Add vertical line to ggplotly plot

Question

I am trying to build a plot combining ggplot2 and plotly. The two vertical lines appear on pure ggplot2, but once I call plotly::ggplotly on it, they go away. How can I make the data also appear on the ggplotly version? If you have a solution using only plot_ly, that would be fine too.

The data:

df <- structure(list(date = structure(c(17226, 17257, 17287, 17318, 
17348, 17379, 17410, 17440, 17471, 17501, 17226, 17257, 17287, 
17318, 17348, 17379, 17410, 17440, 17471, 17501, 17226, 17257, 
17287, 17318, 17348, 17379, 17410, 17440, 17471, 17501), class = "Date"), 
    n = c(253L, 217L, 257L, 166L, 121L, 56L, 68L, 62L, 142L, 
    20L, 174L, 228L, 180L, 158L, 80L, 39L, 47L, 54L, 107L, 12L, 
    93L, 74L, 47L, 49L, 55L, 16L, 52L, 53L, 32L, 3L), act = c("a", 
    "a", "a", "a", "a", "a", "a", "a", "a", "a", "b", "b", "b", 
    "b", "b", "b", "b", "b", "b", "b", "c", "c", "c", "c", "c", 
    "c", "c", "c", "c", "c")), class = "data.frame", row.names = c(NA, 
-30L), .Names = c("date", "n", "act"))

facts_timeline <- structure(list(Date = structure(c(17507, 17293), class = "Date"), 
    ShortDescription = c("Marketing Campaign", "Relevant Fact 1"
    )), row.names = c(NA, -2L), class = c("tbl_df", "tbl", "data.frame"
), spec = structure(list(cols = structure(list(Date = structure(list(
    format = ""), .Names = "format", class = c("collector_date", 
"collector")), Tenant = structure(list(), class = c("collector_character", 
"collector")), ShortDescription = structure(list(), class = c("collector_character", 
"collector")), LongDescription = structure(list(), class = c("collector_character", 
"collector"))), .Names = c("Date", "Tenant", "ShortDescription", 
"LongDescription")), default = structure(list(), class = c("collector_guess", 
"collector"))), .Names = c("cols", "default"), class = "col_spec"), .Names = c("Date", 
"ShortDescription"))

The code to make the plot:

p <- df %>% 
  ggplot(aes(date, n, group = act, color = act)) + 
  geom_line() + 
  geom_vline(data = facts_timeline, aes(xintercept = Date))

Here you can see the two vertical lines:

p

But not here:

ggplotly(p)

Answer 1

Simply set the xintercept to numeric, and everything will work.

p <- df %>% 
  ggplot(aes(date, n, group = act, color = act)) + 
  geom_line() + 
  geom_vline(data = facts_timeline, aes(xintercept = as.numeric(Date)))

p

ggplotly(p)

Answer 2

It is not directly possible to plot vertical lines in plotly, but here is my workaround:

vline_list <- list()
for(i in 1:nrow(facts_timeline)){ 
  vline_list[[i]] <- 
    list(type      = "line",
         fillcolor = line_color, 
         line      = list("black"),
         opacity   = 0.3,
         x0        = facts_timeline$Date[i],
         x1        = facts_timeline$Date[i],
         xref      = "x",
         y0        = 0, 
         y1        = max(df$n),
         yref      = "y")
}

plot_ly(x = ~df$date, y = ~df$n,color = df$act, mode = 'lines') %>% 
  layout(shapes  = vline_list)

With the for loop we iterate through all rows in facts_timeline and create a new line. Tis line has no length of infinity as in the `ggplot. In my example the line is the maximum of the y -axis. You can change this to own needs.

Answer 3

plot_ly(df, 
        x = ~ date, 
        y = ~ n, 
        color = ~act, 
        text = ~act,
        mode = "lines", 
        type = "scatter",
        hoverinfo = "x+y+text") %>% 
    layout(hovermode = "closest",
        xaxis=list(range=c("2017-03-01", "2018-01-01"))) %>%
    add_lines(x=rep(facts_timeline[["Date"]][[1]], 2), 
        y=c(0, 300),
        name=facts_timeline[["ShortDescription"]][[1]],
        inherit=FALSE, 
        hoverinfo = "name",
        line = list(color="#000000")) %>% 
    add_lines(x=rep(facts_timeline[["Date"]][[2]], 2), 
        y=c(0, 300), 
        name=facts_timeline[["ShortDescription"]][[1]],
        inherit=FALSE, 
        hoverinfo = "name",
        line = list(color="#000000"))