Loading... [Link](https://www.luogu.com.cn/problem/CF786B) ## Sol 这是一道线段树优化连边板子题 要用这个的题通常是这样的, 一个点往一个区间内所有点连边,或者一个区间都往一个点连边. 如果你直接暴力连的话时间很差劲,你就需要线段树优化连边. 思路大概是这样的,运用线段树的思想. 线段树的每一个叶子节点对应原图的节点, 非叶节点就对应的是一段区间的点. 这样一来连边就只需要连接线段树上的点. 比如说一个题往一个区间内所有点连边,就是从线段树的叶节点往一个代表整个区间的点连边.这样相当于是从线段树深度比较深的底部往浅的地方连. 一个区间都往一个点连就是往下连. 每次都按照递归在线段树上找一下对应区间的节点,然后连就可以了. 实现的话都可以, 这道题这两种连边都有. 就直接开两个动态开点线段树, 然后共用叶子节点 ## Code ~~不懂的问题看了代码就懂了~~ ```cpp #include <bits/stdc++.h> using namespace std; #define int long long const int SIZE = 4e5 + 5; const int inf = 1e18; int n, q, s, root1, root2, nodeCnt, num; int head[SIZE], dis[SIZE], vis[SIZE], lson[SIZE], rson[SIZE]; struct edge { int to, v, nxt; } edge[SIZE * 10]; namespace ae86 { const int bufl = 1 << 15; char buf[bufl], *s = buf, *t = buf; inline int fetch() { if (s == t) { t = (s = buf) + fread(buf, 1, bufl, stdin); if (s == t) return EOF; } return *s++; } inline int read() { int a = 0, b = 1, c = fetch(); while (!isdigit(c))b ^= c == '-', c = fetch(); while (isdigit(c)) a = a * 10 + c - 48, c = fetch(); return b ? a : -a; } } using ae86::read; inline void addEdge(int u, int v, int d) { edge[++ num].to = v, edge[num].v = d, edge[num].nxt = head[u]; head[u] = num; } inline int buildDown(int l, int r) { if (l == r) return l; int mid = (l + r) >> 1, p = ++ nodeCnt; lson[p] = buildDown(l, mid); rson[p] = buildDown(mid + 1, r); addEdge(p, lson[p], 0); addEdge(p, rson[p], 0); return p; } inline int buildUp(int l, int r) { if (l == r) return l; int mid = (l + r) >> 1, p = ++ nodeCnt; lson[p] = buildUp(l, mid); rson[p] = buildUp(mid + 1, r); addEdge(lson[p], p, 0); addEdge(rson[p], p, 0); return p; } inline void link(int from, int to, int val, int l, int r, int L, int R, int typ) { if (L <= l && r <= R) { if (typ == 1) addEdge(from, to, val); // up -> down else addEdge(to, from, val); // down -> up return; } int mid = (l + r) >> 1; if (L <= mid) link(lson[from], to, val, l, mid, L, R, typ); if (R > mid) link(rson[from], to, val, mid + 1, r, L, R, typ); } inline void dijkstra() { std::priority_queue < pair < int, int > > q; for (int i = 1; i <= nodeCnt; ++ i) dis[i] = inf; memset(vis, 0, sizeof(vis)); q.push(make_pair(0, s)); dis[s] = 0; while (!q.empty()) { int u = q.top().second, v; q.pop(); if (vis[u]) continue; vis[u] = 1; for (int i = head[u]; i; i = edge[i].nxt) { v = edge[i].to; if (dis[v] > dis[u] + edge[i].v) { dis[v] = dis[u] + edge[i].v; q.push(make_pair(-dis[v], v)); } } } } signed main() { // freopen("code.in", "r", stdin); // freopen("code.out", "w", stdout); nodeCnt = n = read(), q = read(), s = read(); root1 = nodeCnt + 1; buildDown(1, n); root2 = nodeCnt + 1; buildUp(1, n); int opt, v, u, l, r, w; while (q --) { opt = read(); if (opt == 1) { u = read(), v = read(), w = read(); addEdge(u, v, w); } else { u = read(), l = read(), r = read(), w = read(); link(opt ^ 3 ? root1 : root2, u, w, 1, n, l, r, opt - 2); } } dijkstra(); for (int i = 1; i <= n; ++ i) printf("%lld ", dis[i] == inf ? -1 : dis[i]); return 0; } ``` 最后修改:2021 年 09 月 04 日 © 允许规范转载 打赏 赞赏作者 支付宝微信 赞 如果觉得我的文章对你有用,请随意赞赏