C doesn't have any built-in Boolean types. What's the best way to use them in C?
-
93C does have boolean type. At least there is one in the most recent standards.Khaled Alshaya– Khaled Alshaya2009-12-17 12:50:54 +00:00Commented Dec 17, 2009 at 12:50
-
12stackoverflow.com/questions/1608318/is-bool-a-native-c-type/…AnT stands with Russia– AnT stands with Russia2009-12-17 18:50:48 +00:00Commented Dec 17, 2009 at 18:50
19 Answers
From best to worse:
Option 1 (C99 and newer)
#include <stdbool.h>
Option 2
typedef enum { false, true } bool;
Option 3
typedef int bool;
enum { false, true };
Option 4
typedef int bool;
#define true 1
#define false 0
Explanation
- Option 1 will work only if you use C99 (or newer) and it's the "standard way" to do it. Choose this if possible.
- Options 2, 3 and 4 will have in practice the same identical behavior. #2 and #3 don't use #defines though, which in my opinion is better.
If you are undecided, go with #1!
7 Comments
<stdbool.h> bool the compiler chooses may be more suitable for the intended purpose of a boolean value than utilizing an int (i.e. the compiler may choose to implement a bool differently than an int). It might also result in stricter type checking at compile time, if you're lucky.<stdbool.h>. For example, they don't satisfy (bool)(13 & 8) == true. With options 2-4, you'd have to write !!(13 & 8) == true instead. This comes up all the time in bit-field testing... think of a macro that returns (13 & 8) with return type bool.A few thoughts on booleans in C:
I'm old enough that I just use plain ints as my boolean type without any typedefs or special defines or enums for true/false values. If you follow my suggestion below on never comparing against boolean constants, then you only need to use 0/1 to initialize the flags anyway. However, such an approach may be deemed too reactionary in these modern times. In that case, one should definitely use <stdbool.h> since it at least has the benefit of being standardized.
Whatever the boolean constants are called, use them only for initialization. Never ever write something like
if (ready == TRUE) ...
while (empty == FALSE) ...
These can always be replaced by the clearer
if (ready) ...
while (!empty) ...
Note that these can actually reasonably and understandably be read out loud.
Give your boolean variables positive names, ie full instead of notfull. The latter leads to code that is difficult to read easily. Compare
if (full) ...
if (!full) ...
with
if (!notfull) ...
if (notfull) ...
Both of the former pair read naturally, while !notfull is awkward to read even as it is, and becomes much worse in more complex boolean expressions.
Boolean arguments should generally be avoided. Consider a function defined like this
void foo(bool option) { ... }
Within the body of the function, it is very clear what the argument means since it has a convenient, and hopefully meaningful, name. But, the call sites look like
foo(TRUE);
foo(FALSE):
Here, it's essentially impossible to tell what the parameter meant without always looking at the function definition or declaration, and it gets much worse as soon if you add even more boolean parameters. I suggest either
typedef enum { OPT_ON, OPT_OFF } foo_option;
void foo(foo_option option);
or
#define OPT_ON true
#define OPT_OFF false
void foo(bool option) { ... }
In either case, the call site now looks like
foo(OPT_ON);
foo(OPT_OFF);
which the reader has at least a chance of understanding without dredging up the definition of foo.
9 Comments
a == b work?a and b count up from zero, I'd recommend a > 0 == b > 0 instead. If you insist on taking advantage of the truthiness of arbitrary non-zero values, !!var yields the boolean 0/1 value equivalent to var, so you could write !!a == !!b, although quite a few readers will find it confusing.!a == !b is also sufficient for testing equality, non-zeros become zero, and zeros become one.!!a as "convert non-boolean a to its equivalent truth value", whereas I'd read !a as "logically invert the boolean variable a". In particular, I'd look for some specific reason the logical inversion was desired.A boolean in C is an integer: zero for false and non-zero for true.
See also Boolean data type, section C, C++, Objective-C, AWK.
1 Comment
Here is the version that I used:
typedef enum { false = 0, true = !false } bool;
Because false only has one value, but a logical true could have many values, but technique sets true to be what the compiler will use for the opposite of false.
This takes care of the problem of someone coding something that would come down to this:
if (true == !false)
I think we would all agree that that is not a good practice, but for the one time cost of doing "true = !false" we eliminate that problem.
[EDIT] In the end I used:
typedef enum { myfalse = 0, mytrue = !myfalse } mybool;
to avoid name collision with other schemes that were defining true and false. But the concept remains the same.
[EDIT] To show conversion of integer to boolean:
mybool somebool;
int someint = 5;
somebool = !!someint;
The first (right most) ! converts the non-zero integer to a 0, then the second (left most) ! converts the 0 to a myfalse value. I will leave it as an exercise for the reader to convert a zero integer.
