stack wrong makeunique suggestion

[sylvaticus]

When trying to stack a df where a column name is "variable", the stack command suggests to use "makeunique", but then this lead to an error:

julia> df = DataFrame(
           foo      = ["a","a","b"],
           variable = ["a","a","a"],
           v1       = [1,1.5,2],
           v2       = [10,11.5,13.5]
       )
3×4 DataFrame
 Row │ foo     variable  v1       v2      
     │ String  String    Float64  Float64 
─────┼────────────────────────────────────
   1 │ a       a             1.0     10.0
   2 │ a       a             1.5     11.5
   3 │ b       a             2.0     13.5

julia> stack(df)
ERROR: ArgumentError: Duplicate variable names: :variable. Pass makeunique=true to make them unique using a suffix automatically.
Stacktrace:
  [1] make_unique!(names::Vector{Symbol}, src::Vector{Symbol}; makeunique::Bool)


julia> stack(df,makeunique=true)
ERROR: MethodError: no method matching stack(::DataFrame, ::Vector{Int64}, ::InvertedIndex{Vector{Int64}}; makeunique::Bool)
This error has been manually thrown, explicitly, so the method may exist but be intentionally marked as unimplemented.

Likely the message should say instead to specify the column to host the values to stack with the keyword argument variable_name.

[bkamins]

Agreed!