I am building a shiny app which has a lot of conditional panel. I have a back button in the app itself to navigate between the conditional panel. I would like to disable the web browsers back button as clicking that button goes to previous webpage(away from my app). Is there a way to do this?
You can write some javascript to do this. Here I have two examples, note that I only tested this on Chrome
Example 1 will return a message upon activation of the back button within the browser
rm(list = ls())
library(shiny)
jscode <- 'window.onbeforeunload = function() { return "Please use the button on the webpage"; };'
ui <- basicPage(
mainPanel(tags$head(tags$script(jscode)))
)
server <- function(input, output,session) {}
runApp(list(ui = ui, server = server))
Example 2 will disable navigation altogether. Personally I don't like this method as people might be annoyed that your site doesn't offer standard navigation functionalities
rm(list = ls())
library(shiny)
jscode2 <- "history.pushState(null, null, document.title);
window.addEventListener('popstate', function () {
history.pushState(null, null, document.title);});"
ui <- basicPage(
mainPanel(tags$head(tags$script(jscode2)))
)
server <- function(input, output,session) {}
runApp(list(ui = ui, server = server))
Related
I would like to integrate captcha verification into my shiny app (like this). But it has an unexpected behavior.
If this line:
recaptchaUI("test", sitekey = "6LeIxAcTAAAAAJcZVRqyHh71UMIEGNQ_MXjiZKhI")
is inside the UI, my app works fine (I mean, the validation is triggered when the checkbox and the button are clicked)
However, if it is inside renderUI, the validation is triggered only by clicking the checkbox (there is no need to click the button)
Here is my minimal working example:
library(shiny)
library(shinyCAPTCHA)
ui <- fluidPage(
# With this line, the captcha works perfectly
recaptchaUI("test", sitekey = "6LeIxAcTAAAAAJcZVRqyHh71UMIEGNQ_MXjiZKhI"),
uiOutput("body")
)
server <- function(input, output, session) {
result <- callModule(recaptcha, "test", secret = "6LeIxAcTAAAAAGG-vFI1TnRWxMZNFuojJ4WifJWe")
output$body <- renderUI({
fluidPage(
# Inside renderUI, it does not work properly.
#recaptchaUI("test", sitekey = "6LeIxAcTAAAAAJcZVRqyHh71UMIEGNQ_MXjiZKhI"),
uiOutput("humansOnly")
)
})
output$humansOnly <- renderUI({
req(result()$success)
tags$h1("For human eyes only!")
})
}
shinyApp(ui, server)
I have an app with several modules, that's why I need to use renderUI. What is the error in my code?
I appreciate any help!
I use the package shinyjs to allow the user to reload the page by clicking on "Reload the page" button in Tab 2 and I would like to stay on Tab 2 after reloading the page. But after realoading, the page is taking to Tab1 instead of Tab2
How can we fix it ? This is my code below :
library(shiny)
library(shinyjs)
jscode <- "shinyjs.refresh = function() { history.go(0); }"
ui <- fluidPage(
useShinyjs(),
extendShinyjs(text = jscode, functions = "refresh"),
tabsetPanel(
tabPanel("Tab 1"),
tabPanel("Tab 2", actionButton("mybutton", "Reload the page",
onclick ="javascript:window.location.reload(true)")))
)
server <- function(input, output, session) {
}
shinyApp(ui = ui, server = server)
Some help would be appreciated
You can resort to plain JavaScript. The idea is that you provide an id to tabsetPanel such that you can use updateTabsetPanel in the server.
All you have to do then is to let the reloaded page know that you want to start on the second tab. As a refresh typically resets all of the inputs and reactives (you are aware of that, right?), you cannot use reactives and you have to rely on another way of communication. The query string would be one possibility.
With these ingredients you can:
Write a simple JavaScript function that reloads the page and adds a parameter to the url.
Parse the query string in the server and if the parameter is found, react on it accordingly.
library(shiny)
js <- HTML("
function doReload(tab_index) {
let loc = window.location;
let params = new URLSearchParams(loc.search);
params.set('tab_index', tab_index);
loc.replace(loc.origin + loc.pathname + '?' + params.toString());
}")
ui <- fluidPage(
tags$head(tags$script(js, type ="text/javascript")),
tabsetPanel(
tabPanel("Tab 1", value = "tab1"),
tabPanel("Tab 2", value = "tab2",
actionButton("mybutton", "Reload the page",
onclick = "doReload('tab2')")),
id = "tabsets"
)
)
server <- function(input, output, session) {
observe({
params <- parseQueryString(session$clientData$url_search)
if ("tab_index" %in% names(params)) {
updateTabsetPanel(session, "tabsets", selected = params$tab_index)
}
})
}
shinyApp(ui = ui, server = server)
Very simple question: In my shiny UI, I have two buttons, A and B
on the click of button B I would like button A to be hidden, but I don't think updateActionButton has this capability. So how is this accomplished?
Thanks in advance
Dean built wonderful shinyjs package that has this functionality. Note that I added toggle instead of hide but you can switch if you like
rm(list = ls())
library(shiny)
library(shinyjs)
ui <- fluidPage(
useShinyjs(),
actionButton("hide","a"),
actionButton("b","b")
)
server <- shinyServer(function(input,output){
observeEvent(input$hide,{
toggle("b")
})
})
runApp(list(ui = ui, server = server))
I am trying to make an app that allow user to input data and then let shiny server to calculate some results, for example, I could make shiny generate a plot or a data table.
