Eachcol get name property

General Usage
1

How can I get the name property when using eachcol?

1 Like
2

You are talking about DataFrames.eachcol right? It seems that in a loop you just get an array.

julia> dfq = DataFrame(a = [3, 4], b = ["ali", "baba"])
2×2 DataFrame
│ Row │ a     │ b      │
│     │ Int64 │ String │
├─────┼───────┼────────┤
│ 1   │ 3     │ ali    │
│ 2   │ 4     │ baba   │

julia> for col in eachcol(dfq)
       println(col, ' ', typeof(col))
       end
[3, 4] Array{Int64,1}
["ali", "baba"] Array{String,1}

So what about:

julia> for name in names(dfq)
       col = dfq[name]
       println(name, ": ", col)
       end
a: [3, 4]
b: ["ali", "baba"]
3

You can also iterate through pairs(eachcol(df))

julia> for (name, col) in pairs(eachcol(df))
       @show name
       end
5 Likes
4

I did something similar. I want to know how to access the name property.

5

Ua! Hard to get it. Don’t know how it works, but yes, this works!

6

Here’s how it works:

julia> using DataFrames

julia> df = DataFrame(rand(10, 10));

julia> @which pairs(eachcol(df))
pairs(itr::DataFrames.DataFrameColumns) in DataFrames at /home/nils/.julia/packages/DataFrames/cdZCk/src/abstractdataframe/iteration.jl:200

so the pairs method being called is implemented here:

Base.pairs(itr::DataFrameColumns) = Base.Iterators.Pairs(itr, keys(itr))

and the DataFrameColumns object comes from here:

eachcol(df::AbstractDataFrame) = DataFrameColumns(df)
4 Likes