保留Plotly默认悬停模板 移除悬停中的迹名称

使用R语言的plotly包绘制瀑布图时，添加第二个text迹后，悬停文本默认会显示迹名称框。希望保留默认悬停模板的样式，但移除迹名称框，但自行尝试的自定义hovertemplate无法还原默认的三角标记，且对非relative类型的柱形显示过多冗余文本。

原瀑布图代码：

library(plotly)
library(dplyr)

data <- tibble(x = c("Sales", "Consulting", "Net revenue", "New Intermediate",
                     "Purchases", "Other expenses", "Profit before tax"), 
               measure = c("relative", "relative", "total", "absolute", "relative", 
                           "relative", "total"),
               text = c("+60", "+80", "", "", "-40", 
                        "-20", "Total"),
               y = c(60, 80, 0, 120, -40, -20, 0)) %>% 
  mutate(x = factor(x, x))


(fig <- data %>% 
    plot_ly() %>% 
    add_text(x = factor("Sales", levels = data$x),
           y = 80, text = "Some Text") %>%
    add_trace(type = "waterfall", measure = ~measure,
              x = ~x, textposition = "outside", y= ~y, text =~text,
              connector = list(line = list(color= "rgb(63, 63, 63)"))) %>%
    layout(title = "Profit and loss statement 2018",
           xaxis = list(title = ""),
           yaxis = list(title = ""),
           autosize = TRUE,
           showlegend = FALSE))

尝试的自定义hovertemplate代码（效果不符合预期）：

data %>% 
  plot_ly() %>% 
  add_trace(type = "waterfall", measure = ~measure,
            x = ~x, textposition = "outside", y= ~y, text =~text,
            connector = list(line = list(color= "rgb(63, 63, 63)")),
            hovertemplate = "(%{x},%{y})<br>%{text}<br>%{delta}<br>Initial: %{initial}<extra></extra>") %>%
  add_text(x = factor("Sales", levels = data$x),
           y = 80, text = "Some Text") %>% 
  layout(title = "Profit and loss statement 2018",
         xaxis = list(title = ""),
         yaxis = list(title = ""),
         autosize = TRUE,
         showlegend = TRUE)

plotly瀑布图的默认悬停模板会根据measure的类型动态显示内容，因此需要针对不同measure值定制对应的hovertemplate，再通过ifelse逻辑动态分配，同时添加<extra></extra>来隐藏迹名称框。

核心自定义模板逻辑 #

根据不同measure类型，匹配默认显示内容：

• relative：显示x轴标签、数值变化（delta）、初始值（initial）以及最终值
• total：仅显示x轴标签和当前总值
• absolute：显示x轴标签和当前数值

对应的hovertemplate设置如下：

hovertemplate = ifelse(
  data$measure == "relative",
  "%{x}<br>%{delta}<br>Initial: %{initial}<br>Final: %{y}<extra></extra>",
  ifelse(
    data$measure == "total",
    "%{x}<br>%{y}<extra></extra>",
    "%{x}<br>%{y}<extra></extra>"
  )
)

完整修正代码 #

library(plotly)
library(dplyr)

data <- tibble(x = c("Sales", "Consulting", "Net revenue", "New Intermediate",
                     "Purchases", "Other expenses", "Profit before tax"), 
               measure = c("relative", "relative", "total", "absolute", "relative", 
                           "relative", "total"),
               text = c("+60", "+80", "", "", "-40", 
                        "-20", "Total"),
               y = c(60, 80, 0, 120, -40, -20, 0)) %>% 
  mutate(x = factor(x, x))

# 生成动态hovertemplate
custom_hover <- ifelse(
  data$measure == "relative",
  "%{x}<br>%{delta}<br>Initial: %{initial}<br>Final: %{y}<extra></extra>",
  ifelse(
    data$measure == "total",
    "%{x}<br>%{y}<extra></extra>",
    "%{x}<br>%{y}<extra></extra>"
  )
)

(fig <- data %>% 
    plot_ly() %>% 
    add_text(x = factor("Sales", levels = data$x),
             y = 80, text = "Some Text",
             # 给text迹也设置空的extra，避免其悬停显示名称
             hovertemplate = "%{text}<extra></extra>") %>%
    add_trace(type = "waterfall", measure = ~measure,
              x = ~x, textposition = "outside", y= ~y, text =~text,
              connector = list(line = list(color= "rgb(63, 63, 63)")),
              hovertemplate = custom_hover) %>%
    layout(title = "Profit and loss statement 2018",
           xaxis = list(title = ""),
           yaxis = list(title = ""),
           autosize = TRUE,
           showlegend = FALSE))

说明 #

1. 动态模板完美匹配默认悬停的内容逻辑，不会出现冗余文本
2. <extra></extra>标签会隐藏迹名称框，同时保留plotly默认的三角标记样式
3. 额外给add_text的迹也设置了空的hovertemplate，避免其悬停时显示多余名称

内容的提问来源于stack exchange，提问作者thothal