Related
Here I asked an similar question and got a working answer. But the solution does not work if 'actionButton' of sub segment is replace by 'selectInput'. On each selection of selectInput creates two outputs. Please help.. Thanks....
library(shiny)
ui <- fluidPage(
verbatimTextOutput("txt",placeholder = T), #"It is Created for Testing"
actionButton("addSeg", "Add a Segment"),
uiOutput("myUI")
)
server <- function(input, output, session) {
alld <- reactiveValues()
alld$ui <- list()
# Action to add new Segment
observeEvent(input$addSeg,{
new_id <- length(alld$ui) + 1
sub_name <- paste0("addSub_", new_id)
alld$ui[[new_id]] <- list(selectInput(sub_name,"Add a variable", choices = c("V1","V2"), selected = NULL))
observeEvent(input[[sub_name]], {
new_text_id <- length(alld$ui[[new_id]]) + 1
alld$ui[[new_id]][[new_text_id]] <- HTML(paste0("Variable ",input[[sub_name]]," added<br>"))
}, ignoreInit = TRUE)
})
output$myUI <- renderUI({alld$ui})
output$txt <- renderText({class(alld$ui)})
}
shinyApp(ui, server)
This behaviour occurs because the custom UI element is re-rendered every time a new element is added to the list. Once you click "V2" and the new text element is added, the selectInput itself re-renders and resets to V1, which is noticed by the observer you've created.
The following might be a solution for you:
observeEvent(input$addSeg,{
new_id <- length(alld$ui) + 1
sub_name <- paste0("addSub_", new_id)
alld$ui[[new_id]] <- list(
selectInput(sub_name,
"Add a variable",
choices = c("", "V1","V2"),
selected = "")
)
observeEvent(input[[sub_name]], {
if (input[[sub_name]] == "") return()
new_text_id <- length(alld$ui[[new_id]]) + 1
alld$ui[[new_id]][[new_text_id]] <- HTML(paste0("Variable ",input[[sub_name]]," added<br>"))
}, ignoreInit = TRUE)
})
What I've done here is add an empty option to your selectInputs, and a condition to the corresponding observer that it shouldn't do anything if the input is empty. This way, I'm harnessing the "resetting" behaviour to be useful instead of annoying.
To filter a data.frame with lots of variables I created a selectizeInput which allows you to select one of the columns of the data. This then creates another selectizeInput for the selected variable which can be used for subsetting the data. The selected value of the second selectizeInput is rendered below.
This is what it looks like
I want to render the selected values of these inputs so that these can be deleted by the user by clicking the black cross. Also a selection of var2 should not be deleted when the Filter selectizeInput is changed to var1.
So it should look like this (assuming the user previously selected value z in var2 and then value a in var1.
Anyone knows a good solution in shiny?
This is the code:
library(shiny)
data <- data.frame(var1 = c("a", "b"), var2 = c("y", "z"))
ui <- fluidPage(
selectizeInput("filter", label = "Filter",
multiple = FALSE, choices = c("var1", "var2")),
uiOutput("filter_var"),
uiOutput("selected_filter_value")
)
server <- function(input, output) {
observeEvent(input$filter, {
# dynamically generate selectizeInput for filter
output$filter_var <- renderUI({
selectizeInput(input$filter, label = input$filter,
choices = data[input$filter], multiple = TRUE)
})
})
# show selected filter values
# selected filter values should stay when choosing new input filter variable
# these should be deletable
observeEvent(input[[input$filter]], {
output$selected_filter_value <- renderUI({
textOutput("text_out")
})
output$text_out <- renderText({
paste0(input$filter, ": ", input[[input$filter]])
})
})
}
shinyApp(ui, server)
Well, I had to rearrange quite a lot and this whole problem is more about finding the right implementation for your case.
You can probably deduct most of it just looking at the code at the end of this post.
Main things explained: You didn't actually say what deleting means to you. So I just assumed you wanted the cells to not appear in the select boxes anymore. For that, I excluded NAs and replaced cells with an NA to show that they are deleted.
I rearranged the select values, such that we actually can delete certain cells, giving row and column names instead of just their values.
And most important, the buttons you wanted to create are dynamic UI elements with dynamic observers, which are then addressed to delete the certain cell.
Note: This solution is not optimal, since I specifically aimed to stay just on the R side of shiny. You can achieve a much more elegant and resource saving solution if you use JavaScript and shiny's custom messages.
Also: I did not address your request to let the selected values visible if the first select box changes. But this is a rather small issue, if you reconsider your setup. And I didn't want to diverge too much from you original code to not be confusing.
