Tree

Time Limit: 16000/8000 MS (Java/Others)    Memory Limit: 65536/65536 K (Java/Others)
Total Submission(s): 920    Accepted Submission(s): 388

Problem Description
You are given a tree with N nodes which are numbered by integers 1..N. Each node is associated with an integer as the weight.

Your task is to deal with M operations of 4 types:

1.Delete an edge (x, y) from the tree, and then add a new edge (a, b). We ensure that it still constitutes a tree after adding the new edge.

2.Given two nodes a and b in the tree, change the weights of all the nodes on the path connecting node a and b (including node a and b) to a particular value x.

3.Given two nodes a and b in the tree, increase the weights of all the nodes on the path connecting node a and b (including node a and b) by a particular value d.

4.Given two nodes a and b in the tree, compute the second largest weight on the path connecting node a and b (including node a and b), and the number of times this weight occurs on the path. Note that here we need the strict second largest weight. For instance, the strict second largest weight of {3, 5, 2, 5, 3} is 3.

 
Input
The first line contains an integer T (T<=3), which means there are T test cases in the input.

For each test case, the first line contains two integers N and M (N, M<=10^5). The second line contains N integers, and the i-th integer is the weight of the i-th node in the tree (their absolute values are not larger than 10^4).

In next N-1 lines, there are two integers a and b (1<=a, b<=N), which means there exists an edge connecting node a and b.

The next M lines describe the operations you have to deal with. In each line the first integer is c (1<=c<=4), which indicates the type of operation.

If c = 1, there are four integers x, y, a, b (1<= x, y, a, b <=N) after c.
If c = 2, there are three integers a, b, x (1<= a, b<=N, |x|<=10^4) after c.
If c = 3, there are three integers a, b, d (1<= a, b<=N, |d|<=10^4) after c.
If c = 4 (it is a query operation), there are two integers a, b (1<= a, b<=N) after c.

All these parameters have the same meaning as described in problem description.

 
Output
For each test case, first output "Case #x:"" (x means case ID) in a separate line.

For each query operation, output two values: the second largest weight and the number of times it occurs. If the weights of nodes on that path are all the same, just output "ALL SAME" (without quotes).

 
Sample Input
2
3 2
1 1 2
1 2
1 3
4 1 2
4 2 3
7 7
5 3 2 1 7 3 6
1 2
1 3
3 4
3 5
4 6
4 7
4 2 6
3 4 5 -1
4 5 7
1 3 4 2 4
4 3 6
2 3 6 5
4 3 6
 