[EDIT]
It is my style to use the explicit setting of a value in an enum when the specific value is required even if the default value would be the same. Example: Because false needs to be zero I use false = 0, rather than false,
[EDIT] Show how to limit the size of enum when compiling with gcc:
typedef __attribute__((__packed__)) enum { myfalse = 0, mytrue = !myfalse } mybool;
That is, if someone does:
struct mystruct {
mybool somebool1;
mybool somebool2;
mybool somebool3;
mybool somebool4;
}
the size of the structure will be 4 bytes rather than 16 bytes.
10 Comments
! can only return values 0 and 1, so true = !false will always assign value 1. This method does not give any extra safety over typedef enum { false, true } bool;.if(!value)), but that exception is not applicable in this specific case.If you are using a C99 compiler it has built-in support for bool types:
#include <stdbool.h>
int main()
{
bool b = false;
b = true;
}
1 Comment
_Bool. It's not an enum. bool, true and false are macros defined in <stdbool.h> that expand to _Bool, 1 and 0 respectively. See en.cppreference.com/w/c/types/booleanFirst things first. C, i.e. ISO/IEC 9899 has had a boolean type for 19 years now. That is way longer time than the expected length of the C programming career with amateur/academic/professional parts combined when visiting this question. Mine does surpass that by mere perhaps 1-2 years. It means that during the time that an average reader has learnt anything at all about C, C actually has had the boolean data type.
For the datatype, #include <stdbool.h>, and use true, false and bool. Or do not include it, and use _Bool, 1 and 0 instead.
There are various dangerous practices promoted in the other answers to this thread. I will address them:
typedef int bool;
#define true 1
#define false 0
This is no-no, because a casual reader - who did learn C within those 19 years - would expect that bool refers to the actual bool data type and would behave similarly, but it doesn't! For example
double a = ...;
bool b = a;
With C99 bool/ _Bool, b would be set to false iff a was zero, and true otherwise. C11 6.3.1.2p1
- When any scalar value is converted to
_Bool, the result is 0 if the value compares equal to 0; otherwise, the result is 1. 59)Footnotes
59) NaNs do not compare equal to 0 and thus convert to 1.
With the typedef in place, the double would be coerced to an int - if the value of the double isn't in the range for int, the behaviour is undefined.
Naturally the same applies to if true and false were declared in an enum.
What is even more dangerous is declaring
typedef enum bool {
false, true
} bool;
because now all values besides 1 and 0 are invalid, and should such a value be assigned to a variable of that type, the behaviour would be wholly undefined.
Therefore iff you cannot use C99 for some inexplicable reason, for boolean variables you should use:
- type
intand values0and1as-is; and carefully do domain conversions from any other values to these with double negation!! - or if you insist you don't remember that 0 is falsy and non-zero truish, at least use upper case so that they don't get confused with the C99 concepts:
BOOL,TRUEandFALSE!
1 Comment
int or unsigned int to hold an enumeration, but I know of nothing in the Standard that would cause an enumeration to behave as anything other than an integer type.typedef enum {
false = 0,
true
} t_bool;
1 Comment
!0 = 1 by the C standard, and !a = 0 for any non-zero value of a. The problem is that any non-zero is considered as true. So if a and b are both "true", it isn't necessarily the case that ` a == b`.C has a boolean type: bool (at least for the last 10(!) years)
Include stdbool.h and true/false will work as expected.
5 Comments
bool to be a macro that expands to _Bool. The difference matters because you can #undef a macro (and that's permitted, at least as a transitional measure), but you can't untypedef a typedef. It doesn't alter the main thrust of your first comment, though.Anything nonzero is evaluated to true in boolean operations, so you could just
#define TRUE 1
#define FALSE 0
and use the constants.
4 Comments
This is just a complement to other answers and some clarification, if you are allowed to use C99.
+-------+----------------+-------------------------+--------------------+
| Name | Characteristic | Dependence in stdbool.h | Value |
+-------+----------------+-------------------------+--------------------+
| _Bool | Native type | Don't need header | |
+-------+----------------+-------------------------+--------------------+
| bool | Macro | Yes | Translate to _Bool |
+-------+----------------+-------------------------+--------------------+
| true | Macro | Yes | Translate to 1 |
+-------+----------------+-------------------------+--------------------+
| false | Macro | Yes | Translate to 0 |
+-------+----------------+-------------------------+--------------------+
Some of my preferences:
_Boolorbool? Both are fine, butboollooks better than the keyword_Bool.- Accepted values for
booland_Boolare:falseortrue. Assigning0or1instead offalseortrueis valid, but is harder to read and understand the logic flow.
Some information from the standard:
_Boolis notunsigned int, but is part of the group unsigned integer types. It is large enough to hold the values0or1.- Do not, but yes, you are able to redefine
booltrueandfalsebut sure is not a good idea. This ability is considered obsolescent and will be removed in future. - Assigning an scalar type (arithmetic types and pointer types) to
_Boolorbool, if the scalar value is equal to0or compares to0it will be0, otherwise the result is1:_Bool x = 9;9is converted to1when assigned tox. _Boolis 1 byte (8 bits), usually the programmer is tempted to try to use the other bits, but is not recommended, because the only guaranteed that is given is that only one bit is use to store data, not like typecharthat have 8 bits available.
