Tell shiny I am talking about a data frame column name not a variable

So I am trying to create a shiny app in which I am plotting a map of India using sf data. I have uploaded a dataset which has multiple columns. So I ask the use to select the column. But when I gave that input to geom_sf(fill = input$select_topic) It firstly thinks the column name is a single char value, so I use geom_sf(fill = .data$input$select_topic) or geom_sf(fill = ,data[[input$select_topic]] ) which gives me error : Can't subset .data outside of a data mask context.

here is the code:

library(shiny)
library(tidyverse)
library(sf)


# joining the main datset and the sf dataset 

map_data <- read.csv("D:\\R Projects\\R experimental\\dashboard\\app-1\\app-1\\data\\edu_data.csv")
map_data <- as.data.frame(map_data)
map_data_2 <- left_join(India, map_data, by = "state_ut")


# Define UI ----
ui <- fluidPage(
  titlePanel("My Shiny App"),
  
  sidebarLayout(position = "right",
    sidebarPanel(
      
      selectInput("select_year", label = "Select Year", 
                  choices = list("2012-13" = 2012,
                                 "2013-14" = 2013, 
                                 "2014-15" = 2014,
                                 "2015-16" = 2015), selected = 2013),
      selectInput("select_topic", label = "Select Category", 
                  choices = list("Drinking Water Facility" = 'water_faci', 
                                 "Girl's Toilet Facility" = 'gtoi_faci',
                                 "Boy's Toilet Facility" = 'btoi_faci',
                                 "Electricity Connection" = 'elec_faci',
                                 "Computer availability" = 'comp_faci',
                                 "Overall Facilities" = 'all_faci',
                                 "Gross Enrollment" = 'gross_enroll',
                                 "Drop-out Rate" = 'drop_rate' 
                                 ), selected = "Drinking Water Facility")
      
    mainPanel(
      h1("Maps"),
      textOutput("selected_year"),
      textOutput("selected_category"),
      plotOutput("map")
      )
  )
  
)

# Define server logic ----
server <- function(input, output) {
  
  output$selected_year <- renderText({
    paste("You have selected the year:", input$select_year)
  })
  
  
  output$selected_category <- renderText({
    paste("You have selected category:", input$select_topic)
  })
  
  })
  output$map <- renderPlot({
    x <- map_data_2 %>% filter(., year == input$select_year)
    
    ggplot(x, aes(label = state_ut)) +
      geom_sf( fill = .data$input$select_topic ) +
      scale_fill_distiller(palette = "GnBu", 
                           direction= 1, 
                           na.value="grey") +
      theme_map() + 
      theme( plot.title = element_text(size=22), 
             legend.position = "left", 
             legend.justification ='left') 
  })  
  
}

# Run the app ----
shinyApp(ui = ui, server = server)

Here is the dataset whose column I need to select

output$map <- renderPlot({
    x <- map_data_2 %>% filter(., year == input$select_year)

    ggplot(x, aes(label = state_ut)) +
      geom_sf( fill = input$select_topic ) +
      scale_fill_distiller(palette = "GnBu", 
                           direction= 1, 
                           na.value="grey") +
      theme_map() + 
      theme( plot.title = element_text(size=22), 
             legend.position = "left", 
             legend.justification ='left') 
  })

will result in a map which has taken the fill values as a column present in the dataset.

Change your fill= in two ways:

• put it within aes(.), and
• use .data[[ input$select_year ]]. (The use of .data$input$select_year cannot work.)

output$map <- renderPlot({
    x <- map_data_2 %>% filter(., year == input$select_year)

    ggplot(x, aes(label = state_ut)) +
      geom_sf(aes(fill = .data[[ input$select_topic ]])) +
      scale_fill_distiller(palette = "GnBu", 
                           direction= 1, 
                           na.value="grey") +
      theme_map() + 
      theme( plot.title = element_text(size=22), 
             legend.position = "left", 
             legend.justification ='left') 
  })

See https://ggplot2.tidyverse.org/articles/ggplot2-in-packages.html#using-aes-and-vars-in-a-package-function-1 for how to reference column names indirectly.