0

In the example Data Frame DF, I have the following columns. 

gender <- c("Male", "Female","Female", "Male")
Location <- c("AB", "BC", "CD", "DE")
hasTV <- c("Yes","Yes","No","No")
Latitude <- c(49.82380908513249,59.478568831926395,59.478568831926395,49.82380908513249)
Longitude <- c(-10.8544921875,-10.8544921875,2.021484375,2.021484375)
DF <- data.frame(gender,Location,hasTV,Latitude,Longitude)

Under UI, i've used radiobuttons to select options from hasTV, checkboxGroupInput to select Gender and selectInput to create a dropdown of Location. I've implemented this in fluidRow as shown below.

sex <- unique(DF$gender)
loc <- unique(DF$Location)
hastv <- unique(DF$hasTV)

radioButtons("radio_hastv",label = "Has TV", choices = hastv, selected = "Yes")
checkboxGroupInput("checkbox_gender", label = "Gender", choices = sex, selected = sex)
selectInput("Location", label = "Location", choices=loc, selected = "AB")
leafletOutput("mymap", height = 415)

In the server function, I have multiple reactive expressions based on the input selected. This is how I've implemented the expressions. 

 filtered_gender <- reactive({
   DF[DF$gender == input$checkbox_gender,]
 })

 filtered_hastv <- reactive({
   DF[DF$hasTV == input$radio_hastv,]
 })

 filtered_loc <- reactive({
   DF[DF$Location == input$loc,]
 })

I've already rendered the leaflet map. However, I'd like my map to change whenever all these three inputs are somehow selected. e.g. if a person selects, gender=Male, location = DE and hasTV=No, the appropriate map with the correct gps is plotted on the map.

So far, i'm only able to update leaflet map using one reactive expression as shown below. 

 observe(leafletProxy("mymap", data = filtered_loc()) %>%
           clearMarkers()%>%
           addMarkers(radius =3)%>%
           label = ~htmlEscape(DF$hasTV)
         )

How do I go about incorporating the other reactive expressions so that the map changes accordingly. Thanks.

andy
  • 1,947
  • 5
  • 27
  • 46

1 Answers1

3

You need to move all those filters in one reactive and use that in your leaflet plot -

library(dplyr)

filtered_data <- reactive({
   DF %>%
     filter(gender %in% input$checkbox_gender,
            hasTV %in% input$radio_hastv,
            Location %in% input$loc
            )
 })

 observe(leafletProxy("mymap", data = filtered_data()) %>%
           clearMarkers()%>%
           addMarkers(radius =3)%>%
           label = ~htmlEscape(hasTV) # note change; using DF here is wrong
         )
Shree
  • 10,835
  • 1
  • 14
  • 36
  • I've implemented it and getting this error, `Data contains 7 rows with either missing or invalid lat/lon values and will be ignored` `Warning: Error in filter_impl: Result must have length 140, not 0"`. Could the N/A's in Lat and Long be the issue? – andy Oct 31 '18 at 02:50
  • When I change `gender` and `hasTV` to be only radiobuttons, it doesn't throw this error. It's a problem with `checkboxGroupInput`. Especially, when I deselect the two options. – andy Oct 31 '18 at 13:15
  • I have edited the answer to use `%in%` instead of `==`. If you deselect all checkbox choices then there is nothing to plot. Also try removing `label` argument in `leafletProxy()`...see if that is causing problems. – Shree Oct 31 '18 at 14:25
  • Thanks! That worked! I have another question, how does one add a reactive expression that will trigger different zoom levels of the map depending on the selected `loc`? – andy Oct 31 '18 at 17:36
  • Not possible to answer that in a comment. You need to ask that as a separate question with a small reproducible example specific to that problem. – Shree Oct 31 '18 at 17:39