Related
I have a data table that has a lot of written text. I would like to organize this in an easy to read way. I have tried a few ideas but they do not seem to work. I am also open to any advice on how I should be organizing the data table.
Here is my code I am trying to control the width of the columns. Right now the width are extremely wide and i would like to narrow selected columns quite a bit.
output$newsfeed = renderDataTable({
datatable(df, rownames = F, extensions = "Scroller",
options = list(deferRender = TRUE,
scrollY = 400,
scrollX = TRUE,
scroller = TRUE,
autoWidth = TRUE,
columnDefs = list(list(width = '50px', targets = list(1,2,3,4,5)))
))
})
Thank you for your help.
Try using width = '10%' for the columns of interest.
library(DT)
library(shiny)
dat <- data.frame(
V1 = c("A", "B"),
V2 = c(
"A cool guy living in US and Canada",
"A cool guy living in New York"
),
V3 = c(
"A cool guy living in US and Canada",
"A cool guy living in New Jersey"
),
V4 = c(
"A cool guy living in US and Canada",
"A cool guy living in California"
),
V5 = c(
"A cool guy living in the US and else where",
"A cool guy living in Texas"
),
stringsAsFactors = FALSE
)
ui <- fluidPage(
DTOutput("table")
)
server <- function(input, output){
output[["table"]] <- renderDT({
datatable(dat, options = list(
deferRender = TRUE,
scrollY = 400,
scrollX = TRUE,
scroller = TRUE,
autoWidth = TRUE,
columnDefs = list(list(width = '10%', targets = c(2,3,4)))
)) # %>% formatStyle(columns = c(2,3), width='20px')
})
}
shinyApp(ui, server)
I guess 'extensions = "Scroller"' conflicts controlling column width.
When I add 'extensions = "Scroller"' to YBS's code then texts is not wrapped.
I am trying to convert cells in a DT:datatable into clickable buttons, which action is append a new tab on my shiny app.
I have been using this post R Shiny: Handle Action Buttons in Data Table as reference but in my case the buttons shows up like this:
My app is a bit more complicated but I will try to re-create the part that does not work
This is an shorter version of my data.frame
mut_genes <- structure(list(acc_num = c("BM0042985", "BM0393251", "BM0673028"), disease = c("Sucrase isomaltase deficiency", "Metachromatic leukodystrophy", "Fatal surfactant deficiency"), gene = c("SI", "ARSA", "ABCA3"), chrom = c("3q25.2-q26.2", "22q13.31-qter", "16p13.3"), genename = c("Sucrase-isomaltase", "arylsulfatase A", "ATP binding cassette subfamily A member 3"), gdbid = c("120377", "119007", "3770735"), omimid = c("609845", "607574", "601615"), amino = c("Leu-Pro", "Glu-Lys", "Met-Ile"), deletion = c(NA_character_, NA_character_, NA_character_), insertion = c(NA_character_, NA_character_, NA_character_), codon = c(341L, 331L, 1L), codonAff = c(341L, 331L, 1L), hgvs = c("1022T>C", "991G>A", "3G>C"), hgvsAll = c("1022TtoC | L341P", "991GtoA | E331K", "3GtoC | M1I"), dbsnp = c("rs267607049", NA, NA), chromosome = c("3", "22", "16"), startCoord = c(165060026L, 50626052L, 2326464L), endCoord = c(165060026L, 50626052L, 2326464L), inheritance = c("AR", "AR", "AR"), gnomad_AC = c(NA_integer_, NA_integer_, NA_integer_), gnomad_AF = c(NA_real_, NA_real_, NA_real_), gnomad_AN = c(NA_integer_, NA_integer_, NA_integer_), mutype = c("missense", "missense", "initiation"), pmid = c("10903344", "12809637", "16641205"), pmidAll = c(NA, NA, "24871971"), base = c("M", "M", "M"), clinvarID = c("1413", NA, NA), clinvar_clnsig = c("Pathogenic", NA, NA), gene_id = c("2073", "190", "10")), row.names = c(NA, -3L), class = c("tbl_df", "tbl", "data.frame"))
library(shiny)
library(shinyjs)
library(tidyverse)
library(DT)
sidebar <- dashboardSidebar(
selectInput("search", label = "Search Options:",
choices = c("General", "Gene", "Mutation", "Reference", "Phenotype"), selected = "Gene"),
sidebarMenu(id="siderbarmenu", sidebarMenuOutput("menu"))
)
header <- dashboardHeader()
body <- dashboardBody(
tags$style(type="text/css",
".shiny-output-error { visibility: hidden; }",
".shiny-output-error:before { visibility: hidden; }",
".shiny-output-error:after { visibility: hidden; }"),
#ui
shinyjs::useShinyjs(),
tabItems(
tabItem("search_general", h1("A was done")),
tabItem(
tabName = "search_exact_gene",
tabsetPanel(
id = "tabs",
tabPanel(
title = "Main Dashboard",
value = "gene1",
fluidRow(
column(12,dataTableOutput("tablafilt_paste_genes"))
)
)
)
)
)
)
ui <- dashboardPage(header, sidebar, body)
And Here is the server part
server <- function(input, output, session) {
output$menu <- renderMenu({
my_general = list(
menuItem("Búsqueda general", tabName="search_general"),
conditionalPanel("input.siderbarmenu == 'search_general'",
textInput(inputId = "search_terms", label = "Search terms"),
selectInput("search_fields", label="Search choices", choices=c("All Fields", "Gene symbol",
"Gene description", "Chromosomal location", "HGNC/OMIM/GDB/Entrez ID", "RefSeq transcript",
"Disease/phenotype", "Gene Ontology"), selected = "Gene symbol"),
actionButton("submit", "Submit query")
)
)
my_gene = list(
menuItem("Búsqueda por gene", tabName="search_exact_gene"),
textInput(inputId = "search_exact_symbol", label = "Exact gene symbol only"),
actionButton("submit3", "Submit query")
)
if(input$search=="General"){
menu = my_general
} else if (input$search=="Gene"){
menu = my_gene
}
sidebarMenu(menu)
})
filtrado <- reactive({
dataset <- input$submit3
glist <- isolate(input$search_exact_symbol)
datos <- filter(mut_genes, gene %in% glist)
shinyInput <- function(FUN, len, id, ...) {
inputs <- character(len)
for (i in seq_len(len)) {
inputs[i] <- as.character(FUN(paste0(id, i), ...))
