codeforces 715B:Complete The Graph
Description
ZS the Coder has drawn an undirected graph of n vertices numbered from 0 to n - 1 and m edges between them. Each edge of the graph is weighted, each weight is a positive integer.
The next day, ZS the Coder realized that some of the weights were erased! So he wants to reassign positive integer weight to each of the edges which weights were erased, so that the length of the shortest path between vertices s and t in the resulting graph is exactly L. Can you help him?
The first line contains five integers n, m, L, s, t (2 ≤ n ≤ 1000, 1 ≤ m ≤ 10 000, 1 ≤ L ≤ 109, 0 ≤ s, t ≤ n - 1, s ≠ t) — the number of vertices, number of edges, the desired length of shortest path, starting vertex and ending vertex respectively.
Then, m lines describing the edges of the graph follow. i-th of them contains three integers, ui, vi, wi (0 ≤ ui, vi ≤ n - 1, ui ≠ vi, 0 ≤ wi ≤ 109). ui and vi denote the endpoints of the edge and wi denotes its weight. If wi is equal to 0 then the weight of the corresponding edge was erased.
It is guaranteed that there is at most one edge between any pair of vertices.
Print "NO" (without quotes) in the only line if it's not possible to assign the weights in a required way.
Otherwise, print "YES" in the first line. Next m lines should contain the edges of the resulting graph, with weights assigned to edges which weights were erased. i-th of them should contain three integers ui, vi and wi, denoting an edge between vertices ui and vi of weight wi. The edges of the new graph must coincide with the ones in the graph from the input. The weights that were not erased must remain unchanged whereas the new weights can be any positive integer not exceeding 1018.
The order of the edges in the output doesn't matter. The length of the shortest path between s and t must be equal to L.
If there are multiple solutions, print any of them.
5 5 13 0 4
0 1 5
2 1 2
3 2 3
1 4 0
4 3 4
YES
0 1 5
2 1 2
3 2 3
1 4 8
4 3 4
2 1 123456789 0 1
0 1 0
YES
0 1 123456789
2 1 999999999 1 0
0 1 1000000000
NO
Here's how the graph in the first sample case looks like :
In the first sample case, there is only one missing edge weight. Placing the weight of 8 gives a shortest path from 0 to 4 of length 13.
In the second sample case, there is only a single edge. Clearly, the only way is to replace the missing weight with 123456789.
In the last sample case, there is no weights to assign but the length of the shortest path doesn't match the required value, so the answer is "NO".
正解:dijkstra
解题报告:
其实这道题也很简单,考场上没想出来真是我傻...
首先那些可以修改的边最小边权为1,则全部改为1,那么如果此时最短路小于等于L,那么至少是有解的。否则无解。
然后依次修改当前还差的边权,直到合法,反正CF机子快,丝毫不虚。
然而我WA了一个晚上,因为我的inf设大了,一加就炸,调了很久才发现...
//It is made by jump~
#include <iostream>
#include <cstdlib>
#include <cstring>
#include <cstdio>
#include <cmath>
#include <algorithm>
#include <ctime>
#include <vector>
#include <queue>
#include <map>
#include <set>
using namespace std;
typedef long long LL;
const int MAXN = ;
const int MAXM = ;
const int MOD = ;
int n,m,L,s,t,ecnt;
int first[MAXN],to[MAXM],next[MAXM],w[MAXM],dis[MAXN];
int tag[MAXM];//!!!!!
int dui[MAXN*],head,tail;
bool in[MAXN],ok; inline int getint()
{
int w=,q=; char c=getchar();
while((c<'' || c>'') && c!='-') c=getchar(); if(c=='-') q=,c=getchar();
while (c>='' && c<='') w=w*+c-'', c=getchar(); return q ? -w : w;
} inline void SPFA(){
//memset(dis,63,sizeof(dis));
for(int i=;i<=n;i++) dis[i]=;//不能设太大!!!不然会炸,WA了好久!!!
dis[s]=; head=tail=; dui[++tail]=s; memset(in,,sizeof(in)); in[s]=;
while(head!=tail) {
head%=n; head++; int u=dui[head]; in[u]=;
for(int i=first[u];i>;i=next[i]) {
int v=to[i];
if(dis[v]>dis[u]+w[i]) {
dis[v]=dis[u]+w[i];
if(!in[v]) { tail%=n; tail++; dui[tail]=v; in[v]=; }
}
}
}
} inline void work(){
//memset(first,-1,sizeof(first));
n=getint(); m=getint(); L=getint(); s=getint()+; t=getint()+;
int x,y,z;ecnt=;
for(int i=;i<=m;i++) {
x=getint()+; y=getint()+; z=getint();
next[++ecnt]=first[x]; first[x]=ecnt; to[ecnt]=y; w[ecnt]=z; if(z==) tag[ecnt]=,w[ecnt]=;
next[++ecnt]=first[y]; first[y]=ecnt; to[ecnt]=x; w[ecnt]=z; if(z==) tag[ecnt]=,w[ecnt]=;
}
SPFA(); if(dis[t]>L) { printf("NO"); return ; }
ok=false;
for(int i=;i<=n;i++) {
for(int j=first[i];j>;j=next[j]) {
if(j & )
if(tag[j]) {
if(dis[t]==L){ ok=true ; break;}
if(dis[t]<L) {
w[j]=w[j^]=L-dis[t]+;
SPFA();
}
}
}
if(ok) break;
}
if(!ok && dis[t]!=L) { printf("NO"); }
else{
printf("YES\n");
for(int i=;i<=n;i++)
for(int j=first[i];j>;j=next[j])
if(j&)
printf("%d %d %d\n",to[j]-,to[j^]-,w[j]);
}
} int main()
{
work();
return ;
}
codeforces 715B:Complete The Graph的更多相关文章
- Codeforces 715B & 716D Complete The Graph 【最短路】 (Codeforces Round #372 (Div. 2))