Code now:
library(shiny)
data <- data.frame(var1 = c("a", "b"), var2 = c("y", "z"))
ui <- fluidPage(
selectizeInput("filter", label = "Filter",
multiple = FALSE, choices = c("var1", "var2")),
uiOutput("filter_var"),
uiOutput("selected_filter_value")
)
server <- function(input, output) {
# Pulled out from original observeEvent
makeSecondInput <- function() {
output$filter_var <- renderUI({
# Names are not enough when wanting to delete data.frame rows (because of duplicates).
# So we instead use row numbers and set the actual values as labels.
choiceData <- na.exclude(data[input$filter])
choices <- rownames(choiceData)
names(choices) <- choiceData[, input$filter]
selectizeInput(input$filter, label = input$filter, selected = input[[input$filter]],
choices = choices, multiple = TRUE)
})
}
observeEvent(input$filter, {
makeSecondInput()
})
# Install a manual trigger to redraw input field, when an option is killed.
trigger <- reactiveVal()
observeEvent(trigger(), ignoreNULL = TRUE, {
makeSecondInput()
})
# Keep track of created observers, so dynamic creation does not wildly stack them up.
observersCreated <- character()
makeButtonObserver <- function(buttonname, colname, rowname) {
# For each delete-button created, install observer to delete data.frame cell.
observeEvent(input[[buttonname]], {
data[rowname, colname] <<- NA
# Force re-evaluation of observer above.
trigger(runif(1))
})
# Track that this button is equipped. (And re-creation of the same button does not add another obs.)
# Note: Observers DON'T get automagically removed after actionButton is no longer in the UI.
observersCreated <<- c(observersCreated, buttonname)
}
observeEvent(input[[input$filter]], {
output$selected_filter_value <- renderUI({
# Could be a list, so splitting that up.
lapply(input[[input$filter]], function(v) {
buttonname <- paste("kill", input$filter, v, sep = "_")
if (!(buttonname %in% observersCreated)) {
makeButtonObserver(buttonname, input$filter, v)
}
span(
paste0(input$filter, ": ", data[v, input$filter]),
actionButton(buttonname, "x")
)
})
})
})
}
shinyApp(ui, server)
This is what I currently have. There are still some issues which I couldn't solve.
Problems:
if I make some selections in input1, then switch from input1 to input2 and unclick one of the selections from input1 and then switch back to input1 these changes will be unmade
the checkboxes are rerendered when a new one is added and in this process sorted which changes the order
Code:
library(shiny)
library(shinyWidgets)
data <- data.frame(var1 = c("a", "b"), var2 = c("y", "z"))
ui <- fluidPage(
selectizeInput("filter", label = "Filter",
multiple = FALSE, choices = c("var1", "var2")),
uiOutput("filter_var"),
uiOutput("selected_filter_value")
)
server <- function(input, output, session) {
values <- reactiveValues(
filter_vals = list(var1 = list(), var2 = list()),
observers = NULL
)
# dynamically generate selectizeInput for variable selected in filter
# set selected values to previous selections
observeEvent(input$filter, {
output$filter_var <- renderUI({
selectInput(input$filter, label = input$filter,
selected = values$filter_vals[[input$filter]],
choices = data[input$filter], multiple = TRUE, selectize = TRUE)
})
})
# store selected values in list
observeEvent(input[[input$filter]], {
values$filter_vals[[input$filter]] <- input[[input$filter]]
})
# we need this because observeEvent is not triggered if input is empty after deleting all selections
observe({
if (is.null(input[[input$filter]])) {
values$filter_vals[[input$filter]] <- list()
}
})
# add an observer for newly created checkbox
# if checkbox is clicked delete entry in list
# keep a list of all existing observers
make_delete_observer <- function(name) {
observeEvent(input[[name]], {
req(input[[name]] == FALSE)
var <- stringr::str_split(name, "_")[[1]][1]
val <- as.integer(stringr::str_split(name, "_")[[1]][2])
values$filter_vals[[var]] <- intersect(values$filter_vals[[var]][-val],
values$filter_vals[[var]])
updateSelectInput(session, var, selected = values$filter_vals[[var]])
})
}
# render selected values which are stored in a list as checkboxes
# add an observeEvent for each checkbox
# store selected values in list
output$selected_filter_value <- renderUI({
req(values$filter_vals[[input$filter]])
req(any(sapply(values$filter_vals, length) > 0))
tag_list <- tagList()
for (i in seq_along(values$filter_vals)) {
for (j in seq_along(values$filter_vals[[i]])) {
new_input_name <- paste0(names(values$filter_vals)[i], "_", j)
new_input <- prettyCheckbox(
inputId = new_input_name, value = TRUE,
label = paste0(names(values$filter_vals)[i], ": ", values$filter_vals[[i]][j]),
icon = icon("close"), status = "danger", outline = FALSE, plain = TRUE
)
# create observer only if it does not exist yet
if (!(new_input_name %in% values$observers)) {
values$observers <- append(values$observers, new_input_name)
make_delete_observer(new_input_name)
}
tag_list <- tagAppendChild(tag_list, new_input)
}
}
tag_list
})
}
shinyApp(ui, server)
I am new to RShiny. I want to populate RShiny dropdowns based previous selections.