}
inputs
}
datos <- mutate(datos, Mutacion=shinyInput(
FUN = actionButton,
len = nrow(datos),
id = 'button_',
label = "Mutacion",
onclick = 'Shiny.onInputChange(\"select_button\", this.id)')
)
return(datos)
})
output$tablafilt_paste_genes <- DT::renderDataTable({
if(is.null(filtrado()))
return()
datos <- filtrado()
DT::datatable(datos, escape = FALSE,
rownames = FALSE,
style = 'bootstrap',
class = 'compact cell-border stripe hover',
filter = list(position = 'top', clear = FALSE),
extensions = c('Buttons', "FixedHeader", "Scroller"),
options = list(
stateSave = FALSE,
ordering = FALSE,
autoWidth = TRUE,
search = list(regex = TRUE, caseInsensitive = TRUE),
columnDefs = list(
list(
className = 'dt-center',
targets = 1:ncol(datos)-1L,
render = JS("function(data, type, row, meta) {",
"return type === 'display' && typeof data === 'string' && data.length > 10 ?",
"'<span title=\"' + data + '\">' + data.substr(0, 10) + '...</span>' : data;",
"}")
)
),
initComplete = JS(
"function(settings, json) {",
"$(this.api().table().header()).css({'font-size': '12px'});",
"}"),
sDom = '<"top">Brtp<"bottom">i', # remove search general box and keep the top filters
scrollX = TRUE,
deferRender=TRUE,
buttons = list('colvis'),
FixedHeader = TRUE,
pageLength = 25,
lengthMenu = list(c(25, 50, 100, -1), list('25', '50', '100', 'All'))
),
callback = JS('table.page(3).draw(false); "setTimeout(function() { table.draw(true); }, 300);"')) %>%
formatStyle(columns = colnames(.$x$data), `font-size` = "15px")
})
}
runApp(shinyApp(ui, server))
My ideal situation is recreate the acc_num column and make it a buttom clickable, but when I have tried with
shinyInput <- function(FUN, len, id, label,...) {
inputs <- character(len)
for (i in seq_len(len)) {
label <- datos$acc_num[i]
inputs[i] <- as.character(FUN(paste0(id, i), label=label, ...))
}
inputs
}
It happens the same, I still see the <button id => even when I set the datatable option escape to false
That's because of the quotes. Your render function generates <span title="<button id = "xxx" ...... and this causes the issue.
You don't want to apply the span to the buttons, so add the regex test !(/button/).test(data) in the conditions:
render = JS("function(data, type, row, meta) {",
"return type === 'display' && typeof data === 'string' && data.length > 10 && !(/button/).test(data) ? ",
"'<span title=\"' + data + '\">' + data.substr(0, 10) + '...</span>' : data;",
"}")
Below is a sample code which takes two inputs: 1) input file and 2) input number of rows. Upon clicking the "Analyze" button the output from the server command return to the "Table" in "Results" tabset. This is a simple example where the command will be executed quickly and switches to the "Results" tabsetpanel.
The below withProgress code only shows the progress bar for the set time and disappears and then the actual code is executed. I would like to show a "Status Message" or "Progress Bar" when the "Analyze" is hit and show as long as the command is run. As long as the progress bar is running the current user (other users can use the app) cannot perform any action from the side bar. Because in the real app, sidebar has more menuItems which does similar tasks like this and each task has a Analyze button. If the user is allowed to browse to sidebar pages and hit Analyze then the app will have overload of performing multiple tasks. Ideally the progress bar functionality should we used with multiple actionButtons.
I read the blogs about async but unable to put right code in the right place. any help is appreciated with a bounty!!
library(shiny)
library(shinydashboard)
sidebar <- dashboardSidebar(width = 200,
sidebarMenu(id = "tabs",
menuItem(
"File", tabName = "tab1", icon = icon("fas fa-file")
)))
body <- tabItem(tabName = "tab1",
h2("Input File"),
fluidRow(
tabPanel(
"Upload file",
value = "upload_file",
fileInput(
inputId = "uploadFile",
label = "Upload Input file",
multiple = FALSE,
accept = c(".txt")
),
checkboxInput('header', label = 'Header', TRUE)
),
box(
title = "Filter X rows",
width = 7,
status = "info",
tabsetPanel(
id = "input_tab",
tabPanel(
"Parameters",
numericInput(
"nrows",
label = "Entire number of rows",
value = 5,
max = 10
),
actionButton("run", "Analyze")
),
tabPanel(
"Results",
value = "results",
navbarPage(NULL,
tabPanel(
"Table", DT::dataTableOutput("res_table"),
icon = icon("table")
)),
downloadButton("downList", "Download")
)
)
)
))
ui <-
shinyUI(dashboardPage(
dashboardHeader(title = "TestApp", titleWidth = 150),
sidebar,dashboardBody(tabItems(body))
))
server <- function(input, output, session) {
file_rows <- reactiveVal()
observeEvent(input$run, {
withProgress(session, min = 1, max = 15, {
setProgress(message = 'Analysis in progress',
detail = 'This may take a while...')
