poj 1002


研究了半天,结果果然是超时了。其实数据处理的时间应该不长,关键是读入数据的时间太长了。还是记下来留作念想吧。

#include "poj.h"


#include <iostream>
#include <string>
#include <vector>
#include <algorithm>
using namespace std;

char Codes[31] = {'0','0','A','D','G','J','M','P','T','W',
				  '0','0','B','E','H','K','N','R','U','X',
				  '0','0','C','F','I','L','O','S','V','Y'};

class Word
{
public:
	string w;
	int count;
	bool operator ==(const Word &value)
    {
		return value.w == this->w;
    }
	bool operator>(const Word& value)
	{
		return this->w > value.w;
	}
	bool operator<(const Word& value)
	{
		return this->w < value.w;
	}
};



string translate(string original)
{
	string result = "";
	for(int i = 0; i < original.length();i++)
	{
		char c = original[i];
		if(c == '-')continue;
		if(c >= '0' && c<= '9'){
			result += c;
		}
		else{
			//find the number c represents
			for(int j = 0; j < 30;j++)
			{
				if(c == Codes[j])
				{
					result += j%10 + '0';
					break;
				}
			}
		}
	}
	return result;
}


void p1002(){
	int i,n;
	cin>>n;
	vector<string> data;
	vector<Word> dumpnumber;
	for(i = 0; i < n; i++)
	{
		string a;
		cin>>a;
		string t = translate(a);
		if( find(data.begin(), data.end(), t) != data.end())
		{
			Word word;
			word.w = t;
			vector<Word>::iterator it = find(dumpnumber.begin(), dumpnumber.end(),word);
			if(it != dumpnumber.end())
			{
				it->count++;
			}
			else
			{
				word.count=2;
				dumpnumber.push_back(word);
			}
		}
		data.push_back(t);
	}
	
	sort(dumpnumber.begin(),dumpnumber.end());
	vector<Word>::iterator it = dumpnumber.begin();
	if(dumpnumber.empty())
	{
		cout<<"No duplicates."<<endl;
	}
	while(it != dumpnumber.end())
	{
		for(i = 0; i < 3; i ++)cout<<it->w[i];
		cout<<"-";
		for(;i<7;i++)cout<<it->w[i];
		cout<<" "<<it->count<<endl;
		it++;
	}
}

终于决定用暴力方法,写了个丑陋的办法,果然AC了,用时1200MS:

#include <iostream>
#include <string>
#include <sstream>
//#include "poj.h"

#using <System.dll>


using namespace System;
using namespace System::Diagnostics;
using namespace std;

char Codes[31] = {'0','0','A','D','G','J','M','P','T','W',
				  '0','0','B','E','H','K','N','R','U','X',
				  '0','0','C','F','I','L','O','S','V','Y'};

int translate(string original)
{
	int result = 0, index = 1000000;
	for(int i = 0; i < original.length();i++)
	{
		char c = original[i];
		if(c == '-')continue;
		if(c >= '0' && c<= '9'){
			result += index * (c - '0');
			index /= 10;
		}
		else{
			//find the number c represents
			for(int j = 0; j < 30;j++)
			{
				if(c == Codes[j])
				{
					result += index * (j%10);
					index /= 10;
					break;
				}
			}
		}
	}
	return result;
}

int count[10000000] = {0};

int main()
{
	
	int i,t,n,flag = -1;
	string a;
	stringstream ss;
	cin>>n;
	Stopwatch ^ stopWatch;
	
	stopWatch = Stopwatch::StartNew();
		

	
	for(i = 0; i < n; i++)
	{
		cin >> a;
		t = translate(a);
		count[t]++;
	}
	for(i = 0; i < 10000000; i++)
	{
		int c = count[i];
		if(c<2)continue;
		if(flag<0)flag = 1;
		ss.clear();
		ss<<i;
		ss>>a;
		while(a.length() < 7) a = '0' + a;
		cout<<a[0]<<a[1]<<a[2]<<"-"<<a[3]<<a[4]<<a[5]<<a[6]<<" "<<c<<endl;
	}
	if(flag<0)cout<<"No duplicates."<<endl;
	

	stopWatch->Stop();
	Int64 ticksThisTime = stopWatch->ElapsedMilliseconds;
	cout<<"time:"<<ticksThisTime<<endl;
	system("PAUSE");
	return 0;
}

同时学学别人家的代码:

