Negative Creep

Candice has inherited some legacy C++ code. It’s legacy enough, for example, that there are about 15 definitions of a boolean, depending which headers you include. It contains, in Candice’s words, “an ambitious attempt to #undef return”.

In short, there are plenty of horrors in the code. But this particular short snippet is the one that drew Candice’s attention.

#define NEGATE_VALUE(x) ( -1 * x )

If only there were some convenient symbol or operator that allowed us to negate a numeric value. Some way we could take, say, a 1, and make it negative, maybe by putting a - in front of it, e.g., -1. And maybe that could also work with variables, so we could write -x to represent the negation of x.

The macro isn’t wrong, it’s just useless in a way that makes me puzzle about what the underlying motivation was. I can’t understand why someone would write this, and even once it’s written, I don’t understand why anyone would use this macro. And yet, it was.