for (i in 1:15) {
setProgress(value = i)
Sys.sleep(0.5)
}
})
system(paste(
"cat",
input$uploadFile$datapath,
"|",
paste0("head -", input$nrows) ,
">",
"out.txt"
),
intern = TRUE)
head_rows <- read.delim("out.txt")
file_rows(head_rows)
})
observeEvent(file_rows(), {
updateTabsetPanel(session, "input_tab", "results")
output$res_table <-
DT::renderDataTable(DT::datatable(
file_rows(),
options = list(
searching = TRUE,
pageLength = 10,
rownames(NULL),
scrollX = T
)
))
})
output$downList <- downloadHandler(
filename = function() {
paste0("output", ".txt")
}, content = function(file) {
write.table(file_rows(), file, row.names = FALSE)
}
)
}
shinyApp(ui = ui, server = server)
Here is a solution based on the (absolutely under-star-ed) library(ipc).
I came across this library due to a question of #Dean Attali, where Joe Cheng mentioned it.
The quick-start guide of the ipc-package gives an example of what you are asking for: AsyncProgress.
Furthermore it provides an example on how to kill a future using AsyncInterruptor.
However, I haven't been able to test it yet.
I worked around the cancel-problem by using #Dean Attali's great package shinyjs to simply start a new session and ignore the old Future (You might be able to improve this, by using AsyncInterruptor).
But nevertheless, I gave your code a Future, dropped your system() cmd because I'm currently running R on Windows and found a way to disable (tribute to #Dean Attali) the analyze button session-wise by giving it session-dependant names:
library(shiny)
library(shinydashboard)
library(ipc)
library(promises)
library(future)
library(shinyjs)
library(datasets)
library(V8)
plan(multiprocess)
header <- dashboardHeader(title = "TestApp", titleWidth = 150)
sidebar <- dashboardSidebar(width = 200,
sidebarMenu(id = "tabs",
menuItem(
"File", tabName = "tab1", icon = icon("fas fa-file")
)))
body <- dashboardBody(useShinyjs(),
fluidRow(column(
12, tabItem(
tabName = "tab1",
h2("Input File"),
textOutput("shiny_session"),
tabPanel(
"Upload file",
value = "upload_file",
fileInput(
inputId = "uploadFile",
label = "Upload Input file",
multiple = FALSE,
accept = c(".txt")
),
checkboxInput('header', label = 'Header', TRUE)
),
box(
title = "Filter X rows",
width = 7,
status = "info",
tabsetPanel(
id = "input_tab",
tabPanel(
"Parameters",
numericInput(
"nrows",
label = "Entire number of rows",
value = 5,
max = 10
),
column(1, uiOutput("sessionRun")),
column(1, uiOutput("sessionCancel"))
),
tabPanel(
"Results",
value = "results",
navbarPage(NULL,
tabPanel(
"Table", DT::dataTableOutput("res_table"),
icon = icon("table")
)),
downloadButton("downList", "Download")
)
)
)
)
)))
ui <- shinyUI(dashboardPage(
header = header,
sidebar = sidebar,
body = body,
title = "TestApp"
))
server <- function(input, output, session) {
output$shiny_session <-
renderText(paste("Shiny session:", session$token))
file_rows <- reactiveVal()
run_btn_id <- paste0("run_", session$token)
cancel_btn_id <- paste0("cancel_", session$token)
output$sessionRun <- renderUI({
actionButton(run_btn_id, "Analyze")
})
output$sessionCancel <- renderUI({
actionButton(cancel_btn_id, "Cancel")
})
paste("Shiny session:", session$token)
observeEvent(input[[run_btn_id]], {
file_rows(NULL)
shinyjs::disable(id = run_btn_id)
progress <- AsyncProgress$new(message = 'Analysis in progress',
detail = 'This may take a while...')
row_cnt <- isolate(input$nrows)
get_header <- isolate(input$header)
future({
fileCon <- file("out.txt", "w+", blocking = TRUE)
linesCnt <- nrow(iris)
for (i in seq(linesCnt)) {
Sys.sleep(0.1)
progress$inc(1 / linesCnt)
writeLines(as.character(iris$Species)[i],
con = fileCon,
sep = "\n")
}
close(fileCon)
head_rows <- read.delim("out.txt", nrows = row_cnt, header=get_header)
progress$close() # Close the progress bar
return(head_rows)
}) %...>% file_rows
return(NULL) # Return something other than the future so we don't block the UI
})
observeEvent(input[[cancel_btn_id]],{
session$reload()
})
observeEvent(file_rows(), {
shinyjs::enable(id = run_btn_id)
updateTabsetPanel(session, "input_tab", "results")
output$res_table <-
DT::renderDataTable(DT::datatable(
req(file_rows()),
options = list(
searching = TRUE,
pageLength = 10,
rownames(NULL),
scrollX = T
)
))
})
output$downList <- downloadHandler(
filename = function() {
paste0("output", ".txt")
},
content = function(file) {
write.table(file_rows(), file, row.names = FALSE)
}
)
}
shinyApp(ui = ui, server = server)
App running:
This question has been answered on a different forum
For future reference, if anyone comes across this question, here's the full answer (I did not come up with this answer, it's by Joe Cheng)
This seems to be the main piece of code you're asking about:
observeEvent(input$run, {
withProgress(session, min = 1, max = 15, {
setProgress(message = 'Analysis in progress',
detail = 'This may take a while...')
