Some geoms always appear below others

[jvgogh]

Rendering of violin plots with box plots (to provide median & stddev) is in wrong order when using ggplotly.

library(ggplot2)
library(plotly)

p <- ggplot(iris,aes(Species,Sepal.Length))+
  geom_violin()+geom_boxplot(width=.1)
q <- ggplotly(p) 

the ggplot (p) will render as expected with violin plots:

the ggplotly (q) will render with the box plot behind the violin plot (and larger than expected):

[ifellows]

This also affects other geoms. Is there a plan to address this issue?

###Sample gaussian distributions with 1, 2 and 3 modes.
df <- data.frame(
  "Distribution" = c(rep("Unimodal", 500),
                     rep("Bimodal", 250),
                     rep("Trimodal", 600)),
  "Value" = c(rnorm(500, 6, 1),
              rnorm(200, 3, .7), rnorm(50, 7, 0.4),
              rnorm(200, 2, 0.7), rnorm(300, 5.5, 0.4), rnorm(100, 8, 0.4))
)
library(ggplot)
library(ggforce)

p <- ggplot(df, aes(Distribution, Value, color=Distribution))+ geom_sina() + geom_boxplot( outlier.alpha=0)
p
p %>% ggplotly()

Note also that the outliers are incorrectly printed (see the black dots) when they should be transparent and the legend is incorrect. Related: #1114

ggplot2:

plotly:

[cpsievert]

It seems the root issue here is some plotly.js traces are always drawn below other traces https://github.com/plotly/plotly.js/issues/3748

These examples should automatically fix themselves if and when plotly.js respects the ordering of traces...

[ifellows]

Thanks for looking into this. Is the outlier printing issue something that would automatically be fixed, or it that different?

[cpsievert]

That's a different problem that I have more control over.