topic Re: Make legend colours consistent while using slicer (R Visuals) in Developer

Make legend colours consistent while using slicer (R Visuals)

fgmh — Thu, 26 Jul 2018 13:54:19 GMT

Hello,

I have created my own R visuals using ggplot2, which automatically chooses a colour scheme for my legend. However, when I use the slicer to filter the data, the colours are reassigned, making it visually a bit disorienting to the user.

For example, before using the slicer:

A - Red

B - Green

C - Blue

D - Orange

After using the slicer to select C and 😧

C - Red

B - Green

Is there a way to make it such that after selecting C and D in the slicer, this is what I get:

C - Blue

D - Orange

Thank you!

Re: Make legend colours consistent while using slicer (R Visuals)

v-chuncz-msft — Sat, 28 Jul 2018 08:48:38 GMT

@fgmh,

Based on my test, you may use scale_manual.

Re: Make legend colours consistent while using slicer (R Visuals)

fgmh — Sat, 28 Jul 2018 10:08:00 GMT

Hello Sam,

Thank you for your suggestion! I realise that I was not very clear in my question, sorry about that!

I am trying to create a Power BI template that is as generic as possible - the number of legend entries and the names of the legend entries will change each time, so I cannot hard code something like:

dd.col <- c("#FF0000FF", "#FF8B00FF", "#E8FF00FF", "#5DFF00FF")
names(dd.col) <- c("A", "B", "C', "D")

and call "scale_fill_manual("Legend", values = dd.col)"

in ggplot

I have tried to create a new colour palette,

# Create color vector
dd.col <- rainbow(length(unique(dataset$LegendEntries)))
names(dd.col) <- unique(dataset$LegendEntries)

Like this, but the colours reset when I choose a subset of information using the Slicer in Power BI.

Would you know of a way to work around this? Thank you!

Regards,

Fel