Returning a dataframe in a reactive function in R Shiny Dashboard

问题

Working with a Shiny Dashboard, ui.r, server.r and several r scripts that contain functions.

Base problem is: I have two data sets that I bring into the server.r and I pass those to a function the prepares/cleanses the data sets and then binds them together and should return a complete dataframe. The dataFrame should be reactive. Here is what I have so far:

In the server.r I load the data prior to the function(input, output, session). Then I have:

################server.r  code #########################
data <- reactive({
   testDF <- prepData(data1, data2)
})

The prepData function does a variety of things but ends with:

####################prepData function return#################
 return(rbind(data1, data2))

If I use something like:

############## server.r code#######################
value = nrow(data()),

Then the code returns the correct value. However I would prefer to just return the data frame as I did with testDF.

Shiny will throw and error in the UI of object 'testDF' not found.

I tried working through: How do I build a reactive dataframe in R / Shiny? using the code:

dataR <- prepData(data1, data2)

makeReactiveBinding(dataR)

This still throws the error. The function is clearly working and has been validated but there must be something I don't understand about the reactive component of utilizing this function. Any help would be appreciated. Thanks!

server.ui

function(input, output, session) { 

### PreProcess the Data

data <- reactive({
  testDF <- prepData(bdata, qdata)
})

#dataR <- prepData(bData, qData)

#makeReactiveBinding(dataR)

### Information Box Populations

output$monthlytransactions = renderInfoBox({
infoBox(
  title = "Payments",
  value = nrow(data()),
  icon = icon("comments-dollar"),
  color = "blue"
)
})

output$monthlyGrossDollars = renderInfoBox({
infoBox(
  title = "Payments",
  value = sum(testDF$GrossAmount),
  icon = icon("comments-dollar"),
  color = "blue"
)
})

 }

prepData

############# FUNCTIONS ##############
prepData <- function(beamData, qlawData){ 

##Processing##
#Join DFs
return(rbind(bData, qData))

回答1:

You can think of reactive expressions as being functions that return a value. So you're getting the object 'testDF' not found error because testDF only exists within the scope of the data reactive object (function).

When you create a reactive expression:

data <- reactive({
  testDF <- prepData(bdata, qdata)
})

what you're actually doing is creating a function that returns a value. In this case, you run prepData(), save the result in the object testDF and then implicitly return that value (since R implicitly returns the value of the last line of a function). You access that value by calling data(), since that's the name of the reactive expression that returns your value, not testDF, since that object is deleted once the function has finished running.

If you want testDF to contain the result of prepData, just change the name of the reactive expression like so:

testDF <- reactive({
    prepData(bdata, qdata)
})

testDF() # this will return the value you want

---

Calling the reactive expression does actually return a data frame which you can subset and work with as normal. So you can work with it just like a data frame (but don't forget the parentheses!):

data()$GrossAmount
data()[data()$GrossAmount > 100, c('name', 'date')]

来源：https://stackoverflow.com/questions/53876948/returning-a-dataframe-in-a-reactive-function-in-r-shiny-dashboard