迷宫问题的C语言求解
1 .Preface
/**
* There have been many data to introduce the algorithm. So I will try to simply explain it and explain the program in detail.
*/
/**
* Prerequisites:
* 1). a survival skill in CPP programing language.
* 2). a curious mind for maze problem.
*
*/
//Image this, you have maze, just as following:
/*
*
* 1 1 1 1 1 1 1 1 1 1
*-> 1 0 0 1 0 0 0 1 0 1
* 1 0 0 1 0 0 0 1 0 1
* 1 0 0 0 0 1 1 0 0 1
* 1 0 1 1 1 0 0 0 0 1
* 1 0 0 0 1 0 0 0 0 1
* 1 0 1 0 0 0 1 0 0 1
* 1 0 1 1 1 0 1 1 0 1
* 1 1 0 0 0 0 0 0 0 1 -->exit
* 1 1 1 1 1 1 1 1 1 1
*/
//the entrance is [1,1], and exit is [8,8].
//How could you find a valid way to get through this?
/*
* when start with [1, 1] , we will arrive some special position, which provide us many paths .For example, if we reside in [1, 1], there have two paths for us, [ 2, 1] and [ 1, 2]. Those speical nodes connect each other and compose a complexity topology network.
*
*
* To solve this problem, we could do as following:
* step 1: just go ahead as one pleases , but record all nodes which has arrived.
* step 2: if arrive a dead end, that mean you was choose a wrong path. So you need to go back the same way, find the last node which you did a choice, and step into another choice.
* step 3: Then repeat step 1 untill arrive the exit.
*
*/
2.Source code
/**
* Now, check this source code. I divided this problem with three parts: a stack, a map, and a boy who provide solve solution. First, let us put that poor little boy into this maze.( brutally )
*/
2. 1 Tool--stack.h
/**
* To traverse the network totally, the little boy must ensure he can go back the same way. So he save all of nodes,which he has been arrived,into a stack. The feature of stack is
* first in, first out. That's what we lack.
*/
/**
* For make this stack more useful, a class template is created.
*/
#ifndef STACK_H
#define STACK_H typedef int INDEX; template < class ELEM>
class STACK{
public:
STACK( int capacity); //capacity
~STACK( void);
/*
* some basic operation function.
*/
bool pop( ELEM &item);
bool push( ELEM item);
/*
* sometimes, we want to visit those elements in the stack simply, instead of
* pop them from stack.
*/
bool reset_v( void); //reset view point
bool pop_v( ELEM &item); //view stack private:
/*
* the bottom of stack
*/
ELEM *base;
/*
* the top of stack.
*/
INDEX top;
/*
* the current position for pop
*/
INDEX cur;
/*
* the current postion for visit stack
*/
INDEX v_cur;
}; #define STACK_MAX 1000 /**
* This is just a simply stack, and even don't consider dynamic extension.
*/
template <class ELEM>
STACK<ELEM>::STACK(int capacity)
{
if( ( capacity<=0)
||( capacity>STACK_MAX))
{
this->base = NULL;
return ;
}
this->base = NULL;
this->base = new ELEM[capacity];
this->top = capacity -1;
this->cur = -1;
this->v_cur = -1;
} template <class ELEM>
STACK<ELEM>::~STACK( void)
{
if( NULL!=this->base)
{
delete [](this->base);
this->base = NULL;
}
} template <class ELEM>
bool STACK<ELEM>::push( ELEM item)
{
if( (NULL==this->base)
||(this->top==this->cur))
return false; this->cur ++;
this->base[this->cur] = item;
return true;
} template <class ELEM>
bool STACK<ELEM>::pop( ELEM &item)
{
if( (NULL==this->base)
||(this->cur<0))
return false; item = this->base[this->cur];
this->cur--; return true;
} /**
* This function is used to visit stack.
*/
template <class ELEM>
bool STACK<ELEM>::pop_v( ELEM &item)
{
if( (NULL==this->base)
||(this->v_cur<0))
return false; item = this->base[this->v_cur];
this->v_cur--; return true;
} /**
* reset the posion of visit at current postion of stack.
* That is necessary before use pop_v().
*/
template <class ELEM>
bool STACK<ELEM>::reset_v( void)
{
this->v_cur = this->cur;
return true;
} #endif
2.2 Tool--map.h
/**
* Obviously, a map is necessary. By the help of the map, the boy could concentrate on hisself's work rather than be busy with some things about map. That make the code is more clear and simple.
