Problem A. Cut The Wire
签到题 略
Problem B. Time-Division Multiplexing
考虑到单个串n<=12n<=12n<=12所有字符串的lcm为2?3?2?5?7?2?3?11=277202*3*2*5*7*2*3*11=27 7202?3?2?5?7?2?3?11=27720,所以一个循环节最长为27720?n27720*n27720?n,不同的字符最多散落在两个循环节内,故总长度应该是27720?n?227720*n*227720?n?2。然后用一个双指针算法求解即可
#include<bits/stdc++.h>
using namespace std;
typedef long long ll;
const int N = 5e5 + 10;int n;
char s[110][14];
int p[110];
int cnt[256];int main() {
int T;scanf("%d", &T);while (T--) {
scanf("%d", &n);memset(cnt, 0, sizeof cnt);int c = 1;for (int i = 0; i < n; i++) {
scanf("%s", s[i]);p[i] = strlen(s[i]);for (int j = 0; j < p[i]; j++) cnt[s[i][j]]++;}int diff = 0;for (int i = 0; i < 256; i++) if (cnt[i]) diff++;string tdm;for (int i = 0; i < 27720 * 2; i++) {
for (int j = 0; j < n; j++) {
tdm.push_back(s[j][i % p[j]]);}}//第二个指针, 不同的个数int j = 0, tot = 0, res = 1e9;memset(cnt, 0, sizeof cnt);for (int i = 0; i < tdm.size(); i++) {
char c = tdm[i];if (!cnt[c]) tot++;cnt[c]++;while (cnt[tdm[j]] > 1) {
cnt[tdm[j]]--;j++;}if (tot == diff) {
res = min(res, i - j + 1);}}printf("%d\n", res);}return 0;
}
Problem F. Power Sum
对于相邻的四个数字,有(x+1)2?(x+2)2?(x+3)2+(x+4)2=4(x + 1)^2 ? (x + 2)^2 ? (x + 3)^2 + (x + 4)^2 = 4(x+1)2?(x+2)2?(x+3)2+(x+4)2=4
故有F(5)=F(1)+"1001"F(5)=F(1)+"1001"F(5)=F(1)+"1001",F(9)=F(5)+"1001"F(9)=F(5)+"1001"F(9)=F(5)+"1001"。构造出四个边界F(1),F(2),F(3),F(4)F(1),F(2),F(3),F(4)F(1),F(2),F(3),F(4),递推即可得到答案
#include<bits/stdc++.h>
using namespace std;
const int N = 1e4 + 10;int main() {
int T;scanf("%d", &T);string s[5] = {
"","1", "0001", "01001", "001" };while (T --) {
string t;int n;scanf("%d", &n);t = s[n % 4 == 0 ? 4 : n % 4];int k = (n - 1) / 4;string b = "1001";while (k) {
if (k & 1) t = t + b;b = b + b;k >>= 1;}printf("%d\n%s\n", strlen(t.c_str()), t.c_str());}return 0;
}
Problem G. Function
注意到g(x)最大值为g(99999)=54g(99999)=54g(99999)=54,因此可以枚举[1,54][1,54][1,54],固定住g(x),然后f(x)f(x)f(x)就是一个二次函数了,这样的话就可以利用抛物线开口的情况进行分类讨论求解。在这里先预处理出满足g(x)=tg(x)=tg(x)=t的所有x,然后用二分就可以快速实现
#include<bits/stdc++.h>
using namespace std;
typedef long long ll;
const int M = 2e5;ll A, B, C, D, N;
int n;
vector<int> gx[55];inline int g(int x) {
int res = 0;while (x) {
res += x % 10;x /= 10;}return res;
}ll calc(ll a, ll b, ll x) {
return a * x * x + b * x;
}
void init() {
for (int i = 1; i <= 1000000; i++) {
gx[g(i)].push_back(i);}
}int main() {
int T;scanf("%d", &T);init();while (T--){
scanf("%lld%lld%lld%lld%lld", &A, &B, &C, &D, &N);ll res = 9e18;for (int i = 1; i <= 54; i++) {
if (gx[i].size() == 0 || gx[i][0]> N) continue;//找到第一个大于N的下标, 则gx[i][r - 1]是满足 <= N的最大数字int r = upper_bound(gx[i].begin(), gx[i].end(), N) - gx[i].begin();ll a = A * i + B, b = C * i * i + D * i;if (a <= 0) res = min(res, min(calc(a, b, gx[i][0]), calc(a, b, gx[i][r - 1])));else{
//对称轴int mid = -b / (2 * a);if (mid <= gx[i][0]) res = min(res, calc(a, b, gx[i][0]));else if (mid >= gx[i][r - 1]) res = min(res, calc(a, b, gx[i][r - 1]));else {
int pos = lower_bound(gx[i].begin(), gx[i].begin() + r, mid) - gx[i].begin();for (int j = -1; j <= 1; j++) {
