Variable Min and Max Program

I'm making a program that measures the max and min of different variables, and the program works find signed and unsigned char and short, but when it comes to int, it doesn't work... seems to loop forever... here's my code for the int part:

Code:

---

#include <stdio.h>

main()
{

    /* SIGNED INT */
    signed int int_signed;
    signed int int_signedP;
   
    for (int_signed = 0; int_signed > (int_signedP = int_signed-1); --int_signed)
        ;
    printf("signed int min: %d\n", int_signed);
   
    for (int_signed = 0; int_signed < (int_signedP = int_signed+1); ++int_signed)
        ;
    printf("signed int max: %d\n", int_signed);
   
    printf("\n");

   
    /* UNSIGNED INT */
    unsigned int int_unsigned;
    unsigned int int_unsignedP;
   
    for (int_unsigned = 0; int_unsigned > (int_unsignedP = int_unsigned-1); --int_unsigned)
        ;
    printf("unsigned int min: %d\n", int_unsigned);
   
    for (int_unsigned = 0; int_unsigned < (int_unsignedP = int_unsigned+1); ++int_unsigned)
        ;
    printf("unsigned int max: %d\n", int_unsigned);
   
    printf("\n");
   
}

---

Any ideas?

This code isn't C because C89/90 doesn't support declarations anywhere except the beginning of a block. Such a feature exists in C99, but C99 doesn't support implicit int, so that argument falls flat. :)

Perhaps it would be safer and easier just to use the macros in limits.h?

Code:

---

#include <stdio.h>
#include <limits.h>

int
main(void)
{
  /* Signed integer */
  printf("Signed int min: %d\n", INT_MIN);
  printf("Signed int max: %d\n", INT_MAX);
  /* Unsigned integer */
  printf("Unsigned int min: %u\n", 0);
  printf("Unsigned int max: %u\n", UINT_MAX);

  return 0;
}

---

Quote:

---

it doesn't work... seems to loop forever...

---

Even a computer will take a while to count up to 2.1 billion or so... patience (or take
Prelude's advice and use the built in macros)

p.s. I don't have the patience to see if your code does eventually give the correct result.

Oh wow. Stupid me, I just had to wait. Thank you.

Quote:

---

Originally Posted by Prelude

This code isn't C because C89/90 doesn't support declarations anywhere except the beginning of a block. Such a feature exists in C99, but C99 doesn't support implicit int, so that argument falls flat. :)

Perhaps it would be safer and easier just to use the macros in limits.h?

---

for the sake of discussion. Why isn't it C, C99 is still C. so is 89/90. in fact if you want to get really pedantic anything from k&r isn't C.
Just curious about that, cause your not the first person i've encountered that seems to dislike the additons in c99. why's that?? ;)

While I can't speak for everyone, it's not that it's a dislike, rather, it's pointing out the differences to people so we don't have to answer:

"Well it's C! Why isn't it working on my compiler?"

Thus, the difference between standard versions is pointed out, because most compilers are not C99 compilant right now. Hell, there are a huge huge number of people still using Turbo C 3.0 that post here! :rolleyes:

Quzah.

>Why isn't it C
Because it uses features from both C89/90 and C99 that are illegal in the other. That's highly, highly suspect. ;)

>C99 is still C. so is 89/90. in fact if you want to get really pedantic anything from k&r isn't C.
C was standardized in 1989. That standard was replaced in 1999. If you want to get really pedantic, anything but C99 isn't C because C99 is the most current standard in effect. Until C99 is widely implemented, however, we're forced to use the common subset of features between C89/90 and C99. That's surprisingly easy because C99 enforces what is considered good style in C89/90. But for the most part around here, we assume that you're either working in that subset or we tolerate a pure C89/90 style while suggesting the better style that C99 requires.

>cause your not the first person i've encountered that seems to dislike the additons in c99. why's that??
I don't see how any part of my post suggests I dislike the additions in C99. Of course, that isn't to say that I don't disapprove of some of the additions to C with the new standard, nor that I disapprove of them all. Like all standards, there are good things and bad things. :)

This is a problem

Code:

---

for (int_unsigned = 0; int_unsigned > (int_unsignedP = int_unsigned-1); --int_unsigned)
        ;

---

int_unsigned is unsigned, or only positive. and you're doing --int_unsigned, and your variable becomes 0xffffffff, which is about 4 thousand millions, positive, and it loops until 0 is reached again.
And here's the big problem!!!:

Code:

---

1.
int_unsigned > (int_unsignedP = int_unsigned-1)
2.
int_unsigned > int_unsigned-1

---

expression 1 is your condition for the for to continue, and expression 2 is a simplification, that shows what's happening. That's ALWAYS true! unless int_unsigned is 0, which makes 0 > 0xffffffff, which is false.

well cool of you guys to answer, Just curious.