Question

I heard some programmers use if(1 == var) instead of if(var == 1) to avoid unintended assignment. Why or in what cases does it cause unintended assignment?

Answer 1

The problem is if you mistype the statement:

if (var = 1)

vs

if (1 = var)

In the first case, the code after the if is executed unconditionally (with no more than a warning from the compiler, which isn’t obliged to produce a warning for you — but the good ones do; if you aren’t using a good compiler, get one!). In the second case, you get a syntax error at compile time, so the problem has to be fixed before the code can compile.

The problem isn’t always as blatant:

if (var = 0)

never executes the code after the if, of course. Often though, you’ll have:

if (var = function(arg1, arg2))

and it won’t be clear whether you intended to assign or compare. You can make it clear to the compiler and code readers (humans) by writing:

if ((var = function(arg1, arg2)) != 0)

or

if (var == function(arg1, arg2))

I don’t use the ‘back-to-front’ comparison technique. I dislike the inverted conditions because they almost invariably read ‘wrong’ to me. I’m not comparing 1 with my variable; I’m comparing my variable with 1. So, even though logically the == operator is commutative, I don’t think commutatively and prefer that ‘riskier’ way. I have not found myself making the assignment vs equality mistake often enough for the issue to be a problem. The compiler warns me if I do make a mistake (and I pay attention to the warning and fix the code so that there isn’t a problem).