Filter Data Range based on User Input

Question

I have a shiny app that plots two variables in a scatter plot and filters based on user input. I am trying to get it so that the user can filter the data based on a range that they specify. However, I only want this filter to be applied if the field is filled out, and return to all data if the field is empty (null). For example, if you launch the code that I have attached, I want the user to be able to check the Specify Data Ranges checkbox and when they put in 4.8 in the min x input box, the data is filtered out so that no data points less than 4.8 are now considered. The sample code that I have prepared is below with my attempt at what I'm trying to do commented out near the middle.

#Check packages to use in library
{
library('shiny') #allows for the shiny app to be used
library('ggvis') #allows for interactive ploting
}

alldata <- iris

#establish options for drop down menus
{
specieschoices <- unique(as.character(alldata$Species))
}
# UI

ui<-fluidPage(
titlePanel("Explorer"),
fluidRow(
column(4,
       wellPanel(
         h4("Apply Filters"),
         selectInput(inputId = "species", label="Select a Species:", choices = sort(specieschoices), selected="setosa", multiple = TRUE, selectize = TRUE)
       )),
column(8,
       ggvisOutput("plot1")
),
column(4,
       wellPanel(
         h4("Data Variables"),
         selectInput(inputId = "x", label="Select x-axis Variable:", choices=as.character(names(alldata[,1:4])),selected='Petal.Length', multiple = FALSE),
         selectInput(inputId = "y", label="Select y-axis Variable:", choices=as.character(names(alldata[,1:4])),selected='Petal.Width', multiple = FALSE)
       )),

column(4,
       wellPanel(
         checkboxInput(inputId = "datarange", label="Specify Data Ranges", value = FALSE),
         conditionalPanel(
           condition = "input.datarange == true",
           wellPanel(
             numericInput(inputId = "minxdata", label="Specify x axis min", value = -9999999999, step = 0.1),
             numericInput(inputId = "maxxdata", label="Specify x axis max", value = 9999999999, step = 0.1),
             numericInput(inputId = "minydata", label="Specify y axis min", value = -9999999999, step = 0.1),
             numericInput(inputId = "maxydata", label="Specify y axis max", value = 9999999999, step = 0.1)
           ))
       ))
))

#SERVER
server<-function(input,output,session)
{

#Set up reactive variables for ranges
filteredData <- reactive({

minX <- input$minxdata
maxX <- input$maxxdata
minY <- input$minydata
maxY <- input$maxydata

# Apply filters
m <- alldata %>% filter(
  `Species` %in% input$species

  ###############THIS IS THE PART THAT I NEED HELP WITH
  #This works as hardcoded with hypothetical user input of x=Petal.Length and y=Petal.Width
  ,
  Petal.Length >= minX,
  Petal.Length <= maxX,
  Petal.Width >= minY,
  Petal.Width <= maxY

  #This is does not work
  # ,
  # as.symbol(input$x) >= minX,
  # as.symbol(input$x) <= maxX,
  # as.symbol(input$y) >= minY,
  # as.symbol(input$y) <= maxY
  #####################################################

)
m <- droplevels(as.data.frame(m))
m
})

vis <- reactive({

#Plot Data with Visualization Customization
xvar <- prop("x", as.symbol(input$x))
yvar <- prop("y", as.symbol(input$y))

p1 = filteredData() %>%
  ggvis(x = xvar, y = yvar) %>%
  layer_points() %>%

  # Specifies the size of the plot
  set_options(width = 800, height = 450, duration = 0)
})

#Actually plots the data
vis %>% bind_shiny("plot1")
}

#Run the Shiny App to Display Webpage
shinyApp(ui=ui, server=server)

UPDATE:

I think I'm on the right track with this, but the min x and min y filters out all data if changed to anything >= 0, and the other filters do not do anything if changed.

  paste0("`", input$x, "`") >= minX,
  paste0("`", input$x, "`") <= maxX,
  paste0("`", input$y, "`") >= minY,
  paste0("`", input$y, "`") <= maxY

See http://stackoverflow.com/questions/38108515/r-shiny-display-full-dataframe-without-filter-input-until-filterinputs-are-chang/38125743#38125743 for two similar examples. — Xiongbing Jin, Jul 15 '16 at 20:28
@warmoverflow I see the similarities between our two issues, but am unclear with how to implement the solution in that post into my solution. I do not use renderPlot like that user uses renderTable so I do not see how I would go about the option to add selections, and I have numeric input that I am trying to filter by not a character so nchar doesn't appear to be applicable either. — User247365, Jul 15 '16 at 21:27
There is one more example at the end of the link that uses numericinput — Xiongbing Jin, Jul 16 '16 at 01:06
I looked at that example as well, but since it filters through a set variable (cyl, vs, and am in that example) and I am trying to filter through a user defined variable (my actual code has many more variables than the petal and sepal length and width for a user to choose from) changing the defined variable like cyl and vs to input$x is unsuccessful. I have edited my sample code to correct this discrepancy in my actual code and the sample code. — User247365, Jul 18 '16 at 15:32

score 0 · Accepted Answer · edited May 23 '17 at 12:14

I have discovered an answer based on this completely unrelated post that approached how they processed data in a different way that worked for me. Downloading png from Shiny (R)

The only change is within filteredData which is posted below.

#Set up reactive variables for ranges
filteredData <- reactive({

minX <- input$minxdata
maxX <- input$maxxdata
minY <- input$minydata
maxY <- input$maxydata
a <- which(names(alldata)==input$x)
xvariable <- as.numeric(alldata[,a])
b <- which(names(alldata)==input$y)
yvariable <- as.numeric(alldata[,b])

# Apply filters
m <- alldata %>% filter(
`Species` %in% input$species,
xvariable >= minX,
xvariable <= maxX,
yvariable >= minY,
yvariable <= maxY
)
m <- droplevels(as.data.frame(m))
m
})

Filter Data Range based on User Input

1 Answers1