CODEWITHSUNDEEP
shinyuser-inputsummary
DS_Guy
DS_Guy

Reputation: 23

Shiny App for generating a summary table of columns based on user input

I am trying to generate a shiny app where a summary table can be generated for a column selected by the user from a drop down of columns. Dataframe used is a sample dataset.



library(shiny)
library(dplyr)

df <- data.frame(grouping_letter = c('A', 'A', 'B', 'B', 'C', 'C'), 
                 grouping_animal = c('Cat', 'Dog', 'Cat', 'Dog', 'Cat', 'Dog'),
                 value = c(1,2,3,4,5,6))

df <- df %>% mutate(
    grouping_letter = as.character(grouping_letter),
    grouping_animal = as.character(grouping_animal))

# Define UI for application that summarizes a table
ui <- fluidPage(

    selectInput(inputId ="column",
                label = "Choose Column for Summary",
                choices = names(df),
                selected = "grouping_letter"),

    tableOutput('table')
)

# Define server logic required to output summary table
server <- function(input, output) {
    groupvar <- reactive({input$column})
    cols <- c('value', unlist(groupvar))
    dt_f <- df[,cols] %>%
        group_by_at(2) %>%
        summarise (value = n(), yield = round(mean(value)*100, 1)) %>%
        mutate(Pct_of_value = paste0(round(100 * value/sum(value), 0), "%"))
    output$table <- renderTable(dt_f)
}

# Run the application 
shinyApp(ui = ui, server = server)

I getting an error: Warning: Error in .subset: invalid subscript type 'list' I am using R 3.6 and RStudio on windows

Upvotes: 2

Views: 1120

Answers (1)

lz100
lz100

Reputation: 7330

There are two problems

  • groupvar is reactive, so you don't want to unlist a reactive value.
  • You need to use groupvar() to get values inside a reactive object. However, in order to run groupvar(), you need to do it within active reactive context: like observe, observeEvent or renderXXX. To give you a not very precise idea is that functions in shiny that allow you to use function({}) expressions.

See my code below, this should be what you want. Notice how I use renderTable.

library(shiny)
library(dplyr)

df <- data.frame(grouping_letter = c('A', 'A', 'B', 'B', 'C', 'C'), 
                 grouping_animal = c('Cat', 'Dog', 'Cat', 'Dog', 'Cat', 'Dog'),
                 value = c(1,2,3,4,5,6))

df <- df %>% mutate(
    grouping_letter = as.character(grouping_letter),
    grouping_animal = as.character(grouping_animal))

# Define UI for application that summarizes a table
ui <- fluidPage(

    selectInput(inputId ="column",
                label = "Choose Column for Summary",
                choices = names(df),
                selected = "grouping_letter"),

    tableOutput('table')
)

# Define server logic required to output summary table
server <- function(input, output) {
    groupvar <- reactive({input$column})

    output$table <- renderTable({
        df[, c('value', groupvar())] %>%
            group_by_at(2) %>%
            summarise (value = n(), yield = round(mean(value)*100, 1)) %>%
            mutate(Pct_of_value = paste0(round(100 * value/sum(value), 0), "%"))
    })
}

# Run the application 
shinyApp(ui = ui, server = server)

Upvotes: 1

Related Questions