For E.g. in the image below,
User first selects the 'route', upon which 'schedule' drop-down gets populated, then user selects 'schedule', then 'trip' drop-down is populated and user selects a 'trip' input.
This is my code:
library(shiny)
library("plotly")
library(lubridate)
require(rgl)
require(akima)
library(dplyr)
library(DT)
data335 <<- read.csv("final335eonly.csv")
#data335[c(2,4,5,8,9,10)] = lapply(data335[c(2,4,5,8,9,10)], as.numeric)
routes <<- as.vector(unique(data335[,'route_no']))
ui <- fluidPage(
titlePanel("Demand Analysis"),
selectInput("routeInput", "Select the route", choices = routes),
selectInput("scheduleInput", "Select the schedule", c("")),
selectInput("tripInput", "Select the trip", c(""))
)
server <- function(input, output, session) {
observeEvent(input$routeInput,
{
x <<- input$routeInput
updateSelectInput(session, "scheduleInput",
choices = data335[data335$route_no == input$routeInput, ]$schedule_no,selected = tail(x, 1)
)
}
)
observeEvent(input$scheduleInput,
{
y <<- input$scheduleInput
updateSelectInput(session, "tripInput",
choices = data335[(data335$route_no == input$routeInput & data335$schedule_no == input$scheduleInput), ]$trip_no,selected = tail(y, 1)
)
}
)
}
shinyApp(ui = ui, server = server)
The input csv file required is here:
Whenever I try to run this seemingly simple code, eventhough the UI appears, when I try to select the inputs in dropdown, RShiny crashes.
Can you please let me know what is causing this?
The problem is happening because you are not giving unique values as choices. data335[data335$route_no == input$routeInput, ]$schedule_no have duplicate values which causes the crash.
Also, you are selecting the value of input$routeInput in your scheduleInput, which is not listed in the choice could be another reason for the crash.
Just commenting the two statements and adding unique to your choices resolves the crash.
Also as #parth pointed out in his comments why are you using <<- everywhere in your code, it not necessary. Although its not the cause of the crash, until and unless you want to share variables between sessions use of <<- inside the server is not a good practice.
Here is your code with the commented section with two selected arguments commented and unique added that works:
library(shiny)
library("plotly")
library(lubridate)
require(rgl)
require(akima)
library(dplyr)
library(DT)
data335 <<- read.csv("final335eonly.csv", stringsAsFactors = FALSE)
routes <<- as.vector(unique(data335[,'route_no']))
ui <- fluidPage(
titlePanel("Demand Analysis"),
selectInput("routeInput", "Select the route", choices = routes),
selectInput("scheduleInput", "Select the schedule", c("")),
selectInput("tripInput", "Select the trip", c(""))
)
server <- function(input, output, session) {
observeEvent(input$routeInput,
{
x <<- input$routeInput
updateSelectInput(session, "scheduleInput",
choices =unique(data335[data335$route_no == input$routeInput, ]$schedule_no),#selected = tail(x, 1)
)
}
)
observeEvent(input$scheduleInput,
{
y <<- input$scheduleInput
updateSelectInput(session, "tripInput",
choices = unique(data335[(data335$route_no == input$routeInput & data335$schedule_no == input$scheduleInput), ]$trip_no),#selected = tail(y, 1)
)
}
)
}
shinyApp(ui = ui, server = server)
Goal: I'm working on a bioinformatics project. I'm currently trying to implement R code that dynamically creates tabPanels (they are essentially carbon copies except for the data output).
Implementation: After doing some research I implemented this solution. It works in a way (the panels that I'm "carbon copying" are created), but the data that I need cannot be displayed.
