For example, sum all the digits of 1253
, 1 2 5 3
which is 11
. This is two digits, so sum those again to get 2
. The final number left is a single digit.
This is what I have so far:
defmodule Kata do
def digital_root(n) do
n |> Integer.digits() |> Enum.reduce(0, &Kernel. /2)
end
end
n = 1253
CodePudding user response:
You can use a multi clause function for this: in the first one, return n
unchanged if it's a single digit. Otherwise, compute the sum of digits (like you've already figured out) and recursively call the function on it again. Also, your reduce can be replaced with Enum.sum/1
.
defmodule Kata do
def digital_root(n) when n < 10, do: n
def digital_root(n) do
n |> Integer.digits() |> Enum.sum() |> digital_root()
end
end
Test:
iex(1)> Kata.digital_root(0)
0
iex(2)> Kata.digital_root(1)
1
iex(3)> Kata.digital_root(9)
9
iex(4)> Kata.digital_root(91)
1
iex(5)> Kata.digital_root(912)
3
iex(6)> Kata.digital_root(9123)
6
iex(7)> Kata.digital_root(1253)
2
CodePudding user response:
def digit_sum(number) when number < 10, do: number
def digit_sum(number) when is_integer(number) do
digit_sum(
for digit <- Integer.digits(number), reduce: 0 do
acc -> acc digit
end
)
end
I like Dogbert's better.
CodePudding user response:
Just for fun, here's a version that doesn't use Integer.digits/2
or Enum
, and instead uses div/2
and rem/2
to calculate the least significant digit for each iteration:
defmodule Kata do
# Function header
def digital_root(to_sum, acc \\ 0)
# Base case. All digits added, acc is a single digit
def digital_root(0, acc) when div(acc, 10) == 0, do: acc
# Finished a round. All digits added, acc is multiple digits. Sum again.
def digital_root(0, acc), do: digital_root(acc, 0)
# Split off the least significant digit, add it, and recurse
def digital_root(to_sum, acc) do
digit = rem(to_sum, 10)
next_to_sum = div(to_sum, 10)
digital_root(next_to_sum, acc digit)
end
end
Output:
iex> Kata.digital_root(1523)
2
In a benchmark, this version is twice as fast as dogbert's, and three times faster than 7stud's answers.
Name ips average deviation median 99th %
adam 14.26 M 70.13 ns ±1898.51% 66.70 ns 83.30 ns
dogbert 7.08 M 141.28 ns ±14510.93% 125 ns 167 ns
sevenstud 4.83 M 206.98 ns ±15193.15% 167 ns 292 ns
Comparison:
adam 14.26 M
dogbert 7.08 M - 2.01x slower 71.14 ns
sevenstud 4.83 M - 2.95x slower 136.85 ns
Operating System: macOS
CPU Information: Apple M1 Pro
Number of Available Cores: 10
Available memory: 16 GB
Elixir 1.13.4
Erlang 24.3.4