for (i in 1:15) {
setProgress(value = i)
Sys.sleep(0.5)
}
})
system(paste(
"cat",
input$uploadFile$datapath,
"|",
paste0("head -", input$nrows) ,
">",
"out.txt"
),
intern = TRUE)
head_rows <- read.delim("out.txt")
file_rows(head_rows)
})
With futures/promises, you need to clearly decide what operations happen inside of the Shiny process, and what operations happen in the future process. In this case, here are the steps that we want to happen, in order:
Show progress message (Shiny process)
Read reactives: input$uploadFile$datapath, input$nrows (Shiny)
Write all but the last nrows to out.txt (future process)
Read out.txt (Could be either, let's say future)
Dismiss progress (Shiny)
Assign result to file_rows (Shiny)
Here's what that looks like:
observeEvent(input$run, {
prog <- Progress$new(session)
prog$set(message = "Analysis in progress",
detail = "This may take a while...",
value = NULL)
path <- input$uploadFile$datapath
nrows <- input$nrows
future({
readLines(path) %>% head(-nrows) %>% writeLines("out.txt")
read.delim("out.txt")
}) %...>%
file_rows() %>%
finally(~prog$close())
})
As long as the future/promise pipeline is the last expression in the observeEvent (which it is in this case, as file_rows() and finally(...) are part of the pipeline) then Shiny will hold off on processing any messages on behalf of the user.
There are two things this solution doesn't address.
Progress messages take a step back; not only are we forced to use the Progress$new() syntax instead of the cleaner withProgress(), but we lost the ability to report on the progress percentage. You can try the new ipc package for a solution to that problem.
This doesn't stop the user from clicking around in the UI; it won't do anything while the async operation is executing, but when the operation is done those interactions will have accumulated in a queue and will be handled in the order that they arrived. If you'd like to actually disable the UI entirely so that they're not able to do anything at all, there's not currently a built-in way to do that in Shiny. Although come to think of it, you might try replacing the use of Progress with showModal(modalDialog(title = "Analysis in progress", "This may take a while...", footer=NULL)); I think that will at least stop mouse clicks.
I am trying to create simple app that acts as a GUI for studying different files having same variables but with different version and content. I am unable to give an app where every time the user opens the app they dont have to enter in their parameters again where they left off. I'd like them to be able to save their parameters and bring them up again when they go back to the app.
I am giving my sample code here, however the number of inputs and plots are far more in the actual app. I want to know if there is any solution to save these dependent inputs and outputs.
library(shiny)
library(pryr)
ui = shinyUI(fluidPage(
# Application title
titlePanel("Example Title"),
# Sidebar structure
sidebarLayout(
sidebarPanel(
textInput("save_file", "Save to file:", value="sample.RData"),
actionButton("save", "Save input value to file"),
uiOutput("load"),
uiOutput("file"),
uiOutput("mytype"),
uiOutput("mysubtype")
),
# Show a plot of the generated distribution
mainPanel(
tabsetPanel(id="tab",
tabPanel(
"Plot",
plotOutput("distPlot"),
checkboxInput(inputId = "density",
label = strong("Show Adjustment Factors"),
value = FALSE),
conditionalPanel(condition = "input.density == true",
sliderInput(inputId = "bandwidth",
label = "Width adjustment: ",
min = 0.5, max = 4, value = 1, step = 0.1),
radioButtons("mycolor", "Color Adjustment: ",
choices = c(Red = "red", Black = "black", Blue = "blue"),selected = "black", inline = TRUE)
)),
tabPanel("Summary",
h3(textOutput("label")),
verbatimTextOutput("summary")
)
))
)
)
)
server = function(input, output, session) {
# render a selectInput with all RData files in the specified folder
output$load <- renderUI({
choices <- list.files("/home/user/Documents/Shiny/", pattern="*.RData")
selectInput("input_file", "Select input file", choices)
})
# render a selectInput with all csv files in the specified folder so that user can choose the version
output$file <- renderUI({
choices.1 <- list.files("/home/user/Documents/Shiny/", pattern="*.csv")
selectInput("input_csv", "Select csv file", choices.1)
})
# Load a csv file and update input
data = eventReactive(input$input_csv, {
req(input$input_csv)
read.csv(paste0("/home/user/Documents/Shiny/",input$input_csv),
header = TRUE,
sep = ",")
})
#Display Type - Types may differ based on file selection
output$mytype <- renderUI({
selectInput("var1", "Select a type of drink: ", choices = levels(data()$Type))
})
#Display SubType - This would be dependent on Type Selection
output$mysubtype <- renderUI({
selectInput("var2", "Select the SubType: ", choices = as.character(data()[data()$Type==input$var1,"Subtype"]))
})
# Save input when click the button
observeEvent(input$save, {
validate(
need(input$save_file != "", message="Please enter a valid filename")
)
mycolor <- input$mycolor
mytype = input$var1
mysubtype = input$var2
density <- input$density
bandwidth <- input$bandwidth
save(bandwidth, density, mycolor, mytype, mysubtype,
file=paste0("/home/user/Documents/Shiny/", input$save_file))
choices <- list.files("/home/user/Documents/Shiny/", pattern="*.RData")
updateSelectInput(session, "input_file", choices=choices)
choices.1 <- list.files("/home/user/Documents/Shiny/", pattern="*.csv")
updateSelectInput(session, "input_csv", choices=choices.1)
})
# Load an RData file and update input
# input$var1, input$var2, input$density, input$bandwidth, input$mycolor),
observeEvent(c(input$input_file),
{
load(paste0("/home/user/Documents/Shiny/",input$input_file))
updateSelectInput(session, "var1", choices = levels(data()$Type), selected = mytype)
updateSelectInput(session, "var2", choices = as.character(data()[data()$Type==mytype,"Subtype"]), selected = mysubtype)
updateCheckboxInput(session, "density", value = density)
updateSliderInput(session, inputId = "bandwidth", value=bandwidth)
updateRadioButtons(session, "mycolor", choices = c(Red = "red", Black = "black", Blue = "blue"), selected = mycolor, inline = TRUE)
})
output$distPlot <- renderPlot({
# generate plot
x = data()[data()$Type == input$var1 & data()$Subtype == input$var2, c("Alcohol_Content","Price")]
plot(x$Alcohol_Content, x$Price, type = "l", xlab = "Alcohol content", ylab = "Price",
main = "Sample Plot",
col="red",
lwd=1.5)
if (input$density)
plot(x$Alcohol_Content, x$Price, type = "p", xlab = "Alcohol content", ylab = "Price",
main = "Sample Plot",
col=input$mycolor,
lwd=input$bandwidth)
})
output$summary <- renderText(summary(data()))
}
shinyApp(ui, server)
The Input csv files would be always stored in
"/home/user/Documents/Shiny/"
The User could just click "Save to
file:" and it should save the user selections inside "sample.RData"
located in same "/home/user/Documents/Shiny/". Hence I want to give a selectinput where user can choose the .RData file also.
