• Codeforces_766_C_(dp)


    C. Mahmoud and a Message
    time limit per test
    2 seconds
    memory limit per test
    256 megabytes
    input
    standard input
    output
    standard output

    Mahmoud wrote a message s of length n. He wants to send it as a birthday present to his friend Moaz who likes strings. He wrote it on a magical paper but he was surprised because some characters disappeared while writing the string. That's because this magical paper doesn't allow character number i in the English alphabet to be written on it in a string of length more than ai. For example, if a1 = 2 he can't write character 'a' on this paper in a string of length 3 or more. String "aa" is allowed while string "aaa" is not.

    Mahmoud decided to split the message into some non-empty substrings so that he can write every substring on an independent magical paper and fulfill the condition. The sum of their lengths should be n and they shouldn't overlap. For example, if a1 = 2 and he wants to send string "aaa", he can split it into "a" and "aa" and use 2 magical papers, or into "a", "a" and "a" and use 3 magical papers. He can't split it into "aa" and "aa" because the sum of their lengths is greater than n. He can split the message into single string if it fulfills the conditions.

    A substring of string s is a string that consists of some consecutive characters from string s, strings "ab", "abc" and "b" are substrings of string "abc", while strings "acb" and "ac" are not. Any string is a substring of itself.

    While Mahmoud was thinking of how to split the message, Ehab told him that there are many ways to split it. After that Mahmoud asked you three questions:

    • How many ways are there to split the string into substrings such that every substring fulfills the condition of the magical paper, the sum of their lengths is n and they don't overlap? Compute the answer modulo 109 + 7.
    • What is the maximum length of a substring that can appear in some valid splitting?
    • What is the minimum number of substrings the message can be spit in?

    Two ways are considered different, if the sets of split positions differ. For example, splitting "aa|a" and "a|aa" are considered different splittings of message "aaa".

    Input

    The first line contains an integer n (1 ≤ n ≤ 103) denoting the length of the message.

    The second line contains the message s of length n that consists of lowercase English letters.

    The third line contains 26 integers a1, a2, ..., a26 (1 ≤ ax ≤ 103) — the maximum lengths of substring each letter can appear in.

    Output

    Print three lines.

    In the first line print the number of ways to split the message into substrings and fulfill the conditions mentioned in the problem modulo 109  +  7.

    In the second line print the length of the longest substring over all the ways.

    In the third line print the minimum number of substrings over all the ways.

    Examples
    input
    3
    aab
    2 3 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
    output
    3
    2
    2
    input
    10
    abcdeabcde
    5 5 5 5 4 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
    output
    401
    4
    3
    Note

    In the first example the three ways to split the message are:

    • a|a|b
    • aa|b
    • a|ab

    The longest substrings are "aa" and "ab" of length 2.

    The minimum number of substrings is 2 in "a|ab" or "aa|b".

    Notice that "aab" is not a possible splitting because the letter 'a' appears in a substring of length 3, while a1 = 2.

    题意:将一个字符串分为多个子串,每一个字符str[i]不能出现在长度大于a[i]的子串中。问一共有多少种分法,其中最长的子串长度,子串数最少的分法的子串数。

    C题,看着一千多个人出,感觉自己应该能做啊,但是最终没有想到是一个dp题。看了题解做出来的。

    思路:遍历字符串,每遍历到一个str[i],令j=i,j递减,判断str[j,i]这个子串是否合法(即每个字符满足a[str[x]-'a']>=i-j+1),若合法dp1[i]=(dp1[i]+dp1[j-1])%M;dp2[i]=min(dp2[i],dp[j-1]+1),最长子串长度在过程中记录一下。

    #include<iostream>
    #include<cstdio>
    #include<cstring>
    using namespace std;
    #define M 1000000007
    
    char str[1005];
    int chara[26];
    int dp1[1005];  //1-i的串分法种数
    int dp2[1005];  //1-i的串最少分为多少段
    
    int main()
    {
        int n;
        while(scanf("%d",&n)!=EOF)
        {
            memset(dp1,0,sizeof(dp1));
            memset(dp2,1,sizeof(dp2));
            dp1[0]=1;
            dp2[0]=0;
            scanf("%s",str+1);
            for(int i=0; i<26; i++)
                scanf("%d",&chara[i]);
            int maxlen=1;
            for(int i=1; i<=n; i++)
            {
                int minc=chara[str[i]-'a'];
                for(int j=i; j>=1; j--)
                {
                    minc=min(minc,chara[str[j]-'a']);
                    if(i-j+1<=minc)
                    {
                        dp1[i]=(dp1[i]+dp1[j-1])%M;
                        //cout<<"i:"<<dp1[i]<<endl;
                        dp2[i]=min(dp2[i],dp2[j-1]+1);
                        if(i-j+1>maxlen)
                            maxlen=i-j+1;
                    }
                    else
                        break;
                }
            }
            printf("%d
    %d
    %d
    ",dp1[n],maxlen,dp2[n]);
        }
        return 0;
    }
  • 相关阅读:
    终于合一起了...
    关于Earley第二篇论文给的建立parse tree的算法的bug
    不作恶
    windows server安装ATI显卡驱动
    不作恶:这次我是一个坚定的五毛党
    看了老刘的文章
    webkit svg高斯模糊的bug
    世界不平坦
    我发现我对人类活动的认识开始有一点点变化了
    css parsing中词法的RegEx(python)
  • 原文地址:https://www.cnblogs.com/jasonlixuetao/p/6379443.html
Copyright © 2020-2023  润新知