如何用R的ggplot与ggforce实现堆叠柱状图局部放大及定制

1. 让Sold柱子显示总和 #

要实现Sold柱子显示所有值的总和，需先对数据按Status分组汇总，再绘图：

library(ggplot2)
library(dplyr)

# 原始数据
df <- data.frame(Product = c("A", "A", "A", "A", "A", "A"),
                 Status = c("Bought", "Sold", "Sold", "Sold", "Sold", "Sold"),
                 Month = c("April", "March", "May", "July", "October", "November"),
                 N = c(48, 12, 11, 1, 2, 3))

# 生成汇总数据：按Product和Status分组求和
df_summary <- df %>%
  group_by(Product, Status) %>%
  summarize(N = sum(N), .groups = "drop")

# 绘制修正后的chart1
(chart1_fixed <- ggplot(df_summary, aes(x = Product, y = N, fill = Status)) +
    geom_col(position = "stack") +
    geom_text(aes(label = N), size = 3, position = position_stack(vjust = 0.5)) +
    theme_bw())

此时Sold柱子会显示所有Sold值的总和（29），与Bought的48堆叠展示。

2. 实现带放大效果的镜像图表 #

要实现「右侧为总和柱（chart1样式）、左侧为Sold详细放大版」的效果，提供两种简便方法：

方法一：使用ggforce的facet_zoom（保留联动放大效果） #

通过side参数控制放大面板位置，同时分别绘制总和数据与详细数据：

library(ggforce)

(chart_final <- ggplot() +
    # 绘制右侧的总和柱（chart1样式）
    geom_col(data = df_summary, aes(x = Product, y = N, fill = Status), position = "stack") +
    geom_text(data = df_summary, aes(x = Product, y = N, label = N), size = 3, position = position_stack(vjust = 0.5)) +
    # 绘制左侧的Sold详细分月柱（仅在放大面板显示）
    geom_col(data = df %>% filter(Status == "Sold"), aes(x = Product, y = N, fill = Month), position = "stack") +
    geom_text(data = df %>% filter(Status == "Sold"), aes(x = Product, y = N, label = N), size = 3, position = position_stack(vjust = 0.5)) +
    theme_bw() +
    # 设置放大范围为0-27.5（覆盖Sold总和），放大面板放在左侧，原图（总和柱）在右侧
    facet_zoom(ylim = c(0, 27.5), side = "left", split = TRUE) +
    # 统一填充色的图例标题
    labs(fill = "分类") +
    # 隐藏facet的标题栏
    theme(strip.background = element_blank(), strip.text = element_blank()))

若需让放大面板在右侧、总和柱在左侧，只需将side = "left"改为side = "right"。

方法二：使用patchwork拼接两个独立图表（更灵活） #

如果不需要联动放大效果，直接用patchwork拼接两个图表，可自由调整左右顺序：

library(patchwork)

# 绘制Sold详细放大图
chart_detail <- ggplot(df %>% filter(Status == "Sold"), aes(x = Product, y = N, fill = Month)) +
  geom_col(position = "stack") +
  geom_text(aes(label = N), size = 3, position = position_stack(vjust = 0.5)) +
  theme_bw() +
  ylim(0, 27.5) + # 固定y轴范围实现放大
  labs(fill = "月份")

# 绘制总和柱图
chart_total <- ggplot(df_summary, aes(x = Product, y = N, fill = Status)) +
  geom_col(position = "stack") +
  geom_text(aes(label = N), size = 3, position = position_stack(vjust = 0.5)) +
  theme_bw() +
  labs(fill = "状态")

# 拼接：左侧放大详细图，右侧总和柱图
chart_detail + chart_total + plot_layout(ncol = 2)

# 镜像效果：右侧放大图，左侧总和柱图
chart_total + chart_detail + plot_layout(ncol = 2)

这种方法可独立控制两个图表的样式，调整自由度更高。

内容的提问来源于stack exchange，提问作者mad10les