The user should also be able to save the inputs on Mainpanel which they would use to modify the chart
Questions:-
Most of the code works fine given above but how can I save #Display Subtype.
What happens if I add one more dependent list like Type and Subtype?
And also if I can get some help on whether the solution would work for multiple select inputs?.
Any help on the code would be really be appreciated.
Dummy Data:-
x = data.frame(Type=rep(c("WINE"), 9), Subtype=rep(c("TABLE WINE RED", "TABLE WINE WHITE", "MONTILLA"), each=3), Alcohol_content= c(14, 11.5, 12, 11, 13.5, 11, 12.5, 12, 11.5), Price = c(30.99, 32.99, 29.99, 33.99, 36.99, 34.99, 119, 32.99, 13.99))
y = data.frame(Type=rep(c("REFRESHMENT"), 9), Subtype=rep(c("CIDER ALL", "SPIRIT", "BEER"), each=3), Alcohol_content= c(5, 5.2, 7, 5.3, 6.9, 5, 5, 6, 5), Price = c(9.99, 9.99, 8.99, 9.95, 3.49, 9.99, 12.99, 13.49, 21.99))
bcl_data1 = rbind(x, y)
write.csv(bcl_data1, "bcl_data1.csv")
There are many more Subtypes under each Type (Wine , Refreshment). I am somehow not able to retrieve the Subtype value through above code, However when I load Sample.RData I can see var2 = my selected value.
I would like to know how save these values please.
Here is a working version of your code. Your problem was the concurrent use of renderUI and updateSelectInput. Everytime you tried to update your selectInput it was re-rendered right away so that the change wasn't visible.
I'd recommend to render the selectInput's in the UI (which I did for "var2") and use updateSelectInput only. (If you really want to continue building your own bookmarks.)
Best regards
library(shiny)
library(pryr)
if(!file.exists("bcl_data1.csv")){
x = data.frame(Type=rep(c("WINE"), 9), Subtype=rep(c("TABLE WINE RED", "TABLE WINE WHITE", "MONTILLA"), each=3), Alcohol_content= c(14, 11.5, 12, 11, 13.5, 11, 12.5, 12, 11.5), Price = c(30.99, 32.99, 29.99, 33.99, 36.99, 34.99, 119, 32.99, 13.99))
y = data.frame(Type=rep(c("REFRESHMENT"), 9), Subtype=rep(c("CIDER ALL", "SPIRIT", "BEER"), each=3), Alcohol_content= c(5, 5.2, 7, 5.3, 6.9, 5, 5, 6, 5), Price = c(9.99, 9.99, 8.99, 9.95, 3.49, 9.99, 12.99, 13.49, 21.99))
bcl_data1 = rbind(x, y)
write.csv(bcl_data1, "bcl_data1.csv")
}
settings_path <- getwd()
# settings_path <- "/home/user/Documents/Shiny/"
ui = shinyUI(fluidPage(
# Application title
titlePanel("Example Title"),
# Sidebar structure
sidebarLayout(
sidebarPanel(
textInput("save_file", "Save to file:", value="sample.RData"),
actionButton("save", "Save input value to file"),
p(),
p(),
uiOutput("load"),
uiOutput("file"),
uiOutput("mytype"),
selectInput("var2", "Select the SubType: ", choices = NULL)
),
# Show a plot of the generated distribution
mainPanel(
tabsetPanel(id="tab",
tabPanel(
"Plot",
plotOutput("distPlot"),
checkboxInput(inputId = "density",
label = strong("Show Adjustment Factors"),
value = FALSE),
conditionalPanel(condition = "input.density == true",
sliderInput(inputId = "bandwidth",
label = "Width adjustment: ",
min = 0.5, max = 4, value = 1, step = 0.1),
radioButtons("mycolor", "Color Adjustment: ",
choices = c(Red = "red", Black = "black", Blue = "blue"),selected = "black", inline = TRUE)
)),
tabPanel("Summary",
h3(textOutput("label")),
verbatimTextOutput("summary")
)
))
)
)
)
server = function(input, output, session) {
# render a selectInput with all RData files in the specified folder
last_save_path <- file.path(settings_path, "last_input.backup")
if(file.exists(last_save_path)){
load(last_save_path)
if(!exists("last_save_file")){
last_save_file <- NULL
}
} else {
last_save_file <- NULL
}
if(!is.null(last_save_file)){
updateTextInput(session, "save_file", "Save to file:", value=last_save_file)
}
output$load <- renderUI({
choices <- list.files(settings_path, pattern="*.RData")
selectInput("input_file", "Select input file", choices, selected = last_save_file)
})
# render a selectInput with all csv files in the specified folder so that user can choose the version
output$file <- renderUI({
choices.1 <- list.files(settings_path, pattern="*.csv")
selectInput("input_csv", "Select csv file", choices.1)
})
# Load a csv file and update input
csv_data = eventReactive(input$input_csv, {