1 Comment
a_is_used*a + b_is_used*b seems to be allowed. It looks quite readable.Since C23
bool and _Bool, and true and false are language keywords for boolean types. bool/_Bool is a type that can hold either the value true or false. Logical operators !, ||, and && can be used.
A scalar-typed value can be implicitly converted to bool. Zero-valued arithmetic types convert to false, as well as null pointers, or anything of type nullptr_t. Any other scalar-type values convert to true. Rules for promotion to higher-rank integer types can be found here.
The macros bool, true, and false from <stdbool.h> are "removed" (from that header specifically), but an implementation can still define such predefined macros for compatibility, and implementations can still use macros to implement them, and if that is done, programs can still undefine and redefine them. The macro __bool_true_false_are_defined is deprecated.
See also https://open-std.org/JTC1/SC22/WG14/www/docs/n2935.pdf.
Since C99
_Bool is a language keyword for a boolean type that can hold either the value 1 or 0. Logical operators !, ||, and && can be used.
When converting scalar types to _Bool, any value that compares equal to zero is converted to 0, and other value is converted to 1. Rules for promotion to higher-rank integer types can be found here.
The standard header <stdbool.h> provides a convenience macro that defines bool as an alias to _Bool, a macro true that expands to the integer-type (not _Bool!) constant 1, a macro false that expands to the integer constant 0, and a macro __bool_true_false_are_defined that expands to the integer constant 1. Programs are allowed to undefine and redefine the macros bool, true, and false.
Prior to C99
You could write something like typedef enum { false, true } bool;.
Comments
You could use _Bool, but the return value must be an integer (1 for true, 0 for false). However, It's recommended to include and use bool as in C++, as said in this reply from daniweb forum, as well as this answer, from this other stackoverflow question:
_Bool: C99's boolean type. Using _Bool directly is only recommended if you're maintaining legacy code that already defines macros for bool, true, or false. Otherwise, those macros are standardized in the header. Include that header and you can use bool just like you would in C++.
Comments
Nowadays C99 supports boolean types but you need to #include <stdbool.h>.
Example:
#include <stdbool.h>
int main()
{
bool arr[2] = {true, false};
printf("%d\n", arr[0] && arr[1]);
printf("%d\n", arr[0] || arr[1]);
return 0;
}
Output:
0
1
2 Comments
Conditional expressions are considered to be true if they are non-zero, but the C standard requires that logical operators themselves return either 0 or 1.
@Tom: #define TRUE !FALSE is bad and is completely pointless. If the header file makes its way into compiled C++ code, then it can lead to problems:
void foo(bool flag);
...
int flag = TRUE;
foo(flag);
Some compilers will generate a warning about the int => bool conversion. Sometimes people avoid this by doing:
foo(flag == TRUE);
to force the expression to be a C++ bool. But if you #define TRUE !FALSE, you end up with:
foo(flag == !0);
which ends up doing an int-to-bool comparison that can trigger the warning anyway.
Comments
It is this:
#define TRUE 1
#define FALSE 0
1 Comment
You can use a char, or another small number container for it.
Pseudo-code
#define TRUE 1
#define FALSE 0
char bValue = TRUE;
2 Comments
int), since on some architectures you get a significant performance hit from having to unpack/mask checks on these variables.If you are using C99 then you can use the _Bool type. No #includes are necessary. You do need to treat it like an integer, though, where 1 is true and 0 is false.
You can then define TRUE and FALSE.
_Bool this_is_a_Boolean_var = 1;
//or using it with true and false
#define TRUE 1
#define FALSE 0
_Bool var = TRUE;
1 Comment
#include <stdbool.h> and use bool, true, and false like the standard wants you to.This is what I use:
enum {false, true};
typedef _Bool bool;
_Bool is a built in type in C. It's intended for boolean values.
1 Comment
stdbool.h are _Bool-typed starting in C23. Using an enum like this subtly breaks comparison operators because you don't get automatic casts to _Bool.You can simply use the #define directive as follows:
#define TRUE 1
#define FALSE 0
#define NOT(arg) (arg == TRUE)? FALSE : TRUE
typedef int bool;
And use as follows:
bool isVisible = FALSE;
bool isWorking = TRUE;
isVisible = NOT(isVisible);
and so on
1 Comment
arg and the expression as a whole: #define NOT(arg) (((arg) == TRUE) ? FALSE : TRUE). However, it would be better to test for falseness (it will give the correct answer even if arg was 23 instead of 0 or 1: #define NOT(arg) (((arg) == FALSE) ? TRUE : FALSE). But the whole expression can be reduced to #define NOT(arg) (!(arg)), of course, which produces the same result.