UVALive - 3942 Remember the Word
input
字符串s 1<=len(s)<=300000
n 1<=n<=4000
word1
word2
...
wordn
1<=len(wordi)<=100
output
由一个或多个word拼成s的种数%20071027.
做法1:dp:单词长度最多为100,d[i]表示到第i个字符结束的种数,则如果d[j]到d[i]这段字符能从trie中找到,d[i]+=d[j],i-100<j<=i
//1225ms
#include <cstdio>
#include <queue>
#include <cstring>
#include <iostream>
#include <cstdlib>
#include <algorithm>
#include <vector>
#include <map>
#include <set>
#include <ctime>
#include <cmath>
#include <cctype>
#define MAX 100000
#define LL long long
#define mod 20071027
struct trie
{
int ch[*MAX][];
bool val[*MAX];
int sz;
trie()
{
sz=;
memset(ch[],,sizeof(ch[]));
}
int idx(char &c) { return c-'a';}
char* insert(char*s)
{
int u=;
for(;*s;s++)
{
int c=idx(*s);
if(!ch[u][c])
{
memset(ch[sz],,sizeof(ch[sz]));
val[sz]=;
ch[u][c]=sz++;
}
u=ch[u][c];
}
val[u]=;
return s;
}
int find(char*s,int n)
{
int u=;
for(int i=;i<n;i++)
{
int c=idx(s[i]);
if(!ch[u][c]) return ;
u=ch[u][c];
}
return val[u];
}
};
trie t;
char s[*MAX+],word[];
int n,maxl,d[*MAX+],cas=;
int main()
{
//freopen("/home/user/桌面/in","r",stdin);
while(scanf("%s",s)==)
{
memset(t.ch[],,sizeof(t.ch[]));
t.sz=;
scanf("%d",&n);
maxl=;
while(n--)
{
scanf("%s",word);
int l=t.insert(word)-word;
maxl=std::max(maxl,l);
}
n=strlen(s);
memset(d,,sizeof(d));
int u=,work=;
for(int i=;i<maxl;i++)
{
int c=t.idx(s[i]);
if(!t.ch[u][c]) break;
u=t.ch[u][c];
if(t.val[u])
{
work=;
d[i]=;
}
}
// for(int i=0;i<=n;i++) printf("%d ",d[i]);printf("\n");
if(work)
{
for(int i=;i<n;i++)
{
for(int j=;j<=maxl;j++)
{
if(i-j>=&&t.find(s+i-j+,j)) d[i]=(d[i]+d[i-j])%mod;
}
}
}
// for(int i=0;i<=n;i++) printf("%d ",d[i]);printf("\n");
printf("Case %d: %d\n",cas++,d[n-]);
}
//printf("time=%.3lf",(double)clock()/CLOCKS_PER_SEC);
return ;
}
dp
做法2:直接模拟:一开始只有一条路找,每个单词结束时分出一条路从头开始
//79ms
#include <cstdio>
#include <queue>
#include <cstring>
#include <iostream>
#include <cstdlib>
#include <algorithm>
#include <vector>
#include <map>
#include <set>
#include <ctime>
#include <cmath>
#include <cctype>
#define MAX 100000
#define LL long long
#define mod 20071027
struct trie
{
int ch[*MAX][];
bool val[*MAX];
int sz;
trie()
{
sz=;
memset(ch[],,sizeof(ch[]));
}
int idx(char &c) { return c-'a';}
void insert(char*s)
{
int u=;
for(;*s;s++)
{
int c=idx(*s);
if(!ch[u][c])
{
memset(ch[sz],,sizeof(ch[sz]));
val[sz]=;
ch[u][c]=sz++;
}
u=ch[u][c];
}
val[u]=;
}
};
trie t;
char s[*MAX+],word[];
int n,cas=,head[][],count[][],num[];
int main()
{
//freopen("/home/user/桌面/in","r",stdin);
while(scanf("%s",s)==)
{
memset(t.ch[],,sizeof(t.ch[]));
t.sz=;
scanf("%d",&n);
while(n--)
{
scanf("%s",word);
t.insert(word);
}
int d=;
num[d]=;
head[d][]=;
count[d][]=;
for(char*p=s;*p;p++,d^=)
{
// printf("%d %d\n",count[d][0],d);
int &idx=num[d^]=;
int c=t.idx(*p);
head[d^][]=;
count[d^][]=;
for(int i=;i<num[d];i++)
{
int &f=head[d][i];
if(t.ch[f][c])//继续走到下一个字母
{
if(t.val[t.ch[f][c]])//走完一个单词,从头开始,总是head[d][0]
{
count[d^][]+=count[d][i];
count[d^][]%=mod;
}
head[d^][idx]=t.ch[f][c];
count[d^][idx++]=count[d][i];
}
}
}
// printf("%d %d\n",count[d][0],d);
printf("Case %d: %d\n",cas++,count[d][]);
}
//printf("time=%.3lf",(double)clock()/CLOCKS_PER_SEC);
return ;
}
模拟
UVALive - 3942 Remember the Word的更多相关文章
- UVALive - 3942:Remember the Word
发现字典里面的单词数目多且长度短,可以用字典树保存 f[i]表示s[i~L]的分割方式,则有f[i]=∑f[i+len(word[j])] 其中word[j]为s[i~L]的前缀 注意字典树又叫前 ...