*/
/**
* Map is a 2D matrix. For a element in the matrix, it compose by three parts:
* X coordinate, Y coordinate and additional data in which we could save some
* attribute information about this node.
*/
#ifndef MAP_H
#define MAP_H #define MAP_MAX 15 typedef int COORDINATE; template <class NODE>
class MAP{
public:
MAP( int width);
~MAP( void);
/*
* get a node which reside in [x,y]. The information of node
* will be write into @nod.
*/
bool cur( COORDINATE x, COORDINATE y , NODE &nod);
/*
* get a node reside in [ x-1, y].
*/
bool left( COORDINATE x, COORDINATE y , NODE &nod);
bool right( COORDINATE x, COORDINATE y, NODE &nod);
bool up( COORDINATE x, COORDINATE y, NODE &nod);
bool down( COORDINATE x, COORDINATE y, NODE &nod);
/*
* set a map node
*/
bool set( COORDINATE x, COORDINATE y, NODE &nod); private:
/*
* point to the map
*/
NODE **p;
int wid;
}; /**
* Init the size of map. Because we don't know any thing about the size,
* nither width, nor height. So we use a trick.
*/
template <class NODE>
MAP<NODE>::MAP( int width)
{
if( width>MAP_MAX)
{
this->p = NULL;
return;
}
this->p = NULL;
#if 0
this->p = ( NODE **)malloc( sizeof(NODE)*width*width);
#else
this->p = new NODE*[width];
for( int i=0; i<width; i++)
this->p[i] = new NODE[width];
#endif
this->wid = width;
} template <class NODE>
MAP<NODE>::~MAP( void)
{
if( NULL!=this->p)
{
#if 0
free (this->p);
#else
for( int i=0; i<this->wid; i++)
delete []this->p[i];
delete []this->p;
#endif
this->p = NULL;
this->wid = 0;
}
} template <class NODE>
bool MAP<NODE>::cur(COORDINATE x, COORDINATE y, NODE &nod)
{
if( (x<0||x>=this->wid)
||(y<0||y>=this->wid))
return false; nod = this->p[x][y];
return true;
} template <class NODE >
bool MAP<NODE >::left( COORDINATE x, COORDINATE y, NODE &nod)
{
if( (x<0||x>=this->wid)
||(y<=0||y>=this->wid))
return false; nod = this->p[ x][y-1];
return true;
} template <class NODE>
bool MAP<NODE>::right( COORDINATE x, COORDINATE y, NODE &nod)
{
if( (x<0||x>=this->wid)
||(y<0||y>=this->wid-1))
return false; nod = this->p[x][y+1];
return true;
} template <class NODE>
bool MAP<NODE>::up( COORDINATE x, COORDINATE y, NODE &nod)
{
if( (x<=0||x>=this->wid)
||(y<0||y>=this->wid))
return false; nod = this->p[x-1][y];
return true;
} template <class NODE>
bool MAP<NODE>::down( COORDINATE x, COORDINATE y, NODE &nod)
{
if( (x<0||x>=this->wid-1)
||(y<0||y>=this->wid))
return false; nod = this->p[x+1][y];
return true;
} template <class NODE>
bool MAP<NODE>::set( COORDINATE x, COORDINATE y, NODE &nod)
{
if( (x<0||x>=this->wid)
||(y<0||y>=this->wid))
return false; if( NULL==this->p)
{
return false;
} this->p[x][y] = nod;
return true;
} #endif
2.3 Operator--boy
#include <stdio.h>
#include <iostream> /**
* To traverse the network totally, the little boy must ensure he can go back the same way.
* So he save all of nodes,which he has been arrived,into a stack. The feature of stack is
* first in, first out. That's what we lack.
*/
#include "../stack.h" /**
* Obviously, a map is necessary. By the help of the map, the boy could
* concentrate on hisself's work rather than be busy with some things about map.
* That make the code is more clear and simple.
*/
#include "map.h" //map node
typedef unsigned char UINT8; #define MAP_WID 10
#define END_X (MAP_WID-2) //8
#define END_Y (MAP_WID-2) //8 #define STACK_DEPTH 200 enum ORIEN{
O_RIGHT,
O_DOWN,
O_LEFT,
O_UP,
O_INVIALID,
O_MAX,
}; enum TERRAIN {
T_NOR = 0,
T_BLOCK = 1,
T_INVALID = 2,
}; /**
* map information, 1 meaning for T_BLOCK. 0 meaning for T_NOR.
