?: operator

[greghop]

coding style issues, which better, how used ?? 1. if (true == v1) { v2 = 0; } else { v2 = 1; } 2. v2 ((true == v1) ? 0 : 1);

[Johann Gerell]

Don't compare a boolean value with true or false. There's no logic in that (sic!). It'll cause more confusion than clarity. For instance:

if(((enabled == true) == true) == true...)

Where dou you stop? It's obviously clearer with

if(enabled)

But, as the question is formulated, me thinks that

v2 = v1 ? 0 : 1;

is best. -- The Blog: Bits and Pieces

[Bob Stanneveld]

Hello, It's just a matter of preference. You can also write:

v2 = !v1;

Since you can use the opposite value of v1, you can omit v2. This can go on endlessly. Choose what comforts you most. ;) Behind every great black man...             ... is the police. - Conspiracy brother Blog[^]

[Gary R Wheeler]

if (v1) {
v2 = 0;
}
else {
v2 = 1;
}

or

v2 = v1 ? 0 : 1;

For myself, I only use the condtional operator (? :) when the result is a simple expression.

---

Software Zen: delete this;

[Tim Smith]

I just want to add, it can also be a very buggy thing to do. Even a C++ 'bool' can have more values than just true and false. But aside from that, if you are working with Windows API, comparing to TRUE is the wrong thing to do since the API are defined as being 0 and != 0 which means that they might return 5219 for TRUE. Tim Smith I'm going to patent thought. I have yet to see any prior art.

[Johann Gerell]

Bob Stanneveld wrote: It's just a matter of preference. You can also write: v2 = !v1; Yes, when v2 is also a bool, but I sensed that the question covered the generic case. -- The Blog: Bits and Pieces

[Bob Stanneveld]

Johann Gerell wrote: Yes, when v2 is also a bool If v2 is an int, true will be changed to 1 and false to 0. Unless v2 is an user defined type, this can work... Behind every great black man...             ... is the police. - Conspiracy brother Blog[^]

[Maxwell Chen]

if(v1) { v2 = 0; }
else { v2 = 1; }

runs a little faster than

v2 = v1 ? 0 : 1;

---

Maxwell Chen

[ThatsAlok]

Maxwell Chen wrote: if(v1) { v2 = 0; } else { v2 = 1; } runs a little faster than v2 = v1 ? 0 : 1; How thats run Faster?

"Opinions are neither right nor wrong. I cannot change your opinion. I can, however, change what influences your opinion." - David Crow

cheers, Alok Gupta VC Forum Q&A :- I/ IV

[Maxwell Chen]

About four years ago I did some test about your question with VC++6. Just now I did again with VC++7.1. My method is as:

void Test_If(bool& v1, int& v2)
{
if(v1) {
v2 = 0;
}
else {
v2 = 1;
}
}

void Test_Op(bool& v1, int& v2)
{
v2 = v1 ? 0 : 1;
}

void CTestComboDlg::OnBnClickedButton1()
{
bool v1 = true;
int v2 = 3;
int i;
DWORD ds, de;

ds = GetTickCount();
for(i = 0; i < 0x10000000; i++) {
	Test\_If(v1, v2);
}
de = GetTickCount();
TRACE("IF: %u \\n", de - ds);

ds = GetTickCount();
for(i = 0; i < 0x10000000; i++) {
	Test\_Op(v1, v2);
}
de = GetTickCount();
TRACE("OP: %u \\n", de - ds);

}

The output is:

IF: 48690
OP: 52576

---

Maxwell Chen

[toxcct]

ThatsAlok wrote: How thats run Faster? maybe because it has to evaluate the ?: operator before.

1. using if :

• evaluate the condition.
• test the if statement
• assign v2

2. using ?: :

• enters the = operator
• enters the ?: operator
• evaluate the condition
• return the right parameter wether the condition is true or false
• assign v2

i feel it like this... but i personnally still use the second case :-D

---

TOXCCT >>> GEII power
[toxcct][VisualCalc]

[ThatsAlok]

Maxwell Chen wrote: he output is: IF: 48690 OP: 52576 Nice, Thanks for same :)

"Opinions are neither right nor wrong. I cannot change your opinion. I can, however, change what influences your opinion." - David Crow

cheers, Alok Gupta VC Forum Q&A :- I/ IV

[Blake Miller]

You should not have to feel a thing. Just DO IT by examining the disassembly. The assembly dump would explain why one is faster than the other.