Plotly: Support for arbitrary coordinate systems

Created on 9 Feb 2017 · 15Comments · Source: ropensci/plotly

It might be possible to support _all_ ggplot2's coordinate systems via plotly.js' cartesian coordinates:

set layout.axisid.showgrid = false
hack together custom graticule using line shapes?

enhancement ggplotly

Source

cpsievert

Most helpful comment

Supporting pie charts (coord_polar) would be awesome!

Seyphaton on 1 Apr 2020

❤9

All 15 comments

Hi @cpsievert, I just wanted to check in on the functionality for polar coordinates; I'm having the same error as #760.

Thank you!

jules32 on 29 Jun 2017

Has this issue been solved - I have code that used to work for a pie chart that will not work. I can get the geom_bar chart working but once I add the coord_polar() it breaks. Any help is appreciated.
thanks

nicolescoggins on 2 Aug 2017

@nicolescoggins can u please provide a minimal example?

cpsievert on 3 Aug 2017

Hi @cpsievert ,

I knocked up a quick minimal example, using airquality.

This draws with ggplot but when I go to use ggplotly, I get the error below.

ggplotly(ggplot(airquality, order = xp) +
  geom_bar(aes(x = Month,
               y = Temp,
               fill = Solar.R)
           ,
           position = "fill",
           stat = "identity",)  +
  coord_polar("y")
)

Error in zero_range(to) : x must be length 1 or 2
In addition: Warning messages:
1: In min(z$x.range) : no non-missing arguments to min; returning Inf
2: In max(z$x.range) : no non-missing arguments to max; returning -Inf
3: In min(z$y.range) : no non-missing arguments to min; returning Inf
4: In max(z$y.range) : no non-missing arguments to max; returning -Inf

gsainsbury86 on 27 Aug 2017

👍6

hi i'm getting the same error while using ggplotly for pie chart.
can anyone pls help?

Asad55555 on 22 Dec 2017

@cpsievert I am having the same issue as @gsainsbury86

DiDeoxy on 28 Aug 2019

Me too

ChrisWoodsSays on 26 Dec 2019

Also having the same issue, my MWE is available here and is as follows

library(ggplot2)
library(plotly)
# Create fake data
df <- data.frame(
           genre=c("Pop", "HipHop", "Latin", "Pop", "Pop", "HipHop"), 
           artist=c("ArianaGrande", "ChrisBrown", "DaddyYankee", "EdSheeran", 
                    "LewisCapaldi", "ShawnMendes")
      )
# Create a ggplot bar plot in polar coordinates
p <- ggplot(data=df) + 
      geom_bar(aes(x=genre, fill=artist)) + coord_polar()
# Wrap it around plotly
ggplotly(p)