求助：优化ggplot2柱状图与折线图美观度并解决绘图异常

需求与问题 #

• 优化需求：
  1. 为App_1、App_2两条折线添加数据点，并在每个点上方标注对应百分比数值
  2. 在Total_Wth蓝色柱状图的中间标注对应数值
• 遇到的问题：使用geom_point时右侧Y轴消失，且新增了第二个重复图例

修正后的完整代码 #

x <- c("Mar_22", "Apr_22", "May_22", "June_22", "July_22", "Aug_22")
Total_Wth <- c(88959, 100396, 117989, 130561, 150282, 148970)
App_1 <- c(0.91, 0.9, 0.91, 0.88, 0.9, 0.9)
App_2 <- c(0.09, 0.1, 0.09, 0.12, 0.10, 0.10)

df <- data.frame(x, Total_Wth, App_1, App_2)
transf_fact <- max(df$Total_Wth)/max(df$App_1)

df %>%
  mutate_at(vars(matches("App")), ~.*transf_fact) %>%
  gather(y_var, val, -x) %>%
  ggplot(aes(x = factor(x, df$x), y = val, fill = y_var)) +
  # 绘制柱状图
  geom_col(data = . %>% filter(y_var == "Total_Wth")) +
  # 标注柱状图中间数值
  geom_text(data = . %>% filter(y_var == "Total_Wth"),
            aes(label = val),
            vjust = 0.5, size = 3) +
  # 绘制折线
  geom_line(data = . %>% filter(y_var != "Total_Wth"),
            aes(color = y_var, group = y_var),
            size = 1.2) +
  # 添加折线数据点
  geom_point(data = . %>% filter(y_var != "Total_Wth"),
            aes(color = y_var, group = y_var),
            size = 3) +
  # 标注折线的百分比数值（转换回原始比例）
  geom_text(data = . %>% filter(y_var != "Total_Wth"),
            aes(label = scales::percent(val/transf_fact), color = y_var),
            vjust = -1, size = 3) +
  # 统一fill和color的图例
  scale_fill_discrete(aesthetics = c("colour", "fill")) +
  # 设置双Y轴
  scale_y_continuous(sec.axis = sec_axis(trans = ~ . / transf_fact, 
                                         name = "%", 
                                         labels = scales::percent),
                     limits = c(0, 170000), 
                     breaks = c(34000 * 0:5)) +
  # 图表标签设置
  labs(x = "月份", y = "总计",
       title = "示例标题",
       subtitle = "示例副标题") +
  theme_bw() +
  theme(legend.position = "bottom",
        legend.title = element_blank(),
        legend.key.size = unit(0.5, 'cm'),
        legend.text = element_text(size=7),
        panel.grid.major.x = element_blank(),
        panel.border = element_blank(),
        plot.title = element_text(hjust = 0),
        plot.subtitle = element_text(size=8, hjust=0, face="italic", color="black"),
        axis.text.x = element_text(size = 10, face = "bold"))

关键修改说明 #

1. 折线数据点与标注：
   • geom_point明确指定过滤后的App类数据，继承折线的color和group映射，避免新增图例
   • 百分比标注通过val/transf_fact转换回原始比例，用scales::percent格式化，vjust=-1让文字显示在点上方
2. 柱状图数值标注：
   • 单独过滤Total_Wth数据，用geom_text在柱子中间（vjust=0.5）标注数值
3. 避免轴与图例异常：
   • 所有图层都严格指定对应的数据子集，确保双Y轴的转换逻辑不被破坏，图例保持统一

内容的提问来源于stack exchange，提问作者Bren85