skip to Main Content

This is rather a simple question. I read other threads, and found that, in order to insert a GGVIS visualization into Shiny, you need:

  1. In ui.R – Call ggvisOutput("EvolucionVisitas")
  2. In server.R – use function bind_shiny("EvolucionVisitas")

I’m having problems plotting the graph for my tab “Evolución Visitas”.
I did both, but i’m failing somewhere.

Nothing prints in my tab: EvoluciónVisitas. Everything else is okay.

Here is my data:

structure(list(date = structure(1:31, .Label = c("2014-12-01", 
"2014-12-02", "2014-12-03", "2014-12-04", "2014-12-05", "2014-12-06", 
"2014-12-07", "2014-12-08", "2014-12-09", "2014-12-10", "2014-12-11", 
"2014-12-12", "2014-12-13", "2014-12-14", "2014-12-15", "2014-12-16", 
"2014-12-17", "2014-12-18", "2014-12-19", "2014-12-20", "2014-12-21", 
"2014-12-22", "2014-12-23", "2014-12-24", "2014-12-25", "2014-12-26", 
"2014-12-27", "2014-12-28", "2014-12-29", "2014-12-30", "2014-12-31"
), class = "factor"), sessions = c(1932L, 1828L, 2349L, 8192L, 
3188L, 3277L, 2846L, 2541L, 5434L, 4290L, 2059L, 2080L, 2111L, 
3776L, 1989L, 1844L, 3641L, 1283L, 1362L, 1568L, 2882L, 1212L, 
957L, 851L, 928L, 1435L, 1115L, 1471L, 1128L, 1022L, 768L), id = 1:31), .Names = c("date", 
"sessions", "id"), row.names = c(NA, -31L), drop = TRUE, class = c("tbl_df", 
"tbl", "data.frame"))

Here my code, thanks.

ui.R

library(shiny)
library(ggvis)

# Define the overall UI
shinyUI(

  # Use a fluid Bootstrap layout
  fluidPage(    

    # Give the page a title
    br(),
    br(),
    titlePanel("Visitas por fuente"),

    # Generate a row with a sidebar
    sidebarLayout(      

      # Define the sidebar with one input



      sidebarPanel(
        dateRangeInput("dates", label = h3("Date range"),
                       start = "2014-12-01", end = "2014-12-31")

      ),


      mainPanel(
        tabsetPanel(
          tabPanel('Visitas por fuente',
                   plotOutput("VisitasFuente")),
          tabPanel('Evolución de las visitas',
                   ggvisOutput("EvolucionVisitas")),
          tabPanel('Comentarios',
                   dataTableOutput("Comentarios"))
        )

    )
  )
))

server.R

library(shiny)
library(ggvis)



Visitas_Por_Fuente <- read.csv("D:\RCoursera\Star-App-2\Visitas_Por_Fuente_Dic.csv")
labelsF = c("Directo", "Email", "Referencias", "SEO", "Social Media", "Campañas", "Adwords")
Visitas_Por_Fuente$date <- as.Date(Visitas_Por_Fuente$date)
ComentariosDic <- read.csv("D:\RCoursera\Star-App-2\ComentariosDic2014.csv",header = TRUE, sep = ";")
ComentariosDic$date <- as.Date(ComentariosDic$date)


shinyServer(


  function(input, output) {



    output$VisitasFuente <- renderPlot({

      # Filter the data based on user selection month     
      date_seq <- seq(input$dates[1], input$dates[2], by = "day")


      VisitasData <- filter(Visitas_Por_Fuente, date %in% date_seq  & Fuentes %in% labelsF)

      VisitasData <- VisitasData %>% group_by(Fuentes) %>%
                                     summarise(sessions = sum(sessions))




      # Bar graph using ggplot2 library 
      ggplot(VisitasData, aes(factor(Fuentes), sessions, fill = Fuentes)) + 
        geom_bar(stat="identity", position = "dodge") +
        geom_text(aes(label = comma(sessions)), position=position_dodge(width=0.9), vjust=-0.25) +
        scale_fill_manual(breaks = c("0", "1", "3", "6", "9", "12", "15"),
                          labels = labelsF,
                          values = c("#E69F00", "#56B4E9", "#009E73", 
                                     "#F0E442", "#0072B2", "#A082F8", "#F072A2"))

