【发布时间】:2020-12-16 11:07:17
【问题描述】:
我想知道如何让我的闪亮应用对任何对 data.frame 的修改做出反应。这在很多方面都很有用。比如:
[server part]
rv <- reactiveValues(
df <- data.frame(...)
)
observeEvent(rv$df, {
...
})
【问题讨论】:
我想知道如何让我的闪亮应用对任何对 data.frame 的修改做出反应。这在很多方面都很有用。比如:
[server part]
rv <- reactiveValues(
df <- data.frame(...)
)
observeEvent(rv$df, {
...
})
【问题讨论】:
这是我想出的一个解决方案的示例。请注意,这只是一个用例(没有模块)。我会告诉你,我成功地将它应用于更复杂的应用程序(> 100 个输入)。技术细节以 cmets 的形式包含在代码中。
library(shiny)
# Simple GUI to let the user play with the app.
ui <- fluidPage(
actionButton("dev","dev"),
actionButton("dev2","dev2")
)
# Here comes the serious business
server <- function(input, output, session) {
# First, I use reactiveValues (clear way to store values)
rv <- reactiveValues(
# And here is our main data frame
df = data.frame(a = 0, b = 1)
)
# Ok, let's get the magic begin ! R recognizes this in the current environment:
makeReactiveBinding("rv$df")
# Also works with things such as:
# makeReactiveBinding(sprint("rv$%s", "df"))
# which opens the way to dynamic UI.
# Then, I get a way to catch my table from a reactive
rdf <- reactive(rv$df)
# And there, I only have to call for this reactive to get data.frame-related event
observe(print(rdf()$b))
# Here are some little ways to interact with the app
# Notice the `<<-` assignment to store new values
# Add values to df$a (expected behavior: print df$b at its value)
observeEvent(input$dev, {
rv$df$a <<- rv$df$a+1
})
# Add values to df$b (expected behavior: print df$b at its new value)
observeEvent(input$dev2, {
rv$df$b <<- rv$df$b+1
})
}
shinyApp(ui, server)
我希望这可能对 Shiny 的一些用户有所帮助。我发现这很容易实现,而且非常有用。
【讨论】: