Colors a pain to define in qbarplot

[vertesy]

Description

Colors a pain to define in qbarplot.
You can't easily provide a color vector as in wbarplot()

Details

qbarplot(xxx, col = 1)

qbarplot(xxx, col = 1)
qbarplot(xxx, col = 2) # has no effect
qbarplot(xxx, palette_use = "aaas", col = 2) # changes color

Alternating colors

qbarplot(xxx, col = 2:1)

To explicitly define colors you need to define it in palette

qbarplot(xxx, col = "brown")
qbarplot(xxx, col = c("brown", "orange")) # Wtf: now it reverts back to default palette and alternating colors

The above is the equivalent of:

qbarplot(xxx, col = 1:2) # and of qbarplot(xxx, col = 1:2, palette_use = "jco")

To explicitly define a SINGLE color you need to define it in palette

qbarplot(xxx, palette = "orange", col = 2) # has no effect

You cannot explicitly define a SET OF COLORS in qbarplot, you need to define it in palette and use col = 1:length(xxx)

qbarplot(xxx, palette = rich.colors(n = length(xxx)))

So the correct way to define a set of colors is:

qbarplot(xxx, palette = rich.colors(n = length(xxx)), col = 1:length(xxx))

This is super weird, and cumbersome, but it is probably due to the fact that qbarplot is a wrapper around ggpubr.

[vertesy]

Solution (not implemented)

qbarplot should repurpose the stupid palette and color options.

col

1. NULL by default
2. If provided, AND == len(xxx)
   palette = rich.colors(n = length(xxx)), col = 1:length(xxx))

[vertesy]

Seems to need a larger revamp

   col = as.character(1:3)[1],
    xlab = ...
    ...) {

# later

  df[["colour"]] <-
    if (length(col) == length(vec)) {
      as.character(col)
    } else if (hline & filtercol != 0) {
      if (filtercol == 1) (df$"value" > hline) else if (filtercol == -1) (df$"value" < hline)
    } else {
      as.character(rep(col, length(vec))[1:length(vec)])
    }