Problem: I'm sure that the way I'm displaying my data is fine. The problem is that I can't use the same output function to display the data as seen here. So let me get to the code...
ui.R
library(shiny)
library(shinythemes)
library(dict)
library(DT)
...# Irrelevant functions removed #...
geneinfo <- read.table(file = "~/App/final_gene_info.csv",
header = TRUE,
sep = ",",
na.strings = "N/A",
as.is = c(1,2,3,4,5,6,7))
ui <- navbarPage(inverse = TRUE, "GENE PROJECT",
theme = shinytheme("cerulean"),
tabPanel("Home",
#shinythemes::themeSelector(),
fluidPage(
includeHTML("home.html")
)),
tabPanel("Gene Info",
h2('Detailed Gene Information'),
DT::dataTableOutput('table')),
tabPanel("File Viewer",
sidebarLayout(
sidebarPanel(
selectizeInput(inputId = "gene", label = "Choose a Gene", choice = genes, multiple = TRUE),
selectInput(inputId = "organism", label = "Choose an Organism", choice = orgs),
selectInput(inputId = "attribute", label = "Choose an Other", choice = attributes),
width = 2),
mainPanel(
uiOutput('change_tabs'),
width = 10))),
tabPanel("Alignment")
)
I'm using uiOutput to generate tabs dynamically on the server side....
server.R
server <- function (input, output, session) {
# Generate proper files from user input
fetch_files <- function(){
python <- p('LIB', 'shinylookup.py', python=TRUE)
system(sprintf('%s %s %s', python, toString(genie), input$organism), wait = TRUE)
print('Done with Python file generation.')
# Fetch a temporary file for data output
fetch_temp <- function(){
if(input$attribute != 'Features'){
if(input$attribute != 'Annotations'){
chosen <- toString(attribute_dict[[input$attribute]])
}
else{
chosen <- toString(input$sel)
extension <<- '.anno'
}
}
else{
chosen <- toString(input$sel)
extension <<- '.feat'
}
count = 0
oneline = ''
f <- paste(toString(genie), toString(input$organism), sep = '_')
f <- paste(f, extension, sep = '')
# Writes a temporary file to display output to the UI
target <- p('_DATA', f)
d <- dict_fetch(target)
temp_file <- tempfile("temp_file", p('_DATA', ''), fileext = '.txt')
write('', file=temp_file)
vectorofchar <- strsplit(toString(d[[chosen]]), '')[[1]]
for (item in vectorofchar){
count = count + 1
oneline = paste(oneline, item, sep = '')
# Only 60 characters per line (Find a better solution)
if (count == 60){
write(toString(oneline), file=temp_file, append=TRUE)
oneline = ''
count = 0
}
}
write(toString(oneline), file=temp_file, append=TRUE)
return(temp_file)
}
# Get the tabs based on the number of genes selected in the UI
fetch_tabs <- function(Tabs, OId, s = NULL){
count = 0
# Add a select input or nothing at all based on user input
if(is.null(s)==FALSE){
selection <- select(s)
x <- selectInput(inputId = 'sel', label = "Choose an Annotation:", choices = selection$keys())
}
else
x <- ''
for(gene in input$gene){
if(count==0){myTabs = character()}
count = count + 1
genie <<- gene
fetch_files()
file_tab <- lapply(sprintf('File for %s', gene), tabPanel
fluidRow(
titlePanel(sprintf("File for %s:", gene)),
column(5,
pre(textOutput(outputId = "file")),offset = 0))
)
addTabs <- c(file_tab, lapply(sprintf('%s for %s',paste('Specific', Tabs), gene), tabPanel,
fluidRow(
x,
titlePanel(sprintf("Attribute for %s:", gene)),
column(5,
pre(textOutput(outputId = OId), offset = 0)))
))
# Append additional tabs every iteration
myTabs <- c(myTabs, addTabs)
}
return(myTabs)
}
# Select the proper file and return a dictionary for selectInput
select <- function(ext, fil=FALSE){
f <- paste(toString(genie), toString(input$organism), sep = '_')
f <- paste(f, ext, sep = '')
f <- p('_DATA', f)
if(fil==FALSE){
return(dict_fetch(f))
}
else if(fil==TRUE){
return(toString(f))
}
}
# Output gene info table
output$table <- DT::renderDataTable(
geneinfo,
filter = 'top',
escape = FALSE,
options = list(autoWidth = TRUE,
options = list(pageLength = 10),
columnDefs = list(list(width = '600px', targets = c(6))))
)
observe({
x <- geneinfo[input$table_rows_all, 2]
if (is.null(x))
x <- genes
updateSelectizeInput(session, 'gene', choices = x)
})
# Output for the File tab
output$file <- renderText({
extension <<- '.gbk'
f <- select(extension, f=TRUE)
includeText(f)
})
# Output for attributes with ony one property
output$attributes <- renderText({
extension <<- '.kv'
f <- fetch_temp()
includeText(f)
})
# Output for attributes with multiple properties (features, annotations)
output$sub <- renderText({
f <- fetch_temp()
includeText(f)
})
# Input that creates tabs and selectors for more input
output$change_tabs <- renderUI({
# Fetch all the appropriate files for output
Tabs = input$attribute
if(input$attribute == 'Annotations'){
extension <<- '.anno'
OId = 'sub'
s <- extension
}
else if(input$attribute == 'Features'){
extension <<- '.feat'
OId = 'sub'
s <- extension
}
else{
OId = 'attributes'
s <- NULL
}
myTabs <- fetch_tabs(Tabs, OId, s = s)
do.call(tabsetPanel, myTabs)
})
}
)
Explanation: Now I'm aware that there's a lot to look at here.. But my problem exists within output$change_tabs (it's the last function), which calls fetch_tabs(). Fetch tabs uses the input$gene (a list of genes via selectizeInput(multiple=TRUE)) to dynamically create a set of 2 tabs per gene selected by the user.