However, due to the space of the UI, I kind of "run out of" space. The input box, and documentation of the app take a whole screen. And when the shiny generate the results it will show at the very bottom of the screen.
Is there a way that I can make shiny pop-up a message box to show the result?
My sudo-code would be:
ui <- fluidPage(
textInput("text", "Name"),
numericInput("age", "Age", 20),
actionButton("demo", "Fill in fields with demo"))
server <- function(input, output, session) {
observeEvent(input$demo, {
****************************
OpenNewPage/MoveScreenDown()
****************************
updateTextInput(session, "text", value = H)
updateNumericInput(session, "age", value = "30")
})}
When clicking the "demo", a message box popup or I can make the screen move to the result part and allow the text to be at the top of the screen.
There are options to show your results in a separated window. But maybe will be easier to have everything on the same window.
You can use the shinyBS library to create a modal window to show the plot. Another option is to use JavaScript to move the scroll to the bottom of the page. I put the two options in the following example, so you can see which one is better for you.
library(shiny)
library(shinyBS)
runApp(list(
ui = shinyUI(fluidPage(
textInput("text", "Name"),
numericInput("age", "Age", 20),
# option 1, using ShinyBS with a modal window
actionButton("demo", "Using a modal"),
# modal window to show the plot
bsModal("largeModalID","Results", "demo", size = "large", plotOutput('plot')),
# Option 2, action button with a JavaScript function to move the scroll to the bottom
# after drawing the plot.
actionButton("demoJS", "Using JS",
# there is a delay to allow the renderPlot to draw the plot and you should
# change it according to the processes performed
onclick = "setTimeout( function() {
$('html, body').scrollTop( $(document).height() );},
300)"),
# to plot the results after click on "Using JS"
uiOutput("plotUI")
)
),
server = shinyServer(function(input, output, session) {
output$plot <- renderPlot({
# simple plot to show
plot(sin, -pi, 2*pi)
})
output$plotUI <- renderUI({
# this UI will show a plot only if "Using JS" is clicked
if (input$demoJS > 0)
# the margin-top attribute is just to put the plot lower in the page
div(style = "margin-top:800px", plotOutput('plot2'))
})
output$plot2 <- renderPlot({
# another simple plot,
plot(sin, -pi, 2*pi)
})
})
))
If you think that the JavaScript option works better for you, you could consider start using the shinyjs library, it includes very useful functions and you can easily add your own JavaScript code to your Shiny Apps.
I have programmed a shiny app in RStudio using R version 3.2.4. The app can be found here: https://koenvandenberge.shinyapps.io/tempConditioning/
Note that there is quite a lot of data to be loaded so it takes a couple of seconds to load.
I have deployed it on my Macbook and it seems as if it only works on Safari and Chromium browsers. It does not seem to work on Chrome or Firefox browsers. With this I mean that the plots which should be generated are not. The drop-down menu is present as it should be, so the app does not crash, but the plots that should be generated by selecting something from the drop-down menu do not.
Is there any way to fix this? You can find the code of my app below:
library(shiny)
library(scales)
load("countMatrix.RData")
countMatrixAllSub = as.data.frame(countMatrix$counts[,-1]) ; rm(countMatrix)
sampleNames = unlist(lapply(strsplit(colnames(countMatrixAllSub),split=".",fixed=TRUE), function(x) x[4]))
sampleNames[28] <- "3c0a"
treat=substr(sampleNames,2,2)
time=substr(sampleNames,3,nchar(sampleNames)-1)
timeC=as.numeric(time)
timeC[timeC==15]=0.25
ui <- shinyUI(fluidPage(
# Application title
titlePanel("Gene expression: conditioning experiment Gust"),
# Sidebar with a
sidebarLayout(
sidebarPanel(
selectInput("gene",
"Pick a gene",
choices = rownames(countMatrixAllSub))
),
# Show a plot of the generated distribution
mainPanel(
plotOutput("genePlot")
)
)
))
# Define server logic required to draw an expression plot
server <- shinyServer(function(input, output) {
output$genePlot <- renderPlot({
par(mar = c(5.1, 4.1, 3, 1))
plot(y=countMatrixAllSub[input$gene,],x=timeC, col=c("black","red")[as.numeric(factor(treat))], pch=19,cex=.6,xaxt="n",xlab="Time", ylab="Expression")
lines(x=unique(timeC[!timeC==0]),y=colMeans(sapply(unique(timeC[!timeC==0]), function(t) as.matrix(countMatrixAllSub[input$gene,treat=="c" & timeC==t]))), col=1)
lines(x=unique(timeC[!timeC==0]),y=colMeans(sapply(unique(timeC[!timeC==0]), function(t) as.matrix(countMatrixAllSub[input$gene,treat=="t" & timeC==t]))), col=2)
axis(1,at=c(0,0.25,1,3,6,9),labels=c("","15m","1h","3h","6h","9h"))
abline(v=c(0,0.25,1,3,6,9),col=alpha("grey",.6))
mtext("Conditioned",side=3, adj=0,col="red")
mtext("Unconditioned",side=3, adj=0.2, col="black")
})
})
# Run the application
shinyApp(ui = ui, server = server)