先排序再找重复?还是找出重复再排序?这是个问题。

我试着换成这种输入方式,但并没有让我的代码更快,反而TLE了。不过多重IF的语句确实比for循环去char数组里找快一些貌似。

但是C语言天然优势使得这段代码只用了800MS左右。

#include <stdio.h>
#include <string.h>
#include <stdlib.h>
#include <ctype.h>

char pn[100000][8];

void main ()
{
  int i, j, n, c;
  int none = 1;

  scanf ("%d", &n);
  for (i = 0; i < n; i++)
    {
      j = 0;
      while (j < 7)
	{
	  if (!isalnum (c = getchar ())) continue;
	  if (c == 'A' || c == 'B' || c == 'C') c = '2';
	  if (c == 'D' || c == 'E' || c == 'F') c = '3';
	  if (c == 'G' || c == 'H' || c == 'I') c = '4';
	  if (c == 'J' || c == 'K' || c == 'L') c = '5';
	  if (c == 'M' || c == 'N' || c == 'O') c = '6';
	  if (c == 'P' || c == 'R' || c == 'S') c = '7';
	  if (c == 'T' || c == 'U' || c == 'V') c = '8';
	  if (c == 'W' || c == 'X' || c == 'Y') c = '9';
	  pn[i][j++] = c;
	}
      pn[i][7] = '\0';
    }

  qsort (pn, n, 8, (int (*) (const void *, const void *)) strcmp);

  j = 0;
  for (i = 1; i < n; i++)
  {
    if (strcmp (pn[i], pn[j]) != 0)
      {
		if (i - j > 1)
		  {
			printf (
			  "%c%c%c-%s %d\n", pn[j][0], pn[j][1], pn[j][2], pn[j] + 3, i - j
			);
			none = 0;
		  }
		j = i;
      }
  }
  if (i - j > 1)
    {
      printf (
	"%c%c%c-%s %d\n", pn[j][0], pn[j][1], pn[j][2], pn[j] + 3, i - j
      );
      none = 0;
    }
  if (none)
    printf ("No duplicates.\n");

  system("PAUSE");
}

这是另一段代码,思路也是先排序,然后一边找重复一边输出:(转自http://blog.chinaunix.net/uid-8059407-id-2034320.html)

#include <stdio.h>
#include <stdlib.h>
#include <string.h>

char map[26] = {'2', '2', '2', '3', '3', '3', '4', '4', '4', 
                '5', '5', '5', '6', '6', '6', '7', '7', '7', 
                '7', '8', '8', '8', '9', '9', '9', '9'};//按键映射表

char phone[100000][9];    //全部电话号码

int n;                    //电话号码的数量

char number[80];        //一个未经处理的电话号码


int main(void)
{
    int loop;                    //循环变量

    int phone_i=0, phone_j=0;    //phone数组的下标

    int tag = 0;                //重复标志,0没有处于一个重复,1有重复

    int count = 1;                //重复次数


    scanf("%d", &n);

    for(phone_i=0; phone_i<n; phone_i++){//格式化所有输入数据为111-1111的形式

        scanf("%s", number);
        for(loop=0; number[loop]!='\0'; loop++){//检查每一个字符

            if(phone_j == 3){//第四个字符必须是'-'

                phone[phone_i][phone_j] = '-';
                phone_j++;
            }
            if(number[loop]>='0' && number[loop]<='9'){//当前字符是数字

                phone[phone_i][phone_j] = number[loop];
            }
            else if(number[loop]>='A' && number[loop]<='Z'){//当前字符是大写字母

                phone[phone_i][phone_j] = map[number[loop]-'A'];
            }
            else if(number[loop] == '-'){//忽略'-'

                continue;
            }
            phone_j++;
        }
        phone_j = 0;
    }
    
    //排序全部电话号码

    qsort(phone, n, 9, strcmp);

    for(loop=1; loop<n; loop++){//检查重复次数,输出结果

        if(strcmp(phone[loop], phone[loop-1]) == 0){
            tag = 1;
            count++;
        }
        else{//没有重复

            if(count > 1){
                printf("%s %d\n", phone[loop-1], count);
            }
            count = 1;
        }
    }
    if(count > 1){//查看最后一条记录是否有重复

        printf("%s %d\n", phone[loop-1], count);
    }
    if(!tag) 
        printf("No duplicates.\n");
    return 0;
}

这个很有借鉴意义,根据输入来确定自定义数据指针的长度:(转自http://blog.csdn.net/hello_cosmos/article/details/8683599)

#include <stdio.h>
#include <stdlib.h>
#include <string.h>

typedef struct __data{
   int telNum;
   int repTim;
} da;

int main(){
	
	da *data;
	int numOfRecord,i;
	char temp[2000];  //数组开小了会WA。。。我了个擦。。。多么变态的测试数据啊。。第二第三次WA都是这个问题
	
	scanf("%d",&numOfRecord);
	
	data=(da *)calloc(numOfRecord,sizeof(da));
	
	for (i=0;i<numOfRecord;i++){
		scanf("%s",temp);
		stdlizeNum(temp,&(data[i]));
	}
	sortPrintf(data,numOfRecord);

  free(data);
  return 0;
}

int stdlizeNum(char *num,da *ans){
  int len=strlen(num);
  int i;
  int k=0;
  