*/
static int map_v[MAP_WID][MAP_WID] = {
1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 0, 0, 1, 0, 0, 0, 1, 0, 1,
1, 0, 0, 1, 0, 0, 0, 1, 0, 1,
1, 0, 0, 0, 0, 1, 1, 0, 0, 1,
1, 0, 1, 1, 1, 0, 0, 0, 0, 1,
1, 0, 0, 0, 1, 0, 0, 0, 0, 1,
1, 0, 1, 0, 0, 0, 1, 0, 0, 1,
1, 0, 1, 1, 1, 0, 1, 1, 0, 1,
1, 1, 0, 0, 0, 0, 0, 0, 0, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
}; /*
* basic element in map was used to record all information about
* a node.
*/
typedef struct {
COORDINATE x;
COORDINATE y;
TERRAIN val;
ORIEN orien;
bool hasComing;
} ELEMENT; /**
* And this is that poor boy. He will provide a algorithm to solve
* this maze problem. Of course, he need some tools: stack and
* map.
*/
class BOY {
public:
BOY( void);
~BOY( void);
/*
* begin to traverse this maze.
*/
bool work( void);
/*
* show information
*/
bool ShowMap( void);
bool ShowStack( void); private:
/*
* two important functions, they compose the core of this algorithm.
*/
bool forward( void);
bool backward( void);
/*
* work for those function above.
*/
bool hasNBranch( ELEMENT &c_elem, ELEMENT &nod, ORIEN &from); //has a new branch
bool isEnd( void); //arrive the end
bool findVisible( ELEMENT &c_elem, ELEMENT &nod, ORIEN& from);
bool wasCome( ELEMENT &c_elem); /*
* tool 1
*/
MAP<ELEMENT> map;
/*
* tool 2
*/
STACK<ELEMENT> stack;
/*
* current position
*/
ELEMENT cur;
}; BOY::BOY(void):map(MAP_WID), stack(STACK_DEPTH)
{
cur.x = 1;
cur.y = 1;
cur.val = T_NOR;
cur.orien = O_INVIALID; ELEMENT tmp_elem;
COORDINATE i,j;
for( i=0; i<MAP_WID; i++)
for( j=0; j<MAP_WID; j++)
{
tmp_elem.x = i;
tmp_elem.y = j;
tmp_elem.orien = O_INVIALID;
tmp_elem.hasComing = false;
tmp_elem.val = ( TERRAIN)map_v[i][j];
this->map.set( tmp_elem.x, tmp_elem.y, tmp_elem);
}
} BOY::~BOY( void)
{} bool BOY::work(void)
{
bool isContinue = true;
while( isContinue)
{
printf("this->cur[ %d, %d]\n", this->cur.x, this->cur.y);
/*
* go ahead until encounter a dead end or arrive the exit.
*/
while( this->forward( ))
{
printf("this->cur[ %d, %d]\n", this->cur.x, this->cur.y);
if( this->isEnd( ) )
return true;
}
/*
* when the boy has encounter a dead end, he need to backtrack.
* find a valid path.
*/
printf("back>\n");
isContinue=this->backward( );
} return false;
} /**
* based on current position, try to forward a step. If success, the previous postion
* will be push in the stack. and update the information of past node as arrived.
* if fail, that meaning current node is a dead end.
*/
bool BOY::forward(void)
{
ELEMENT tmp_elem;
ORIEN tmp_from = O_RIGHT; if( !this->hasNBranch( this->cur, tmp_elem, tmp_from))
{
return false;
}
this->cur.orien = tmp_from;
this->stack.push( this->cur); this->cur = tmp_elem;
this->cur.hasComing = true;
this->map.set( this->cur.x, this->cur.y, this->cur); return true;
} /**
* one of the core function. when the boy arrived a dead end,
* this function will be call . It go back the same way untill find
* a valid node that could give the little boy a new path(or a branch).
*/
bool BOY::backward(void)
{
ELEMENT tmp_elem;
ORIEN tmp_from = O_RIGHT;
while( this->stack.pop( tmp_elem))
{
this->cur = tmp_elem;
if( this->hasNBranch( this->cur, tmp_elem, tmp_from))
{
return true;
}
} return false;
} /**
* check whether @c_elem node has a valid path that deserve to visit.
* as same as other function, all information will be write into @elem
* and @from.