    })

    **############# Evolución de las visitas ##############################################
    #####################################################################################**


    output$EvolucionVisitas <- renderPlot({

      # Filter the data based on user selection month     
      date_seq <- seq(input$dates[1], input$dates[2], by = "day")


      EvolucionVisitas <-  filter(Visitas_Por_Fuente, date %in% date_seq) 


      mysessions <- function(x) {
        if(is.null(x)) return(NULL)
        #notice below the id column is how ggvis can understand which session to show 
        row <- EvolucionVisitas[EvolucionVisitas$id == x$id, ]
        #prettyNum shows the number with thousand-comma separator  
        paste0("Sessions:", "&nbsp;",prettyNum(row$sessions, big.mark=",",scientific=F)) 
      }




      EvolucionVisitas %>% 
        ggvis(x= ~date, y= ~sessions, key := ~id) %>%
        layer_points()  %>%
        add_tooltip(mysessions ,"hover") %>%
        layer_paths() %>%
        add_axis("x", value=c(as.character(EvolucionVisitas$date[1]),as.character(EvolucionVisitas$date[round(length(EvolucionVisitas$date)/2,0)]),
                              as.character(tail(EvolucionVisitas$date, n=1)))) %>%
        bind_shiny("EvolucionVisitas")







    #####################################################################################
    #####################################################################################


    output$Comentarios = renderDataTable({

      date_seq <- seq(input$dates[1], input$dates[2], by = "day")


      ComentariosDic <- filter(ComentariosDic, date %in% date_seq)

      ComentariosDic <- filter(ComentariosDic, !grepl("^$", Comentarios))


    })

})

3

Answers


  1. Where you have the code in server.R starting

      output$EvolucionVisitas <- renderPlot({
    

    you could try wrapping it in a reactive like this, rather than in renderPlot:

    vis <- reactive({
    
    # Filter the data based on user selection month     
          date_seq <- seq(input$dates[1], input$dates[2], by = "day")
    
    
          EvolucionVisitas <-  filter(Visitas_Por_Fuente, date %in% date_seq) 
    
    
          mysessions <- function(x) {
            if(is.null(x)) return(NULL)
            #notice below the id column is how ggvis can understand which session to show 
            row <- EvolucionVisitas[EvolucionVisitas$id == x$id, ]
            #prettyNum shows the number with thousand-comma separator  
            paste0("Sessions:", "&nbsp;",prettyNum(row$sessions, big.mark=",",scientific=F)) 
          }
    
    myvis <-
        ggvis(x= ~date, y= ~sessions, key := ~id) %>%
            layer_points()  %>%
            add_tooltip(mysessions ,"hover") %>%
            layer_paths() %>%
            add_axis("x", 
    
        value=c(as.character(EvolucionVisitas$date[1]),as.character(EvolucionVisitas$date[round(length(EvolucionVisitas$date)/2,0)]),
                                      as.character(tail(EvolucionVisitas$date, n=1)))) 
    
    
        myvis
        })
    

    Then outside of the reactive put :

    vis %>% bind_shiny("EvolucionVisitas")
    

    I think I recall something similar when I was doing a shiny/ggvis – my code is on my github here: https://github.com/jalapic/shinyapps/tree/master/soccerteams it might be of help.

    Login or Signup to reply.
  2. Only troubleshooting the ggvis, your problem is primarily a result of your attempt to customize the x-axis. The ggvis is trying to be smart by interpreting the dates as times. For the purposes of this plot, I believe it would be best to treat them as factors.

    Here is a complete reproducible answer.

    shiny::runGist("https://gist.github.com/cdeterman/0ac102cd68a7987a8a90")
    

    You will notice a few other differences. It is probably best to make your dataset reactive so you can reuse it in multiple places without additional overhead. Also, as @jalapic initially suggested, you want to make your ggvis object reactive so the plot can be dynamic and use the nice tooltips.

    Login or Signup to reply.
  3. Whew… that’s a bit of a mess. I stripped it down to the ggvis part and tried to get it running. Check it out in this gist.

    You can run it with:

    shiny::runGist("https://gist.github.com/corynissen/f75ecae388f81be13436")
    
    Login or Signup to reply.
Please signup or login to give your own answer.
Back To Top
Search