What's Happening: So if the user selects 2 genes then 4 tabs are created. With 5 genes 10 tabs are created... And so on and so forth... Each tab is EXACTLY THE SAME, except for the data.
Roadblocks: BUT... for each tab I'm trying to use the same output Id (since they are EXACTLY THE SAME) for the data that I want to display (textOutput(outputId = "file")). As explained above in the second link, this simply does not work because HTML.
Questions: I've tried researching several solutions, but I would rather not have to implement this solution. I don't want to have to rewrite so much code. Is there any way I can add a reactive or observer function that can wrap or fix my output$file function? Or is there a way for me to add information to my tabs after the do.call(tabsetPanel, myTabs)? Am I thinking about this the right way?
I'm aware that my code isn't commented very well so I apologize in advance. Please feel free to critique my coding style in the comments, even if you don't have a solution. Please and thank you!
I've come up with a very VERY crude answer that will work for now...
Here is the answer from #BigDataScientist
My Issue with BigDataScientist's Answer:
I can't dynamically pass data to the outputs. The output functions are not interpreted until they are needed... So if I wanted to pass the for loop iterator that you created (iter) into the dynamically created outputs, then I wouldn't be able to do that. It can only take static data
My Solution:
I end up taking advantage of sys.calls() solution I found here in order to get the name of the function as a string. The name of the function has the info I need (in this case a number).
library(shiny)
library(shinythemes)
myTabs <<- list()
conv <- function(v1) {
deparse(substitute(v1))
}
ui <- navbarPage(inverse = TRUE, "GENE PROJECT",
theme = shinytheme("cerulean"),
tabPanel("Gene Info",
sidebarLayout(
sidebarPanel(
sliderInput("bins",
"Number of bins:",
min = 1,
max = 5,
value = 3)
),
# Show a plot of the generated distribution
mainPanel(
uiOutput('changeTab')
)
)
)
)
server <- function(input, output) {
observe({
b <<- input$bins
myTabs <<- list()
# Dynamically Create output functions
# Dynamically Create formatted tabs
# Dynamically Render the tabs with renderUI
for(iter in 1:b){
x <<- iter
output[[sprintf("tab%s", iter)]] <- renderText({
temp <- deparse(sys.calls()[[sys.nframe()-3]])
x <- gsub('\\D','',temp)
x <- as.numeric(x)
f <- sprintf('file%s.txt', x)
includeText(f)
})
addTabs <<- lapply(sprintf('Tab %s', iter), tabPanel,
fluidRow(
titlePanel(sprintf("Tabble %s:", iter)),
column(5,
pre(textOutput(outputId = sprintf('%s%s','tab', iter))))))
myTabs <<- c(myTabs, addTabs)
}
myTabs <<- c(myTabs, selected = sprintf('Tab %s', x))
output$changeTab <- renderUI({
do.call(tabsetPanel, myTabs)
})
})
}
# Run the application
shinyApp(ui = ui, server = server)
I think your being a victim of this behavior. Try:
for (el in whatever) {
local({
thisEl <- el
...
})
}
like Joe suggests in the first reply to the Github issue I linked to. This is only necessary if you're using a for loop. lapply already takes el as an argument, so you get this "dynamic evaluation" benefit (for lack of a better name) for free.
For readability, I'm going to quote most of Joe's answer here:
You're the second person at useR that I talked to that was bitten by this behavior in R. It's because all the iterations of the for loop share the same reference to el. So when any of the created reactive expressions execute, they're using whatever the final value of el was.
You can fix this either by 1) using lapply instead of a for loop; since each iteration executes as its own function call, it gets its own reference to el; or 2) using a for loop but introducing a local({...}) inside of there, and creating a local variable in there whose value is assigned to el outside of the reactive.