*/
bool BOY::hasNBranch( ELEMENT &c_elem, ELEMENT &elem, ORIEN &from)
{
ELEMENT tmp_elem;
ORIEN tmp_from = from;
while(1)
{
//find next visible position.
//Y:continue
//N:this node is a ending
if( !this->findVisible( c_elem, tmp_elem, tmp_from))
{
return false;
}
//was coming? //Y:coninue
//N:right way
if( !this->wasCome( tmp_elem))
{//this is a new branch
break;
}
tmp_from =(ORIEN)( tmp_from + 1); //next orientation
} elem = tmp_elem;
from = tmp_from;
return true;
} /**
* arrive the exit of maze ? */
bool BOY::isEnd(void)
{
if( (this->cur.x==END_X)
&&(this->cur.y ==END_Y))
{
return true;
} return false;
} /**
* find a visible path that is not block. It use @c_elem as the current view point,
* if success , write information into @elem and @from.
*/
bool BOY::findVisible( ELEMENT &c_elem, ELEMENT &elem, ORIEN& from)
{
ELEMENT tmp;
/*
* check valid path clockwise.
*/
switch( from)
{
case O_RIGHT:
if( (this->map.right( c_elem.x, c_elem.y, tmp))
&&( tmp.val == T_NOR))
{
elem = tmp;
from = O_RIGHT;
return true;
}
case O_DOWN:
if( (this->map.down( c_elem.x, c_elem.y, tmp))
&&( tmp.val == T_NOR))
{
elem = tmp;
from = O_DOWN;
return true;
}
case O_LEFT:
if( (this->map.left( c_elem.x, c_elem.y, tmp))
&&( tmp.val == T_NOR))
{
elem = tmp;
from = O_LEFT;
return true;
}
case O_UP:
if( (this->map.up( c_elem.x, c_elem.y, tmp))
&&( tmp.val == T_NOR))
{
elem = tmp;
from = O_UP;
return true;
}
default :;
} return false;
} /**
* Though this function is very tiny, it hold a important position
* in the totally algorithm. The upper function will call this to ensure
* whether a node is deserve to visit. By add a series of strategies
* we could improve the algorithm.
*/
bool BOY::wasCome( ELEMENT &c_elem)
{
#if 1
//label
return c_elem.hasComing;
#else
//标准1
if( c_elem.hasComing)
return true; //标准2
ELEMENT elem;
this->stack.reset_v();
while( this->stack.pop_v( elem))
{
if( (c_elem.x==elem.x)
&&(c_elem.y==elem.y))
return true;
} return false; //wasn't coming #endif
} /**
* show the status of map
*/
bool BOY::ShowMap(void)
{
printf("-----------MAP---------------------\n");
COORDINATE i,j;
for( i=0; i<MAP_WID; i++)
{
for( j=0; j<MAP_WID; j++)
{
ELEMENT tmp;
if(!this->map.cur( i, j, tmp))
{
printf("error: [ %d, %d]\n", i, j);
return false;
}
printf("%3d", tmp.val);
}
printf("\n");
} return true;
} /**
* show the status of stack, just visit it and don't pop element from it
*/
bool BOY::ShowStack(void)
{
printf("-----------STACK---------------------\n"); ELEMENT tmp_elem; this->stack.reset_v( );
while( this->stack.pop_v( tmp_elem))
{
printf("[ %d, %d]\n", tmp_elem.x, tmp_elem.y);
}
} int main()
{
BOY boy;
boy.ShowMap( );
boy.work( );
boy.ShowStack( );
return 0;
}
迷宫问题的C语言求解的更多相关文章
- 应用栈解决迷宫问题的C语言实现
题目来自于严蔚敏<数据结构>,参考伪代码实现的程序: #include <stdio.h> #include <malloc.h> //记录通道块在迷宫矩阵当中的横 ...
- Leetcode 20题 有效的括号(Valid Parentheses) Java语言求解
题目描述: 给定一个只包括 '(',')','{','}','[',']' 的字符串,判断字符串是否有效. 有效字符串需满足: 左括号必须用相同类型的右括号闭合. 左括号必须以正确的顺序闭合. 注意空 ...
- s=1+2*3+4*5*6+7*8*9*10+.... C语言求解
#include <stdio.h> /*类似斐波那契数列的计算方式 项 1 2 3 4 1 2*3 4*5*6 7*8*9*10 生成项的起始数字 1 2 4 7 和后一项的差值 1 2 ...
