I recently noticed that the following expression compiles in my compiler (Clang):
float a;
if (0.0 < a < 1.0) { ... }
Does this do wh
Does this do what I expect?
No as it is very simple. < operator is having a left to right associativity it check as follows
(0.0 < a < 1.0 ) is simply
(0.0 < a) < 1.0
in second form
int main()
{
float a = 8.4;
if (0.0 < a && a < 9) //same as ( 0.0 < a) && (a < 9)
printf("x");
}
o/p=x
Because of left-to-right associativity of < operator the expression condition (0.0 < a < 1.0) means ((0.0 < a) < 1.0) == 1 < 1.0 or 0 < 1.0 depending on value of a.
So no, its not identical to if (0.0 < a && a < 1.0) (perhaps you might confusing from Python compassion rule) but in C it will be interpenetrated as I explained above.
A difference you can observe in an example --- 0.0 < a < 1.0 == true when a == 0.0, where as (0.0 < a && a < 1.0) == false, for a == 0.0, below is my code (read comments):
#include<stdio.h>
void print_(int c){
c ? printf("True \n"):
printf("False \n");
}
int main(void){
float a = 0.0f;
print_(0.0f < a < 1.0f); // 0.0 < 0.0 < 1.0f == 0 < 1.0f == True
print_(0.0f < a && a < 1.0f); // 0.0f < 0.0f && ... == False && ... = False
return 0;
}
output:
True
False
Check its working @Ideone
Does this do what I expect?
No, it does
(0.0 < a) < 1.0
and (0.0 < a) yields 0 or 1. So you end up having 0 < 1.0 or 1 < 1.0 evaluated.
Note that in some other languages it does what you expect.
For example in Python,
a < b < c
is called a chained comparison and it is equivalent to:
(a < b) and (b < c)
Maybe we can think in this way(just for fun):
if(0.0 <a < 1.0)
==>
if((0.0 <a)<1.0)
think :
if 0.0 < a == true, then (true <1.0) == false!
else if 0.0 < a == false ,then (false <1.0) == true!
So I can translate if(0.0 <a <1.0) to if(!(0.0<a))
Interesting , right?