How To Format A Number to Precision in Elixir?

There isn't anything like this in Elixir's standard library as far as I know, but you can use :io_lib.format/2 from Erlang to print a float with specific number of digits after the dot:

iex(1)> :io_lib.format("~.3f", [123.101])
['123.101']
iex(2)> :io_lib.format("~.6f", [123.101])
['123.101000']
iex(3)> :io_lib.format("~.3f", [123.0])
['123.000']
iex(4)> :io_lib.format("~.6f", [123.0])
['123.000000']

The returned value is an iolist, which can be converted to a binary using IO.iodata_to_binary/1 if required (many functions that take a binary can accept an iolist directly as well, e.g. IO.puts/1):

iex(5)> :io_lib.format("~.6f", [123.0]) |> IO.iodata_to_binary
"123.000000"

~.0f doesn't work but for that you can simply call trunc/1:

iex(6)> trunc(123.123)
123

Just want to supply a alternative to Dogbert's excellent answer.

It is also possible to use :erlang.float_to_binary/2

ex.

iex(5)> :erlang.float_to_binary(123.101, [decimals: 6])
"123.101000"

iex(6)> :erlang.float_to_binary(123.0, [decimals: 3])
"123.000"

iex(7)> :erlang.float_to_binary(123.101, [decimals: 1])
"123.1"

iex(8)> :erlang.float_to_binary(123.000, [decimals: 0])
"123"

I've used the decimal package for this previously

iex(6)> 123 |> Decimal.new() |> Decimal.round(3) |> Decimal.to_string()
"123.000"

In my case the data was already in Decimal format after being queried from the database.

These answers are mostly odd. First, she didn't say anything about converting to a string. Second, there is certainly no need to jump into Erlang land!

In order:

Starting number: 123.101

Ending number: 123.101000 # Adding 3 digits to the precision of a float.

iex> 123.101 |> Decimal.from_float() |> Decimal.round(6)
#Decimal<123.101000>


Starting number: 123

Ending number: 123.000 # Adding 3 digits to the precision of an integer.

iex> 123 |> Decimal.from_float() |> Decimal.round(3)    
#Decimal<123.000>


Starting number: 123.101

Ending number: 123.1 # removing precision

iex> 123.101 |> Decimal.from_float() |> Decimal.round(1)
#Decimal<123.1>


Starting number: "123.000" # must be a string as 123.000 == 123.0

Ending number: 123 # removing precision

iex> "123.000" |> Decimal.new() |> Decimal.round(0)
#Decimal<123>

Once you have a number in Decimal, you can do what you want, e.g. convert to a float, convert to a string, etc.

iex> 123.101 |> Decimal.new() |> Decimal.round(6) |> Decimal.to_string()
"123.101000"

iex> 123.101 |> Decimal.new() |> Decimal.round(6) |> Decimal.to_float() 
123.101

iex> 123.101 |> Decimal.new() |> Decimal.round(0) |> Decimal.to_integer()
123

As an aside, I recommend working only in Decimal until you've finished any data manipulation, e.g. using Decimal.mult(num1, num2), Decimal.div(num1, num2), etc. Decimal rocks!