Sample Output
Case #1:
ALL SAME
1 2
Case #2:
3 2
1 1
3 2
ALL SAME
/*
hdu 5002 (动态树lct) problem:
给你一棵树树,主要包含四个操作:
1 x y u v:断开x,y之间的边 连接上u,v
2 x y w:将x->y之间的点权全部置为w
3 x y w:将x->y之间的点权全部加上w
4 x y:查询x->y之间第二大的 solve:
只是需要维护下第二大值,其它直接套模板 hhh-2016-08-20 17:21:29
*/
#pragma comment(linker,"/STACK:124000000,124000000")
#include <algorithm>
#include <iostream>
#include <cstdlib>
#include <cstdio>
#include <cstring>
#include <vector>
#include <map>
#define lson ch[0]
#define rson ch[1]
#define ll long long
#define clr(a,b) memset(a,b,sizeof(a))
#define key_val ch[ch[root][1]][0]
using namespace std;
const int maxn = 300100;
const int INF = 0x3f3f3f3f; struct Node* null;
struct Node
{
Node* ch[2] ;
Node* fa;
int Size ;
int mMax ;
int sMax ;
int max_num ;
int Max_num ;
int val ;
int add ;
int same ;
int rev;
void newnode(int v)
{
val = v ;
mMax = v ;
sMax = -INF ;
Max_num = 1 ;
max_num = 0 ;
Size = 1 ;
add = 0 ;
same = -INF ;
fa = ch[0] = ch[1] = null ;
rev = 0;
}
void update_rev()
{
if(this == null)
return ;
swap(ch[0],ch[1]);
rev ^= 1;
}
void update_add(int v)
{
if(this == null )return ;
add += v;
mMax += v,val += v;
if(sMax != -INF) sMax += v;
} void update_same(int v)
{
if(this == null) return ;
same = v;
add = 0,val = v,mMax = v;
sMax = -INF,Max_num = Size,max_num = 0;
}
void cal(int val,int num)
{
if ( val == -INF ) return ;
if ( val < sMax ) return ;
if ( val > mMax )
{
sMax = mMax ;
max_num = Max_num ;
mMax = val ;
Max_num = num ;
}
else if ( val == mMax )
{
Max_num += num ;
}
else if ( val > sMax )
{
sMax = val ;
max_num = num ;
}
else max_num += num ;
}
void push_up () {
Size = ch[0]->Size + 1 + ch[1]->Size ;
mMax = sMax = -INF ;
max_num = Max_num = 0 ;
cal ( val , 1 ) ;
cal ( ch[0]->mMax , ch[0]->Max_num ) ;
cal ( ch[0]->sMax , ch[0]->max_num ) ;
cal ( ch[1]->mMax , ch[1]->Max_num ) ;
cal ( ch[1]->sMax , ch[1]->max_num ) ;
} void push_down()
{
if(rev)
{
ch[0]->update_rev();
ch[1]->update_rev();
rev = 0;
}
if(same != -INF)
{
ch[0]->update_same(same);
ch[1]->update_same(same);
same = -INF;
}
if(add)
{
ch[0]->update_add(add);
ch[1]->update_add(add);
add = 0;
}
} void link_child ( Node* to , int d )
{
ch[d] = to;
to->fa = this ;
} int isroot()
{
return fa == null || this != fa->ch[0] && this != fa->ch[1] ;
}
void down()
{
if ( !isroot () ) fa->down () ;
push_down () ;
}
void Rotate ( int d )
{
Node* f = fa ;
Node* ff = fa->fa ;
f->link_child ( ch[d] , !d ) ;
if ( !f->isroot () )
{
if ( ff->ch[0] == f ) ff->link_child ( this , 0 ) ;
else ff->link_child ( this , 1 ) ;
}
else fa = ff ;
link_child (f,d) ;
f->push_up () ;
} void splay ()
{
down () ;
while ( !isroot () ) {
if ( fa->isroot () ) {
this == fa->ch[0] ? Rotate ( 1 ) : Rotate ( 0 ) ;
} else {
if ( fa == fa->fa->ch[0] ) {
this == fa->ch[0] ? fa->Rotate ( 1 ) : Rotate ( 0 ) ;
Rotate ( 1 ) ;
} else {
this == fa->ch[1] ? fa->Rotate ( 0 ) : Rotate ( 1 ) ;
Rotate ( 0 ) ;
}
}
}
push_up () ;
} void access()
{
Node* now = this ;
Node* x = null ;
while ( now != null )
{
now->splay () ;
now->link_child ( x , 1 ) ;
now->push_up () ;
x = now ;
now = now->fa ;
}
splay () ;
} void make_root()
{
access();
update_rev();
} void cut()
{
access();
ch[0]->fa = null;
ch[0] = null;
push_up();
}
Node* find_root ()
{
access () ;
Node* to = this ;
while ( to->ch[0] != null )
{
to->push_down () ;
to = to->ch[0] ;
}
return to ;
}
void cut(Node* to)
{
to->make_root();
cut();
} void link(Node* to)
{
to->make_root();
to->fa = this;
}
void make_same(Node* to,int val)
{
to->make_root();
access();
update_same(val);
}
void make_add(Node* to,int val)
{
to->make_root();
access();
update_add(val);
}
void query(Node* to)
{
to->make_root();
access(); if(!max_num)
printf("ALL SAME\n");
else
printf("%d %d\n",sMax,max_num);
}
};
Node memory_pool[maxn];
Node* now;
Node* node[maxn]; void Clear()
{
now = memory_pool;
now->newnode(-INF);
null = now ++;
null->Size = 0;
} int main()
{
int T,n,cas = 1,m;
int x,y,a,b,c;
int ob;
// freopen("in.txt","r",stdin);
scanf("%d",&T);
while(T--)
{
Clear();
scanf("%d%d",&n,&m);
printf("Case #%d:\n",cas++);
for(int i = 1; i <= n; i++)
{
scanf("%d",&x);
now->newnode(x);
node[i] = now++;
} for(int i = 1; i < n; i++)
{
scanf("%d%d",&a,&b);
node[a]->link(node[b]); }
for(int i= 1; i <= m; i++)
{
scanf("%d",&ob);
if(ob == 1)
{
scanf("%d%d%d%d",&x,&y,&a,&b);
node[x]->cut(node[y]);
node[a]->link(node[b]);
}
else if(ob == 2)
{
scanf("%d%d%d",&x,&y,&c);
node[x]->make_same(node[y],c);
}
else if(ob == 3)
{
scanf("%d%d%d",&x,&y,&c);
node[x]->make_add(node[y],c);
}
else if(ob == 4)
{
scanf("%d%d",&x,&y);
node[x]->query(node[y]); }
}
}
return 0;
}

  

hdu 5002 (动态树lct)的更多相关文章

  1. hdu 5398 动态树LCT

    GCD Tree Time Limit: 5000/2500 MS (Java/Others)    Memory Limit: 65536/65536 K (Java/Others)Total Su ...

  2. hdu 5314 动态树

    Happy King Time Limit: 10000/5000 MS (Java/Others)    Memory Limit: 262144/262144 K (Java/Others)Tot ...

