CodeForces 923C Perfect Security
C. Perfect Security
time limit per test3.5 seconds
memory limit per test512 megabytes
inputstandard input
outputstandard output
Alice has a very important message M consisting of some non-negative integers that she wants to keep secret from Eve. Alice knows that the only theoretically secure cipher is one-time pad. Alice generates a random key K of the length equal to the message’s length. Alice computes the bitwise xor of each element of the message and the key (, where denotes the bitwise XOR operation) and stores this encrypted message A. Alice is smart. Be like Alice.
For example, Alice may have wanted to store a message M = (0, 15, 9, 18). She generated a key K = (16, 7, 6, 3). The encrypted message is thus A = (16, 8, 15, 17).
Alice realised that she cannot store the key with the encrypted message. Alice sent her key K to Bob and deleted her own copy. Alice is smart. Really, be like Alice.
Bob realised that the encrypted message is only secure as long as the key is secret. Bob thus randomly permuted the key before storing it. Bob thinks that this way, even if Eve gets both the encrypted message and the key, she will not be able to read the message. Bob is not smart. Don’t be like Bob.
In the above example, Bob may have, for instance, selected a permutation (3, 4, 1, 2) and stored the permuted key P = (6, 3, 16, 7).
One year has passed and Alice wants to decrypt her message. Only now Bob has realised that this is impossible. As he has permuted the key randomly, the message is lost forever. Did we mention that Bob isn’t smart?
Bob wants to salvage at least some information from the message. Since he is not so smart, he asks for your help. You know the encrypted message A and the permuted key P. What is the lexicographically smallest message that could have resulted in the given encrypted text?
More precisely, for given A and P, find the lexicographically smallest message O, for which there exists a permutation π such that for every i.
Note that the sequence S is lexicographically smaller than the sequence T, if there is an index i such that Si < Ti and for all j < i the condition Sj = Tj holds.
Input
The first line contains a single integer N (1 ≤ N ≤ 300000), the length of the message.
The second line contains N integers A1, A2, …, AN (0 ≤ Ai < 230) representing the encrypted message.
The third line contains N integers P1, P2, …, PN (0 ≤ Pi < 230) representing the permuted encryption key.
Output
Output a single line with N integers, the lexicographically smallest possible message O. Note that all its elements should be non-negative.
Examples
inputCopy
3
8 4 13
17 2 7
output
10 3 28
inputCopy
5
12 7 87 22 11
18 39 9 12 16
output
0 14 69 6 44
inputCopy
10
331415699 278745619 998190004 423175621 42983144 166555524 843586353 802130100 337889448 685310951
226011312 266003835 342809544 504667531 529814910 684873393 817026985 844010788 993949858 1031395667
output
128965467 243912600 4281110 112029883 223689619 76924724 429589 119397893 613490433 362863284
Note
In the first case, the solution is (10, 3, 28), since , and . Other possible permutations of key yield messages (25, 6, 10), (25, 3, 15), (10, 21, 10), (15, 21, 15) and (15, 6, 28), which are all lexicographically larger than the solution.
上传一份推倒思路,我懒得打字了!
#include<stdio.h>
#include<algorithm>
using namespace std;
int a[300005];
typedef struct Trie
{
int sum;
Trie *next[2];
}Trie;
void Create(Trie *root, int x)
{
int now, i;
Trie *p = root;
for(i=30;i>=0;i--)
{
now = 0;
if(x&(1<<i))
now = 1;
if(p->next[now]==NULL)
{
Trie *temp = new Trie;
temp->next[0] = temp->next[1] = NULL;
temp->sum = 0;
p->next[now] = temp;
}
p = p->next[now];
p->sum++;
}
}
int Query(Trie *root, int x)
{
int now, i, ans = 0;
Trie *p = root;
for(i=30;i>=0;i--)
{
now = 0;
if(x&(1<<i))
now = 1;
if(p->next[now]==NULL || p->next[now]->sum==0)
p = p->next[now^1], ans += (1<<i);
else
p = p->next[now];
p->sum--;
}
return ans;
}
int main(void)
{
int n, i, x;
Trie *root = new Trie;
root->next[0] = root->next[1] = NULL;
root->sum = 0;
scanf("%d", &n);
for(i=1;i<=n;i++)
scanf("%d", &a[i]);
for(i=1;i<=n;i++)
{
scanf("%d", &x);
Create(root, x);
}
for(i=1;i<=n;i++)
printf("%d ", Query(root, a[i]));
puts("");
return 0;
}
CodeForces 923C Perfect Security的更多相关文章
- Codeforces 948D Perfect Security(字典树)
题目链接:Perfect Security 题意:给出N个数代表密码,再给出N个数代表key.现在要将key组排序,使key组和密码组的亦或所形成的组字典序最小. 题解:要使密码组里面每个数都找到能使 ...