B. Complete The Graph time limit per test 4 seconds memory limit per test 256 megabytes input standa ...
- 【Codeforces】716D Complete The Graph
D. Complete The Graph time limit per test: 4 seconds memory limit per test: 256 megabytes input: sta ...
- Codeforces 1009D:Relatively Prime Graph
D. Relatively Prime Graph time limit per test 2 seconds memory limit per test 256 megabytes input st ...
- 【codeforces 716D】Complete The Graph
[题目链接]:http://codeforces.com/problemset/problem/716/D [题意] 给你一张图; 这张图上有一些边的权值未知; 让你确定这些权值(改成一个正整数) 使 ...
- CodeForces 715B Complete The Graph 特殊的dijkstra
Complete The Graph 题解: 比较特殊的dij的题目. dis[x][y] 代表的是用了x条特殊边, y点的距离是多少. 然后我们通过dij更新dis数组. 然后在跑的时候,把特殊边都 ...
- CF715B. Complete The Graph
CF715B. Complete The Graph 题意: 给一张 n 个点,m 条边的无向图,要求设定一些边的边权 使得所有边权都是正整数,最终 S 到 T 的最短路为 L 1 ≤ n ≤ 100 ...
- 算法:图(Graph)的遍历、最小生成树和拓扑排序
背景 不同的数据结构有不同的用途,像:数组.链表.队列.栈多数是用来做为基本的工具使用,二叉树多用来作为已排序元素列表的存储,B 树用在存储中,本文介绍的 Graph 多数是为了解决现实问题(说到底, ...
- 译:Local Spectral Graph Convolution for Point Set Feature Learning-用于点集特征学习的局部谱图卷积
标题:Local Spectral Graph Convolution for Point Set Feature Learning 作者:Chu Wang, Babak Samari, Kaleem ...
- Codeforces 715B. Complete The Graph 最短路,Dijkstra,构造
原文链接https://www.cnblogs.com/zhouzhendong/p/CF715B.html 题解 接下来说的“边”都指代“边权未知的边”. 将所有边都设为 L+1,如果dis(S,T ...
随机推荐
- JMeter学习(六)集合点
JMeter也有像LR中的集合点,本篇就来介绍下JMeter的集合点如何去实现. JMeter里面的集合点通过添加定时器来完成. 注意:集合点的位置一定要在Sample之前. 集合点:简单来理解一下, ...
- 经典71道Android试题及答案
本文为开发者奉献了70道经典Android面试题加答案--重要知识点几乎都涉及到了,你还等啥,赶紧收藏吧!! 1. 下列哪些语句关于内存回收的说明是正确的? (b) A. 程序员必须创建一个线程来释放 ...
- 一篇文章告诉你为何GitHub估值能达20亿美元
软件开发平台GitHub今日宣布,已获得硅谷多家知名风投2.5亿美元融资,这也让其融资总额达到了3.5亿美元,此轮融资对GitHub的估值约为20亿美元. GitHub有何特别之处? GitHub创立 ...
- WP老杨解迷:如何获得更多的应用评价和解读内容刷新
Windows Phone的市场评论功能研究的时间比较长,只是这一功能,估计就能写一篇论文,我曾搞过多款评论数超高的游戏,其中<少林塔防>是重量级的作品,至今稳坐最高评分第一把交椅,如果不 ...
- [转]Windows网络编程学习-面向连接的编程方式
直接附上原文链接:windows 网络编程学习-面向连接的编程方式
- [转]World Wind学习总结一
WW的纹理,DEM数据,及LOD模型 以earth为例 1. 地形数据: 默认浏览器纹理数据存放在/Cache/Earth/Images/NASA Landsat Imagery/NLT Landsa ...
- Oracle的if else if
前段时间写Oracle存储过程就遇到问题.原来写成这样if 1=2 then null;elseif 1=3 then nullend if;在PL/SQL编辑环境下elseif没有变色,说明不是 ...
- 图像相似度算法的C#实现及测评
近日逛博客的时候偶然发现了一个有关图片相似度的Python算法实现.想着很有意思便搬到C#上来了,给大家看看. 闲言碎语 才疏学浅,只把计算图像相似度的一个基本算法的基本实现方式给罗列了出来,以至于在 ...
- Matlab中的fread函数
Matlab中fread函数用法 "fread"以二进制形式,从文件读出数据. 语法1:[a,count]=fread(fid,size,precision) 语法2:[a, ...
- Orchard常见问题
本文链接:http://www.cnblogs.com/souther/p/4543299.html 什么是Orchard Orchard是一个免费,开源,注重社区的项目,其目标是提供ASP.NET平 ...