  3. HDU 4718 The LCIS on the Tree (动态树LCT)

    The LCIS on the Tree Time Limit: 6000/3000 MS (Java/Others)    Memory Limit: 65535/65535 K (Java/Oth ...

  4. 动态树LCT小结

    最开始看动态树不知道找了多少资料,总感觉不能完全理解.但其实理解了就是那么一回事...动态树在某种意思上来说跟树链剖分很相似,都是为了解决序列问题,树链剖分由于树的形态是不变的,所以可以通过预处理节点 ...

  5. bzoj2049-洞穴勘测(动态树lct模板题)

    Description 辉辉热衷于洞穴勘测.某天,他按照地图来到了一片被标记为JSZX的洞穴群地区.经过初步勘测,辉辉发现这片区域由n个洞穴(分别编号为1到n)以及若干通道组成,并且每条通道连接了恰好 ...

  6. [模板] 动态树/LCT

    简介 LCT是一种数据结构, 可以维护树的动态加边, 删边, 维护链上信息(满足结合律), 单次操作时间复杂度 \(O(\log n)\).(不会证) 思想类似树链剖分, 因为splay可以换根, 用 ...

  7. 动态树LCT(Link-cut-tree)总结+模板题+各种题目

    一.理解LCT的工作原理 先看一道例题: 让你维护一棵给定的树,需要支持下面两种操作: Change x val:  令x点的点权变为val Query x y:  计算x,y之间的唯一的最短路径的点 ...

  8. SPOJ OTOCI 动态树 LCT

    SPOJ OTOCI 裸的动态树问题. 回顾一下我们对树的认识. 最初,它是一个连通的无向的无环的图,然后我们发现由一个根出发进行BFS 会出现层次分明的树状图形. 然后根据树的递归和层次性质,我们得 ...

  9. BZOJ 2002: [Hnoi2010]Bounce 弹飞绵羊 (动态树LCT)

    2002: [Hnoi2010]Bounce 弹飞绵羊 Time Limit: 10 Sec  Memory Limit: 259 MBSubmit: 2843  Solved: 1519[Submi ...

随机推荐

  1. auto_prepend_file与auto_append_file使用方法

    auto_prepend_file与auto_append_file使用方法 如果需要将文件require到所有页面的顶部与底部. 第一种方法:在所有页面的顶部与底部都加入require语句. 例如: ...

  2. JAVA_SE基础——29.构造函数

    黑马程序员入学Blog... jvm创建Java对象时候需要调用构造器,默认是不带参数的.在构造器中,你可以让jvm帮你初始化一些参数或者执行一系列的动作. 它是对象创建中执行的函数,及第一个被执行的 ...

  3. linux下面的打包压缩命令

    tar命令 tar [-cxtzjvfpPN] 文件与目录 ....linux下面压缩之前要把一堆文件打个包再压缩,即使只有一个文件也需要打个包.例子:tar czvf 1.tar.gz hello. ...

  4. vscode使用shell

    https://stackoverflow.com/questions/42606837/how-to-use-bash-on-windows-from-visual-studio-code-inte ...

  5. Python内置函数(29)——slice

    英文文档: class slice(stop) class slice(start, stop[, step]) Return a slice object representing the set ...

  6. 说说Java代理模式

    代理实现可以分为静态代理和动态代理. 静态代理 静态代理模式其实很常见,比如买火车票这件小事:黄牛相当于是火车站的代理,我们可以通过黄牛买票,但只能去火车站进行改签和退票.在代码实现中相当于为一个委托 ...

  7. vue2.0+koa2+mongodb实现注册登录

    前言 前段时间和公司一个由技术转产品的同事探讨他的职业道路,对我说了一句深以为然的话: "不要把自己禁锢在某一个领域,技术到产品的转变,首先就是思维上的转变.你一直做前端,数据的交互你只知道 ...

  8. IDE-Android Studio -FAQ-使用习惯(不断更新 欢迎留言)

    摘要: 从ecplise工具切换到android studio后遇到了很多问题,起初亦非常痛苦,城墙内外阅博无数才得以解决.所以把当时遇到的问题记录下来,方便后来人学习. 另如果有遇到未纪录的问题欢迎 ...

  9. 使用 C#/.NET Core 实现单体设计模式

    本文的概念内容来自深入浅出设计模式一书 由于我在给公司做内培, 所以最近天天写设计模式的文章.... 单体模式 Singleton 单体模式的目标就是只创建一个实例. 实际中有很多种对象我们可能只需要 ...

  10. 排序技巧——双关键字排序(快速排序,sort)

    一个萌新的成长之路 Background 在做题过程中,我们常会遇到对双关键字排序的情况,如:当分数相等时,序号小的在前. 这时我们可以通过定义cmp函数作为sort的参数进行排序. Solution ...