  ans->repTim=0;
  ans->telNum=0;
  for (i=0;i<len;i++){
     switch(num[i]){
     case 'A':
     case 'B':
     case 'C':
     	 ans->telNum+=2;
     	 ans->telNum*=10;
     break;
     case 'D':
     case 'E':
     case 'F':
     	 ans->telNum+=3;
     	 ans->telNum*=10;
     break;
     case 'G':
     case 'H':
     case 'I':
     	 ans->telNum+=4;
     	 ans->telNum*=10;
     break;
     case 'J':
     case 'K':
     case 'L':
     	 ans->telNum+=5;
     	 ans->telNum*=10;
     break;
     case 'M':
     case 'N':
     case 'O':
     	 ans->telNum+=6;
     	 ans->telNum*=10;
     break;
     case 'P':
     case 'R':
     case 'S':
     	 ans->telNum+=7;
     	 ans->telNum*=10;
     break;
     case 'T':
     case 'U':
     case 'V':
     	 ans->telNum+=8;
     	 ans->telNum*=10;
     break;
     case 'W':
     case 'X':
     case 'Y':
     	 ans->telNum+=9;
     	 ans->telNum*=10;
     break;
     case '-':
     break;
     default:
     	 ans->telNum+=(int)(num[i]-'0');
     	 ans->telNum*=10;
     break;
     }
  }
  ans->telNum/=10;
}
int sortCmp_fun(const void *a,const void *b){
   
   return ((*(da *)a).telNum)-((*(da *)b).telNum);

}
int sortPrintf(da *input,int len){
	int i,offOn;
	
	qsort(input,len,sizeof(da),sortCmp_fun);
  
  for (i=0;i<len;i++){
     if (input[i].telNum==input[i-1].telNum){
        input[i].repTim=input[i-1].repTim+1;
        input[i-1].repTim=0;
     }
  }

  offOn=0;
  for (i=0;i<len;i++){
     if (input[i].repTim!=0){
     	  offOn=1;
        printf("%03d-%04d %d\n",input[i].telNum/10000,input[i].telNum%10000,input[i].repTim+1);  //"%d"这里不会出'0',而"%0d"会打印出来,第一次WA就是这里出错了
     }
  }
  if (offOn==0){
  	printf("No duplicates.\n");
  }

}


这里也有一些调试和借鉴的地方:(转自http://blog.csdn.net/wmmthu/article/details/7505850)

#include <stdio.h>
#include <string.h>
#include <stdlib.h>
#define UNCODE(a) (a>='A'?((a>'Q'?(a-'A'-1):(a-'A'))/3+2):a-'0')
//#define DEBUG

int uncode(char* s)
{
	int i;
	int result=0;
	for(i=0;i<strlen(s);i++)
	{
		if(s[i]=='-') continue;
		result=result*10+UNCODE(s[i]);
	}
	return result;
}

int cmp(const void* a,const void* b)
{
	return *(int*)a-*(int*)b;
}

int main()
{
	#ifdef DEBUG
		freopen("data.dat","r",stdin);
		//freopen("out.dat","w",stdout);
	#endif
	int* store;
	char s[50];
	int time;
	int n=0;
	int i;
	int start,value;
	int dup=0;
	scanf("%d",&time);
	store=(int*)malloc(sizeof(int)*(time));
	for(n=0;n<time;n++)
	{
		scanf("%s",s);
		store[n]=uncode(s);
	}
 	qsort(store,time,sizeof(store[0]),cmp);
	value=store[0];
	start=0;
	for(n=0;n<time;n++)
	{
		if(store[n]!=value){ 
			if((n-start)>1){	
				printf("%03d-%04d %d\n",store[start]/10000,store[start]%10000,n-start);
				dup=1;
			}
			start=n;
			value=store[n];
		}
	}
	if((n-start)>1){	
				printf("%03d-%04d %d\n",store[start]/10000,store[start]%10000,n-start);
				dup=1;
			}
	if(!dup)
	{
		printf("No duplicates.\n");
	}
	free(store);
return 1;
}

/* Copyright Derek Kisman (ACM ICPC ECNA 1999) */

char map[26] = {'2', '2', '2', '3', '3', '3', '4', '4', '4',
				'5', '5', '5', '6', '6', '6', '7', 0, '7', '7', '8', '8', '8', '9', '9', '9', 0};

char ph[100000][9];
int nph;

char buf[1000];

main() {
	int i, j, k, x, y, z, n;
	char ch;

	memset( ph, 0, sizeof(ph) );
	scanf( " %d", &nph );
	for( i = 0; i < nph; i++ ) {
		scanf( " %s", buf );
		x = 0;
		for( j = 0; buf[j]; j++ ) {
			if( buf[j] == '-' ) continue;
			if( buf[j] >= 'A' && buf[j] <= 'Z' ) buf[j] = map[buf[j]-'A'];
			ph[i][x++] = buf[j];
			if( x == 3 ) ph[i][x++] = '-';
		}
	}
	qsort( ph, nph, 9, strcmp );
	x = 1; z = 0;
	for( i = 1; i < nph; i++ ) {
		if( strcmp( ph[i-1], ph[i] ) ) {
			if( x > 1 ) {
				printf( "%s %d\n", ph[i-1], x );
				z = 1;
			}
			x = 1;
		} else {
			x++;
		}
	}
	if( x > 1 ) {
		printf( "%s %d\n", ph[i-1], x );
		z = 1;
	}
	if( !z ) printf( "No duplicates.\n" );
}


评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值