如何在facet_wrap的堆叠条形图顶部添加Revenue总计文本？

问题说明 #

需要给facet_wrap生成的每个堆叠条形图顶部，添加对应年份的Revenue数值作为总计文本，数值与数据中metric列的"Revenue"值一致。

当前代码 #

prof %>% 
  filter(metric != "Revenue") %>% 
  pivot_longer(!1:2, names_to = "group") %>%
  mutate(total_2024 = sum(value * (year == 2024)), .by = group) %>%
  mutate(group = reorder(group, -total_2024)) %>% 
  mutate(year_group_total = sum(value), .by = c(year, group)) %>% 
  ggplot() + 
  aes(x = factor(year), y = value, fill = metric) +
  geom_bar(stat = "identity", position = "stack") +
  geom_text(aes(label = round(value, 0)), 
            position = position_stack(vjust = 0.5),
            size = 3, color = "white") + 
  facet_wrap(~ group, nrow = 1)

数据结构 #

structure(list(metric = c("Semi-direct/Indirect OPEX", "Direct OPEX", 
"Cogs", "EBITDA", "Semi-direct/Indirect OPEX", "Direct OPEX", 
"Cogs", "EBITDA", "Revenue", "Revenue"), year = c(2024, 2024, 
2024, 2024, 2023, 2023, 2023, 2023, 2023, 2024), `B2C Mobile` = c(897.784296273134, 
1553.75734792605, 2174.93374732592, 5755.54423659222, 998.273127452503, 
1491.49788927451, 2577.32537561267, 5188.04527286322, 10255.1416652029, 
10382.0196281173), `B2B Mobile` = c(556.814329695538, 910.678821511126, 
1206.88678722175, 2974.74324198596, 551.422958887446, 886.952093722743, 
1291.87657757128, 3091.66275796777, 5821.91438814924, 5649.12318041437
), `B2C Fiber` = c(517.558827512705, 495.164268332664, 2047.61667685037, 
471.94510699589, 412.859413724063, 454.877885907187, 1936.82417093636, 
500.145306066416, 3304.70677663402, 3532.28487969163), `HFC B2C` = c(252.327289621926, 
194.055038683145, 892.79869969144, 305.281096059764, 183.864737312406, 
177.365375913805, 1020.84928874347, 403.547767116329, 1785.62716908601, 
1644.46212405627), `B2B Fiber` = c(130.704875738789, 181.150718557421, 
315.08402928116, 259.876581766672, 184.022162044549, 203.280264535604, 
290.095491871566, 288.017740790265, 965.415659241984, 886.816205344042
), Wholesale = c(335.66507766929, 81.3907132734683, 359.637328238667, 
726.393721768013, 606.085398971104, -161.245692311656, 379.926654798916, 
721.092000854812, 1545.85836231318, 1503.08684094944), Other = c(582.212942299726, 
460.192151400037, 486.308969368716, 660.596166217528, 884.601699952072, 
349.083787813294, 545.586302785773, 509.765926491606, 2289.03771704274, 
2189.31022928601)), class = c("tbl_df", "tbl", "data.frame"), row.names = c(NA, 
-10L))

当前效果 #

解决方案 #

核心思路是先从原始数据中提取每个group和year对应的Revenue数值，再将其作为文本标注在每个堆叠条形的顶部。

修改后完整代码 #

# 提取Revenue数据并整理格式
revenue_data <- prof %>%
  filter(metric == "Revenue") %>%
  pivot_longer(!1:2, names_to = "group", values_to = "revenue") %>%
  select(-metric)

# 主数据处理+绘图
prof %>% 
  filter(metric != "Revenue") %>% 
  pivot_longer(!1:2, names_to = "group") %>%
  mutate(total_2024 = sum(value * (year == 2024)), .by = group) %>%
  mutate(group = reorder(group, -total_2024)) %>% 
  # 合并Revenue数据
  left_join(revenue_data, by = c("group", "year")) %>%
  ggplot() + 
  aes(x = factor(year), y = value, fill = metric) +
  geom_bar(stat = "identity", position = "stack") +
  # 条形内部数值标签
  geom_text(aes(label = round(value, 0)), 
            position = position_stack(vjust = 0.5),
            size = 3, color = "white") + 
  # 添加顶部Revenue总计文本
  geom_text(aes(y = revenue, label = round(revenue, 0)),
            size = 3.5, color = "black",
            vjust = -0.5) +
  facet_wrap(~ group, nrow = 1) +
  # 调整y轴范围，避免顶部文本被截断
  expand_limits(y = max(revenue_data$revenue) * 1.1)

关键步骤说明 #

1. 提取并整理Revenue数据：从原始数据中筛选出metric为"Revenue"的行，转成长格式后得到每个分组（group）和年份（year）对应的Revenue值。
2. 合并数据：将Revenue数据与主数据通过group和year关联，确保每个行都能拿到对应年份的Revenue数值。
3. 添加顶部文本：用geom_text指定y轴位置为revenue，标签为四舍五入后的Revenue值，通过vjust = -0.5将文本放在条形顶部上方。
4. 调整y轴范围：使用expand_limits扩展y轴上限，避免顶部的总计文本被图表边界截断。

内容的提问来源于stack exchange，提问作者HoelR