POJ 2528 Mayor's posters

#ifdef _DEBUG
#pragma warning(disable : 4996)
#endif
#include <iostream>
#include <string>
#include <vector>
#include <stack>
#include <queue>
#include <deque>
#include <set>
#include <map>
#include <algorithm>
#include <functional>
#include <sstream>
#include <utility>
#include <cstring>
#include <cstdio>
#include <cstdlib>
#include <ctime>
#include <cmath>
#include <cctype>
#define CLEAR(a, b) memset(a, b, sizeof(a))
#define CLOSE() ios::sync_with_stdio(false)
#define IN() freopen("in.txt", "r", stdin)
#define OUT() freopen("out.txt", "w", stdout)
#define PF(a) printf("%d\n", a)
#define SF(a) scanf("%d", &a)
#define SFF(a, b) scanf("%d%d", &a, &b)
#define FOR(i, a, b) for(int i = a; i < b; ++i)
#define LL long long
#define MAXN 10005
#define MAXM 100005
#define MOD 1000007
#define INF 2147483648
using namespace std;
//-------------------------CHC------------------------------//
//POJ 2528 Mayor's posters

//思路：由于区间更新后都PushDown到了最底层，所以查询也得到最底层逐个计数，比较耗时。

#define Lson cur<<1, l, mid
#define Rson cur<<1|1, mid+1, r

int n, m;
int low[MAXN], high[MAXN];
int col[MAXN << 4];
int cnt[MAXN << 4];
vector<int> v;

void PushDown(int cur) {
	col[cur << 1] = col[cur << 1 | 1] = col[cur];
	col[cur] = 0;
}

void Update(int L, int R, int c, int cur, int l, int r) {
	if (L <= l && r <= R) {
		col[cur] = c;
		//printf("col = %d\n", col[cur]);
		return;
	}
	if (col[cur]) PushDown(cur);
	int mid = (l + r) >> 1;
	if (L <= mid) Update(L, R, c, Lson);
	if (R > mid) Update(L, R, c, Rson);
}

void Query(int cur, int l, int r) {
	if (l == r) {
		if(col[cur]) cnt[col[cur]]++;
		//printf("col = %d\n", col[cur]);
		return;
	}
	if(col[cur]) PushDown(cur);
	int mid = (l + r) >> 1;
	Query(Lson);
	Query(Rson);
}

void Hash() {
	v.clear();
	FOR(i, 0, n) v.push_back(low[i]), v.push_back(high[i]);
	sort(v.begin(), v.end()); 
	int t = v.size();
	FOR(i, 1, t) {
		if (v[i] - v[i - 1] > 1) v.push_back(v[i - 1] + 1);
	}
	sort(v.begin(), v.end());
	v.erase(unique(v.begin(), v.end()), v.end());
	//for (auto i : v) printf("%d ", i); puts("");
	m = v.size();
	FOR(i, 0, n) {
		int L, R;
		L = find(v.begin(), v.end(), low[i]) - v.begin();
		R = find(v.begin(), v.end(), high[i]) - v.begin();
		//printf("(%d, %d)\n", L + 1, R + 1);
		Update(L+1, R+1, i + 1, 1, 1, m);
	}
}

int main() {
	int T;
	SF(T);
	while (T--) {
		CLEAR(col, 0);
		CLEAR(cnt, 0);
		SF(n);
		FOR(i, 0, n) SFF(low[i], high[i]);
		Hash();
		Query(1, 1, m);
		int ans = 0;
		FOR(i, 1, m + 1) if (cnt[i]) ans++;
		PF(ans);
	}
	return 0;
}

仅仅是更改了查询方式，就快了很多了。
存在部分点，没有pushdown到叶子节点。
(还是有点不清楚，感觉并没有太多的点能直接返回的诶)

void query(int cur, int l, int r) {
	if (res[cur]) cnt[res[cur]] = 1;
	else {
		if (l == r) return;
		int mid = Mid(l, r);
		PushDown(cur);
		query(lson);
		query(rson);
	}
}