Use "label" as a prefix when printing from dataframes using "group"

[sylvaticus]

Consider the following example:

using DataFrames, Plots, StatPlots
df = DataFrame(
  product = ["orange","orange","orange","orange","banana","banana","banana","banana"],
  year = [2010,2011,2012,2013,2010,2011,2012,2013],
  prod = [120,150,170,160,100,130,165,158],
  cons = [70,90,100,95,   80,95,110,120]
)
plotlyjs() 
mycolors = [:blue, :red ]
fruits_plot = plot(df, :year, :prod, group=:product, linestyle = :solid, linewidth=3, linecolor=:match, color_palette = mycolors)
fruits_plot = plot!(df, :year, :cons, group=:product, linestyle = :dot, linewidth=3, linecolor=:match, color_palette = mycolors)

This print (forget here the colour issue):

It would be nice to have the possibility to label the first group of series differently than the second one, while still retaining the individual serie label.
If I add to the first group of series label="Production of " or to the second one label="Consumption of " this will be substituted to the one coming from the dataframe using group.
This proposal is to consider label as a prefix and concatenate it with the label coming from the dataframe when group is used.
Alternatively a label_suffix and/or a label_postfix attribute could be introduced.

[mkborregaard]

I can see how it's helpful in the current example, but not that it is a general functionality we need. It would be nice with a general design for how legends should work that considers all the possibilities, but this does not seem generally useful.

In the current example, you can just do label = ("Production of " * sort(unique(df[:product]))') to get the behaviour you want.

[sylvaticus]

Hello, in Julia 0.6 this workaround doesn't work any more.. nor using reshape(sort(unique(df[:product])), (1, :)) works, as the object produced
(1×2 Base.ReshapedArray{String,2,DataArrays.DataArray{String,1},Tuple{}}) is not the one expected by plots..

[daschw]

label = "Production of " .* reshape(sort(unique(df[:product])), (1, :)) works for me.

[sylvaticus]

yep that works, thank you..