Support for plotly about ggnewscale HOT 4 CLOSED

I have no idea how plotly works, but form what little I've used it, I think it basically hand-crafts translations of each ggplot2 geom to the corresponding plotly mark. If that's the case, then I don't think it'd be possible because the way ggnewscale works is by creating new geoms with new aesthetics for each new scale.

from ggnewscale.

I've skimmed the relevant source code.

For this specific case you can use:

library(ggplot2)
library(ggnewscale)
# Equivalent to melt(volcano)
topography <- expand.grid(x = 1:nrow(volcano),
                          y = 1:ncol(volcano))
topography$z <- c(volcano)

# point measurements of something at a few locations
set.seed(42)
measurements <- data.frame(x = runif(30, 1, 80),
                           y = runif(30, 1, 60),
                           thing = rnorm(30))

# Defines a the to_basic method to the NewGeomContour class and renames the relevant variables. 
# Very hacky. Don't know how it will generalise. Will break if you add more scales.  
to_basic.NewGeomContour <- function(data, prestats_data, layout, params, p, ...) {
  if (!"fill" %in% names(data)) data$fill <- NA
  colnames(data)[colnames(data) == "colour_new"] <- "colour"
  colnames(data)[colnames(data) == "colour_new_plotlyDomain"] <- "colour_plotlyDomain"

  plotly:::prefix_class(data, "GeomPath")
}


ggplot(mapping = aes(x, y)) +
  geom_contour(data = topography, aes(z = z, color = stat(level))) +
  scale_color_viridis_c(option = "D") +
  new_scale_color() +
  geom_point(data = measurements, size = 3, aes(color = thing)) +
  scale_color_viridis_c(option = "A") -> p

plotly::ggplotly(p)

from ggnewscale.

I have no idea how plotly works, but form what little I've used it, I think it basically hand-crafts translations of each ggplot2 geom to the corresponding plotly mark. If that's the case, then I don't think it'd be possible because the way ggnewscale works is by creating new geoms with new aesthetics for each new scale.

I understand the same as you do, that's why I assumed that it should not be so easy, but just in case I was asking you.

My concern arose from this: ggplotly no me grafica el mismo grafico que ggplot

from ggnewscale.

 if (!"fill" %in% names(data)) data$fill <- NA
  colnames(data)[colnames(data) == "colour_new"] <- "colour"
  colnames(data)[colnames(data) == "colour_new_plotlyDomain"] <- "colour_plotlyDomain"

I tried something very similar, an automatic wrapper of each new geom to the plotly function, obviously it solves the warning issue but does not take into account the logic of ggnewscale.

Thanks for your time Elio

from ggnewscale.