TLDR -为什么这个自定义函数不起作用？

library(ggplot2)
dat <- data.frame(date = c("2023-06-01", "2023-06-02", "2023-06-03"),
                  shots = c(8, 12, 10),
                  passes = c(340, 410, 390))

cust_ggplot <- function(df, date, metric){
  
  min_date <- min({{ date }}) # invalid 'type' (closure argument)
  # min_date <- min(!! ensym(date))  # Error in !ensym(x): invalid argument type.
  # min_date <- min(df[[date]]) # works if date is enclosed in quotes like this dat |> cust_title("date", "shots")
                                # But I want it to look like tidyverse function with no quotes needed.
  
}

dat |> 
  cust_ggplot(date, shots)

我想模仿tidyverse风格，不需要使用dat |> cust_title("date", "shots")

完整走查：我试图把这个简单的ggplot代码转换成一个函数。

### Data
dat <- data.frame(date = c("2023-06-01", "2023-06-02", "2023-06-03"),
                  shots = c(8, 12, 10),
                  passes = c(340, 410, 390)
)

### Working Code
dat |>
  ggplot(aes(x = date, y = shots)) +
  geom_col() +
  ggtitle(
    paste(
      "Shot Totals from",
      min(dat$date),
      "-",
      max(dat$date)
    )
  )

我可以让情节工作刚刚好通过

### Turn it into a function

# This portion works
cust_ggplot <- function(df, date, metric){
  
  df |>
    ggplot(aes(x = {{ date }}, y = {{ metric }} )) +
    geom_col() 
  
}

dat |> 
  cust_ggplot(date, shots) +
  ggtitle("Shot Totals")

但是当我试图找到最小或最大日期传递给副标题时，我的尝试都不起作用。

cust_ggplot <- function(df, date, metric){
  
  # min_date <- min({{ date }}) invalid 'type' (closure argument)
  # min_date <- min(!! ensym(date))  #Error in !ensym(x): invalid argument type.
  # min_date <- min(df[[date]]) # works if date is enclosed in quotes like this dat |> cust_title("date", "shots")
                                # But I want it to look like tidyverse function with no quotes needed.
  
  df |>
    ggplot(aes(x = {{ date }}, y = {{ metric }} )) +
    geom_col() 
  
}