int t = pos + j;if (t >= 0 && t < r) res = min(res, calc(a, b, gx[i][t]));}}}}printf("%lld\n", res);}return 0;
}
Problem I. Command Sequence
对于每一个点位,看看之前遍历过的序列之中有多少个点和其相同,这样就找到了一个序列,这一点可以用一个hash表来实现
#include<bits/stdc++.h>
using namespace std;
typedef long long ll;
const int N = 1e5, M = 2e5;map<ll, int> S;
char s[N];
int n;ll get(int x, int y) {
x += N, y += N;return (ll)x * M + y;
}int main() {
int T;scanf("%d", &T);while (T--){
scanf("%d", &n);scanf("%s", s);int x = 0, y = 0;S.clear();S[get(x,y)]++;ll res = 0;for (int i = 0; i < n; i++) {
if (s[i] == 'U') x--;else if (s[i] == 'D') x++;else if (s[i] == 'L') y--;else y++;res += S[get(x, y)];S[get(x, y)]++;}printf("%lld\n", res);}return 0;
}
Problem K. Shooting Bricks
贴一个原题链接。。
P1174 打砖块
#include <iostream>
#include <cstring>
using namespace std;
const int N = 210;int f[N][N][2];
int g[N][N], v[N][N][2];
bool st[N][N];
int n, m, k;int main()
{
#ifdef ONLINE_JUDGE
#else freopen("in.txt", "r", stdin);
#endifcin >> n >> m >> k;for(int i = 1; i <= n; i++)for(int j = 1; j <= m; j++){
char c;cin >> g[i][j] >> c;if(c == 'Y') st[i][j] = true;}for(int i = 1; i <= m; i++) {
for(int j = n, cnt = 0; j ; j--) {
if(st[j][i]) v[i][cnt][1] += g[j][i];else {
cnt++;v[i][cnt][1] = v[i][cnt][0] = v[i][cnt - 1][1] + g[j][i];} }}for(int i = 1; i <= m; i++)for(int j = 0; j <= k; j++)for(int l = 0; l <= min(n, j); l++){
f[i][j][1] = max(f[i][j][1], f[i - 1][j - l][1] + v[i][l][1]);//从前面借if(l) f[i][j][0] = max(f[i][j][0], f[i - 1][j - l][1] + v[i][l][0]);if(j > l) f[i][j][0] = max(f[i][j][0], f[i - 1][j - l][0] + v[i][l][1]);}cout << f[m][k][0];return 0;
}
Problem L. remove
一道贪心题,首先可以先构造一个序列ana_nan?,然后再套公式,用快速幂来求解答案,题目要求mod2642^{64}264,那么用unsigned long long
即可
首先,可以知道当i>=lcm(pi)i >= lcm(p_i)i>=lcm(pi?)的时候,a[i]=0a[i]=0a[i]=0, 所以,只需要处理[1, min((lcm{ pip_ipi?}, n)]这个区间的所有数字即可。
然后,在有解的情况下,a[i]是单调不减的,可以自行证明
如果我们想要求一个x,如果它有一个质因子p在给定的质数集合之中,那么,x可以转移的范围是[x+1,x+p-1](逆着考虑)。所以可以先类似于素数筛,把[1, min((lcm{ pip_ipi?}, n)]这个区间中p的倍数都标记上。对于每个位置i,f[i](pk∣i)f[i](p_k|i)f[i](pk?∣i) 为位置i可以转移到最远位置所用的质数,最后枚举每个i,那么i+f[i]?1i+f[i]-1i+f[i]?1就是i能转移到的最大右边界,如果i+f[i]?1>=ri+f[i]-1 >= ri+f[i]?1>=r,则让r一直往后走就行,g[r]=g[i]+1g[r]=g[i] + 1g[r]=g[i]+1 (r > i)
#include <iostream>
#include <cstring>
#include <algorithm>
using namespace std;
const int N = 2e6 + 10, M = 1e5 + 10;
typedef unsigned long long ULL;
typedef long long LL;
int n, m;
int a[N], p[N], f[N], g[N];ULL qmi(ULL a, ULL b) {
ULL res = 1;while(b) {
if(b & 1) res = res * a;a *= a;b >>= 1;}return res;
}
int main()
{
int T;scanf("%d", &T);while(T--) {
scanf("%d%d", &n, &m);fill(f, f + n + 1, 0);for(int i = 1; i <= m; i++) scanf("%d", p + i);LL mul = 1;for(int i = 1; i <= m; i++) {
mul *= p[i];if(mul > n) {
mul = n + 1;break;}}for(int i = 1; i <= m; i++) {
for(int j = 0; j < mul; j += p[i]) {
f[j] = max(f[j], p[i]);}}g[0] = 0;for(int i = 0, r = 1; i < mul; i++) {
while(r <= i + f[i] - 1 && r < mul) g[r++] = g[i] + 1; }ULL res = 0;for(int i = 1; i < mul; i++) res += (ULL)g[i] * qmi(23333, n - i);cout << res << endl;}return 0;
}