HYSBZ - 2157树链剖分

【题目描述】
HYSBZ - 2157树链剖分

【题目分析】
这道题给出的是边权而不是点权，但是我们分析这个树就会发现每个节点都只有一个父亲，也就是每条边的边权都可以存放在儿子节点上，然后在遍历路径的时候我们在从前往后遍历，但是注意最后一条链的链首不要算在内（因为我们只算边权，链首存储的值不在路径上）
接下来就是无穷无尽的代码了，我自己写的一直一直wa，我看了一天了也没有看出哪里有问题，借鉴大佬的代码：
【AC代码】

#include<iostream>
#include<cstdio>
#include<cstring>
#include<cstdlib>
#include<algorithm>using namespace std;const int maxn = 120000;
#define ll long long
#define ls now<<1
#define rs now<<1|1
inline int read()
{int x = 0, f = 1;char ch = getchar();while (ch < '0' || ch > '9') { if (ch == '-') f = -1; ch = getchar(); }while (ch >= '0' && ch <= '9') { x = x * 10 + ch - 48; ch = getchar(); }return x * f;
}
struct node
{ll f, t, v;
}e[maxn << 1];
struct tree
{ll l, r, sum, maxx, minn, add, mul , num;
}tre[maxn << 2];
ll n, m, tot, cnt;
ll head[maxn], nxt[maxn << 1], used[maxn], dep[maxn], fa[maxn], sz[maxn], w1[maxn];
ll w2[maxn] , son[maxn] , top[maxn] , num[maxn];
inline void buildnode(ll a, ll b , ll c)
{tot++;e[tot].f = a;e[tot].t = b;e[tot].v = c;nxt[tot] = head[a];head[a] = tot;
}
inline void pushup(ll now)
{tre[now].sum = tre[ls].sum + tre[rs].sum;tre[now].maxx = max(tre[ls].maxx, tre[rs].maxx);tre[now].minn = min(tre[ls].minn, tre[rs].minn);
}
inline void build(ll now, ll l, ll r)
{tre[now].l = l;tre[now].r = r;tre[now].num = r - l + 1;if (l == r){tre[now].sum = w2[l];tre[now].maxx = w2[l];tre[now].minn = w2[l];return;}ll mid = (l + r) >> 1;build(ls, l, mid);build(rs, mid + 1 , r);pushup(now);
}
inline void pushdown(ll now)
{if (tre[now].mul){tre[ls].sum *= -1;tre[rs].sum *= -1;ll t = tre[ls].maxx;tre[ls].maxx = -tre[ls].minn;tre[ls].minn = -t;t = tre[rs].maxx;tre[rs].maxx = -tre[rs].minn;tre[rs].minn = -t;}tre[ls].mul = (tre[ls].mul + tre[now].mul) % 2;tre[rs].mul = (tre[rs].mul + tre[now].mul) % 2;tre[now].mul = 0;
}
inline void update1(ll now, ll x ,  ll p)
{if (tre[now].l == tre[now].r){tre[now].sum = p;tre[now].maxx = p;tre[now].minn = p;return;}pushdown(now);ll mid = (tre[now].l + tre[now].r) >> 1;if (x <= mid) update1(ls, x, p);if (x > mid) update1(rs, x, p);pushup(now);
}
inline void update2(ll now, ll l, ll r)
{if (l <= tre[now].l && tre[now].r <= r){tre[now].sum = -tre[now].sum;tre[now].mul = (tre[now].mul + 1) % 2;ll t = tre[now].maxx;tre[now].maxx = -tre[now].minn;tre[now].minn = -t;return;}pushdown(now);ll mid = (tre[now].l + tre[now].r) >> 1;if (l <= mid) update2(ls, l, r);if (r > mid) update2(rs, l, r);pushup(now);
}
inline void dfs1(ll x, ll fat)
{dep[x] = dep[fat] + 1;fa[x] = fat;sz[x] = 1;int k = -1;for (int i = head[x]; i; i = nxt[i]){int u = e[i].t;if (u == fa[x]) continue;w1[u] = e[i].v;dfs1(u, x);sz[x] += sz[u];if (sz[u] > k) k = sz[u], son[x] = u;}
}
inline void dfs2(ll x , ll topx)
