Shiny App新冠数据展示求助：筛选与排序功能实现

Hey there! Let's work through fixing your Shiny app issue—since you're new to R and Shiny, I'll keep this step-by-step and easy to follow.

First, Let's Diagnose the Core Issues #

From what you described:

1. Your dropdown works, but charts won't render → This is almost certainly a problem with how you're handling reactive data in the server logic.
2. Running ggplot alone shows missing parts of the chart → Likely a date formatting issue or missing date entries in your raw data.

Let's Build a Working Solution #

I'll share a complete, annotated example that matches your requirements, then break down the key parts you might have missed.

Step 1: Prep Your Data First

First, make sure your date column (Day) is formatted as a proper date type—this fixes the "missing middle" issue in ggplot. Also, we'll fill in any missing dates so your line charts are continuous:

library(tidyverse)
library(lubridate)

# Load your long-format data (replace with your file path)
covid_data <- read.csv("your_covid_data.csv") %>%
  # Convert Day to date type (adjust format if your dates are in a different style, e.g., "%m/%d/%Y")
  mutate(Day = ymd(Day)) %>%
  # Group by county to fill missing dates
  group_by(County) %>%
  # Add any missing dates between the first and last entry
  complete(Day = seq(min(Day), max(Day), by = "day")) %>%
  # Replace NA values with 0 for cases/recoveries/deaths
  replace_na(list(`positive cases` = 0, recoveries = 0, deaths = 0)) %>%
  ungroup()

Step 2: Full Shiny App Code

Here's the complete UI and server logic that will work with your data:

library(shiny)

# UI Section
ui <- fluidPage(
  titlePanel("COVID-19 County Data Dashboard"),
  sidebarLayout(
    # Dropdown menu for county selection
    sidebarPanel(
      selectInput(
        inputId = "selected_county",
        label = "Choose a County:",
        choices = unique(covid_data$County),
        selected = unique(covid_data$County)[1] # Default to first county
      )
    ),
    # Main panel with three tabbed charts
    mainPanel(
      tabsetPanel(
        tabPanel("Confirmed Cases", plotOutput("cases_plot")),
        tabPanel("Recoveries", plotOutput("recoveries_plot")),
        tabPanel("Deaths", plotOutput("deaths_plot"))
      )
    )
  )
)

# Server Logic
server <- function(input, output) {
  # Reactive filtered data: updates automatically when user picks a new county
  filtered_data <- reactive({
    # req() ensures we don't run this until the user selects a county
    req(input$selected_county)
    covid_data %>% filter(County == input$selected_county)
  })

  # Render Confirmed Cases Chart
  output$cases_plot <- renderPlot({
    filtered_data() %>%
      ggplot(aes(x = Day, y = `positive cases`)) +
      geom_line(color = "#e74c3c", linewidth = 1) +
      labs(title = "Daily Confirmed COVID-19 Cases",
           x = "Date",
           y = "Number of Cases") +
      theme_minimal() +
      theme(axis.text.x = element_text(angle = 45, hjust = 1)) # Rotate dates to avoid overlap
  })

  # Render Recoveries Chart
  output$recoveries_plot <- renderPlot({
    filtered_data() %>%
      ggplot(aes(x = Day, y = recoveries)) +
      geom_line(color = "#2ecc71", linewidth = 1) +
      labs(title = "Daily COVID-19 Recoveries",
           x = "Date",
           y = "Number of Recoveries") +
      theme_minimal() +
      theme(axis.text.x = element_text(angle = 45, hjust = 1))
  })

  # Render Deaths Chart
  output$deaths_plot <- renderPlot({
    filtered_data() %>%
      ggplot(aes(x = Day, y = deaths)) +
      geom_line(color = "#34495e", linewidth = 1) +
      labs(title = "Daily COVID-19 Deaths",
           x = "Date",
           y = "Number of Deaths") +
      theme_minimal() +
      theme(axis.text.x = element_text(angle = 45, hjust = 1))
  })
}

# Run the app
shinyApp(ui = ui, server = server)

Key Fixes & Explanations #

1. Reactive Data: The filtered_data() reactive expression is the heart of the app. It automatically updates whenever the user selects a new county, so all three charts use the latest filtered data—no need to repeat filtering code for each plot.
2. req() Function: This prevents Shiny from trying to render charts before the user picks a county, which avoids confusing error messages.
3. Date Formatting: Converting Day to a Date type tells ggplot to treat it as a continuous variable, so your X-axis will show all dates without gaps. The complete() step fills in any missing dates in your raw data, making the line chart smooth.
4. Column Names with Spaces: When referencing positive cases (a column name with a space), wrap it in backticks ` so R knows it's a single column.

Testing the ggplot Code Alone #

If you want to test your ggplot code outside Shiny, just filter manually and check the date type:

# Pick a county to test
test_county <- "Your County Name"
test_data <- covid_data %>% filter(County == test_county)

# Verify date type (should say "Date")
class(test_data$Day)

# Plot
ggplot(test_data, aes(x = Day, y = `positive cases`)) +
  geom_line(color = "red") +
  theme_minimal()

This should show a continuous line without missing sections now that your data is properly formatted.

内容的提问来源于stack exchange，提问作者Nathan May