Access data created from reactive function to define reactiveValues in Shiny

Question

I'm exploring the possibilities with interactive ggplot2 in shiny. Inspired by this I created a shiny app that exclude points from a dataset and plots the data where the excluded points are of a different color.

app.R

library(shiny)
library(ggplot2)
server<-function(input, output) {
  data <- reactive({
    set.seed(10)
    df=data.frame(x=rnorm(100),y=rnorm(100))
    df
  })

  vals<-reactiveValues(keeprows=rep(TRUE, 100))

  output$plot1 <- renderPlot({
    df=data()
    keep=df[vals$keeprows, ,drop=FALSE]
    exclude=df[!vals$keeprows, ,drop=FALSE]
    plot=ggplot(data=keep,aes(x,y))+geom_point()+theme_bw()+
      geom_point(data=exclude,fill=NA,col="black",alpha=0.75,shape=21)
    plot
  })

  observeEvent(input$plot1_click,{
    df=data()
    res <- nearPoints(df, input$plot1_click, allRows = TRUE,threshold=5)
    vals$keeprows <- xor(vals$keeprows, res$selected_)

  })

}

ui <- fluidPage(

  titlePanel("Reactive test"),


  mainPanel(
    plotOutput("plot1",click="plot1_click")
  )
)

shinyApp(ui = ui, server = server)

This works perfectly, but now I want to be able to define vals with:

vals<-reactiveValues(keeprows=rep(TRUE,nrow(CustomDataInput))

In the case of my example, I tried accessing number of rows from the data created in data():

vals<-reactiveValues(keeprows=rep(TRUE,nrow(data()))

This gives me an error because I tried to access a reactive variable in a non-reactive environment. Is there a way to access the data created in a reactive function to define reactiveValues?

Thank you for your time!

Shiva Shiva · Accepted Answer · 2015-07-21T13:24:05

The error pretty much addresses the problem. The correct way to do this is as follows.

library(shiny)
library(ggplot2)
server<-function(input, output) {

  vals <- reactiveValues()
  data <- reactive({
    set.seed(10)
    df=data.frame(x=rnorm(100),y=rnorm(100))
    vals$keeprows = rep(TRUE,nrow(df))
    df
  })

  #vals<-reactiveValues(keeprows=rep(TRUE,100))
  output$plot1 <- renderPlot({
    df=data()
    keep=df[vals$keeprows, ,drop=FALSE]
    exclude=df[!vals$keeprows, ,drop=FALSE]
    plot=ggplot(data=keep,aes(x,y))+geom_point()+theme_bw()+
      geom_point(data=exclude,fill=NA,col="black",alpha=0.75,shape=21)
    plot
  })

  observeEvent(input$plot1_click,{
    df=data()
    res <- nearPoints(df, input$plot1_click, allRows = TRUE,threshold=5)
    vals$keeprows <- xor(vals$keeprows, res$selected_)

  })

}

ui <- fluidPage(

  titlePanel("Reactive test"),


  mainPanel(
    plotOutput("plot1",click="plot1_click")

  )
)

shinyApp(ui = ui, server = server)

Declare the vals variable before hand and use that in reactive() function to send variables to vals as shown above. You should be fine.

Access data created from reactive function to define reactiveValues in Shiny

app.R

1 Answers