req(input$input_csv)
read.csv(file.path(settings_path,input$input_csv),
header = TRUE,
sep = ",")
})
#Display Type - Types may differ based on file selection
output$mytype <- renderUI({
req(csv_data())
selectInput("var1", "Select a type of drink: ", choices = unique(csv_data()$Type))
})
#Display SubType - This would be dependent on Type Selection
observeEvent(input$var1, {
req(csv_data())
req(input$var1)
updateSelectInput(session, "var2", "Select the SubType: ", choices = as.character(csv_data()[csv_data()$Type==input$var1,"Subtype"]), selected = isolate(input$var2))
})
# Save input when click the button
observeEvent(input$save, {
validate(
need(input$save_file != "", message="Please enter a valid filename")
)
last_save_file <- input$save_file
save(last_save_file, file=last_save_path)
mycolor <- input$mycolor
mytype = input$var1
mysubtype = input$var2
density <- input$density
bandwidth <- input$bandwidth
save(bandwidth, density, mycolor, mytype, mysubtype,
file=file.path(settings_path, input$save_file))
})
# Load an RData file and update input
observeEvent(input$input_file, {
req(input$input_file)
load(file.path(settings_path, input$input_file))
updateSelectInput(session, "var1", choices = unique(csv_data()$Type), selected = mytype)
updateSelectInput(session, "var2", choices = mysubtype, selected = mysubtype)
updateCheckboxInput(session, "density", value = density)
updateSliderInput(session, "bandwidth", value = bandwidth)
updateRadioButtons(session, "mycolor", choices = c(Red = "red", Black = "black", Blue = "blue"), selected = input$mycolor)
})
output$distPlot <- renderPlot({
req(csv_data())
req(input$var1)
req(input$var2)
# generate plot
x = csv_data()[csv_data()$Type == input$var1 & csv_data()$Subtype == input$var2, c("Alcohol_content", "Price")]
if(nrow(x) > 0){
x <- x[order(x$Alcohol_content), ]
plot(x$Alcohol_content, x$Price, type = "l", xlab = "Alcohol content", ylab = "Price",
main = "Sample Plot",
col="red",
lwd=1.5)
if (input$density)
plot(x$Alcohol_content, x$Price, type = "p", xlab = "Alcohol content", ylab = "Price",
main = "Sample Plot",
col=input$mycolor,
lwd=input$bandwidth)
}
})
output$summary <- renderText(summary(csv_data()))
}
shinyApp(ui, server)
I have seen that this problem has happened to other people, but their solutions have not worked for me. I have my app.R file and a .RData file with the required inputs in the same ECWA_Strategic_Planning_Tool directory. When I run:
library(rsconnect)
rsconnect::deployApp('C:/Users/mikialynn/Documents/Duke/Spring2017/MP/GISTool/Final/ECWA_Strategic_Planning_Tool')
I get the following error on the web page that opens up:
ERROR: An error has occurred. Check your logs or contact the app author for clarification.
However, I cannot find anything wrong. I install all of my packages, I use relative pathways etc. I am pasting all of the code from my app below. If anyone can spot what I'm doing wrong, I would greatly appreciate it!
library(shiny)
library(leaflet)
library(sp)
library(rgdal)
library(rstudioapi) # For working directory
library(raster)
library(RColorBrewer)
library(rgeos) #Maybe use gSimplify to simplify polygon
library(DT) #To make interactive DataTable
library(plotly) #For pie chart
library(ggplot2) # for layout
# Set Working Directory
setwd(dirname(rstudioapi::getActiveDocumentContext()$path))
# Load R Workspace
load('Shiny.Strategies.RData')
# UI variables
neigh.names <- levels(merge.proj$View)
neigh.default <- c("Urban7")
dt.names <- c('PARCEL_ID', 'PIN', 'OWNER_NAME', 'SITE_ADDRE', 'OWNER_ADDR',
'SUM_ACRE', 'LANDUSE_DE', 'LAND_VALUE', 'TOTAL_VALU', 'SALE_PRICE',
'Pluvial_WtScore', 'Rest_WtScore', 'GI_WtScore', 'SC_WtScore',
'UNCWI_WtScore', 'Total_Score', 'View')
dt.default <- c('PARCEL_ID', 'Pluvial_WtScore', 'Rest_WtScore',
'GI_WtScore', 'SC_WtScore', 'UNCWI_WtScore', 'Total_Score', 'View')
# Build UI
ui <- fluidPage(
titlePanel("ECWA Strategic Planning Tool"),
HTML('<br>'),
column(2,
HTML("<strong>Instructions:</strong><br/><br/>"),
HTML("<p>1) Select weights for parameters and click 'Run' to
initiate tool.<br/><br/>
2) Use rightside panel to adjust Table and Map Settings.<br/>
<br/>
3) Use search/sort functions of Table to identify parcels.
