配 )

    问在第一个串中删掉几个字符能否得到第二个串.注意在第二个串中不连续的单词在第一个串中也必须不连续. 一组数据: Input: abababbbbababbb aba ab Output: I HAVE ...

  2. poj2406 Power Strings(kmp)

    poj2406 Power Strings(kmp) 给出一个字符串,问这个字符串是一个字符串重复几次.要求最大化重复次数. 若当前字符串为S,用kmp匹配'\0'+S和S即可. #include & ...

  3. POJ 2406 Power Strings(KMP)

    Description Given two strings a and b we define a*b to be their concatenation. For example, if a = & ...

  4. LightOJ 1258 Making Huge Palindromes(KMP)

    题意 给定一个字符串 \(S\) ,一次操作可以在这个字符串的右边增加任意一个字符.求操作之后的最短字符串,满足操作结束后的字符串是回文. \(1 \leq |S| \leq 10^6\) 思路 \( ...

  5. codeM编程大赛E题 (暴力+字符串匹配(kmp))

    题目大意:S(n,k)用k(2-16)进制表示1-n的数字所组成的字符串,例如S(16,16)=123456789ABCDEF10: 解题思路: n最大50000,k最大100000,以为暴力会超时. ...

  6. 经典串匹配算法(KMP)解析

    一.问题重述 现有字符串S1,求S1中与字符串S2完全匹配的部分,例如: S1 = "ababaababc" S2 = "ababc" 那么得到匹配的结果是5( ...

  7. Leetcode28--->字符串的匹配(KMP)

    题目: 题目的本质是给定两个字符串str1,str2,求str1中的str2串开始的地方,即字符串的匹配,KMP算法 思路:时间复杂度为O(m + n),空间复杂度为O(n),原串的长度为m,子串的长 ...

  8. 题解0012:剪花布条(KMP)

    信奥一本通1465 KPM例题 题目链接:http://ybt.ssoier.cn:8088/problem_show.php?pid=1465 题目描述:给出花布条和小饰条(字符串),求花布条中能剪 ...

  9. BZOJ 3796 Mushroom追妹纸 哈希+二分(+KMP)

    先把两个串能匹配模式串的位置找出来,然后标记为$1$(标记在开头或末尾都行),然后对标记数组求一个前缀和,这样可以快速查到区间内是否有完整的一个模式串. 然后二分子串(答案)的长度,每次把长度为$md ...

随机推荐

  1. C#中Math.Round()的中国式用法

    C#中的Math.Round()并不是使用的"四舍五入"法.而是(银行家算法),即:四舍六入五取偶.事实上这也是IEEE的规范,因此所有符合IEEE标准的语言都应该采用这样的算法. ...

  2. NEC 框架规范 css function

    /* function */.f-cb:after,.f-cbli li:after{display:block;clear:both;visibility:hidden;height:0;overf ...

  3. CSS基础全荟

    一.CSS概述 1.css是什么?? 层叠样式表 2.css的引入方式 1.行内样式   在标签上加属性style="属性名1:属性值1;属性名2:属性值2;..." 2.内嵌式  ...

  4. python计算MD5

    python有自带的MD5模块hashlib,用起来简单很多.Python Hashlib模块的使用说明 http://docs.python.org/2/library/hashlib.htmlfd ...

  5. spring-boot整合ehcache实现缓存机制

    EhCache 是一个纯Java的进程内缓存框架,具有快速.精干等特点,是Hibernate中默认的CacheProvider. ehcache提供了多种缓存策略,主要分为内存和磁盘两级,所以无需担心 ...

  6. python爬虫 爬取steam热销游戏

    好久没更新了啊...最近超忙 这学期学了学python 感觉很有趣 就写着玩~~~ 爬取的页面是:https://store.steampowered.com/search/?filter=globa ...

  7. linux文件IO操作篇 (一) 非缓冲文件

    文件IO操作分为 2 种 非缓冲文件IO 和 缓冲文件IO 它们的接口区别是 非缓冲 open() close() read() write() 缓冲 fopen() fclose() fread() ...

  8. stm32+lwip(二):UDP测试

    我是卓波,很高兴你来看我的博客. 系列文章: stm32+lwip(一):使用STM32CubeMX生成项目 stm32+lwip(二):UDP测试 stm32+lwip(三):TCP测试 stm32 ...

  9. python中矢量化字符串方法

  10. SAPFiori

    最新SAP Fiori常用事务代码持续更新中...谢谢支持   注意: 以 / 开头的事务码需要加/N或/O进入,否则进不去   SEGW:  创建Gateway Service   /UI2/FLP ...