app.R

# ###### Instructions ##############################
# 
# The below app explores the build-in R dataset "mtcars". In this tutorial you
# will expand this app.
# 
# Run the app by either using the runApp() command or by hitting "Run App" in the
# upper right corner of the script editor. Explore the interactive behaviour of 
# the app and how it is implemented codewise.
#
#
# ### 1st task ###
# 
# Improve the layout of the tab "Scatter Plots" by moving the input elements into
# a side bar and the plot into the main panel. You can do this using the 
# sidebarLayout function.
# 
# 
# ### 2nd task ###
# 
# Add a check box to the sidebar panel that allows the user to add a regression
# line to the plot.
#
# Hint: You can add geom_smooth(method = "lm") to the ggplot object to add a
# regression line to the plot.
# 
# 
# ### 3rd task ###
# 
# The "Data" tab is blank. Implement the following:
# The tab should display the mtcars data in a table. The sidebar panel should allow
# user to filter the data for specific levels of the variables cyl and gear (e.g.  
# drop down menus or radio buttons).
#
#

###Setup
if(!require("shiny")) install.packages("shiny")
if(!require("ggplot2")) install.packages("ggplot2")
if(!require("dplyr")) install.packages("dplyr")
if(!require("shinythemes")) install.packages("shinythemes")



###### Shiny App ################################


### UI
ui <- fluidPage(
  theme = shinytheme(theme = "sandstone"), 
	titlePanel("mtcars Shiny Tutorial"),
  #Github link
  tags$a(href = "https://github.com/meiradania/shinyApp", 
		   "Click here to see the R source code", target = "_blank"),
	tabsetPanel(
		tabPanel("Scatter Plots",
				sidebarLayout(
				  sidebarPanel(
				    #Select axes variables
				    selectInput(inputId = "yvar", label = "y Axis",
				                choices = c("mpg", "disp", "hp", "drat", "wt", "qsec"),
				                selected = "mpg"),
				    
				    selectInput(inputId = "xvar", label = "x Axis",
				                choices = c("mpg", "disp", "hp", "drat", "wt", "qsec"),
				                selected = "disp"),
				    
				    #Check box that allows the user to add a regression line to the plot.
				    checkboxInput(inputId = "regressionline", label = "Regression Line")

				  ),
				  
				  mainPanel(
				    #Scatter plot output
				    plotOutput(outputId = "scatter", height = "400px")
				  )
				)	
		),
		tabPanel("Data",
		         sidebarLayout(
		           sidebarPanel(
		             # cyl filter
		             selectInput(inputId = "filtercyl", label = "Filter - Number of cylinders:",
		                         choices = unique(mtcars$cyl), selected = 4),
		             # gear filter
		             selectInput(inputId = "filtergear", label = "Filter - 	Number of forward gears:",
		                         choices = unique(mtcars$gear), selected = 4),
		             
		             # Button
		             actionButton(inputId = "button", label = "Apply filters"),
		             actionButton(inputId = "buttonReset", label = "Reset filters")
		           ),
		           mainPanel(
		             dataTableOutput(outputId = "data")
		             )
		         )
		)
	),
	#Source link
	tags$a(href = "https://stat.ethz.ch/R-manual/R-devel/library/datasets/html/mtcars.html", 
		   "Source: Henderson and Velleman (1981)", target = "_blank")
)


### Server function
server <- function(input, output) {
	#Create scatterplot
	output$scatter <- renderPlot({
	  if(input$regressionline)
	    plot <- ggplot(mtcars, aes_(y = as.name(input$yvar), x = as.name(input$xvar))) +
	      geom_point(size = 2.5) + geom_smooth(method = "lm")  
	  else
	    plot <- ggplot(mtcars, aes_(y = as.name(input$yvar), x = as.name(input$xvar))) +
	      geom_point(size = 2.5) 
		plot
	})
	
	#Add data table code here
	output$data <- renderDataTable(mtcars)
	observeEvent(input$button, 
	             { 
	               valuecyl <- input$filtercyl
	               valuegear <- input$filtergear
	               output$data <- renderDataTable(mtcars[mtcars$cyl == valuecyl 
	                                                     & mtcars$gear == valuegear ,])
	             })
	observeEvent(input$buttonReset, 
	             output$data <- renderDataTable(mtcars))
}

### Create Shiny object
shinyApp(ui = ui, server = server)