flexdashboard中ggplotly结合geom_vline不显示竖线求助

手动构造的事件日期（event tibble）能正常生成带蓝色虚线竖线的交互式图表，但从Excel导入并转换为相同POSIXct格式的fecha.evento tibble，无法显示竖线。两者数据结构一致，已尝试修改变量类型、调整代码，问题仍未解决。

正常显示的代码 #

library(openair)
library(tidyverse)
library(plotly)

aq <- importAURN(site = c("my1", "nott"), year = 2000,
                 pollutant = c("nox", "no2", "o3"))
aq <- aq %>% pivot_longer(cols = 2:4, names_to = "Parameter", values_to = "Value")

event <- tibble(date = c("2000-03-23 11:00:00","2000-05-29 11:40:00","2000-08-22 12:00:00"))
event$date <- as.POSIXct(event$date,format="%Y-%m-%d %H:%M", tz = "UTC")

b <- ggplot(aq,
       aes(x = date, 
           y = Value, 
           colour = site,
           group = site)) + 
  geom_line() + 
  geom_vline(data = event ,
             aes(xintercept = as.numeric(date)),
             linetype = 2, 
             colour = "blue",
             show.legend = TRUE) +
  theme_bw() +
  facet_grid(Parameter ~ .,
             scales = "free")

ggplotly(b,dynamicTicks = TRUE) %>% 
  layout(legend = list(orientation = "h", y = -0.2))

无法显示竖线的代码 #

fecha.evento <- tibble(date = as.character(evento.q$`Toma conocimiento`))
fecha.evento$date <- as.POSIXct(fecha.evento$date,format="%Y-%m-%d %H:%M:%S", tz = "UTC")

a <- ggplot(quintero,
       aes(x = date,
           y = dato_val_ug,
           colour = Ubicación, 
           group = Ubicación)) + 
  geom_line() + 
  geom_vline(data = fecha.evento,
             aes(xintercept = as.numeric(date)),
             linetype = 2, 
             colour = "blue",
             show.legend = TRUE) +
  theme_bw() +
  facet_grid(Parámetro ~ .,
             scales = "free")

ggplotly(a,dynamicTicks = TRUE) %>% 
  layout(legend = list(orientation = "h", y = -0.2))

数据结构 #

> str(event)
tibble [3 × 1] (S3: tbl_df/tbl/data.frame)
 $ date: POSIXct[1:3], format: "2023-03-23 11:00:00" "2023-05-02 11:40:00" "2023-04-22 12:00:00"

> str(fecha.evento)
tibble [10 × 1] (S3: tbl_df/tbl/data.frame)
 $ date: POSIXct[1:10], format: "2023-03-23 11:00:00" "2023-03-29 11:40:00" "2023-03-30 12:00:00" "2023-03-31 10:30:00" ...

排查方向与解决建议 #

• 检查日期范围匹配：用range(quintero$date)和range(fecha.evento$date)对比，确认事件日期是否在主数据的x轴区间内，超出范围的竖线不会显示。
• 强制时区一致性：用lubridate::force_tz(fecha.evento$date, tzone = "UTC")重置时区，确保和quintero$date完全一致，避免隐含时区差异。
• 移除as.numeric()转换：将geom_vline的aes(xintercept = as.numeric(date))改为aes(xintercept = date)，直接使用POSIXct类型更稳定。
• 排查分面影响：先去掉facet_grid测试竖线是否显示，若正常再检查各分面的时间范围是否包含事件日期。
• 验证日期转换有效性：用head(fecha.evento$date)查看转换结果，确认无NA值，若存在则调整日期格式字符串匹配Excel导出的格式。

内容的提问来源于stack exchange，提问作者Camilo Montes