Shiny modules import data and hide html if success - r

Question

I would like to know how it is possible with callModule() to hide html if my data are imported with a reactive boolean (boolmark)
Here's my code
ui.R
shinyUI(
column(width=9,
div( style = "width:100% ; max-width: 1200px; height: 100%" ,
conditionalPanel(condition = 'output.boolmark', tags$(HTML code),
conditionalPanel(condition = '!output.boolmark', code))
div(id="pass",style = "word-wrap: break-word;",
column(width = 3,
csvFileInput("datafile", "User data (.csv format)"))
server.R
shinyServer(
csvf <- callModule(csvFile, "datafile",stringsAsFactors = FALSE))
csvmodules.R
csvFileInput <- function(id, label = "CSV file") {
ns <- NS(id)
tagList(
fileInput(ns("file"),multiple = T, label),
checkboxInput(ns("heading"), "Has heading"),
selectInput(
ns("quote"),
"Quote",
c(
"None" = "",
"Double quote" = "\"",
"Single quote" = "'"
)))}
csvFile <- function(input, output, session, stringsAsFactors) {
showmark <<- T # Boolean uses to hide or show the mardkwon serving to load data
userFile <- reactive({
# If no file is selected, don't do anything
validate(need(input$file, message = FALSE))})
output$boolmark <- reactive({
showmark
})
observe({
print(showmark)})
outputOptions(output,"boolmark",suspendWhenHidden=F)
csv <- lapply(
csvtest,
FUN = function (x)
fread(data.table = F,check.names = F,
header = T,sep = ";",dec = ","
)
)
observe({
showmark <<-F
print(showmark)
}) # modify and lock the bool value to false
output$boolmark <- reactive({
showmark
})
return (csv)}
)}
Well, this code is working outside the module but the html is not hidden when using the modules
Why and how to fix it ?

Answer 1

As already mentioned in the comments the problem is that conditionalPanel needs to know in which namespace to look for the input- and output ids appearing in the condition argument. This can be done by passing the namespace function to conditionalPanel.
ns <- NS("datafile")
conditionalPanel("!output.boolmark", ns, ...)
A simplified and fixed version of your app could look like this:
library(shiny)
myModuleUI <- function(id) {
ns <- NS(id)
checkboxInput(ns("checkbox"), "checkbox")
}
myModule <- function(input, output, session) {
output$value <- reactive({ input$checkbox })
outputOptions(output, "value", suspendWhenHidden = FALSE)
}
shinyApp(
fluidPage(
myModuleUI("myModule"),
conditionalPanel("output.value", ns = NS("myModule"), "checkbox is TRUE")
),
function(input, output, session) {
callModule(myModule, "myModule")
}
)
Normally, it is not a good practice to refer to the module by id from outside the module so I would use the following adaptation.
myModuleUI <- function(id) {
ns <- NS(id)
checkboxInput(ns("checkbox"), "checkbox")
}
myModule <- function(input, output, session) {
return(reactive({input$checkbox}))
}
shinyApp(
fluidPage(
myModuleUI("myModule"),
conditionalPanel("output.value", "checkbox is TRUE")
),
function(input, output, session) {
output$value <- callModule(myModule, "myModule")
outputOptions(output, "value", suspendWhenHidden = FALSE)
}
)
The general idea behind this is that you either want to implement the logic inside your module (call conditionalPanel in myModuleUI) or return a value from myModule that gives the main app the possibility to work with the reactive objects generated by the module.