- Codeforces 948D Perfect Security
Perfect Security 题意:给你一个A[i]数组, 再给你一个B[i]数组, 现在用选取 B[i] 数组中的一个 去和 A[i] 数组里的一个元素去进行异或操作, B[i]数组的元素只能用 ...
- Codeforces 948D Perfect Security 【01字典树】
<题目链接> 题目大意: 给定两个长度为n的序列,可以改变第二个序列中数的顺序,使得两个序列相同位置的数异或之后得到的新序列的字典序最小. 解题分析: 用01字典树来解决异或最值问题.因为 ...
- 2018.12.08 codeforces 948D. Perfect Security(01trie)
传送门 01trie板子题. 给出两个数列,允许把第二个数列重新排列. 求使得两个数列每个位置对应的数的异或值和成为最小值的每个位置的异或和. 把第二个数列插入到01trie里面然后对于第一个数列中的 ...
- 【CodeForces】947 C. Perfect Security 异或Trie
[题目]C. Perfect Security [题意]给定长度为n的非负整数数组A和数组B,要求将数组B重排列使得A[i]^B[i]的字典序最小.n<=3*10^5,time=3.5s. [算 ...
- 01Trie树 CF923C Perfect Security
CF923C Perfect Security 上下各n个数,求一种排列p,使上面的数i异或pi成为新的数i,求方案另字典序最小,输出该结果 01Trie树. 记录每个节点经过多少次. 每一次查询的时 ...
- Codeforces 923 C. Perfect Security
http://codeforces.com/contest/923/problem/C Trie树 #include<cstdio> #include<iostream> us ...
- 923c C. Perfect Security
Trie树. 要求字典序最小,所以由前到后贪心的选择.建一个trie树维护b数列. #include<cstdio> #include<algorithm> #include& ...
- codeforces 14A - Letter & codeforces 859B - Lazy Security Guard - [周赛水题]
就像title说的,是昨天(2017/9/17)周赛的两道水题…… 题目链接:http://codeforces.com/problemset/problem/14/A time limit per ...
随机推荐
- 【c++】C++中const用法总结
1. const常量,如const int max = 100; 优点:const常量有数据类型,而宏常量没有数据类型.编译器可以对前者进行类型安全检查,而对后者只进行字符替换,没有类型安全 ...
- Android 开源框架ViewPageIndicator 和 ViewPager 仿网易新闻clientTab标签
之前用JakeWharton的开源框架ActionBarSherlock和ViewPager实现了对网易新闻clientTab标签的功能,ActionBarSherlock是在3.0下面的机器支持Ac ...
- socket的bind函数是不是只能绑定本地IP,不能绑定外网IP么?
参考: https://bbs.csdn.net/topics/391024376 别瞎猜测. 所谓bind,就是指绑定本地接受端口. 指定ip,是为了分辨多ip主机. --------------- ...
- linuxc线程信号-pthread_cond_wait理解
pthread_cond_wait(pthread_cond_t *cond, pthread_mutex_t*mutex)函数 传入的參数mutex用于保护条件,由于我们在调用pthread_con ...
- 【Mongodb教程 第九课 】MongoDB 删除文档
remove() 方法 MongoDB的 remove() 方法用于从集合中删除文档.remove() 方法接受两个参数.第一个是删除criteria ,第二是justOne标志: deletion ...
- sqlserver中All、Any和Some用法与区别
转自:http://blog.csdn.net/gyc1105/article/details/8063624 SQLServer中有三个关键字可以修改比较运算符:All.Any和Some,其中Som ...
- OO模式-Composite
组合模式也叫做"部分-总体"模式,这样事实上定义也就非常明显了,正好和数据结构的知识相相应.把对象组合成树形结构以表示"部分-总体"的层次结构. 先看类图: w ...
- CSS中:overflow:hidden的作用
功能1.隐藏溢出 在IE6下,当子容器的宽高超出父容器时,父容器就会被撑开来. 要想解决这个问题,在父容器中除定义宽和高的值以外,还必须写overflow:hidden,这样就能把子容器的其它内容隐 ...
- HDU 5544 Ba Gua Zhen dfs+高斯消元
Ba Gua Zhen Problem Description During the Three-Kingdom period, there was a general named Xun Lu wh ...
- GoodUI:页面布局的技巧和设计理念
http://goodui.org/ 中文翻译:http://www.cnblogs.com/Wayou/p/goodui.html 一年了,小小少年从幼年期过渡到成长期要开始加速冲刺了.毕竟钻头就是 ...