修改R Markdown Flexdashboard实现分组/子分组的页面、标签与图表

原代码基于Stack Overflow用户@z-lin的实现，可生成按物种划分页面、变量对划分标签的Flexdashboard。现需适配含group/subgroup/x/y列的数据集，实现以下功能：

• 按group字段生成独立页面
• 每个页面内按subgroup字段生成标签页
• 每个标签页展示对应子分组的x-y散点图与数据表格

修改后代码运行时报错，提示缺少x、y变量，需修正代码以实现预期功能。

数据生成代码 #

set.seed(123)
df <- data.frame(
  group = rep(c("GroupA", "GroupB", "GroupC"), each = 100),
  subgroup = rep(c("Sub1", "Sub2"), each = 50, times = 3),
  x = rnorm(300),
  y = rnorm(300) + 0.5 * rep(c(1, -1), each = 50, times = 3)
)

报错的修改后代码 #

---
title: "Grouped Flexdashboard"
output: flexdashboard::flex_dashboard
---

```{r setup, include=FALSE}
library(flexdashboard)
library(ggplot2)
library(dplyr)

df <- read.csv("data.csv") # 实际使用时替换为数据路径
groups <- unique(df$group)

`r lapply(groups, function(g) {
cat(paste0("\n# ", g, "\n\n"))

subgroups <- unique(df$subgroup[df$group == g])

lapply(subgroups, function(sg) {
cat(paste0("## ", sg, "\n\n"))

# 筛选当前分组和子分组的数据
sub_data <- df %>% filter(group == g, subgroup == sg)

# 散点图
cat("### 散点图\n")
print(ggplot(sub_data, aes(x = x, y = y)) + geom_point())

# 数据表格
cat("### 数据表格\n")
print(DT::datatable(sub_data))

cat("\n")

})
})`

## 修正后的代码
错误原因：`lapply`循环中生成动态内容时，ggplot和DT无法正确捕获循环环境中的`x`/`y`变量，直接用`cat`拼接markdown的方式存在环境变量传递问题。修正方案如下：

```r
---
title: "Grouped Flexdashboard"
output: flexdashboard::flex_dashboard
---

```{r setup, include=FALSE}
library(flexdashboard)
library(ggplot2)
library(dplyr)
library(DT)

set.seed(123)
df <- data.frame(
  group = rep(c("GroupA", "GroupB", "GroupC"), each = 100),
  subgroup = rep(c("Sub1", "Sub2"), each = 50, times = 3),
  x = rnorm(300),
  y = rnorm(300) + 0.5 * rep(c(1, -1), each = 50, times = 3)
)

groups <- unique(df$group)

`r lapply(groups, function(g) {

local({
current_group <- g
cat(paste0("\n# ", current_group, "\n\n"))

subgroups <- unique(df$subgroup[df$group == current_group])

lapply(subgroups, function(sg) {
  # 捕获当前subgroup变量
  local({
    current_subgroup <- sg
    cat(paste0("## ", current_subgroup, "\n\n"))
    
    sub_data <- df %>% filter(group == current_group, subgroup == current_subgroup)
    
    # 生成散点图代码块
    cat("### 散点图\n")
    cat("```{r echo=FALSE}\n")
    cat("ggplot(sub_data, aes(x = x, y = y)) + geom_point() + ggtitle('", current_subgroup, "散点图')\n")
    cat("```\n\n")
    
    # 生成数据表格代码块
    cat("### 数据表格\n")
    cat("```{r echo=FALSE}\n")
    cat("datatable(sub_data, caption = '", current_subgroup, "数据')\n")
    cat("```\n\n")
  })
})

})
})`

### 关键修改点
1. 用`local()`包裹循环内的代码，确保每个`group`和`subgroup`变量在独立环境中被捕获，避免循环变量覆盖导致的引用错误。
2. 将ggplot和DT的代码通过`cat`生成独立的R代码块，而非直接打印对象，确保Flexdashboard能正确解析渲染内容。
3. 为图表和表格添加标题/说明，提升可读性。

内容的提问来源于stack exchange，提问作者Jacek Kotowski