AutoHotkey Community

Great explanation Lexikos. One thing: Why can't number format be retained internally when floats are operated on? Or is this essentially what v2 does as v1 relies on SetFormat?

What do you mean? 0.00 + 1 = 1.00? Well, 0.00 is a string. To perform math on it, AutoHotkey has to convert it into a number (i.e. in binary format). The output of any math operation is either an empty string (error) or a binary number (string format is irrelevant). Floating-point numbers are stored in binary form on both v1 and v2, but how you get a binary number in the first place differs slightly. Numeric literals in v1 are always SYM_OPERAND (which is either just a string, or a string with a cached binary integer), whereas numeric literals in v2 are either SYM_INTEGER (binary integer) or SYM_FLOAT (binary floating-point number).

I think I understand now. To return x as a string representation, you could do the following:
This is effectively the same as using return (x) in my old example because return (expression) "forces" an expression, thus, returning a string. But, as you said, the value returned wouldn't be a pure number. Please correct me if I'm misunderstanding this.