为什么我的Shiny plot下载应用程序不能在IE中使用？

Question

我已经看到许多文章使用JavaScript在IE中下载文件，但是由于我对JavaScript的了解非常少，我将不胜感激。下面我们有一个闪亮的应用程序，该程序在Firefox中可以完美运行，但是在IE中尝试时，根本不会出现下载窗口。

library(shiny)
library(plotly)
library(lubridate)

d <- data.frame(X1 = rnorm(50,mean=50,sd=10), 
                X2 = rnorm(50,mean=5,sd=1.5), 
                Y = rnorm(50,mean=200,sd=25))

ui <-fluidPage(
  title = 'Download Plotly',
  sidebarLayout(

    sidebarPanel(
      htmlOutput("text"),
      selectInput("plot_download", "Select plot to download", choices=list("plot1","plot2")),
      actionButton('download_plot', "Download")

    ),

    mainPanel(

      plotlyOutput('regPlot'),
      plotlyOutput('regPlot2'),
      tags$script('

                  document.getElementById("download_plot").onclick = function() {
                  var plot = $("#plot_download").val();
                  if(plot == "plot1"){
                  var gd = document.getElementById("regPlot");
                  }else{
                  var gd = document.getElementById("regPlot2");
                  }
                  Plotly.Snapshot.toImage(gd, {format: "png"}).once("success", function(url) {
                  var a = window.document.createElement("a");
                  a.href = url; 
                  a.type = "image/png";
                  a.download = "plot.png";
                  document.body.appendChild(a);
                  a.click();
                  document.body.removeChild(a);                      
                  });
                  }
                  ')
    )
  )
)

server <- function(input, output, session) {

  output$text <- renderUI({
    eopm <- Sys.Date() - days(day(Sys.Date()))
    HTML(paste(eopm))
  })

  regPlot11 <- reactive({
    plot_ly(d, x = d$X1, y = d$X2, mode = "markers")
  })
  output$regPlot <- renderPlotly({
    regPlot11()
  })

  regPlot222 <- reactive({
    plot_ly(d, x = d$X1, y = d$X2, mode = "markers")
  })
  output$regPlot2 <- renderPlotly({
    regPlot222()
  })

}

shinyApp(ui = ui, server = server)

我如何使其在IE中工作？

[编辑]

在我的帖子下发表评论后，我设法借助navigator.msSaveBlob做到了这一点：

library(shiny)
library(plotly)
library(lubridate)

d <- data.frame(X1 = rnorm(50,mean=50,sd=10), 
                X2 = rnorm(50,mean=5,sd=1.5), 
                Y = rnorm(50,mean=200,sd=25))

ui <-fluidPage(
  title = 'Download Plotly',
  sidebarLayout(

    sidebarPanel(
      htmlOutput("text"),
      selectInput("plot_download", "Select plot to download", choices=list("plot1","plot2")),
      actionButton('download_plot', "Download")),

    mainPanel(

      plotlyOutput('regPlot'),
      plotlyOutput('regPlot2'),
      tags$script('document.getElementById("download_plot").onclick = function() {
                  var plot = $("#plot_download").val();
                  if(plot == "plot1"){
                  var gd = document.getElementById("regPlot");
                  }else{
                  var gd = document.getElementById("regPlot2");
                  }
                  if (navigator.msSaveBlob) { 
                   Plotly.Snapshot.toImage(gd, {format: "png"}).once("success", function(url){
                      var blob = new Blob([url],{type:"image/png"});
                      navigator.msSaveBlob(blob, "plot.png");
                  } else {
                      Plotly.Snapshot.toImage(gd, {format: "png"}).once("success", function(url) {
                      var a = window.document.createElement("a");
                      a.href = url; 
                      a.type = "image/png";
                      a.download = "plot.png";
                      document.body.appendChild(a);
                      a.click();
                      document.body.removeChild(a);                      
                      });
                  }

                  }
                  '))))

server <- function(input, output, session) {

  output$text <- renderUI({
    eopm <- Sys.Date() - days(day(Sys.Date()))
    HTML(paste(eopm))
  })

  regPlot1 <- reactive({
    plot_ly(d, x = d$X1, y = d$X2, mode = "markers")
  })
  output$regPlot <- renderPlotly({
    regPlot1()
  })

  regPlot2 <- reactive({
    plot_ly(d, x = d$X1, y = d$X2, mode = "markers")
  })
  output$regPlot2 <- renderPlotly({
    regPlot2()
  })

}

shinyApp(ui = ui, server = server)

但是我无法从Plotly.Snapshot.toImage函数检索对象，我认为它是url，但是正如您在我的代码中看到的那样，它不起作用... 我将不胜感激！