How to put javascript function in json rendered by e_inspect()?

[rdatasculptor]

Consider this chart:

USArrests |>
  e_charts(Assault) |>
  e_scatter(Murder) |>
  e_tooltip(formatter = htmlwidgets::JS("
          function(params){
            return('Assault: ' + params.value[0] + 
          '<br />Murder: ' + params.value[1])
          }
        ") )

The chart contains tooltips:

So what if I change the chart first in a json by using e_inspect() and change it back again to an echart by using echarts_from_json()?
like this:

USArrests |>
  e_charts(Assault) |>
  e_scatter(Murder) |>
  e_tooltip(formatter = htmlwidgets::JS("
          function(params){
            return('Assault: ' + params.value[0] + 
          '<br />Murder: ' + params.value[1])
          }
        ") ) |> 
  e_inspect(json = TRUE) |>
  echarts_from_json()

this is the result:

Obviously the tooltip javascript function is not recogized anymore.

Any ideas about how to deal with this? Is there a way (in R) to store the javascript function of the tooltip in the json in such a way that it can retrieved properly after changing it back to an echart?

[munoztd0]

As always nice catch/riddle @rdatasculptor. Just added new feature in #561 so just install the github version of echarts4r and add the jswrapper = TRUE flag

library(echarts4r)

plot <- USArrests |>
  e_charts(Assault) |>
  e_scatter(Murder) |>
  e_tooltip(
    formatter = htmlwidgets::JS("function(params){return('Assault: ' + params.value[0] +'<br />Murder: ' + params.value[1])}        ")
    ) |> 
  e_inspect(json = TRUE, pretty =TRUE) |>
  echarts_from_json(jswrapper = TRUE)

plot

Let me know how it goes !

[rdatasculptor]

great work @munoztd0 !!! it works like a charm :) You solved this one really fast. Very cool.

Another (entirely different) issue is that the json doesn't contain the renderer parameter if you set it as "svg". Do you think that can be added to the json as well somehow? Maybe I will file another issue for that.

[munoztd0]

Can you please open a new issue for this

[rdatasculptor]

@munoztd0 Take a look at what @helgasoft did: helgasoft/echarty#3 (comment)
Very interesting approach. I will dig into this. Not sure yet if echarts4r can benefit from it.

Edit. Don't think it is much different from your approach 🧐, but cool anyway!