题意:给你一个圆形和很多个矩形,然后要你求圆形的圆周有多少被矩形覆盖。

思路:比赛的时候是有思路的了,不过一直在调别的题,最后剩下30分钟肯定来不及敲。想法是这样的,要是我们可以求出每个矩形覆盖了圆周的哪些区间,我们最后就对这些区间排序然后求区间和就好了,但是问题是怎么知道哪些区间是要的,哪些区间是不要的呢? 首先我们对矩形的四条线段和矩形求交,把所有的交点求出来,然后将交点用atan2转化成极角(我用的区间是[0,2pi]),实际上直接用极角肯定也没问题吧),然后排序,排序之后我们会发现我们要的区间一定是相邻的两个角度的点对应的区间,但是有可能这些区间是不是我们要的,判断的条件就是两个角度的对应的弧段的中点如果在矩形里面的话我们就将这个区间保留。

如此一来就将所有的区间弄了出来,接下来就是求线段的并的总长度了,for一下就好。

但是后来WA了一发,原因是如果圆完全被矩形包含的话,我们解不出交点,会被默认答案是0,所以上面的方法只有在有交点的时候成立,所以判的时候加多一个判是否被完全包含,被完全包含最后直接输出2*pi*r就好,代码写的略长= =

#pragma warning(disable:4996)
#include<iostream>
#include<cstring>
#include<cstdio>
#include<algorithm>
#include<vector>
#include<cmath>
#include<string>
#define maxn 150
#define eps 1e-7
using namespace std; const double pi = acos(-1.0); struct Point
{
double x, y;
Point(double xi, double yi) :x(xi), y(yi){}
Point(){}
Point operator + (const Point &b){
return Point(x + b.x, y + b.y);
}
Point operator - (const Point &b){
return Point(x - b.x, y - b.y);
}
double ang(){
double res=atan2(y,x);
if (res < 0) return res + 2*pi;
else return res;
}
Point rot(double ag){
return Point(x*cos(ag) - y*sin(ag), x*sin(ag) + y*cos(ag));
}
}; struct Seg
{
Point st, ed;
Seg(Point sti, Point edi) :st(sti), ed(edi){}
Seg(){}
}; struct Inter
{
double l, r;
Inter(double li, double ri) :l(li), r(ri){}
Inter(){}
bool operator < (const Inter &b) const{
if (l == b.l) return r < b.r;
else return l < b.l;
}
}; int dcmp(double x){
return (x>eps) - (x < -eps);
} double ctr;
int n; vector<double> getInterSection(Seg seg){
vector<double> res;
if (seg.st.x == seg.ed.x){
double xx = seg.st.x;
double y1 = seg.st.y, y2 = seg.ed.y;
if (dcmp(abs(xx) - ctr) > 0) return res;
else if (dcmp(abs(xx) - ctr) == 0){
if (0 >= y1 && 0 <= y2){
res.push_back(Point(xx, 0).ang());
}
}
else{
double dy = sqrt(ctr*ctr - xx*xx);
if (dy >= y1&&dy <= y2){
res.push_back(Point(xx, dy).ang());
}
dy = -dy;
if (dy >= y1&&dy <= y2){
res.push_back(Point(xx, dy).ang());
}
}
}
else{
double yy = seg.st.y;
double x1 = seg.st.x, x2 = seg.ed.x;
if (dcmp(abs(yy) - ctr) > 0) return res;
else if (dcmp(abs(yy) - ctr) == 0){
if (0 >= x1 && 0 <= x2){
res.push_back(Point(0, yy).ang());
}
}
else{
double dx = sqrt(ctr*ctr - yy*yy);
if (dx >= x1&&dx <= x2){
res.push_back(Point(dx, yy).ang());
}
dx = -dx;
if (dx >= x1&&dx <= x2){
res.push_back(Point(dx, yy).ang());
}
}
}
return res;
} bool withinRect(double xx,double yy, double li, double lj, double ri, double rj){
return xx >= li&&xx <= ri&&yy >= lj&&yy <= rj;
} int main()
{
while (cin >> ctr >> n){
double li, lj, ri, rj;
Seg seg;
vector<double> tot;
vector<double> tmp;
vector<Inter> inter; bool flag = false;
for (int i = 0; i < n; i++){
scanf("%lf%lf%lf%lf", &li, &lj, &ri, &rj);
if (flag) continue;
if (li <= -ctr && ctr <= ri && lj <= -ctr&&ctr <= rj){
flag = true; continue;
}
tot.clear();
seg.st = Point(li, lj); seg.ed = Point(li, rj);
tmp = getInterSection(seg);
tot.insert(tot.end(), tmp.begin(), tmp.end()); seg.st = Point(ri, lj); seg.ed = Point(ri, rj);
tmp = getInterSection(seg);
tot.insert(tot.end(), tmp.begin(), tmp.end()); seg.st = Point(li, lj); seg.ed = Point(ri, lj);
tmp = getInterSection(seg);
tot.insert(tot.end(), tmp.begin(), tmp.end()); seg.st = Point(li, rj); seg.ed = Point(ri, rj);
tmp = getInterSection(seg);
tot.insert(tot.end(), tmp.begin(), tmp.end()); sort(tot.begin(), tot.end());
int siz = tot.size();
for (int k = 0; k < siz; k++){
if (dcmp(tot[k] - tot[(k + 1) % siz]) == 0) continue;
Point mid;
if (dcmp(tot[(k + 1) % siz] - tot[k])>0) mid = Point(ctr, 0).rot((tot[k] + tot[(k + 1) % siz]) / 2);
else mid = Point(ctr, 0).rot((tot[k] + tot[(k + 1) % siz]-2*pi) / 2);
if (withinRect(mid.x, mid.y, li, lj, ri, rj)){
if (dcmp(tot[(k + 1) % siz] - tot[k]) >= 0){
inter.push_back(Inter(tot[k], tot[(k + 1) % siz]));
}
else{
inter.push_back(Inter(0, tot[(k+1)%siz]));
inter.push_back(Inter(tot[k % siz], 2 * pi));
}
}
}
}
sort(inter.begin(), inter.end());
double ans = 0;
if (flag) {
printf("%.3lf\n", 2 * pi*ctr); continue;
}
if (inter.size() == 0) {
printf("%.3lf\n", ans); continue;
}
double lt = inter[0].l, rt = inter[0].r;
for (int i = 1; i < inter.size(); i++){
if (inter[i].l <= rt){
rt = max(rt, inter[i].r);
}
else{
ans += rt - lt;
lt = inter[i].l, rt = inter[i].r;
}
}
ans += rt - lt;
printf("%.3lf\n", ans*ctr);
}
return 0;
}