The goal
I am working on a Shiny app that allows the user to upload their own data and focus on the entire data or a subset by providing data filtering widgets described by the below graph
The select input "Variable 1" will display all the column names of the data uploaded by the user and the selectize input "Value" will display all the unique values of the corresponding column selected in "Variable 1". Ideally, the user will be able to add as many such rows ("Variable X" + "Value") as possible by some sort of trigger, one possibility being clicking the "Add more" action button.
A possible solution
After looking up online, I've found one promising solution given by Nick Carchedi pasted below
ui.R
library(shiny)
shinyUI(pageWithSidebar(
# Application title
headerPanel("Dynamically append arbitrary number of inputs"),
# Sidebar with a slider input for number of bins
sidebarPanel(
uiOutput("allInputs"),
actionButton("appendInput", "Append Input")
),
# Show a plot of the generated distribution
mainPanel(
p("The crux of the problem is to dynamically add an arbitrary number of inputs
without resetting the values of existing inputs each time a new input is added.
For example, add a new input, set the new input's value to Option 2, then add
another input. Note that the value of the first input resets to Option 1."),
p("I suppose one hack would be to store the values of all existing inputs prior
to adding a new input. Then,", code("updateSelectInput()"), "could be used to
return inputs to their previously set values, but I'm wondering if there is a
more efficient method of doing this.")
)
))
server.R
library(shiny)
shinyServer(function(input, output) {
# Initialize list of inputs
inputTagList <- tagList()
output$allInputs <- renderUI({
# Get value of button, which represents number of times pressed
# (i.e. number of inputs added)
i <- input$appendInput
# Return if button not pressed yet
if(is.null(i) || i < 1) return()
# Define unique input id and label
newInputId <- paste0("input", i)
newInputLabel <- paste("Input", i)
# Define new input
newInput <- selectInput(newInputId, newInputLabel,
c("Option 1", "Option 2", "Option 3"))
# Append new input to list of existing inputs
inputTagList <<- tagAppendChild(inputTagList, newInput)
# Return updated list of inputs
inputTagList
})
})
The downside
As pointed by Nick Carchedi himself, all the existing input widgets will undesirably get reset every time when a new one is added.
A promising solution for data subsetting/filtering in Shiny
As suggested by warmoverflow, the datatable function in DT package provides a nice way to filter the data in Shiny. See below a minimal example with data filtering enabled.
library(shiny)
shinyApp(
ui = fluidPage(DT::dataTableOutput('tbl')),
server = function(input, output) {
output$tbl = DT::renderDataTable(
iris, filter = 'top', options = list(autoWidth = TRUE)
)
}
)
If you are going to use it in your Shiny app, there are some important aspects that are worth noting.
Filtering box type
For numeric/date/time columns: range sliders are used to filter rows within ranges
For factor columns: selectize inputs are used to display all possible categories
For character columns: ordinary search boxes are used
How to obtain the filtered data
Suppose the table output id is tableId, use input$tableId_rows_all as the indices of rows on all pages (after the table is filtered by the search strings). Please note that input$tableId_rows_all returns the indices of rows on all pages for DT (>= 0.1.26). If you use the DT version by regular install.packages('DT'), only the indices of the current page are returned
To install DT (>= 0.1.26), refer to its GitHub page
Column width
If the data have many columns, column width and filter box width will be narrow, which makes it hard to see the text as report here
Still to be solved
Despite some known issues, datatable in DT package stands as a promising solution for data subsetting in Shiny. The question itself, i.e. how to dynamically append arbitrary number of input widgets in Shiny, nevertheless, is interesting and also challenging. Until people find a good way to solve it, I will leave this question open :)
Thank you!
are you looking for something like this?
