Create interactive webmap with markers in R using Shiny, Leaflet and rCharts
Asked Answered
V

2

8

I am trying to create an interactive webmap in R to display storms using Shiny, Leaflet and rCharts (the structure is loosely based on the http://ramnathv.github.io/bikeshare app).

The idea is that the user selects one storm name at a time (df$Name) and the markers for that storm (lat/long) are displayed in a Leaflet map (with zoom in/out function).

I cannot get Leaflet to load a new map and markers for each separate storm name.

This is what the data (TCs.Rda) look like (sample data file uploaded here):

 Year   Name   Wind   lat    long
 2010   BONNIE 15     30.100 -91.000
 2010   FIVE   25     30.000 -88.900
 2010   FIVE   25     30.400 -88.800
 2010   FIVE   25     30.800 -88.600

server.R

library(shiny); library(rCharts); library(leaflet)    
load("TCs.Rda") 
name <- sort(unique(data$Name)) 
shinyServer(function(input, output){      
     dataset <- reactive({  df <- data[data$Name == input$name, ]  })    
     output$add <- renderText({paste("Storm name:", input$name)})  
     output$Controls <- renderUI({  
         list(selectInput("name", "Select storm", name, selected=name[1])) })  
     output$myChart <- renderMap({
         df <- dataset()
         map <- Leaflet$new()
         map$setView(c(35, -80),zoom=5) 
         map$tileLayer(provider='Stamen.TonerLite')          
         for (i in 1:nrow(df)) {
         map$marker(c(df[i, "lat"], df[i, "long"]), bindPopup=df[i, "Wind"])}  
         map$set(dom='myChart')
         map$enablePopover(TRUE)
         map$fullScreen(TRUE)
         map      
       })  
    })

ui.R

library(shiny); library(rCharts); library(leaflet)
shinyUI(pageWithSidebar( 
  headerPanel("Storm tracks"),  
  sidebarPanel(h3("My subtitle"), uiOutput("Controls")),  
  mainPanel(                     
     tabsetPanel(        
        h3(textOutput("add")),             
        tabPanel("map", tags$style(".leaflet {height: 400px;}"),  # I can only get this to work with a tabset panel, but ideally both the textOutput and map would go directly in the mainPanel 
        showOutput("myChart", "leaflet")))
      )
  ))
Vibrissa answered 10/5, 2015 at 12:33 Comment(0)
V
6

I figured it out myself eventually! Another post (Shiny renders a responsive rCharts leaflet map once, but is blank if you change the input variable) indicated that the line of code "map$set(dom='myChart')" could be preventing Leaflet from reloading a new map on each selection.

I didn't think that could be the issue here (my example is a little different and does not use geojson), but apparently it was.

This is my working code, in case it helps anyone else -

server.R

library(shiny);library(rCharts);library(leaflet)
load("TCs.Rda")
name <- sort(unique(data$Name)) 
shinyServer(function(input, output){
   dataset <- reactive({df<- data[data$Name == input$name, ]})    
   output$add <- renderText({paste("Storm name:", input$name)})   
   output$Controls <- renderUI({list(selectInput("name", "Select storm", name, selected=name[1]) ) })   
output$myChart <- renderMap({
   df <- dataset()
   map <- Leaflet$new()
   map$setView(c(35, -80),zoom=3)
   map$tileLayer(provider='Stamen.TonerLite')       
   for (i in 1:nrow(df)) {map$marker(c(df[i, "lat"], df[i, "long"]))}        
   map$fullScreen(TRUE)
   map      
  })  
})

ui.R

library(shiny); library(rCharts); library(leaflet)
shinyUI(pageWithSidebar( 
   headerPanel("Storm tracks"),  
   sidebarPanel(h3("My subtitle"), uiOutput("Controls")),  
   mainPanel(      
      tabsetPanel(        
        tabPanel("map", h3(textOutput("add")),
             tags$style(".leaflet {height: 400px;}"), 
             showOutput("myChart", "leaflet")))
  )
  ))

And the result looks like this:

enter image description here

Vibrissa answered 12/5, 2015 at 8:21 Comment(0)
C
2

I strongly recommend rstudio's leaflet.

I've an introductory tutorial on it and why it's so awesome here: http://www.r-bloggers.com/the-leaflet-package-for-online-mapping-in-r/

Also, I'm using it to build an interact planning tool for the DfT: https://github.com/npct/pct

Good luck with it!

Catullus answered 13/5, 2015 at 9:6 Comment(0)

© 2022 - 2024 — McMap. All rights reserved.