大数运算

#include <iostream>
#include <stdio.h>
#include <cstring>
#include<cassert>
#include <cmath>

using namespace std;

const int MaxShort = 100;//最多有MaxShort * 2个字节//如果要表示64位int，MaxShort = 4
const int baseNum = pow(2.0,16.0); 
const int Shift = 16;
typedef unsigned short Unshort;

typedef struct BigNum
{
	int m_len;
	Unshort m_data[MaxShort];
}BigNum;

BigNum CreateAbsBigNum(char *src)//低位放前，高位放后；
{
	BigNum bg;
	Unshort *data = bg.m_data;
	int b = 0;
	while (src[0]!='\0')
	{
		int i = 0;
		int reminder = 0;
		
		while (src[i]!='\0')
		{
			int divident = src[i] - '0' + reminder * 10;
			reminder = divident % baseNum;
			src[i] = divident / baseNum + '0';
			++i;
		}
		data[b++] = reminder;
		
		while (*src =='0')
			++src;
	}
	bg.m_len = b;
	return bg;
	
}

BigNum AbsMinus(BigNum &bg1,BigNum &bg2)//bg1 >bg2
{
	int left = bg1.m_len;
	int right = bg2.m_len;
	

	int carry = 0;
	BigNum bg;
	
	for (int i=0;i<right;++i)
	{
		int tmp = bg1.m_data[i] - bg2.m_data[i] - carry;
		if(tmp<0)
		{
			carry = 1;
			bg.m_data[i] = baseNum + tmp;
		}
		else
		{
			bg.m_data[i] = tmp;
			carry = 0;
		}
	}

	for (int i= right;i<left;++i)
	{
		int tmp = bg1.m_data[i] - carry;
		if (tmp <0)
		{
			carry = 1;
			bg.m_data[i] = baseNum +tmp;
		}
		else
		{
			bg.m_data[i] = tmp;
			carry = 0;
		}
	}
	bg.m_len = left;
	while (bg.m_len>=0 && bg.m_data[bg.m_len -1]==0)
	{
		--bg.m_len;
	}

	return bg;

}

BigNum AbsAdd(BigNum &bg1,BigNum &bg2)
{
	int left = bg1.m_len;
	int right = bg2.m_len;

	int size = left;
	int maxLen = right;
	BigNum *bgmax = &bg2;
	if(size>right)
	{
		size = right;
		maxLen = left;
		bgmax = &bg1;
	}

	int carry = 0;
	BigNum bg;
	for (int i=0;i<size;++i)
	{
		int tmp = bg1.m_data[i] + bg2.m_data[i] + carry;
		//carry = tmp /baseNum;
		carry = tmp>>Shift;
		//bg.m_data[i] = tmp &(1<<Shift - 1);
		bg.m_data[i] = tmp & 0xffff;
	}
	
	for (int i = size;i<maxLen;++i)
	{
		int tmp = bgmax->m_data[i] + carry;
		carry = tmp>>Shift;
		//carry = tmp /baseNum;
		//bg.m_data[i] = tmp &(1<<Shift - 1);
		bg.m_data[i] = tmp & 0xffff;
	}

	if (carry)
	{
		bg.m_data[maxLen] = 1;
		++maxLen;
	}
	bg.m_len = maxLen;
	return bg;
}

BigNum AbsMul(BigNum &bg1,BigNum &bg2)
{
	BigNum bg;
	memset(bg.m_data,0,sizeof(bg.m_data));
	int i,j;
	for (i=0;i<bg1.m_len;++i)
	{
		size_t carry = 0;
		for (j=0;j <bg2.m_len;++j)
		{
			size_t tmp =  bg.m_data[i+j] + bg1.m_data[i] * bg2.m_data[j] + carry;
			bg.m_data[i+j] = tmp %baseNum;
			carry = tmp /baseNum;
		}
		if (carry >0)
		{
			bg.m_data[i+j] += carry; 
		}
	}
	if(bg.m_data[bg1.m_len+bg2.m_len -1] > 0)
		bg.m_len = bg1.m_len+bg2.m_len;
	else
		bg.m_len = bg1.m_len+bg2.m_len -1;

	return bg;



	
}

void GetAbsDecNum(BigNum &bg,char *tgt)
{
	int size = bg.m_len;
	Unshort *data = bg.m_data;
	int b=0;
	
	while (size>0)
	{
		int i = size - 1;
		int reminder = 0;

		while (i>=0)
		{
			int divident = data[i] + reminder * baseNum;
			reminder = divident % 10;
			data[i] = divident /10;
			--i;
		}
		tgt[b++] = reminder + '0';

		while(size>0 && data[size -1]==0)
			--size;
	}
	tgt[b] = '\0';
}

void Reverse(char *src)
{
	int size = strlen(src);
	int beg = size/2 -1;
	for (;beg>=0;--beg)
	{
		int sym = size -beg -1;
		char tmp = src[beg];
		src[beg] = src[sym];
		src[sym] = tmp;
	}
}

int main()
{
	char t[100];
	memset(t,0,sizeof(t));
	char s[] = "6430006453235007";
	BigNum bg = CreateAbsBigNum(s);
	//GetAbsDecNum(bg,t);
	//Reverse(t);
	//cout<<t<<endl;

	memset(t,0,sizeof(t));
	char s1[]="1234567932211123423411431421246";
	BigNum bg1 = CreateAbsBigNum(s1);
	//GetAbsDecNum(bg1,t);
	//Reverse(t);
	//cout<<t<<endl;
	BigNum bg2 = AbsMul(bg1,bg);

	
	memset(t,0,sizeof(t));
	GetAbsDecNum(bg2,t);
	Reverse(t);
	cout<<t<<endl;
}

主要思想就是用一个单位譬如char，int,short表示一个进制。譬如，对short而言，最大能表示的进制是2^16

还差除法没有写，主要是参考：http://blog.chinaunix.net/uid-26750235-id-3129246.html

可以考虑用一个int表示进制为10000，这样更方便