R语言ggplot绘制Waffle图表时0值导致图例无法合并的解决方法

核心原因 #

当数据包含0值时，ggplot会自动过滤无数据的类别，但scale_label_pictogram与scale_color_manual的映射若未同步保留所有类别，就会导致图例无法合并。

解决方案步骤 #

• 先将类别列转换为包含所有水平的因子，确保0值对应的类别不会被自动丢弃：

  df_1$category <- factor(df_1$category, levels = c("类别1", "类别2", "类别3")) # 替换为你的实际类别列表
  

• 在两个scale中都设置drop = FALSE，强制保留所有类别映射：

  scale_label_pictogram(
    name = NULL, # 统一图例名称，避免重复
    values = c("类别1" = "icon1", "类别2" = "icon2", "类别3" = "icon3"),
    drop = FALSE
  ) +
  scale_color_manual(
    name = NULL,
    values = c("类别1" = "#ff0000", "类别2" = "#00ff00", "类别3" = "#0000ff"),
    drop = FALSE
  )
  

• 若图标与颜色未自动对齐，用guides()统一图例样式：

  guides(
    label = guide_legend(order = 1),
    color = guide_legend(order = 1)
  )
  

完整示例代码 #

library(tidyverse)
library(waffle)

# 构造含0值的示例数据
df_1 <- tibble(
  category = c("A", "B", "C"),
  value = c(5, 0, 3)
)

# 转换为因子并锁定所有类别水平
df_1$category <- factor(df_1$category, levels = c("A", "B", "C"))

# 绘制waffle图并合并图例
ggplot(df_1, aes(fill = category, values = value)) +
  geom_waffle(n_rows = 3, color = "white", size = 1.25) +
  scale_label_pictogram(
    name = NULL,
    values = c("A" = "car", "B" = "bike", "C" = "bus"),
    drop = FALSE
  ) +
  scale_fill_manual(
    name = NULL,
    values = c("A" = "#2c3e50", "B" = "#e74c3c", "C" = "#3498db"),
    drop = FALSE
  ) +
  guides(
    fill = guide_legend(override.aes = list(label = c("car", "bike", "bus")))
  ) +
  theme_void()

关键说明 #

• drop = FALSE是核心，确保所有类别无论有无数据都保留在图例中
• 统一两个scale的name参数（如都设为NULL），让ggplot识别它们属于同一图例
• 若图标与颜色错位，可通过guides()的override.aes手动指定每个类别的图标，强制合并

内容的提问来源于stack exchange，提问作者Zhiqiang Wang