Adding paths to scatterplot map

A few months ago I wrote code to add overlaid paths to a scatterplot map, and it worked fine. (The paths were unrelated to underlying scatterplot.) I tried running my code again (v 4.7.0) and it’s now broken. Plotly apparently wants the add_paths function to take data arguments with lengths equal to the number of scatterplot points. However, I want to draw superimposed lines that are unrelated to the underlying scatterplot. I’ve modified the “Basic Scatter on Map” example from to demonstrate this issue:

df <- read.csv('')

# geo styling
g <- list(
  scope = 'usa',
  projection = list(type = 'albers usa'),
  showland = TRUE,
  landcolor = toRGB("gray95"),
  subunitcolor = toRGB("gray85"),
  countrycolor = toRGB("gray85"),
  countrywidth = 0.5,
  subunitwidth = 0.5

p <- plot_geo(df, lat = ~lat, lon = ~long) %>%
    text = ~paste(airport, city, state, paste("Arrivals:", cnt), sep = "<br />"),
    color = ~cnt, symbol = I("square"), size = I(8), hoverinfo = "text"
  ) %>%
  colorbar(title = "Incoming flights<br />February 2011") %>%
    title = 'Most trafficked US airports<br />(Hover for airport)', geo = g

p # works fine

# latitude and longitude data for drawing line 
line.lng = seq(-116,-55,length.out = 25) = seq(56,48,length.out = 25)

p %>% add_paths(x=~line.lng,,
# throws error, saying that the lat/lon data supplied to 
# add_paths has to be the same length as the number of data points
# on the underlying scatterplot - this did not happen in an earlier version

Any suggestions would be greatly appreciated.