- UVALive - 3942 Remember the Word[树状数组]
UVALive - 3942 Remember the Word A potentiometer, or potmeter for short, is an electronic device wit ...
- UVALive - 3942 Remember the Word[Trie DP]
UVALive - 3942 Remember the Word Neal is very curious about combinatorial problems, and now here com ...
- 【暑假】[实用数据结构]UVAlive 3942 Remember the Word
UVAlive 3942 Remember the Word 题目: Remember the Word Time Limit: 3000MS Memory Limit: Unknown ...
- UVALive 3942 Remember the Word 字典树+dp
/** 题目:UVALive 3942 Remember the Word 链接:https://vjudge.net/problem/UVALive-3942 题意:给定一个字符串(长度最多3e5) ...
- UVALive 3942 Remember the Word
题意:给出一个由S个不同单词组成的字典和一个长字符串.把这个字符串分解成若干个单词的连接(单词可以重复 使用),有多少种方法? Sample Input abcd 4 a b cd ab Sample ...
- Remember the Word UVALive - 3942(dp+trie)
题意: 给S个不同的单词和一个长字符串 问将其分解为若干个单词有多少种方法(单词可重复使用) 解析: dp[i]表示在这个字符串中以某个位置i为起点的 的一段子字符串 则这个子字符串若存在某个前缀恰好 ...
- UVALive 3942 Remember The Word (Tire)
状态是DAG,因此方案用dp统计,dp[i] = sum(dp[i+len(x)]),x是以i开头的前缀且是单词,关键在于快速判断一个前缀是不是单词,可用Trie. 每一次转移的复杂度是O(maxle ...
- UVALive - 3942 Remember the Word (Trie + DP)
题意: 给定一篇长度为L的小写字母文章, 然后给定n个字母, 问有多少种方法用这些字母组成文章. 思路: 用dp[i]来表达[i , L]的方法数, 那么dp[i] 就可以从dp[len(x) + i ...
随机推荐
- 部署WEB应用的三种方式[转]
一.基本部署 Tomcat安装目录下有一个webapps目录,该目录存放所有的WEB应用程序,Tomcat会自动管理该目录下的所有WEB应用.因此,最简单的部署方式就是将要部署的WEB应用直接拷贝到T ...
- 应用 Valgrind 发现 Linux 程序的内存问题(转)
Valgrind 概述 体系结构 Valgrind 是一套Linux下,开放源代码(GPL V2)的仿真调试工具的集合.Valgrind由内核(core)以及基于内核的其他调试工具组成.内核类似于一个 ...
- html readonly和disabled的区别
今天我创建了一个html表单,其中一个字段是sequence number, 这个字段是用ajax异步从后台获取自动生成的,我不希望用户修改这个值,于是我使用了如下html代码来处理: <for ...
- kettle 连接Hadoop 遇错
kettle从windows中往hdfs中写文件
- sql标识符和格式
数据库名是一个标识符,表名也是一个标识符,在SQL SERVER中标识符分为两类: (1).常规标识符;(2).分隔标识符两者重要的区别:常规标识符必须严格遵守命名的规定,而分隔标识符则可以不遵守命名 ...
- 启动tomcat时报错:java.lang.OutOfMemoryError: PermGen space
1.修改myeclipse.ini 在Myeclipse安装目录下G:\MyEclipse8.5\Genuitec\MyEclipse 8.5有一个myeclipse.ini配置文件,设置如下: -v ...
- mac 文件路径问题
Finder栏显示访问路径 操作步骤: 打开“终端”(应用程序->实用工具),输入以下两条命令: defaults write com.apple.finder _FXShowPosixPath ...
- hibernate的批量更新、批量删除
hibernate的批处理API:session.createQuery(hql).executeUpdate(),如果有参数则在执行之前设置参数. 批量更新示例: @Test public void ...
- HDU 5813 Elegant Construction
构造.从a[i]最小的开始放置,例如放置了a[p],那么还未放置的,还需要建边的那个点 需求量-1,然后把边连起来. #pragma comment(linker, "/STACK:1024 ...
- vbs和qtp一些脚本
********************************************************************************************** 用vbs调 ...