How to force plotly R to plot missing values with category axis - r-plotly

Question

I want to plot a simple bar chart in plotly R and by default it will skip the NA observations. In ggplot we have a parameter to disable the default NA removal, or we can customize the axis limits. However I cannot get it done in plotly.
dt_plot <- data.frame(categories = letters[1:10], values = c(rep(NA_integer_, 3), 1:5, rep(NA_integer_, 2)))
plot_ly(data = dt_plot) %>%
add_bars(x = ~categories, y = ~values)
I want to show the x axis as the consistent letters 1:10, because I'm actually including this plot in a shiny app with dynamic data selection. Some data have values in all x values, some only have values in a subset. I want to keep the plot to be consistent and always show complete x values.
There is a similar question here, but the answer doesn't apply to my case, because I'm using a category type x axis:
https://plot.ly/r/reference/#layout-xaxis
If the axis type is "category", it should be numbers, using the scale where each category is assigned a serial number from zero in the order it appears.
I tried different range combinations but it didn't work. plotly seemed always remove NA observations first.
There is a related issue where Carson Sievert suggested a hack, but it didn't really work me either.
# this do show all the x values but the plot is wrong
layout(xaxis = list(type = "category", tickvals = 1:10/10, ticktext = letters[1:10], range = c(0, 1)))
By inspecting the plot object, it looks like the NA data is removed before constructing the plot:
{
"visdat": {
"7abc7354f619": ["function () ", "plotlyVisDat"]
},
"cur_data": "7abc7354f619",
"attrs": {
"7abc7354f619": {
"alpha_stroke": 1,
"sizes": [10, 100],
"spans": [1, 20],
"x": {},
"y": {},
"type": "bar",
"inherit": true
}
},
"layout": {
"margin": {
"b": 40,
"l": 60,
"t": 25,
"r": 10
},
"xaxis": {
"domain": [0, 1],
"automargin": true,
"range": [0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9],
"title": "categories",
"type": "category",
"categoryorder": "array",
"categoryarray": ["a", "b", "c", "d", "e", "f", "g", "h", "i", "j"]
},
"yaxis": {
"domain": [0, 1],
"automargin": true,
"title": "values"
},
"hovermode": "closest",
"showlegend": false
},
"source": "A",
"config": {
"showSendToCloud": false
},
"data": [
{
"x": ["d", "e", "f", "g", "h"],
"y": [1, 2, 3, 4, 5],
"type": "bar",
"marker": {
"color": "rgba(31,119,180,1)",
"line": {
"color": "rgba(31,119,180,1)"
}
},
"error_y": {
"color": "rgba(31,119,180,1)"
},
"error_x": {
"color": "rgba(31,119,180,1)"
},
"xaxis": "x",
"yaxis": "y",
"frame": null
}
],
...

Answer 1

There is info in reference: https://plotly.com/r/reference/layout/xaxis/#layout-xaxis-range, but not so obvious.
But first of all: categorical variable needs to be factor.
Second you need to specify range from 0 to number of levels -1. To better looking use +-0.5 on ends
dt_plot <- data.frame(categories = letters[1:10], values = c(rep(NA_integer_, 3), 1:5, rep(NA_integer_, 2)))
dt_plot$categories <- as.factor(dt_plot$categories)
plot_ly(data = dt_plot) %>%
add_bars(x = ~categories, y = ~values)%>%
layout(xaxis = list(range = list(-0.5, 9.5)))