r - 向堆积条形图添加标签 - ggvis

Question

有没有办法在 ggvis 中将标签附加到堆叠条形图？

示例代码:

plotData = data.frame(v1    = c("ABC", "ABC", "ABC", "XYX","XYX","XYX"), 
                      v2    = c("A"  , "B"  , "C"  , "X"  ,"Y"  ,"Z"  ),
                      value = c(30   ,  60  , 10   ,  20  , 40  , 40  ))

plotData %>%
  ggvis(y = ~v1, fill = ~v2) %>%
  compute_stack(stack_var = ~value, group_var = ~ v1) %>% 
  layer_rects(x = ~stack_lwr_, x2 = ~stack_upr_, height = band()) %>%
  add_axis("y", title = "Variable") %>%
  add_axis("x", title = "Percentage") 

绘图输出:

首选输出 - 我将如何修改上面的图表以将标签添加到各个堆栈？ :

Answer 1

这不是一个理想的解决方案，但我还是会发布它，因为它可能会为您指明前进的方向。

基本上，我创建了一个新列来指定标签的位置，并使用它将文本放置到图表上。

不过，我还没有想出如何将 % 放到标签上。

library(ggvis)

plotData = data.frame(v1    = c("ABC", "ABC", "ABC", "XYX","XYX","XYX"), 
                      v2    = c("A"  , "B"  , "C"  , "X"  ,"Y"  ,"Z"  ),
                      value = c(30   ,  60  , 10   ,  20  , 40  , 40  ))

library(dplyr)

## This can be shortened, but I'm leaving it as is to show the steps
plotData <- plotData %>%
  group_by(v1) %>%
  mutate(cs = cumsum(value)) %>%
  mutate(pos = ifelse(is.na(cs - lag(cs)), cs, (cs - lag(cs)))) %>%
  mutate(pos = cs - (pos/2)) %>%
  ungroup

plotData %>%
  ggvis(y = ~v1, fill = ~v2) %>%
  compute_stack(stack_var = ~value, group_var = ~ v1) %>% 
  layer_rects(x = ~stack_lwr_, x2 = ~stack_upr_, height = band()) %>%
  add_axis("y", title = "Variable") %>%
  add_axis("x", title = "Percentage") %>%
  layer_text(x = prop("x", ~pos),
             y = prop("y", ~v1, scale = "ycenter"),
             text:=~value, fill := "white", fontSize:=20)