在 ggvis 中使用 layer_points + layer_boxplot 时添加交互性

Question

当我还想添加工具提示时，我无法让 layer_points() 和 layer_boxplot() 一起工作。

require(ggvis)

### Example Data
mtc<-mtcars
mtc$label <- rownames(mtcars)   #unique Identifier


## Tooltip but not boxplot - works fine
mtc  %>% 
  ggvis(~cyl, ~qsec, key:= ~label) %>% 
  layer_points(size := 15) %>% 
  set_options(width = "auto") %>%
  add_tooltip(function(data) data$label)
})  

## No tooltip - boxplot + points  - works fine
mtc  %>% 
  ggvis(~cyl, ~qsec) %>% 
  layer_points(size := 15) %>% 
  layer_boxplots(size := 0)   %>%
  set_options(width = "auto")

## Attempt at combining boxplot, points, tooltip - fails
mtc  %>% 
  ggvis(~cyl, ~qsec, key:= ~label) %>% 
  layer_points(size := 15) %>% 
  layer_boxplots(size := 0)   %>%
  set_options(width = "auto") %>%
  add_tooltip(function(data) data$label)

#Error in eval(expr, envir, enclos) : object 'label' not found
#Plot is produced but no interactivity

有没有人知道如何将箱线图叠加到layer_points() 加上工具提示（基本上识别异常值是我的目标）。

另一种答案是如何为layers_boxplot()自动生成的异常值添加交互性

Answer 1

看来我已经解决了 - 他们 key 必须在 layer_points() 范围内

mtc  %>% 
  ggvis(~cyl, ~qsec) %>% 
  layer_points(key:= ~label, size := 15) %>% 
  layer_boxplots(size := 0)   %>%
  set_options(width = "auto") %>%
  add_tooltip(function(data) data$label)