Proper way to initiate an array with `missing`

General Usage
1

pretty simple question with no documentation that I could find. is this the right way to get a 3x1 array of missings that at some point I may want to fill with with Int?

julia> x = Union{Int64, Missings.Missing}[missing for i in 1:3]
3-element Array{Union{Int64, Missings.Missing},1}:
 missing
 missing
 missing

julia> x[2] = 1
1

the other way I could think is a bit contrived, given that I would like to initiate to missing

julia> x=Vector{Union{Int64, Missings.Missing}}(3)
3-element Array{Union{Int64, Missings.Missing},1}:
 #undef
 #undef
 #undef

julia> x[2] = 1
1

julia> x[1] = missing
missing

julia> x
3-element Array{Union{Int64, Missings.Missing},1}:
    missing
   1       
 #undef

cheers

2 
julia> Vector{Union{Int,Missing}}(missing, 3)
3-element Array{Union{Missing, Int64},1}:
 missing
 missing
 missing
3 Likes
3

Perhaps

convert(Array{Union{Missing, Int}}, fill(missing, 3, 1))

or

fill!(Array{Union{Missing, Int}}(3, 1), missing)

but it is not much better. I wish fill would take a type argument.

4

Either

using Missings
missings(Int64, 10)

or:

using DataArrays
DataArray(Int64, 10)

if you want a DataArray

8 Likes
5

How about

missings(Int64, 3)

This should work for any type.

6

i like that most, thanks! had no idea how to use the Missings.jl package.

8

A related question, will Missing in a DataFrame, produce an empty box?

9

What does “empty box” mean? Also I don’t really see how the question is connected, better to start a new thread than to post in a 4-year old one.

10

A box that contains no value. Maybe it should be it’s own topic.

11

Leaving the “box” aside, why wasn’t the quoted answer accepted as the more Julian solution? Thank you.

12

It would be nice to see

missings([T=Float64,] dims::Tuple)

added to Base