ZOJ3238 Water Ring(计算几何)的更多相关文章

  1. 【计算几何】Water Testing

    Water Testing 题目描述 You just bought a large piece of agricultural land, but you noticed that – accord ...

  2. 一点公益商城开发系统模式Ring Buffer+

    一个队列如果只生产不消费肯定不行的,那么如何及时消费Ring Buffer的数据呢?简单的方案就是当Ring Buffer"写满"的时候一次性将数据"消费"掉. ...

  3. OpenCASCADE Ring Type Spring Modeling

    OpenCASCADE Ring Type Spring Modeling eryar@163.com Abstract. The general method to directly create ...

  4. 使用Ring Buffer构建高性能的文件写入程序

    最近常收到SOD框架的朋友报告的SOD的SQL日志功能报错:文件句柄丢失.经过分析得知,这些朋友使用SOD框架开发了访问量比较大的系统,由于忘记关闭SQL日志功能所以出现了很高频率的日志写入操作,从而 ...

  5. [AlwaysOn Availability Groups]AlwaysOn Ring Buffers

    AlwaysOn Ring Buffers 一些AlwaysOn的诊断信息可以从SQL Server ring buffers.或者从sys.dm_os_ring_buffers.ring buffe ...

  6. [LeetCode] Pacific Atlantic Water Flow 太平洋大西洋水流

    Given an m x n matrix of non-negative integers representing the height of each unit cell in a contin ...

  7. [LeetCode] Trapping Rain Water II 收集雨水之二

    Given an m x n matrix of positive integers representing the height of each unit cell in a 2D elevati ...

  8. [LeetCode] Water and Jug Problem 水罐问题

    You are given two jugs with capacities x and y litres. There is an infinite amount of water supply a ...

  9. [LeetCode] Trapping Rain Water 收集雨水

    Given n non-negative integers representing an elevation map where the width of each bar is 1, comput ...

随机推荐

  1. 6.24 AppCan移动开发者大会,我爱我家即将闪亮登场!

    6.24 AppCan移动开发者大会进入倒计时,报名通道即将关闭! “6月24日, 2016AppCan移动开发者大会即将召开,以“平台之上,应用无限”为主题,1500位行业精英汇聚在此,重磅新品发布 ...

  2. hdu 2544 最短路

    题目连接 http://acm.hdu.edu.cn/showproblem.php?pid=2544 最短路 Description 在每年的校赛里,所有进入决赛的同学都会获得一件很漂亮的t-shi ...

  3. Repair the database using DBCC CHECKDB

    So now if you want to place AdventureWorks2008R2 sample database in a single-user mode, then write t ...

  4. Azure SQL 数据库与新的数据库吞吐量单位DTU

    azure中新的数据库吞吐量单位 (Database Throughput Unit, DTU) 是什么,以及用户如何通过它来了解新服务级别可以提供的服务内容.DTU 对于提供预测性更强的性能体验起着 ...

  5. Golang的iota的特性

    Golang的iota的特性: 1. iota在每个ConstBlock中自动归0. 2. iota在每个ConstSpec后自动增1. 换言之: iota是ConstBlock中ConstSpec的 ...

  6. openssl AES加密以及padding

    好习惯,先上代码再说事 加密 void AesEncrypt(unsigned char* pchIn, int nInLen, unsigned char *ciphertext, int & ...

  7. Oracle之sql追踪

    select * from v$sqlarea t where t.sql_text like '%_070%' order by t.LAST_ACTIVE_TIME desc SELECT * F ...

  8. R语言实战读书笔记1—语言介绍

    第一章 语言介绍 1.1 典型的数据分析步骤 1.2 获取帮助 help.start() help("which") help.search("which") ...

  9. ios coreData使用

    ios中的coredata的使用(转) 分类: ios2013-07-15 18:12 27288人阅读 评论(1) 收藏 举报 Core Data数据持久化是对SQLite的一个升级,它是ios集成 ...

  10. UISegmentedControl swift

    // // ViewController.swift // UILabelTest // // Created by mac on 15/6/23. // Copyright (c) 2015年 fa ...