library(shiny)
LHSchoices <- c("X1", "X2", "X3", "X4")
#------------------------------------------------------------------------------#
# MODULE UI ----
variablesUI <- function(id, number) {
ns <- NS(id)
tagList(
fluidRow(
column(6,
selectInput(ns("variable"),
paste0("Select Variable ", number),
choices = c("Choose" = "", LHSchoices)
)
),
column(6,
numericInput(ns("value.variable"),
label = paste0("Value ", number),
value = 0, min = 0
)
)
)
)
}
#------------------------------------------------------------------------------#
# MODULE SERVER ----
variables <- function(input, output, session, variable.number){
reactive({
req(input$variable, input$value.variable)
# Create Pair: variable and its value
df <- data.frame(
"variable.number" = variable.number,
"variable" = input$variable,
"value" = input$value.variable,
stringsAsFactors = FALSE
)
return(df)
})
}
#------------------------------------------------------------------------------#
# Shiny UI ----
ui <- fixedPage(
verbatimTextOutput("test1"),
tableOutput("test2"),
variablesUI("var1", 1),
h5(""),
actionButton("insertBtn", "Add another line")
)
# Shiny Server ----
server <- function(input, output) {
add.variable <- reactiveValues()
add.variable$df <- data.frame("variable.number" = numeric(0),
"variable" = character(0),
"value" = numeric(0),
stringsAsFactors = FALSE)
var1 <- callModule(variables, paste0("var", 1), 1)
observe(add.variable$df[1, ] <- var1())
observeEvent(input$insertBtn, {
btn <- sum(input$insertBtn, 1)
insertUI(
selector = "h5",
where = "beforeEnd",
ui = tagList(
variablesUI(paste0("var", btn), btn)
)
)
newline <- callModule(variables, paste0("var", btn), btn)
observeEvent(newline(), {
add.variable$df[btn, ] <- newline()
})
})
output$test1 <- renderPrint({
print(add.variable$df)
})
output$test2 <- renderTable({
add.variable$df
})
}
#------------------------------------------------------------------------------#
shinyApp(ui, server)
Now, I think that I understand better the problem.
Suppose the user selects the datasets::airquality dataset (here, I'm showing only the first 10 rows):
The field 'Select Variable 1' shows all the possible variables based on the column names of said dataset:
Then, the user selects the condition and the value to filter the dataset by:
Then, we want to add a second filter (still maintaining the first one):
Finally, we get the dataset filtered by the two conditions:
If we want to add a third filter:
You can keep adding filters until you run out of data.
You can also change the conditions to accommodate factors or character variables. All you need to do is change the selectInput and numericInput to whatever you want.
If this is what you want, I've solved it using modules and by creating a reactiveValue (tmpFilters) that contains all selections (variable + condition + value). From it, I created a list with all filters (tmpList) and from it I created the proper filter (tmpListFilters) to use with subset.
This works because the final dataset is "constantly" being subset by this reactiveValue (the tmpFilters). At the beginning, tmpFilters is empty, so we get the original dataset. Whenever the user adds the first filter (and other filters after that), this reactiveValue gets updated and so does the dataset.
Here's the code for it:
library(shiny)
# > MODULE #####################################################################
## |__ MODULE UI ===============================================================
variablesUI <- function(id, number, LHSchoices) {
ns <- NS(id)
tagList(
fluidRow(
column(
width = 4,
selectInput(
inputId = ns("variable"),
label = paste0("Select Variable ", number),
choices = c("Choose" = "", LHSchoices)
)
),
column(
width = 4,
selectInput(
inputId = ns("condition"),
label = paste0("Select condition ", number),
choices = c("Choose" = "", c("==", "!=", ">", ">=", "<", "<="))
)
),
column(
width = 4,
numericInput(
inputId = ns("value.variable"),
label = paste0("Value ", number),
value = NA,
min = 0
)
)
)
)
}
## |__ MODULE SERVER ===========================================================
filter <- function(input, output, session){
reactive({
req(input$variable, input$condition, input$value.variable)
fullFilter <- paste0(
input$variable,
input$condition,
input$value.variable
)
return(fullFilter)
})
}
# Shiny ########################################################################
## |__ UI ======================================================================
ui <- fixedPage(
fixedRow(
column(
width = 5,
selectInput(
inputId = "userDataset",
label = paste0("Select dataset"),
choices = c("Choose" = "", ls("package:datasets"))
),
h5(""),
actionButton("insertBtn", "Add another filter")
),
column(
width = 7,
tableOutput("finalTable")
)
)
)
## |__ Server ==================================================================
server <- function(input, output) {
### \__ Get dataset from user selection ------------------------------------
originalDF <- reactive({
req(input$userDataset)
tmpData <- eval(parse(text = paste0("datasets::", input$userDataset)))
if (!class(tmpData) == "data.frame") {
stop("Please select a dataset of class data.frame")
}
tmpData
})