- Leetcode 239题 滑动窗口最大值(Sliding Window Maximum) Java语言求解
题目链接 https://leetcode-cn.com/problems/sliding-window-maximum/ 题目内容 给定一个数组 nums,有一个大小为 k 的滑动窗口从数组的最左侧 ...
- Leetcode 703题数据流中的第K大元素(Kth Largest Element in a Stream)Java语言求解
题目链接 https://leetcode-cn.com/problems/kth-largest-element-in-a-stream/ 题目内容 设计一个找到数据流中第K大元素的类(class) ...
- LeetCode 225题用队列实现栈(Implement Stack using Queues) Java语言求解
链接 https://leetcode-cn.com/problems/implement-stack-using-queues/ 思路 首先演示push()操作:将元素依次进入队1,进入时用top元 ...
- LeetCode 232题用栈实现队列(Implement Queue using Stacks) Java语言求解
题目链接 https://leetcode-cn.com/problems/implement-queue-using-stacks/ 题目描述 使用栈实现队列的下列操作: push(x) -- 将一 ...
- Leetcode 206题 反转链表(Reverse Linked List)Java语言求解
题目描述: 反转一个单链表. 示例: 输入: 1->2->3->4->5->NULL 输出: 5->4->3->2->1->NULL 迭代解 ...
- Leetcode 142题 环形链表 II(Linked List Cycle II) Java语言求解
题目描述: 给定一个链表,返回链表开始入环的第一个节点. 如果链表无环,则返回 null. 为了表示给定链表中的环,我们使用整数 pos 来表示链表尾连接到链表中的位置(索引从 0 开始). 如果 p ...
随机推荐
- SpringMVC高级参数绑定(数组和List)
本节内容: 参数绑定之数组 将表单的数据绑定到List 复制下上篇博客中的工程,作为今天开发的工程. 一.参数绑定之数组 1. 需求 在商品列表页面选中多个商品,然后删除. 2. 需求分析 功能要求商 ...
- Mybatis输入映射和输出映射
本节内容: 输入参数映射 输出映射 resultMap Mapper.xml映射文件中定义了操作数据库的sql,每个sql是一个statement,映射文件是mybatis的核心. 一.环境准备 复制 ...
- 【转】Git命令大全(非常齐全)
$ git init // 初始化一个Git仓库$ git status // 查看仓库的状态$ git add . // 将所有修改添加到暂存区$ git add * // Ant风格添 ...
- 启动DELPHI2010出现 EditorLineEnds.ttr 错误的解决方法
在网上找到了很多方法.其实解决这个问题的方法, 最简单的就是把 EditorLineEnds.ttr 改名为 EditorLineEnds.ttf 然后,安装它, 安装完成后就OK了.
- MVC中使用Web API和EntityFramework
在ASP.NET MVC中使用Web API和EntityFramework构建应用程序 最近做了一个项目技术预研:在ASP.NET MVC框架中使用Web API和EntityFramework ...
- 【LOJ】#2269. 「SDOI2017」切树游戏
题解 把所有的数组一开始就FWT好然后再IFWT回去可以减小常数 从13s跑到0.7s-- 可以参照immortalCO的论文,感受一下毒瘤的动态动态DP 就是用数据结构维护线性递推的矩阵的乘积 由于 ...
- 【LOJ】#2128. 「HAOI2015」数字串拆分
题解 题中给的函数可以用矩阵快速幂递推 我们记一个数组dp[i](这个数组每个元素是一个矩阵)表示从1到i所有的数字经过拆分矩阵递推的加和 转移方法是 \(dp[i] = \sum_{j = 0}^{ ...
- ajax jqplot ssh实现图表的持续更新
实现功能: 数据库有新数据图表会更新到 数据库查询使用ssh框架中的hibernate 想法: 画图表的ajaxautoruncopy.jsp利用ajax收到7-13.jsp传过来的数据 7-13.j ...
- 把.html转换成.jsp中jqplot画图表不能正常显示,出错的心得
在做这个的时候,明明html中是完全可行的,如下图: 但后缀名改成.jsp后竟出现如下情况: 这太坑爹了吧,我的图呢! 哎,又要自己找代码问题了,无奈! 先给出我还没修改前的代码吧,关于里面的.js, ...
- 5,EasyNetQ-Send Receive
而发布/订阅和请求/响应模式是位置透明的,因为您不需要指定消息的消费者所在的位置,发送/接收模式专门用于通过命名队列进行通信. 它也不会对可以通过队列发送的消息的类型做任何假设. 这意味着您可以通过同 ...