codeforces 278B New Problem

B. New Problem

time limit per test

2 seconds

memory limit per test

256 megabytes

input

standard input

output

standard output

Coming up with a new problem isn't as easy as many people think. Sometimes it is hard enough to name it. We'll consider a title original if it doesn't occur as a substring in any titles of recent Codeforces problems.

You've got the titles of n last problems — the strings, consisting of lowercase English letters. Your task is to find the shortest original title for the new problem. If there are multiple such titles, choose the lexicographically minimum one. Note, that title of the problem can't be an empty string.

A substring s[l... r] (1 ≤ l ≤ r ≤ |s|) of string s = s₁s₂... s_|s| (where |s| is the length of string s) is string s_ls_l + 1... s_r.

String x = x₁x₂... x_p is lexicographically smaller than string y = y₁y₂... y_q, if either p < q and x₁ = y₁, x₂ = y₂, ... , x_p = y_p, or there exists such number r (r < p, r < q), that x₁ = y₁, x₂ = y₂, ... , x_r = y_r and x_r + 1 < y_r + 1. The string characters are compared by their ASCII codes.

Input

The first line contains integer n (1 ≤ n ≤ 30) — the number of titles you've got to consider. Then follow n problem titles, one per line. Each title only consists of lowercase English letters (specifically, it doesn't contain any spaces) and has the length from 1 to 20, inclusive.

Output

Print a string, consisting of lowercase English letters — the lexicographically minimum shortest original title.

Sample test(s)

Input

5
threehorses
goodsubstrings
secret
primematrix
beautifulyear

Output

j

Input

4
aa
bdefghijklmn
opqrstuvwxyz
c

Output

ab

Note

In the first sample the first 9 letters of the English alphabet (a, b, c, d, e, f, g, h, i) occur in the problem titles, so the answer is letter j.

In the second sample the titles contain 26 English letters, so the shortest original title cannot have length 1. Title aa occurs as a substring in the first title.

题目叫你找一个最小的新标题，要求是标题不能与原有标题重复。刚开始一直搞错了，以为只是找两个长度的字符串。后来发现是整个字符串里面去寻找。这个时候，就可以用一个strstr函数来达到目标。strstr函数是用来搜索字符串的，即Strstr（str1，str2），在1中寻找2是否出现，如果不出现就返回NULL了，否则就返回字符串的其余部分。

这样一来就简单了。

#include<stdio.h>
#include<string.h>
int main()
{
    char zimu[]={
        'a','b','c','d','e','f','g','h','i','j','k','l','m','n','o','p','q','r','s',
        't','u','v','w','x','y','z'
    };
    int n,i,j,k,l[40],hash[200],f,mark;
    char str[40][40],yuan[2];
    while(scanf("%d",&n)!=EOF)
    {    
        memset(hash,0,sizeof(hash));
        f=0;mark=0;
        for(i=1;i<=n;i++)
        {
            scanf("%s",&str[i]);
            l[i]=strlen(str[i]);
            for(j=0;j<l[i];j++)
            hash[str[i][j]-'a']++;
        }
        for(i=0;i<26;i++)
        if(hash[i]==0){
            printf("%c\n",zimu[i]);
          f=1;
            break;
        }
        if(f)continue;
        for(i='a';i<='z';i++)
        {for(j='a';j<='z';j++)
        {
            mark=0;
            yuan[0]=i;yuan[1]=j;
            yuan[2]='\0';
    
            for(k=1;k<=n;k++)
                {if(strstr(str[k],yuan)!=NULL)
                {mark=1;
                break;
                }
                }
            
            if(mark==0){printf("%c%c\n",yuan[0],yuan[1]);break;}
        
        }
        if(mark==0)break;
        }
    }

return 0;
}