On 2 May 2016 at 18:20, Rob Stewart wrote:

On May 1, 2016 11:25:39 PM EDT, Jeremy Murphy < jeremy.william.murphy@gmail.com> wrote:

...

Dear list entities,

is there such a convention or precedent for naming an iterated binary operation?

For example, Boost.Functional/Hash has an iterated hash() called hash_range(). I don't mind the "_range" suffix but I prefer the "iterated_" prefix based on the mathematical definition: https://en.wikipedia.org/wiki/Iterated_binary_operation

The case in point is gcd() in Boost.Math: we want to provide an iterated gcd but we can't quite decide how to name it. I like the "iterated_" prefix but I would rather go with whatever might be the convention.

Given that the definition you linked noted only a couple of cases using different names for iterated variants and a couple of cases that use the same name as for the non-iterated function, it isn't clear that there is a case for preferring that modifier. (I'm no mathematician, so I have no experience to draw upon here.)

Yes, I haven't searched far and wide for examples yet. I like the "iterated_" prefix because it seems the most universal or generic in meaning: it's not actually specific to C++ even though we have "iterators". That's just my personal preference, not what I think is necessarily the best answer. In C++, given overloading, it isn't necessary to use a different name for

different arguments. Therefore, I wonder whether "gcd" is sufficient for the iterated case.

Indeed, this is an interesting point and I thought about it for some time already. My (strong) feeling is that functions with different semantics should have different names, but I don't have a formal or strong argument to support this. I know we can use the same name but it doesn't seem like a good idea apart from minimizing the number of names in a namespace. A variadic template version with the same name makes sense to me, though. Cheers. Jeremy

On 3 May 2016 at 05:19, Rob Stewart wrote:

On May 2, 2016 5:30:40 AM EDT, Jeremy Murphy < jeremy.william.murphy@gmail.com> wrote:

...

A variadic template version with the same name makes sense to me, though.

There's some detail missing. How is your iterated GCD function different from a variadic gcd()?

Only in the type signature. It's fairly obvious but I'll make it explicit. Variadic gcd takes n things of type T and returns a T: f(T, ..., T) -> T whereas "iterated" gcd takes a thing (Range) or things (iterators) that denote a range of Ts: g(U) -> T g(U, U) -> T So yes, in the implementation, variadic gcd is an iterated binary operation (with special case of n=1). But the semantics are different because of how the input is interpreted, which for me is sufficient reason to give them different names. So it's probably more accurate that I say it's gcd over an iterator range that I think should have a different name to 'plain' gcd (which is iterated in a variadic version). Which makes the "_range" suffix all the more sensible. Cheers. Jeremy