Consider the following snippet:

int max = 10;
int a = 0;
while (true) {
 // do a thing that may result in an early return 

 if (++a >= max) {
 break;
 }
}

throw new RuntimeException("It ran out of attempts");

There are a few WTFs in the above. The loop’s a bit weird, the flow of control seems to be in a few places… but at the heart of it is a bit of codey code – ++a >= max

In this case does it definitely do 10 attempts? or is it maybe 9? or 11?

General case:

If you can’t, at a glance, determine what the logical expression is doing, it’s too complex.

As it happens, I think this does 10 attempts:

• ++a is a prefix increment, which adds 1 to a
• The >= means a cannot be 10 or more and that is being done after the increment
• a starts as 0
• the first attempt happens when it’s 0, the 10th happens when it’s 9
• So it does 10 attempts

… probably.

While programming languages allow us shortcuts to do things and several ways to express the same things, clearly a for (i=0; i<10; i++) style expression would be instantly recognisable for what it is. The codey code is almost always better refactored.

It seems like loops are prone to codey code.