我正在开发一个闪亮的应用程序，以在闪亮的UI上给定的时间间隔（比如说1秒）中一个接一个地显示XML文件中的单词和图像。如果用户选择了文件，它将读取文件并在UI上显示第一个单词，然后在1秒后第一个单词消失并在UI上显示第二个单词，然后在1秒后第二个单词消失并在UI上显示第三个单词，等等上。如果文件中有图像，则它也应显示在UI上，并且1秒钟后也应消失。 我编写了一个代码，以便在间隔的每一秒内从一个句子中得到一个单词。现在它显示在R控制台上，但我想在闪亮的UI中显示这些单词和图像。 以下是问题的说明（出于清洁原因，这不是我的实际代码，但仅作为说明）： 这是我的代码：
library(shiny)
library(tokenizers)
library(magick)
library(stringr)
library(qdap)
library(xml2)
ui <- dashboardPage(skin = 'purple',
dashboardHeader(title = "Document Reader"),
dashboardSidebar(width = 200,
sidebarMenu(
menuItem("Dashboard", tabName = "dashboard", icon = icon("dashboard"))
)
),
dashboardBody(skin = 'purple',
tabItems(
# First tab content
tabItem(tabName = "dashboard",
h2("Dashboard tab content"),
fluidRow(
box(title = "Uploading File",
fileInput("file1", "Choose a Document:"),
actionButton(inputId = "submit",label = "Read")
),
column(8,verbatimTextOutput("opt1"))
)
)
)
)
)
server <- function(input, output, session) {
dw <- eventReactive(input$submit,{
filepath <- input$file1$datapath
doc <- read_xml(filepath)
nodeset <- xml_children(doc)
for(i in 1:length(nodeset)){
wpm <- as.integer(xml_attr(nodeset[i], "wpm"))
#print(paste("WPM", wpm))
data2 <-xml_find_all(nodeset[i], ".//p | .//localImage")
line <- xml_text(data2)
image <- xml_attr(data2, "descr")
image <- image[!is.na(image)]
for(j in line){
#print(j)
if ( j == ""){
tiger = image_read(image)
sink(tempfile())
print(tiger)
sink()
}
else{
words = unlist(strsplit(j, ' '))
words = words[words !=""]
words = str_remove_all(words, "\n")
total = length(words)
for (w in 1:total){
len = length(words[[total]])
for (w1 in 1:len){
if (str_detect(words[[w]][w1],'[.]')){
print(words[[w]][w1])
Sys.sleep(1)
} else {
print(words[[w]][w1])
Sys.sleep(0.5)
}
}
}
}
}
}
})
output$opt1 <- renderUI({
dw()
})
}
shinyApp(ui=ui, server=server)
这是我的XML文件：
<?xml version="1.0"?>
<data>
<section wpm = "200">
<p>History</p>
<p>The society that sparked change at Cambridge</p>
<p>The ancient university owes its scientific status to an unexpected source, finds Georgina Ferry.</p>
<localImage id="0" caption = "The Cambridge Observatory was founded in 1823, four years after the Cambridge Philosophical Society" descr="/home/i9/DocumentReader/DocumentReader/image/Image1.png"/>
<p>We conclude this section by discussing the problem of classification,since it will serve as a prototypical problem for a significant part of this book. It occurs frequently in practice: for instance, when performing spam filtering, we are interested in a yes/no answer as to whether an e-mail contains relevant information or not.</p>
</section>
</data>
如何在闪亮的用户界面上显示这些单词。
答案 0 :(得分：0)
我找到一种解决方案，这要感谢Google并混合使用不同的来源：
ui <- dashboardPage(skin = 'purple',
dashboardHeader(title = "Document Reader"),
dashboardSidebar(width = 200,
sidebarMenu(
menuItem("Dashboard", tabName = "dashboard", icon = icon("dashboard"))
)
),
dashboardBody(skin = 'purple',
tabItems(
# First tab content
tabItem(tabName = "dashboard",
h2("Dashboard tab content"),
fluidRow(
box(title = "Uploading File",
fileInput("text", "Enter Text:"),
actionButton(inputId = "submit",label = "Read"))
,box(verbatimTextOutput("var"),column(8,uiOutput("opt1"))
),
box(actionButton(inputId = "display",label = "display")
)
)
))))
server <- function(input, output, session) {
rv <- reactiveValues(words = character(0), n = 0,
sentence = "")
# action button
# output$var <- renderText({
# if(input$display < length(input_file1()))
# return(input_file1()[input$display])
# })
#
observe({
if(input$submit){
rv$sentence <- unlist(strsplit(readLines(input$text$datapath), ' '))
}
# Re-execute this reactive expression immediately after
# it finishes (though you can also specify a longer time:
# e.g. 1000 means re-execute after 1000 milliseconds have
# passed since it was last executed)
invalidateLater(1000, session)
isolate({
if (rv$n >= length(rv$sentence)) {
return()
} else {
rv$n <- rv$n + 1
rv$words <- rv$sentence[rv$n]
}
})
})
output$var <- renderPrint({
return(rv$words)
})
}
shinyApp(ui, server)
总结：
reactiveEvent()和observe() invalidateLater来重做一个Observ（）真的很有用来源：