Thread: Handling big numbers in C

Hey, I've been searching for a way of handling big numbers in C for a long time now... I want to handle numbers up to a trillion.

I have found the gmp library, which seems to do exactly what I want, but my compiler (Digital Mars) dosen't have it by default. I went to the gmp website and downloaded the zip file containaing the library but I don't know how to install it... I searched all over the net for a tutorial on how to install the library and found a few of them, but I don't understand the instructions... They seem to use some Unix comands while installing the library =S

So any help at installing the gmp library or pointing me towards another way of handling big numbers in C would be deeply appreciated.

Thanks in advance =)

I have used gmp, please specify where do you stop in the installation.

robatino

Sorry, I said zip when they're infact .tar.gz files... I forgot that.

I use windows, and thought that .tar.gz files could also be used in windows.

Mortissus

I just download the file containing the library, didn't know what to do after that.

Looking through the gmp-discuss mailing list archives (http://gmplib.org/#MAILINGLISTS), I found this:

Native Windows Binaries for GMP-4.2.1

robatino wich of these files should I download? There are a lot of them =(

I'm using Digital Mars.

I'm not familiar with Windows, but you could just download the pre-built binaries for your system, whatever that is.

Your compiler doesn't give you a 64 bit long long type?

I had an assignment once about this subject. I used linked lists to do the job, the huge integer was given from a file or from stdin and then firstly i saved every digit in nodes including the sign and after this i created add,sub, mul, div, mod etc. I have writen to many lines of code, i advise you to make a given library work as its a hard thing to do.

brewbuck

Yes it does, but when I said a trillion I meant much bigger numbers...

I'm doing Project Euler and a lot of it's challenges involves dealing with ridiculously big numbers... Such as this one:



(I don't if even gmp could handle 28433 * 2^7830457+1, there's probably a smarter way for calculating it's last ten digits, but there are other problems wich for example require you to calculate the max digital sum of a^b where a, b < 100.....)

Lala if you if your using Digital mars, in order to use the GMP you need copying the lib file to the Lib folder under \dm\lib and copy the header file on to the \dm\include folder. And that then include the header in your main code.

Hope you can star using the library.

ssharish2005

Sorry for the delay. To install GMP try this:

tar xvf gmp-4.2.1.tar.gz
cd gmp-4.2.1
./configure
make
make install

Good luck, any problems just ask

robatino

I've tried following your instructions but my code didn't work... Am I on the right track?

Code:

#include <stdio.h>

int main ()
{
    unsigned long long number = 28433;

    for (int i = 0; i <= 7830457; i++)
    {
        number = (number << 1) &#37; 10000000000;
    }

    number++;

    printf ("%d",  number);
    return 0;
}

ssharish2005

I've copied all the .lib files to the lib directory and all .h files to the include directory but when I tried to compile a code that was in project euler's forum I get a bunch of errors...

Code:

 Error 42: Symbol Undefined ___gmpz_clear
 Error 42: Symbol Undefined ___gmpz_get_str
 Error 42: Symbol Undefined ___gmpz_add
 Error 42: Symbol Undefined ___gmpz_ui_pow_ui
 Error 42: Symbol Undefined ___gmpz_init

Inside the zip file was a dll file wich I didn't copy to any folders... Maybe the error is there.

Mortissus

I don't know where and how to use these comands

E &#233; sempre bom encontrar um brasileiro por ai =)