{num[x] = ++cnt;w2[cnt] = w1[x];top[x] = topx;if (!son[x]) return;dfs2(son[x], topx);for (int i = head[x]; i; i = nxt[i]){int u = e[i].t;if (u == fa[x] || u == son[x]) continue;dfs2(u, u);}
}
inline int querys(ll now, ll l, ll r)
{if (l <= tre[now].l && tre[now].r <= r)return tre[now].sum;pushdown(now);ll mid = (tre[now].l + tre[now].r) >> 1, ans = 0;if (l <= mid) ans += querys(ls, l, r);if (r > mid) ans += querys(rs, l, r);return ans;
}
inline ll query_max(ll now, ll l , ll r)
{if (l <= tre[now].l &&tre[now].r <= r)return tre[now].maxx;pushdown(now);ll mid = (tre[now].l + tre[now].r) >> 1, ans = -1e9;if (l <= mid) ans = max(ans, query_max(ls, l, r));if (r > mid) ans = max(ans, query_max(rs, l, r));return ans;
}
inline ll query_min(ll now, ll l , ll r)
{if (l <= tre[now].l &&tre[now].r <= r)return tre[now].minn;pushdown(now);ll mid = (tre[now].l + tre[now].r) >> 1, ans = 1e9;if (l <= mid) ans = min(ans, query_min(ls, l, r));if (r > mid) ans = min(ans, query_min(rs, l, r));return ans;
}
inline ll qline(ll x, ll y)
{ll ans = 0;while (top[x] != top[y]){if (dep[top[x]] < dep[top[y]]) swap(x , y);ans += querys(1, num[top[x]], num[x]);x = fa[top[x]];}if (dep[x] > dep[y]) swap(x, y);ans += querys(1, num[x] + 1, num[y]);return ans;
}
inline void cline(ll x, ll y)
{while (top[x] != top[y]){if (dep[top[x]] < dep[top[y]]) swap(x , y);update2(1, num[top[x]], num[x]);x = fa[top[x]];}if (dep[x] > dep[y]) swap(x, y);update2(1, num[x] + 1, num[y]);
}
inline ll qpmax(ll x, ll y)
{ll ans = -1e9;while (top[x] != top[y]){if (dep[top[x]] < dep[top[y]]) swap(x , y);ans = max(ans, query_max(1, num[top[x]], num[x]));x = fa[top[x]];}if (dep[x] > dep[y]) swap(x, y);ans = max(ans, query_max(1, num[x] + 1, num[y]));return ans;
}
inline ll qpmin(ll x, ll y)
{ll ans = 1e9;while (top[x] != top[y]){if (dep[top[x]] < dep[top[y]]) swap(x , y);ans = min(ans, query_min(1, num[top[x]], num[x]));x = fa[top[x]];}if (dep[x] > dep[y]) swap(x, y);ans = min(ans, query_min(1, num[x] + 1, num[y]));return ans;
}
string s1 = "C", s2 = "N", s3 = "SUM", s4 = "MAX", s5 = "MIN";
int main()
{//freopen("1.in","r",stdin);n = read();for (int i = 1; i <= n - 1; i++){int a, b, c;a = read(); b = read(); c = read();buildnode(a + 1, b + 1, c);buildnode(b + 1, a + 1, c);}dfs1(1 , 0);dfs2(1 , 1);build(1 , 1 , n);m = read();while (m--){string s;int a, b, k;cin >> s >> a >> b;a += 1 , b += 1;if (s == s1){a -= 1 , b -= 1;if (dep[e[a * 2 - 1].f] > dep[e[a * 2 - 1].t])k = e[a * 2 - 1].f;else k = e[a * 2 - 1].t;update1(1, num[k], b);}if (s == s2) cline(a, b);if (s == s3) printf("%lld\n", qline(a, b));if (s == s4) printf("%lld\n", qpmax(a, b));if (s == s5) printf("%lld\n", qpmin(a, b));}return 0;
}

我自己写的也贴上，方便日后检查

#include<cstdio>
#include<cstring>
#include<cmath>
#include<climits>
#include<cstdlib>
#include<algorithm>
#include<queue>
#include<vector>
#include<set>using namespace std;typedef long long ll;const int MAXN=120005;
int fa[MAXN],A[MAXN],val[MAXN],pos[MAXN];
int siz[MAXN],son[MAXN],h[MAXN],top[MAXN];
int cnt=0,n,m;
int Sum[MAXN<<2],Max[MAXN<<2],Min[MAXN<<2],lazy[MAXN<<2];
struct node
{int u,v,w;
}e[MAXN<<1];
int head[MAXN<<1],nxt[MAXN<<1];
int tot=0;void AddEdge(int u,int v,int w)
{tot++;e[tot].u=u; e[tot].v=v; e[tot].w=w;nxt[tot]=head[u]; head[u]=tot;
}void dfs1(int u,int f)
{int i,v;siz[u]=1;son[u]=0;fa[u]=f;h[u]=h[f]+1;for(i=head[u];i;i=nxt[i]){v=e[i].v;if(v!=f){val[v]=e[i].w;dfs1(v,u);siz[u]+=siz[v];if(siz[son[u]]<siz[v]) son[u]=v;}}
}
void dfs2(int u,int f,int k)
{int i,v;top[u]=k;pos[u]=++cnt;A[cnt]=val[u];if(!son[u]) return; if(son[u]) dfs2(son[u],u,k);for(i=head[u];i;i=nxt[i]){v=e[i].v;if(v!=f&&v!=son[u]) dfs2(v,u,v);}
}void pushup(int k)
{Sum[k]=Sum[k<<1]+Sum[k<<1|1];Max[k]=max(Max[k<<1],Max[k<<1|1]);Min[k]=min(Min[k<<1],Min[k<<1|1]);
}void pushdown(int k)
{if(lazy[k]){Sum[k<<1]*=-1; Sum[k<<1|1]*=-1;swap(Max[k<<1],Min[k<<1]);Max[k<<1]*=-1; Min[k<<1]*=-1;swap(Max[k<<1|1],Min[k<<1|1]);Max[k<<1|1]*=-1; Min[k<<1|1]*=-1;lazy[k<<1]^=1; lazy[k<<1|1]^=1;lazy[k]=0;}
}void build(int k,int l,int r)
{if(l==r){Sum[k]=Max[k]=Min[k]=A[l];return;}int mid=(l+r)>>1;build(k<<1,l,mid);build(k<<1|1,mid+1,r);pushup(k);
}void PointChange(int k,int l,int r,int x,int v)
{if(l==r && l==x){Sum[k]=Max[k]=Min[k]=v;return;}pushdown(k);int mid=(l+r)>>1;if(x<=mid) PointChange(k<<1,l,mid,x,v);else PointChange(k<<1|1,mid+1,r,x,v);pushup(k);
}void IntervalChange(int k,int l,int r,int L,int R)
{if(l>=L && r<=R){Sum[k]*=-1; swap(Max[k],Min[k]);Max[k]*=-1; Min[k]*=-1;lazy[k]^=1;return;}pushdown(k);int mid=(l+r)>>1;if(L<=mid) IntervalChange(k<<1,l,mid,L,R);if(R>mid) IntervalChange(k<<1|1,mid+1,r,L,R);pushup(k);
}int IntervalSum(int k,int l,int r,int L,int R)
{if(L<=l && r<=R){return Sum[k];}int mid=(l+r)/2;pushdown(k);int ret=0;if(L<=mid) ret+=IntervalSum(k<<1,l,mid,L,R);if(R>mid) ret+=IntervalSum(k<<1|1,mid+1,r,L,R);return ret;
}int IntervalMax(int k,int l,int r,int L,int R)
{if(L<=l && r<=R){return Max[k];}int mid=(l+r)/2;pushdown(k);int ret=INT_MIN;if(L<=mid) ret=max(IntervalSum(k<<1,l,mid,L,R),ret);if(R>mid) ret=max(IntervalSum(k<<1|1,mid+1,r,L,R),ret);return ret;
}int IntervalMin(int k,int l,int r,int L,int R)
{if(L<=l && r<=R){return Min[k];}int mid=(l+r)/2;pushdown(k);int ret=INT_MAX;if(L<=mid) ret=min(IntervalSum(k<<1,l,mid,L,R),ret);if(R>mid) ret=min(IntervalSum(k<<1|1,mid+1,r,L,R),ret);return ret;
}int FindSum(int u,int v)
{int ans=0;while(top[u]!=top[v]){if(h[top[u]]<h[top[v]]) swap(u,v);ans+=IntervalSum(1,1,n,pos[top[u]],pos[u]);u=fa[top[u]];}if(h[u]<h[v]) swap(u,v);ans+=IntervalSum(1,1,n,pos[v]+1,pos[u]);return ans;
}int FindMax(int u,int v)
{int ans=INT_MIN;while(top[u]!=top[v]){if(h[top[u]]<h[top[v]]) swap(u,v);ans=max(IntervalMax(1,1,n,pos[top[u]],pos[u]),ans);u=fa[top[u]];}if(h[u]<h[v]) swap(u,v);ans=max(IntervalMax(1,1,n,pos[v]+1,pos[u]),ans);return ans;
}int FindMin(int u,int v)
{int ans=INT_MAX;while(top[u]!=top[v]){if(h[top[u]]<h[top[v]]) swap(u,v);ans=min(IntervalMin(1,1,n,pos[top[u]],pos[u]),ans);u=fa[top[u]];}if(h[u]<h[v]) swap(u,v);ans=min(IntervalMin(1,1,n,pos[v]+1,pos[u]),ans);return ans;
}void update(int u,int v)
{while(top[u]!=top[v]){if(h[top[u]]<h[top[v]]) swap(u,v);IntervalChange(1,1,n,pos[top[u]],pos[u]);u=fa[top[u]];}if(h[u]<h[v]) swap(u,v);IntervalChange(1,1,n,pos[v]+1,pos[u]);
}int main()
{int u,v,w,idx;char cmd[10];scanf("%d",&n);//for(int i=1;i<=n;i++) scanf("%d",&val[i]);for(int i=1;i<n;i++){scanf("%d%d%d",&u,&v,&w);AddEdge(u+1,v+1,w); AddEdge(v+1,u+1,w);}dfs1(1 , 0);dfs2(1 ,0, 1);build(1,1,n);scanf("%d",&m);while(m--){scanf("%s",cmd);if(cmd[0]=='C'){scanf("%d%d",&idx,&w);u=e[idx*2-1].u; v=e[idx*2-1].v;if(h[u]<h[v]){PointChange(1,1,n,pos[v],w);}else{PointChange(1,1,n,pos[u],w);}}else if(cmd[0]=='N') {scanf("%d%d",&u,&v);update(u+1,v+1);}else if(cmd[0]=='S'){scanf("%d%d",&u,&v);printf("%d\n",FindSum(u+1,v+1));}else if(cmd[1]=='A'){scanf("%d%d",&u,&v);printf("%d\n",FindMax(u+1,v+1));}else if(cmd[1]=='I'){scanf("%d%d",&u,&v);printf("%d\n",FindMin(u+1,v+1));}}return 0;
}

