[Webinar] Streamline your web hosting managementRegister Today

x
• Status: Solved
• Priority: Medium
• Security: Public
• Views: 264

Long Integers

Hi all

How can I create a long integer number.
The integer must have a length from 0 to
300'000'000'000'000'000'000.

It must be in C++ on the WINDOWS Platform

Thanks

0
new_user
• 3
1 Solution

Commented:
there are several approaches to this.  You can use a class/structure that stores several ordinary integers to represent a huge integer that has sufficient bits to represent the number.  You will then have to provide arithmetic functions for the class that perform the appropriate calculations on the numbers.   Another approach is to use BCD.  In this case you woudl have a class that stores a string that contains the digits represented.  Again you would have to write the mathmatical operations or the class.  BCD is slower in computation and takes up more memory, however it is easier to program and to use (you can check its value while debugging much more easily)  it is also much faster to convert to and from ASCII, so it is faster to enter and display.

continues
0

Commented:
For the first method, you will need to store 3 32-bit integers to have appropriate precission, like

class LongInt
{
int Low;
int  Mid;
int Hgh;
public:
class LongInt(int i = 0) : Low(i), Mid(0), Hgh(0) {};
class LongInt(const LongInt &Src) : Low(Src.Low), Mid(Src.Mid), Hgh(Src.Hgh) {};

LongInt &operator += (const LongInt &Rht)
{
// Note I beleive this works as the right way to set the carry
// but its been years.   You might need to do a little testing.
int Carry;
int Old

Old = Low;
Low += Rht.Low;
Carry = Low < Old:1:0;

Old = Low;
Mid += Rht.Mid += Carry;
Carry = Low < Old:1:0;

Hgh += Rht.Hgh += Carry;
};
};

LongInt operator += (const LongInt &Lft,const LongInt &Rht)
{
LongInt Result = Lft;
Result += Rht;
return Result;
}
0

Commented:
Hopefully that gives you enough to get started.  If you need more help I can provide it.  First you need to figure out which method you wnt to use.  Personally, I prefer the BCD route (I can probably help you more that way too) mostly because I can then use the data type more easily.  I.e. while debugging I can see what value is stored.  There is no easy way to check the value stored in the binary form, since the debugger shows 3 ints, not 1 int, and you can't even plug them into your calculator to find the right value, because it will exceed the precission of the calculator.
0

Author Commented: