As the title says, make a classic FizzBuzz function or method without using if/else
(or equivalents like ternaries, ?:
, sneaky).
Specifically:
- The function should accept one argument, assume it will always be a positive integer.
- The function should return a string (or something coercible into a string in loosely typed languages) according to the following rules:
- If the given number is divisible by
3
, then returnFizz
- If the given number is divisible by
5
, then returnBuzz
- If the given number is divisible by both
3
and5
, then return the combinationFizzBuzz
- If the given number is none of those things, then return the given number
- If the given number is divisible by
The expected outputs for the first fifteen numbers in order is:
1,2,Fizz,4,Buzz,Fizz,7,8,Fizz,Buzz,11,Fizz,13,14,FizzBuzz
Hard mode
Do this without "secret" conditionals like ||
and &&
(in loosely typed languages like JavaScript) or null
coalescing operators like ??
or null
safe operators like ?.
or &.
.
Also no looping constructs that could be abused into a conditional like while
or for
.
Hint number 1
I tagged functional
on this post because functional programming can be used to solve this.
Hint number 2
I tagged oop
on this post because the original object oriented concepts of message passing can be used to solve this.
Post below with your answers!
Think this is impossible? I'll post my own answers in both FP and OOP styles next week.
(If you were hoping for the next installment of my With Only CSS series, styling radio buttons, that's coming on Monday, follow me so you won't miss it!)
Latest comments (48)
A bit late to the party...
Welcome to the party!
Very nice! I feel like this is actually pretty similar to my own Object-oriented approach (if you wanted to read):
Unconditional Fizzbuzz: an Object-oriented approach
Nathan Kallman ・ Jul 6 '20 ・ 3 min read
Has someone done this yet? It's really very fast.
There's a ternary there, I admit.
Nice! But
test
uses a ternary?:
(which is just a different way to spellif
in my opinion)If you can get rid of that I think you'll have a solution!
I can't bring myself to be terribly interested in re-implementing language features like boolean or ternary. I loved seeing some of the really outlandish solutions like that un-sane RegEx that was posted. I enjoyed seeing the contortions done to avoid conditionals, too, but that's not something I'm into for it's own sake. I'm happy to fail at it.
There is a meaningful distinction between ternary and
if
.if
is a wilderness with few boundaries, as a form of Many-Valued Logic with an arbitrary number of values. Ternary is a specific subset of Many-Valued Logic; Three-Valued Logic. The Three-Valued Logic of ternary operators provides specific limitations that are not necessarily implied by the N-Valued Logic of theif
statement. In most programming languages ternary is an operator, andif
is a control structure block. Operators can be used in expressions, which is generally not true ofif
statements. Ternary is usually slightly computationally more expensive thanif
but has some advantages in terms of limiting complexity.In real-world code I'm going to write an
if
(withoutelse if
orelse
if I can help it) about 95% of the time. It is both more readable and more performant in most cases. But when I need an expression that can make a decision between boolean and a maybe, I reach for ternary.If you wanted to argue that
switch
is basicallygoto
but potentially worse due to unintended fallthrough, then we'd have something to agree on.The magic of the fizzbuzz problem is, of course, that it helps us understand in more granular detail that our design decisions, and design non-decisions, come with unavoidable trade-offs. It is a meditation on (a horrible term)
non-functional requirements
.How about this:
Nice!
This could probably get simplified (I've seen the one-liner of this one), but I still wanted to post mine
I am posting for a friend who found some absolutely mind blowing ideas.
First a one-liner that extends my destructuring idea.
secondly a solution built around a regex from hell/heaven
holy COW did I find out a lot of things about object destructuring while making this. Thanks for the challenge, Nathan. It was so great to notice how instinctively I reach for a conditional.
I'm glad you liked it! I think its good to do a little self reflection on the way we do things every once in a while.
And WOW is that an impressive use of destructuring in JS. I knew about each of those things independently; I never have thought about putting them all together like that!
I'm thinking of an absolutely egregious version using try and catch, but I'm going to go ahead and assume that try/catch would be a "surrogate conditional"
😂 it probably is, but now I want to see it implemented with just try/catch!
Well done! Clever use of
Math.sign
to select either the first or second element from an arraySlightly shorter without bothering to call out for variable potential conditions and using anons:
Happy Juneteenth! I love everyone's approaches here, it's fun to see the way each person thinks through the problem.
I'll be posting my answers on Monday (along with the next installment of With Only CSS) so encourage your friends to give this a shot over the weekend (or take another go yourself)!
Follow me and react with a 🔖 so you remember to come back and see what I post here.
Drop a 🦄 or ❤️ if you'd be interested in some follow up posts going more in depth into each (OO/FP) solution: how they work; how they are similar; and how they are different.
Thanks everyone who took the time to post a solution to my little challenge; I'll see you all on Monday!
You can see both solutions in action on this codepen (planning one final post doing a compare/contrast between the two approaches)
And here is my object-oriented approach (deeper article on it to come soon)
Similarly to the functional approach; the main part of this solution is defining "true" and "false" as objects. But now instead of parameters, the objects have the same interface and return one or the other of the attributes set on the object.
As promised, here is my functional approach to solving this (deeper article on it to come soon)
Similar to Heiker's solution above; the main part of this solution is defining "true" and "false" as functions taking the same two parameters but returning one or the other.
Still not perfect, but this one comes with a special thanks to Mr. Kevlin Henney
Nice! I like how it uses
false
andtrue
as keys on an object to select the resulting string.If you can get rid of the
||
usage, then this will meet the hard mode requirements...Well Mr. Henney has a great answer:
edit: but it still has a conditional...
Excellent! I think you've checked off all the boxes. Thanks for submitting!
I think your
.get
is a secret conditional. (It will return the value at the givenkey
or the default IF the value is undefined)If you can show how to implement
.get
without anif
then I think this is a great answer!You are referring to the following clause:
My understanding is that this applies to the usage of such operators as conditionals when they are applied to non-boolean types. For example:
In my code, it is used strictly on Boolean values, which I don't think was against the rules.
It uses it in a strictly boolean sense, which I'll allow. What isn't allowed is the loose/early-return way JS can use
&&
; a contrived example that would not pass hard mode:(an easy litmus test: does changing the order of the
&&
expression change the result?)Some comments may only be visible to logged-in visitors. Sign in to view all comments.