Member Avatar for raistie_1

Hi all, I am currently trying to generate a number from a specified range
using a function

int number_range(int from, int to) {
    return (from + rand() % (to-from+1));
}

but negative numbers cause the program to crash.
How can i go about doing this so it can take numbers of all kinds?

  • 5 Contributors
  • 4 Replies
  • 280 Views
  • 3 Hours Discussion Span
  • Latest Post Latest Post by RainbowMatrix

Dani AI

Generated

The crash you see is not because rand() cannot handle negative numbers — it is caused by undefined/incorrect arithmetic around the range. If (to - from + 1) becomes zero, negative, or overflows a signed int, the right-hand operand to % is invalid and you get UB (or a crash). was right to check the ordering; 's abs idea does not address ordering or overflow; 's compact formula only works when min <= max and the subtraction does not overflow; showed seeding, but seeding should happen once at program startup (not inside each call).

A safer approach:

  • Validate or swap inputs so min <= max.
  • Compute the range in an unsigned 64-bit type to avoid signed overflow.
  • Use rejection sampling rather than a single rand() % range to avoid bias and to ensure the modulus operand is positive.
  • Build a wide-enough random candidate by combining rand() calls (portable) or use a platform API like arc4random_uniform / random() / C++11 <random> when available.
  • Seed the RNG once at program start (srand(time(NULL))) if using rand().

Example (portable C, avoids overflow and reduces bias):

#include <stdlib.h>
#include <limits.h>
#include <stdint.h>

int number_range_safe(int a, int b) {
    int min = a, max = b;
    if (min > max) { int t = min; min = max; max = t; }

    unsigned long long range = (unsigned long long)max - (unsigned long long)min + 1ULL;
    if (range <= 1ULL) return min;

    unsigned int rand_bits = 0;
    unsigned int tmp = RAND_MAX;
    while (tmp) { rand_bits++; tmp >>= 1; }

    unsigned long long limit = (ULLONG_MAX / range) * range;
    unsigned long long r;
    do {
        r = 0ULL;
        unsigned int filled = 0;
        while (filled < 64) {
            r = (r << rand_bits) | (unsigned long long)rand();
            filled += rand_bits;
        }
    } while (r >= limit);

    return (int)(min + (r % range));
}

Troubleshooting tips: test with extremes (INT_MIN, INT_MAX) and with swapped inputs; add assertions to catch bad arguments. If you need high-quality or cryptographic randomness, use a proper library or OS-provided secure RNG rather than rand().

Recommended Answers

All 4 Replies

Member Avatar for np complete

It can make problem if from > to. Check it.

Else try this.

ans = to - from ;

flag = rand() % ans + 1 ;

return (from + ans) ;
Member Avatar for delta_frost

You can use the abs function from the math.h header file to get the aboslute value of the generated random number and then operate as necessary.

Member Avatar for WaltP

The correct form is:

(rand() % (max-min+1) + min;

Member Avatar for RainbowMatrix

Refer to my die's application and u solve ur issue youself:

#include <iostream>
#include <math.h>
#include <cstdlib>
#include <ctime>
using namespace std;

int main()
{
  int any,any2;

  cout<<"Enter a random number  : ";
  cin>>any;

     srand(time(0));
     any2=1+(rand()%6);
    cout <<"The random number is : "<<any2 << endl;
    if(any==any2){
    //your oode
    }
    else {//your oode
    }
}
Be a part of the DaniWeb community

We're a friendly, industry-focused community of developers, IT pros, digital marketers, and technology enthusiasts meeting, networking, learning, and sharing knowledge.