Show your blog’s visitors on map with Shiny

You may have interest about who visited your web site, and it’s better if you can see them on map, like this one. In this blog, I will show you how to create a such kind of shiny-app to track your visitors.

A shiny-app is a web site which allow users to interactively input and get output. The idea of creating a visitor-tracking shiny-app is: when users open your blog site, a shiny-app will also be opened (which can be embedded in your web site), and once the shiny-app be opened, it will save user’s IP related information to your server.

JavaScript code for getting the IP related information:

  var I = 0
  var tt = setInterval(function(){
    var L = $("#log");
    $.get("", function(e) {
      L.val(Date() + "  ipInfo|" + e.ip + "," + + "," + e.loc);
    }, "jsonp");
    if(L.val().length > 0) {
  }, 100)

Where #log is a textInput input of shiny for storing the IP related information.

The following part in above is used to get the IP related information and store it into “#log”, and tell shiny-app that the value of “#log” has changed. So, we can write some code in “server.R” to save the value of “#log” into server.

    $.get("", function(e) {
      L.val(Date() + "  ipInfo|" + e.ip + "," + + "," + e.loc);
    }, "jsonp");

And the following part is used to stop continue fetching IP information:

    if(L.val().length > 0) {

In server.R

The code for storing IP information:

    msg <- gsub(".*ipInfo", paste(date(), "ipInfo"), input$log)
    if (nchar(msg) > 20) {
      write(msg, "www/visitors.txt", append  = TRUE)

The code for map:

  output$myMap <- renderMap({
    Dat <- rDat()
    Map <- Leaflet$new()
    Map$addParams(width="100%", height="800px;", layerOpts = list(maxZoom = 18, zoomControl = FALSE))
    Map$addParams(bounds = list(c(min(Dat$Lat), min(Dat$Lon)) - 10, c(max(Dat$Lat), max(Dat$Lon)) + 10))
    Points <- as.list(rep(NA, nrow(Dat)))
    for (i in 1:nrow(Dat)) {
      Points[[i]] <- c(lat = Dat$Lat[i], lon = Dat$Lon[i], 
                       value = paste(Dat$City[i], as.character(Dat$Time[i]), sep = ": "))
    Map$addParams(cluster = Points)

where Map$addParams(cluster = Points) creates an object called “spec.cluster”. We need to add the following JavaScript code at the end of /usr/local/lib/R/site-library/rCharts/libraries/leaflet/layouts/chart.html to show the cluster points.

 if (spec.cluster != undefined) {
    var markers = L.markerClusterGroup();
		for (var i = 0; i < spec.cluster.length; i++) {
			var a = spec.cluster[i];
			var title = a.value;
			var marker = L.marker(new L.LatLng(, a.lon), { title: title });

We also need include some JavaScript libraries needed into /usr/local/lib/R/site-library/rCharts/libraries/leaflet/external, they are:

  • MarkerCluster.css

  • MarkerCluster.Default.css

  • leaflet.markercluster-src.js

To include above libraries, we need to modify the file /usr/local/lib/R/site-library/rCharts/libraries/leaflet/config.yml as:

    - external/leaflet.css
    - external/MarkerCluster.css 
    - external/MarkerCluster.Default.css

    - external/leaflet.js
    - external/leaflet.markercluster-src.js

All scripts (includes demo data) can be downloaded here. Remember backup the leaflet folder before replacing it with the one in download folder (www/leaflet).

Written on November 10, 2015