Codeforces Round #223 (Div. 2) C

C. Sereja and Prefixes
time limit per test
1 second
memory limit per test
256 megabytes
input
standard input
output
standard output

Sereja loves number sequences very much. That's why he decided to make himself a new one following a certain algorithm.

Sereja takes a blank piece of paper. Then he starts writing out the sequence in m stages. Each time he either adds a new number to the end of the sequence or takes l first elements of the current sequence and adds them c times to the end. More formally, if we represent the current sequence as a1, a2, ..., an, then after we apply the described operation, the sequence transforms into a1, a2, ..., an[, a1, a2, ..., al] (the block in the square brackets must be repeated c times).

A day has passed and Sereja has completed the sequence. He wonders what are the values of some of its elements. Help Sereja.

Input

The first line contains integer m (1 ≤ m ≤ 105) — the number of stages to build a sequence.

Next m lines contain the description of the stages in the order they follow. The first number in the line is a type of stage (1 or 2). Type 1 means adding one number to the end of the sequence, in this case the line contains integer xi (1 ≤ xi ≤ 105) — the number to add. Type 2 means copying a prefix of length li to the end ci times, in this case the line further contains two integers li, ci (1 ≤ li ≤ 105, 1 ≤ ci ≤ 104)li is the length of the prefix, ci is the number of copyings. It is guaranteed that the length of prefix li is never larger than the current length of the sequence.

The next line contains integer n (1 ≤ n ≤ 105) — the number of elements Sereja is interested in. The next line contains the numbers of elements of the final sequence Sereja is interested in. The numbers are given in the strictly increasing order. It is guaranteed that all numbers are strictly larger than zero and do not exceed the length of the resulting sequence. Consider the elements of the final sequence numbered starting from 1from the beginning to the end of the sequence.

Please, do not use the %lld specifier to read or write 64-bit integers in С++. It is preferred to use the cincout streams or the %I64dspecifier.

Output

Print the elements that Sereja is interested in, in the order in which their numbers occur in the input.

Sample test(s)
input
6
1 1
1 2
2 2 1
1 3
2 5 2
1 4
16
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
output
1 2 1 2 3 1 2 1 2 3 1 2 1 2 3 4

题目大意:

维护一个初始为空的序列,支持两种操作,共 m 个(1 <= m <= 10 ^ 5):

1 将一个数插入到数列的尾端(插入的数不大于10 ^ 5)

2 将数列的前 l 位复制 c 次,贴到数列的尾端(1 <= l <= 10 ^ 5, 1 <= c <= 10 ^ 4)

然后进行 n (1 <= n <= 10 ^ 5)次查询,每次给出一个数 x ,输出最终数列的第x位


#include <iostream>
#include <stdio.h>
#include <string>
#include <string.h>
#include <algorithm>
#include <stdlib.h>
#include <vector>
using namespace std;
typedef long long LL ;
const int Max_N = 100008 ;
LL Len[Max_N] ;
int n  ;
int type[Max_N] ;
LL L_X[Max_N] ,C[Max_N] ;
int find_id(LL x){     /*二分查找区间段*/
    int Left = 1 ;
    int Right = n ;
    int Mid  ,ans ;
    while(Left <= Right){
          Mid = (Left + Right) >> 1 ;
          if(Len[Mid] >= x){
               ans = Mid ;
               Right = Mid - 1 ;
          }
          else
               Left = Mid + 1 ;
    }
    return ans ;
}

int answer(LL id){     /*递归查找值*/
    int index = find_id(id) ;
    if(type[index] == 1)
       return L_X[index] ;
    LL re = id - Len[index - 1] ;
    re %= L_X[index] ;
    if(re == 0)
       re = L_X[index] ;
    return answer(re) ;
}

int main(){
    int i , m;
    LL id ;
    scanf("%d",&n) ;
    Len[0] = 0 ;
    for(i = 1 ; i <= n ; i++){
        scanf("%I64d",&type[i]) ;
        if(type[i] == 1){
            scanf("%I64d",&L_X[i]) ;
            Len[i] = Len[i-1] + 1 ;
        }
        else{
            scanf("%I64d%I64d",&L_X[i],&C[i]) ;
            Len[i] = Len[i-1] + L_X[i]*C[i]  ;
        }
    }
    scanf("%d",&m) ;
    while(m--){
         scanf("%I64d",&id) ;
         printf("%d ",answer(id)) ;
    }
    return 0  ;
}
原文地址:https://www.cnblogs.com/liyangtianmen/p/3524375.html