Why you should (sometimes) reset errno manually!

test_errno.c

#include <stdio.h>
#include <limits.h>
#include <stdlib.h>
#include <errno.h>
#include <assert.h>


int main(void) {

    unsigned long number = -1;
    char *endptr = NULL;

    /* Checking that errno is set to 0 at the initialisation of our program */
    assert(errno == 0);

    /* Trying to open a file with an invalid mode */
    /* This set errno to EINVAL value */
    fopen("something", "invalid");
    assert(errno == EINVAL);

    /* the missing line! */
    /* Uncomment it to fix the behavior of this snippet */
    /* errno = 0;*/

    number = strtoul("0", &endptr, 0);
    /* Trying to catch errors according to the man page of strtoul */

    /* Extract of the manpage:
     * The strtoul(), strtoull(), strtoumax() and strtouq() functions return
     * either the result of the conversion or, if there was a leading minus
     * sign, the negation of the result of the conversion, unless the origi-
     * nal (non-negated) value would overflow; in the latter case, strtoul()
     * returns ULONG_MAX, strtoull() returns ULLONG_MAX, strtoumax() returns
     * UINTMAX_MAX, and strtouq() returns ULLONG_MAX.  In all cases, errno
     * is set to ERANGE.  If no conversion could be performed, 0 is returned
     * and the global variable errno is set to 
     * EINVAL (the last feature is not portable across all platforms).
     */

    if ((number == 0 && errno == EINVAL) || (number == ULONG_MAX && errno == ERANGE)) {
        printf("wrong error detected\n");
    } else {
        printf("computation succeeded\n");
    }
}