Select row to display Total Score Chart.<br/><br/>
4) Input View and Parcel ID from Table to Map settings to
identify parcel in Map.<br/><br/>
5) When satisfied with weights, click 'Export Shapefile' to
save shapefile of all parcels.<p/><br/>"),
HTML("<strong>Calculate Parcel Scores: </strong><br/>"),
helpText('The sum of the weights must equal to 1.'),
sliderInput(inputId = "weightPluvial", label = "Weight for Pluvial
Flooding",
value = 0.20, min = 0, max = 1),
sliderInput(inputId = "weightRest", label = "Weight for
Restoration",
value = 0.20, min = 0, max = 1),
sliderInput(inputId = "weightGI", label = "Weight for Green
Infrastructure",
value = 0.20, min = 0, max = 1),
sliderInput(inputId = "weightSC", label = "Weight for City
Stormwater Controls",
value = 0.20, min = 0, max = 1),
sliderInput(inputId = "weightUNCWI", label = "Weight for UNCWI",
value = 0.20, min = 0, max = 1),
actionButton("run", "Run"),
actionButton("export", "Export Shapefile")),
column(8,
HTML("<h3><strong>Table Summary</strong></h3>"),
HTML("<br>"),
dataTableOutput("table")),
column(2,
HTML("<p><br><br></p>"),
HTML("<h4>Table Settings:</h4>"),
checkboxGroupInput(inputId = 'show_vars', label = 'Select column(s)
to display in Table:', choices = dt.names, selected = dt.default),
HTML("<strong>Total Score Chart:</strong>"),
helpText("Please select Table row to display pie chart."),
plotlyOutput("pie")
),
fluidRow(
column(8, offset = 2,
HTML("<br>"),
HTML("<h3><strong>Map Display</strong></h3>"),
leafletOutput("map", height = 800),
HTML("<br><br>")),
column(2,
HTML("<p><br><br><br></p>"),
HTML("<h4>Map Settings:</h4>"),
checkboxGroupInput(inputId = 'show_neigh', label = 'Select
View(s) to display in Map:', choices = neigh.names,
selected = neigh.default),
HTML("<br>"),
sliderInput("range", "Select score range to display in Map:", min
= 0.0, max= 10.0, value = as.numeric(c("0.0", "10.0")), step = 0.1),
HTML("<br>"),
HTML("<strong>Parcel Zoom:</strong>"),
helpText("The View and Score Range must contain the parcel of
interest to execute zoom."),
numericInput('parcel','Enter Parcel ID',0)
)
))
# SERVER
server <- function(input, output) {
defaultData <-
eventReactive(input$run, {
# Multiply by Weights
merge.proj#data$Pluvial_WtScore <-
round(merge.proj#data$Pluvial_Score*input$weightPluvial, digits = 1)
merge.proj#data$Rest_WtScore <-
round(merge.proj#data$Rest_Score*input$weightRest, digits = 1)
merge.proj#data$GI_WtScore <-
round(merge.proj#data$GI_Score*input$weightGI, digits = 1)
merge.proj#data$SC_WtScore <-
round(merge.proj#data$SC_Score*input$weightSC, digits = 1)
merge.proj#data$UNCWI_WtScore <-
round(merge.proj#data$UNCWI_Score*input$weightUNCWI, digits = 1)
# Find Total Score
merge.proj#data$Total_Score <- merge.proj#data$Pluvial_WtScore +
merge.proj#data$Rest_WtScore + merge.proj#data$GI_WtScore +
merge.proj#data$SC_WtScore + merge.proj#data$UNCWI_WtScore
return(merge.proj)
})
# Subset by neighborhood
neighData <- reactive ({
merge.proj <- defaultData()
merge.proj[merge.proj$View%in%input$show_neigh,]
})
# Plot with leaflet
# Palette for map
colorpal <- reactive({
merge.proj <- neighData()
colorNumeric(palette = "YlOrRd",
domain = merge.proj$Total_Score)
})
# Pop Up Option for map
# popup <- paste0("<strong>Parcel ID: </strong>",
# merge.proj#data$PARCEL_ID,
# "<br><strong>Total Score: </strong>",
# merge.proj#data$Total_Score)
# Label Option for map
labels <- reactive({
merge.proj <- neighData()
sprintf("<strong>Parcel ID: </strong>%s<br/><strong>Total Score:
</strong>%g",
merge.proj$PARCEL_ID,
merge.proj$Total_Score) %>% lapply(htmltools::HTML)
})
# Render Default Map
output$map <- renderLeaflet ({
merge.proj <- neighData()
pal <- colorpal()
lab <- labels()
leaflet() %>%
#addProviderTiles(provider='Esri.WorldImagery') %>%
# setView(zoom =) %>%
addTiles() %>%
addPolygons(
#data = merge.proj[input$show_neigh,, drop = FALSE],
data=merge.proj,
fillColor = ~pal(Total_Score),
weight = 1,
opacity = 1,
color = "white",
dashArray = "3",
fillOpacity = 0.7,
highlight = highlightOptions(
weight = 3,
color = "#666",
dashArray = "",
fillOpacity = 0.7,
bringToFront = TRUE),
# popup= popup) %>%
label = lab,
labelOptions = labelOptions(
style = list("font-weight" = "normal", padding = "3px 8px"),
textsize = "15px",
direction = "auto")) %>%
addLegend(position = "bottomleft",pal = pal, opacity = 0.7, values =
merge.proj$Total_Score, title = "<strong>Total Score</strong>")
})
# Build Data Table
output$table <- renderDataTable({
merge.proj <- defaultData()
table.dat <- merge.proj[, c('PARCEL_ID', 'PIN', 'OWNER_NAME',
'SITE_ADDRE', 'OWNER_ADDR', 'SUM_ACRE', 'LANDUSE_DE', 'LAND_VALUE',