### \__ Get the column names -----------------------------------------------
columnNames <- reactive({
req(input$userDataset)
tmpData <- eval(parse(text = paste0("datasets::", input$userDataset)))
names(tmpData)
})
### \__ Create Reactive Filter ---------------------------------------------
tmpFilters <- reactiveValues()
### \__ First UI Element ---------------------------------------------------
### Add first UI element with column names
observeEvent(input$userDataset, {
insertUI(
selector = "h5",
where = "beforeEnd",
ui = tagList(variablesUI(paste0("var", 1), 1, columnNames()))
)
})
### Update Reactive Filter with first filter
filter01 <- callModule(filter, paste0("var", 1))
observe(tmpFilters[['1']] <- filter01())
### \__ Other UI Elements --------------------------------------------------
### Add other UI elements with column names and update the filter
observeEvent(input$insertBtn, {
btn <- sum(input$insertBtn, 1)
insertUI(
selector = "h5",
where = "beforeEnd",
ui = tagList(variablesUI(paste0("var", btn), btn, columnNames()))
)
newFilter <- callModule(filter, paste0("var", btn))
observeEvent(newFilter(), {
tmpFilters[[paste0("'", btn, "'")]] <- newFilter()
})
})
### \__ Dataset with Filtered Results --------------------------------------
resultsFiltered <- reactive({
req(filter01())
tmpDF <- originalDF()
tmpList <- reactiveValuesToList(tmpFilters)
if (length(tmpList) > 1) {
tmpListFilters <- paste(tmpList, "", collapse = "& ")
} else {
tmpListFilters <- unlist(tmpList)
}
tmpResult <- subset(tmpDF, eval(parse(text = tmpListFilters)))
tmpResult
})
### \__ Print the Dataset with Filtered Results ----------------------------
output$finalTable <- renderTable({
req(input$userDataset)
if (is.null(tmpFilters[['1']])) {
head(originalDF(), 10)
} else {
head(resultsFiltered(), 10)
}
})
}
#------------------------------------------------------------------------------#
shinyApp(ui, server)
# End
If you are looking for a data subsetting/filtering in Shiny Module :
filterData from package shinytools can do the work. It returns an expression as a call but it can also return the data (if your dataset is not too big).
library(shiny)
# remotes::install_github("ardata-fr/shinytools")
library(shinytools)
ui <- fluidPage(
fluidRow(
column(
3,
filterDataUI(id = "ex"),
actionButton("AB", label = "Apply filters")
),
column(
3,
tags$strong("Expression"),
verbatimTextOutput("expression"),
tags$br(),
DT::dataTableOutput("DT")
)
)
)
server <- function(input, output) {
x <- reactive({iris})
res <- callModule(module = filterDataServer, id = "ex", x = x, return_data = FALSE)
output$expression <- renderPrint({
print(res$expr)
})
output$DT <- DT::renderDataTable({
datatable(data_filtered())
})
data_filtered <- eventReactive(input$AB, {
filters <- eval(expr = res$expr, envir = x())
x()[filters,]
})
}
shinyApp(ui, server)
You can also use lazyeval or rlang to evaluate the expression :
filters <- lazyeval::lazy_eval(res$expr, data = x())
filters <- rlang::eval_tidy(res$expr, data = x())
You need to check for existing input values and use them if available:
# Prevent dynamic inputs from resetting
newInputValue <- "Option 1"
if (newInputId %in% names(input)) {
newInputValue <- input[[newInputId]]
}
# Define new input
newInput <- selectInput(newInputId, newInputLabel, c("Option 1", "Option 2", "Option 3"), selected=newInputValue)
A working version of the gist (without the reset problem) can be found here: https://gist.github.com/motin/0d0ed0d98fb423dbcb95c2760cda3a30
Copied below:
ui.R
library(shiny)
shinyUI(pageWithSidebar(
# Application title
headerPanel("Dynamically append arbitrary number of inputs"),
# Sidebar with a slider input for number of bins
sidebarPanel(
uiOutput("allInputs"),
actionButton("appendInput", "Append Input")
),
# Show a plot of the generated distribution
mainPanel(
p("This shows how to add an arbitrary number of inputs
without resetting the values of existing inputs each time a new input is added.
For example, add a new input, set the new input's value to Option 2, then add
another input. Note that the value of the first input does not reset to Option 1.")
)
))
server.R
library(shiny)
shinyServer(function(input, output) {
output$allInputs <- renderUI({
# Get value of button, which represents number of times pressed (i.e. number of inputs added)
inputsToShow <- input$appendInput
# Return if button not pressed yet
if(is.null(inputsToShow) || inputsToShow < 1) return()
# Initialize list of inputs
inputTagList <- tagList()
# Populate the list of inputs
lapply(1:inputsToShow,function(i){
# Define unique input id and label
newInputId <- paste0("input", i)
newInputLabel <- paste("Input", i)
# Prevent dynamic inputs from resetting
newInputValue <- "Option 1"
if (newInputId %in% names(input)) {
newInputValue <- input[[newInputId]]
}
# Define new input
newInput <- selectInput(newInputId, newInputLabel, c("Option 1", "Option 2", "Option 3"), selected=newInputValue)
# Append new input to list of existing inputs
inputTagList <<- tagAppendChild(inputTagList, newInput)
})
# Return updated list of inputs
inputTagList
})
})
(The solution was guided on Nick's hints in the original gist from where you got the code of the promising solution)