Hello,
Why is it better to do the following
if (NULL != my_pointer)
then
if (my_pointer != NULL)?
My boss says that if my_pointer is NULL and C evaluates a null
pointer first it will crash, but isn't it true that either way a null
pointer will be evaluated?
Thanks a lot.
neither of the examples you posted would create a problem. a pointer is just a variable that holds an address. the problem arises when you try to access data at an invalid address (by dereferencing the pointer).
Code:#include <cmath> #include <complex> bool euler_flip(bool value) { return std::pow ( std::complex<float>(std::exp(1.0)), std::complex<float>(0, 1) * std::complex<float>(std::atan(1.0) *(1 << (value + 2))) ).real() < 0; }
Hmm... never heard that one before. I tried it, and the program didn't crash. Just do what the boss says. Keep the person who gives you your paycheck happy. ;-)
You sure your boss wasn't thinking of the difference between NULL = my_pointer and my_pointer = NULL?
In the stone age, compilers didn't let you know that you might be screwing up if you accidentally wrote thisOriginally Posted by alois_rone
instead of thisCode:if(my_pointer = NULL)So writing it the other way around was a way to make up for using poor tools.Code:if(my_pointer == NULL)
Last edited by Dave_Sinkula; 02-18-2006 at 08:44 PM. Reason: Which is just another way of stating what OnionKnight mentioned.
7. It is easier to write an incorrect program than understand a correct one.
40. There are two ways to write error-free programs; only the third one works.*
..... even now, "if (my_pointer = NULL)" can in some (admittedly rare) cases be what the programmer intends. So, few compilers will flag this construct as an error by default, but they may give a warning. A few too many programmers ignore warnings or tweak compiler settings to turn warnings off. Those same people then complain about the language or about the tools when they get in trouble .......
MSVC wont give you an error unless you increase the warning level to max. And honestly, why would I want the compiler to give me a warning every time I did something like:
Code:if(i = foo()) { ... }
Because it could easily and perhaps more clearly be written like this?
To each his own. Tailor your tools to your style and whatever best allows you to write clear, correct, maintainable code. I have my preferences; so does everybody else. I want lint's "Boolean test of assignment" diagnostic to appear because it does catch bugs, and I generally write code to avoid it.Code:error = foo(); /* do something */ if ( error ) /* test result of doing something and act appropriately */ { ... }
7. It is easier to write an incorrect program than understand a correct one.
40. There are two ways to write error-free programs; only the third one works.*
Thanks so much for the responses. My boss is a fresh grad... and I am a senior engineer, but new to C, so maybe I shouldn't take ALL his comments too seriously.
It's good to check things for yourself, get as much information on things as you can. Being informed is always a good thing.I agree. To illustrate the point, we have a "author / teacher" here who most of us agree is full of ........ most of the time.
Quzah.
Hope is the first step on the road to disappointment.
I've seen that construct CONSTANT != variable or CONSTANT == variable before and it is usually used as a way of having your compiler catch something like this:
If you look closely, you will see that becuase of a typo, the if statement is really setting the value of myvar to CONSTANT, instead of checking for equality (==).Code:/* Mistake 1 -- compiler won't catch it */ if (myvar = CONSTANT) { /* do something here */ }
If you flip the expression around
the compiler will gak on it because you cannot change the value of a constant.Code:/* Mistake 2 -- compiler will */ if (CONSTANT = myvar) {
Because the first mistake up there is perfectly legal C, it'll cause funky runtime errors that might be hard to find.
Personally, I never use it -- causes too much confusion with new programmers <grin>
Thanks again. Yeah I think that's the real reason behind ('=' instead
of '=='), although my boss got the wrong explanationYou guys are very helpful!
