13

I'm plotting results from various British elections in Leaflet and ran into a bit of a problem with legends.

For the various results in the general election I'm using the same colour function with different domain for the different data (the yellow-> purple scale in the picture)

This is created with (for the first two as examples):

labvotescols <- colorNumeric(
  c("Yellow", "Purple"),
  domain = Westminster$LabourVotes,
  ukipvotescols <- colorNumeric(
    c("Yellow", "Purple"),
    domain = Westminster$UKIPVotes,

and so on...

Currently I have the legend 

map = map %>% addLegend("bottomright", pal = ukipvotescols, values = Westminster$UKIPVotes,
                    title = "(e.g.) % voting UKIP at GE2015",
                    opacity = 1)

as one example of this, but really I'd like to get rid of all the values on the legend and just have "less" at the yellow end and "more" at the purple end. Is this possible? map picture showing the scale

I tried playing around and then googling but to no avail.

amonk
  • 1,769
  • 2
  • 18
  • 27
Robert Hickman
  • 869
  • 1
  • 6
  • 22
  • I have a feeling that given that the breaks/labels etc are computed by `addLegend` for a gradient, it won't let you put a string in there. – Akhil Nair Jul 02 '16 at 14:38
  • I gave a similar answer [here](https://stackoverflow.com/a/40277470/5977215) and [here](https://stackoverflow.com/a/36098407/5977215) . Maybe something you can pull together from those solutions. – SymbolixAU Sep 22 '17 at 05:54

1 Answers1

16

You could change it from a yellow -> purple scale and make your own scale:

map %>%
  addLegend("bottomright", 
            colors = c("#FFC125",  "#FFC125", "#8A4117", "#7D0552", "#571B7E"),
            labels = c("less", "", "", "", "more"),
            title = "(e.g.) % voting UKIP at GE2015",
            opacity = 1)

If you get the correct colors then it should look similar. Not the answer you were looking for, but it's a good workaround. Your output would look like this:

enter image description here

Spend more time looking for better color transition and you could get a legend that looks similar to the yellow-purple color pallet you have up top.

bathyscapher
  • 1,615
  • 1
  • 13
  • 18
jpf5046
  • 729
  • 7
  • 32
  • 2
    `mypal <- colorNumeric(c("Yellow", "Purple"), domain = df$values, na.color = "Black") ` for the palette , and `map = map %>% addLegend("bottomright", pal = mypal, values = df$values, title = "quant data title", opacity = 1, labFormat = function(type, cuts, p) { n = length(cuts) cuts[n] = "more" for (i in 2:(n-1)){cuts[i] = ""} cuts[1] = "less" paste0(cuts[-n], cuts[-1])}) ` for the legend, was working for me. I'll leave it here in case anyone comes looking. I'm sure there's some problem with it but so far its been holding up. – Robert Hickman Jul 07 '16 at 11:48
  • 1
    @John Friel, Good answer, OP should accept and close this question out. – nate Oct 10 '18 at 20:33