如何解决R语言Plotly中线图坐标轴零线消失问题？

问题 #

在R语言中使用Plotly绘制折线图，数据包含NA值，已实现折线从非NA值处开始绘制，但出现两个问题：

• x轴零线消失，即使设置zeroline = TRUE也无法显示；
• 尝试添加x为月份范围、y=0的轨迹补全零线，给该轨迹设置hovermode = 'none'后，悬停一月或十二月时仍会显示“trace 0 : 0 hL”的提示信息。

附实现代码：

library(data.table)
library(plotly)

dataBeverage <- data.table(
    month = c(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12),
    volume = c(NA, NA, NA, 3000, 2450, 3200, 4678, 2435, 1342, 1731, 2340, 3500))

p <- plot_ly(data = dataBeverage, x = ~month, y = ~volume, type = 'scatter', mode = 'lines', line = list(color='#17BECF', shape = 'spline'))%>%
    layout(
        hovermode = 'x unified', separators = ', ',
        xaxis = list(
            range = c(min(dataBeverage$month),max(dataBeverage$month)),
            title = '',
            tickangle=-40,
            ticktext = list('January', 'February', 'March', 'April', 'May', 'June', 'July', 'August', 'September', 'Octobre', 'November', 'December'),
            tickvals = list(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12),
            showgrid = FALSE
        ),
        yaxis = list(
            title = list(text = 'Montly volume', standoff = 15L),
            ticksuffix = ' hL',
            tickformat = ',d'
        ))
p

解决方案 #

1. 恢复x轴零线 #

Plotly会自动根据y轴数据范围调整可视区域，由于你的数据y值全为正数且最小值远大于0，默认不会显示y=0的零线。只需手动指定y轴范围包含0，并开启zeroline即可：
在layout的yaxis参数中添加以下配置：

yaxis = list(
    title = list(text = '月度销量', standoff = 15L),
    ticksuffix = ' hL',
    tickformat = ',d',
    zeroline = TRUE,  # 开启零线显示
    range = c(0, max(dataBeverage$volume, na.rm = TRUE))  # 强制y轴从0开始
)

无需额外添加y=0的轨迹，就能直接显示x轴对应的零线。

2. 解决额外轨迹的悬停提示问题 #

如果必须通过添加轨迹补全零线，注意单个轨迹的悬停控制参数是hoverinfo = 'none'，而非全局布局的hovermode。同时可关闭该轨迹的图例显示：

# 添加零线轨迹
p <- p %>% add_trace(
    x = ~month, y = 0, type = 'scatter', mode = 'lines',
    line = list(color = '#000000'),
    hoverinfo = 'none',  # 关闭该轨迹的悬停提示
    showlegend = FALSE  # 隐藏图例项
)

修改后的完整代码 #

library(data.table)
library(plotly)

dataBeverage <- data.table(
    month = c(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12),
    volume = c(NA, NA, NA, 3000, 2450, 3200, 4678, 2435, 1342, 1731, 2340, 3500))

p <- plot_ly(data = dataBeverage, x = ~month, y = ~volume, type = 'scatter', mode = 'lines', line = list(color='#17BECF', shape = 'spline')) %>%
    layout(
        hovermode = 'x unified', separators = ', ',
        xaxis = list(
            range = c(min(dataBeverage$month),max(dataBeverage$month)),
            title = '',
            tickangle=-40,
            ticktext = list('一月', '二月', '三月', '四月', '五月', '六月', '七月', '八月', '九月', '十月', '十一月', '十二月'),
            tickvals = list(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12),
            showgrid = FALSE
        ),
        yaxis = list(
            title = list(text = '月度销量', standoff = 15L),
            ticksuffix = ' hL',
            tickformat = ',d',
            zeroline = TRUE,
            range = c(0, max(dataBeverage$volume, na.rm = TRUE))
        ))
p

内容的提问来源于stack exchange，提问作者ju0325