R语言DT包中能否保存表格编辑内容？

Hey there! Great question—this is a super common point of confusion when working with DT's editing feature. Let me break this down for you:

Why the default edit doesn't update your data.frame #

DT renders your table in the browser (frontend), while your original data.frame lives in your R session (backend). These are separate environments, so edits you make in the visual table don't automatically sync back to the R object by default. That's why your manual df[1,2] <- 76 style edits still work directly, but DT's frontend edits don't carry over unless you set up a bridge between the two.

How to sync edits to your underlying data (the part you might have missed) #

You need to use Shiny's reactive framework to capture the edit events from the DT table and update your data.frame accordingly. Here's a simple, working example:

library(shiny)
library(DT)

ui <- fluidPage(
  DTOutput("editable_table")
)

server <- function(input, output, session) {
  # Store your data in a reactive value (so it can be updated dynamically)
  reactive_data <- reactiveVal(data.frame(
    Number = c(10, 20, 30),
    Text = c("apple", "banana", "cherry"),
    stringsAsFactors = FALSE
  ))
  
  # Render the editable DT table
  output$editable_table <- renderDT({
    datatable(reactive_data(), editable = TRUE)
  })
  
  # Capture edit events and update the underlying data
  observeEvent(input$editable_table_cell_edit, {
    edit_info <- input$editable_table_cell_edit
    row_idx <- edit_info$row
    col_idx <- edit_info$col
    new_value <- edit_info$value
    
    # Update the reactive data with the new value
    updated_df <- reactive_data()
    updated_df[row_idx, col_idx] <- new_value
    reactive_data(updated_df)
    
    # Optional: Print the updated data to the console to verify
    cat("Updated data:\n")
    print(reactive_data())
  })
}

shinyApp(ui, server)

When you run this Shiny app, any edits you make in the table will instantly update the reactive_data object (which acts as your editable data.frame).

What if you're using R Markdown instead of a standalone Shiny app? #

You can use the same approach with the runtime: shiny option in your R Markdown document:

---
title: "DT Editing in R Markdown"
runtime: shiny
---

```{r setup, include=FALSE}
library(DT)

# Reactive storage for our data
reactive_df <- reactiveVal(data.frame(
  ID = 1:5,
  Value = c(100, 200, 300, 400, 500)
))

# Render editable table
renderDT({
  datatable(reactive_df(), editable = TRUE)
})

# Sync edits to reactive data
observeEvent(input$table_cell_edit, {
  info <- input$table_cell_edit
  updated <- reactive_df()
  updated[info$row, info$col] <- info$value
  reactive_df(updated)
})

# Show the updated data below the table
renderPrint({
  reactive_df()
})

## Is this a feature that could be enhanced?
DT's design intentionally separates frontend rendering from backend R data—since it's primarily an HTML table library, it doesn't assume you're working in a Shiny environment by default. That said, if you think the sync process could be simplified (e.g., a built-in function to auto-sync edits to a `data.frame`), you could absolutely submit a feature request to the DT GitHub repository. For now, though, using Shiny's reactive tools is the standard, supported way to achieve the "edit + save" workflow you're looking for.

内容的提问来源于stack exchange，提问作者pssguy