求助：基于ggplot与Shiny补充类Plotly交互式散点图两项功能

Got it, let's add those two missing interactive features to your Shiny app! I've updated your code to include both clicking legend items to hide/show groups, and hover tooltips that display the exact X/Y values. Here's the full working code:

library(ggplot2)
library(ggrepel)
library(shiny)
library(plotly) # Added for enhanced interactivity

ui <- fluidPage(
  plotlyOutput("graph", height = 450, width = 900)
)

server <- function(input, output) {
  # Reactive values to track zoom ranges and visible groups
  ranges <- reactiveValues(x = NULL, y = NULL)
  visible_groups <- reactiveValues(selected = unique(Data$Group))
  
  # Creating sample Data
  Data <- data.frame(
    Group = c("Group 2","Group 1","Group 3","Group 2","Group 5","Group 4","Group 6", 
              "Group 7","Group 4","Group 3","Group 1","Group 5","Group 6","Group 7", 
              "Group 2","Group 4","Group 6","Group 7","Group 3","Group 1"),
    Fruit = c("apple","apple","apple","mango","apple","apple","apple","apple","mango","mango", 
              "mango","mango","mango","mango","orange","orange","orange","orange","orange", 
              "orange"),
    Percentage = c(68.46846847,77.35849057,72.72727273,26.12612613,76.31578947,62.79069767, 
                   71.05263158,69.23076923,30.23255814,25,20.75471698,23.68421053,23.68421053, 
                   23.07692308,5.405405405,6.976744186,5.263157895,7.692307692,2.272727273, 
                   1.886792453)
  )
  
  # Reactive filtered data based on visible groups
  filtered_data <- reactive({
    Data[Data$Group %in% visible_groups$selected, ]
  })
  
  # Handle legend click events to toggle group visibility
  observeEvent(event_data("plotly_legendclick"), {
    clicked_group <- event_data("plotly_legendclick")$name
    if(clicked_group %in% visible_groups$selected){
      visible_groups$selected <- visible_groups$selected[visible_groups$selected != clicked_group]
    } else {
      visible_groups$selected <- c(visible_groups$selected, clicked_group)
    }
  })
  
  # Handle double-click to zoom to selected brush area
  observeEvent(event_data("plotly_doubleclick"), {
    brush <- event_data("plotly_selected")
    if(!is.null(brush)){
      ranges$x <- c(min(brush$x), max(brush$x))
      ranges$y <- c(min(brush$y), max(brush$y))
    } else {
      ranges$x <- NULL
      ranges$y <- NULL
    }
  })
  
  # Render the interactive plot
  output$graph <- renderPlotly({
    p <- ggplot(filtered_data(), aes(y = reorder(Fruit, Percentage), x = Percentage, color = Group,
                                     text = paste("Fruit:", Fruit, "<br>Percentage:", round(Percentage, 2), "%"))) +
      geom_point(size = 4) +
      scale_x_continuous(limits = c(0, 100)) +
      theme(panel.grid.major.y = element_line(color = "gray90", size = 0.7),
            panel.background = element_blank(),
            strip.background = element_blank(),
            panel.border = element_rect(color = "black", fill = NA, size = 1)) +
      labs(y = "", x = "% of people", color = "")
    
    # Convert to plotly and add zoom logic
    ggplotly(p, tooltip = "text") %>%
      layout(dragmode = "select", # Enable box selection
             xaxis = list(range = ranges$x),
             yaxis = list(range = ranges$y))
  })
}

shinyApp(ui = ui, server = server)

Key Changes Explained: #

1. Mouse Hover Tooltips:

   • Added a text aesthetic in the ggplot call that formats hover content to show both the fruit name and rounded percentage value.
   • Used ggplotly(p, tooltip = "text") to render the plot with custom, clean hover tooltips that display this formatted text.

2. Legend Click to Hide/Show Groups:

   • Created a visible_groups reactive value to track which groups are currently displayed.
   • Added an observer for plotly_legendclick events: when a legend item is clicked, the code toggles that group's visibility by adding/removing it from the tracked list.
   • A filtered_data reactive filters the original dataset to only include visible groups before plotting.

3. Retained Box-Select + Double-Click Zoom:

   • Set dragmode = "select" in the plotly layout to enable box selection for zooming.
   • Added an observer for plotly_doubleclick events: when you double-click after selecting a box, the plot zooms to that area; double-clicking without a selection resets the zoom.

This implementation gives you all three requested interactive features, with a Plotly-like experience while keeping your original ggplot styling intact.

内容的提问来源于stack exchange，提问作者Vedha Viyash