HYSBZ - 2157树链剖分相关推荐

BZOJ 2157 「国家集训队」旅游（树链剖分，线段树，边权转点权）【BZOJ计划】
整理的算法模板合集: ACM模板点我看算法全家桶系列!!! 实际上是一个全新的精炼模板整合计划题目链接 https://hydro.ac/d/bzoj/p/2157 是 hydro 的 BZOJ ...
树链剖分入门+HYSBZ - 1036树的统计Count
今天学习了树链剖分,记录一下. [题目背景] HYSBZ - 1036树的统计Count [题目分析] 题目要求求任意结点之间路径的和以及路径上最大的结点,还有可能修改.如果正常做可能会很复杂(我也不 ...
HDU 3966 POJ 3237 HYSBZ 2243 HRBUST 2064 树链剖分
树链剖分是一个很固定的套路一般用来解决树上两点之间的路径更改与查询思想是将一棵树分成不想交的几条链并且由于dfs的顺序性给每条链上的点或边标的号必定是连着的那么每两个点之间的路径都可以拆成几 ...
树链剖分——杨子曰算法
树链剖分--杨子曰算法先搞一道题,给一个n个结点的树(n<=30000),每个结点上有一个权值,然后是m个询问(m<=200000),对于每个询问k,a,b,如果k为1,则输出从a到b路 ...
树链剖分入门——[kuangbin]树链剖分
树链剖分的本质就是将一棵树拆分成一段一段连续的区间,然后放在一起就可以用一棵单独的线段树处理区间问题,只需要将树上节点和线段树节点的对应关系求好就可以很方便的互相转换,而树上两点之间路径的相关问题就可 ...
【ZJOI2008】树的统计（树链剖分）
传送门 Solution: 就是树链剖分入门题啦~ // luogu-judger-enable-o2 #include<bits/stdc++.h> #define N 30005 #d ...
树链剖分+线段树 HDOJ 4897 Little Devil I（小恶魔）
题目链接题意: 给定一棵树,每条边有黑白两种颜色,初始都是白色,现在有三种操作: 1 u v:u到v路径(最短)上的边都取成相反的颜色 2 u v:u到v路径上相邻的边都取成相反的颜色(相邻即仅有一 ...
P1505 [国家集训队]旅游树链剖分
题目链接题意:树上更新某一点权值,更新两点简单路径权值,查询最大,最小,和思路:思路应该比较简单,就是树链剖分后用线段树维护区间最大最小以及区间和. 但是本题比较特殊的是给的边权,转化为点权即可. ...
【模板】树链剖分 P3384
题目链接 //部分转自:https://www.luogu.org/problemnew/solution/P3384 初学树链剖分,感觉这个模板题还是容易理解的,但是实在是码量很大的. 知识点: 重 ...

HYSBZ - 2157树链剖分

HYSBZ - 2157树链剖分相关推荐

最新文章

热门文章