defining object macros

Question

Hello,

gcc 4.4.4 c89

I have seen this in some code I am maintaining.

#define FALSE 0
#define TRUE (!FALSE)

is there any difference in writing the following:

#define FALSE 0
#define TRUE 1

Its been used in some functions like this:

if(condition failed) {
    return FALSE;
}
else {
    return TRUE;
}

Many thanks for any suggestions,

Answer 1

The C "if" statement interprets zero as false,and absolutely anything else at all as true. So it's not really important what value you use for TRUE, though 1 would be conventional.

But (!0) == 1 anyway so this should not change anything.

>> cat joe.c
#include <stdlib.h>

#define FALSE 0
#define TRUE (!FALSE)

int main (int argc, char * argv[]) {
  printf("false %d\ntrue %d\n", FALSE, TRUE);
  return 0;
}
>> cc joe.c
>> ./a.out
false 0
true 1
>>

Answer 2

One difference is, in the (!FALSE) case, if you change:

#define FALSE 0

to

#define FALSE 1

Your program still "works", without modifying TRUE... however, it's unlikely that would be safe anyways, because I'm sure your program relies on constructs such as

if (funtion_that_returns_FALSE()) {
}

Which would break if you changed the definition of FALSE. Don't change it :)

Answer 3

Edit: I read the question backwards. Sorry. Since TRUE is !FALSE and FALSE is 0, then TRUE is 1.

The compiler will evaluate FALSE to 0 anyhow, since it will see (!1) which doesn't make sense to compute at run time.

It's just silly.

Answer 4

The best way to do this is:

#ifndef FALSE
#define FALSE 0
#elif FALSE != 0
#undef FALSE
#define FALSE REWRITE_PROGRAM("FALSE")
#endif

#ifndef TRUE
#define TRUE 1
#elif TRUE != 1
#undef TRUE
#define TRUE REWRITE_PROGRAM("TRUE")
#endif

Where REWRITE_PROGRAM is a macro which calls a function (providing the line number and source file in advance) that, using advanced neural networks, will intelligently rewrite your code so as to avoid conflicting definitions of TRUE or FALSE where TRUE is defined to be anything other than 1 and FALSE to be anything other than 0. After it generates the corrected code and builds it, it will execute the newly built version of the application and exit.

As a backup, the rewrite AI will also make sure that its new assumptions are correct by adding more code with REWRITE_PROGRAM should it ever encounter a case where its assumptions are incorrect.

Ultimately the program will rewrite itself repeatedly until it is correct, and this way you don't have to worry about how TRUE and FALSE are defined, conflicts, and whether those definitions will change in the future.