'TOTAL_VALU', 'SALE_PRICE', 'Pluvial_WtScore', 'Rest_WtScore', 'GI_WtScore',
'SC_WtScore', 'UNCWI_WtScore', 'Total_Score', 'View')]
datatable(data = table.dat#data[, input$show_vars, drop = FALSE],
options = list(lengthMenu = c(5, 10, 20, 30), pageLength = 20), rownames =
FALSE)
})
# Plot-ly
output$pie <- renderPlotly({
merge.proj <- defaultData()
names <- c('Pluvial', 'Rest', 'GI', 'SC', 'UNCWI')
colors <- c('rgb(128,133,133)', 'rgb(211,94,96)', 'rgb(144,103,167)',
'rgb(114,147,203)', 'rgb(171,104,87)')
selectedrowindex <-
input$table_rows_selected[length(input$table_rows_selected)]
selectedrowindex <- as.numeric(selectedrowindex)
df <- data.frame(merge.proj[selectedrowindex, c('Pluvial_WtScore',
'Rest_WtScore', 'GI_WtScore', 'SC_WtScore', 'UNCWI_WtScore')])
vector <- unname(unlist(df[1,]))
if (!is.null(input$table_rows_selected)) {
par(mar = c(4, 4, 1, .1))
plot_ly(labels = names, values = vector, type = 'pie',
textposition = 'inside',
textinfo = 'label+percent',
insidetextfont = list(color = '#FFFFFF'),
hoverinfo = 'text',
text = ~paste('Score:', vector),
marker = list(colors = colors,
line = list(color = '#FFFFFF', width = 1)),
#The 'pull' attribute can also be used to create space between the sectors
showlegend = FALSE) %>%
layout(#title = '% Total Score',
xaxis = list(showgrid = FALSE, zeroline = FALSE, showticklabels = FALSE),
yaxis = list(showgrid = FALSE, zeroline = FALSE, showticklabels = FALSE))
}
else {return(NULL)}
})
# Update map to parcel score slider
# Subset data
filteredData <- reactive({
merge.proj <- neighData()
merge.proj[merge.proj#data$Total_Score >= input$range[1] &
merge.proj#data$Total_Score <= input$range[2],]
})
# New Palette
colorpal2 <- reactive({
merge.proj <- filteredData()
colorNumeric(palette = "YlOrRd",
domain = merge.proj$Total_Score)
})
# Pop Up Option
# popup <- paste0("<strong>Parcel ID: </strong>",
# merge.proj#data$PARCEL_ID,
# "<br><strong>Total Score: </strong>",
# merge.proj#data$Total_Score)
# Label Option
labels2 <- reactive({
merge.proj <- filteredData()
sprintf("<strong>Parcel ID: </strong>%s<br/><strong>Total Score:
</strong>%g",
merge.proj$PARCEL_ID,
merge.proj$Total_Score) %>% lapply(htmltools::HTML)
})
#Leaflet Proxy
observe({
merge.proj <- filteredData()
pal2 <- colorpal2()
lab2 <- labels2()
leaf <- leafletProxy("map", data = filteredData()) %>%
clearShapes() %>%
addPolygons(
fillColor = ~pal2(Total_Score),
weight = 1,
opacity = 1,
color = "white",
dashArray = "3",
fillOpacity = 0.7,
highlight = highlightOptions(
weight = 3,
color = "#666",
dashArray = "",
fillOpacity = 0.7,
bringToFront = TRUE),
# popup= popup) %>%
label = lab2,
labelOptions = labelOptions(
style = list("font-weight" = "normal", padding = "3px 8px"),
textsize = "15px",
direction = "auto"))
if(input$parcel>0){
sub.dat <- merge.proj[merge.proj$PARCEL_ID==input$parcel,]
zx <- mean(extent(sub.dat)[1:2])
zy <- mean(extent(sub.dat)[3:4])
leaf <- leaf %>%
setView(lng=zx,lat=zy,zoom=16)
}
leaf
})
#Update Legend
observe({
proxy <- leafletProxy("map", data = filteredData())
pal2 <- colorpal2()
proxy %>% clearControls()
proxy %>% addLegend(position = "bottomleft",pal = pal2, opacity = 0.7,
values = ~Total_Score, title = "<strong>Total Score</strong>")
})
# Export new shapefile
#make so that user can choose name and allow overwrite
observeEvent(input$export, {
merge.proj <- defaultData()
writeOGR(merge.proj, dsn = "Data", layer = "Strategies_Output", driver =
"ESRI Shapefile")
})
}
shinyApp(ui = ui, server = server)
Issue resolved! My initial suspicion was correct; it had to do with the .rdata file. It also relates to shinyapp.io's servers which run on a Linux based server. From my reading, Linux only handles lowercase file paths and extensions. The reason why it worked for the .csv file is because it's pretty common to have the file extension saved in all lowercase. This was not the case for the .RData file. Using the RStudio IDE and the physical "Save Workspace" button, the default file extension is .RData (case sensitive). I couldn't rename the file extension (for some reason, I'm not the most tech-savvy person). Similar to the load() function, there's the save() function. Previously, I used the save() file as follows (note the capitalized .RData at the end):
save(df_training_separated_with_models, file = "sample_data_with_models.RData")
However, using the same function in all lowercase fixes the issue:
save(df_training_separated_with_models, file = "sample_data_with_models.rdata")
Hope this helps any other poor soul with the same issue that is scouring the internet and other forums.
Cheers!