Set default zoom in plotly

5
votes

I have a time series I'm plotting in R using ggplotly to auto-convert my ggplot2 graph plotly. My time series goes back 20 years, but when it's brought up I only want it to display the most recent 4 years of data. I've used

layout(ggplotly_object, xaxis=list(range=c(min_date,max_date)))

This does not appear to even be working to limit the date ranges, which I'm setting using lubridate to subtract 4 years from the maximum date.

I have not found any documentation on changing the default zoom of a plotly plot to a limited range of data while still allowing the user to zoom out and pan to past data. Any tips would be appreciated

2
rplotggplot2plotly
just subset your dataset for the range you want – MLavoie

2 Answers

4
votes

The date axis is in measured in milliseconds, so you need to convert to this unit first. Here's an example:

library(plotly)
library(lubridate)

set.seed(42)

# Dummy data
t1 <- ymd_hms("2006-03-14 12:00:00")
t2 <- ymd_hms("2016-03-14 12:00:00") 
df <- data.frame(t = seq(t1, t2, by = 'week'), 
                  y = rexp(522, rate = 0.25))

# Full plot
p <- plot_ly(df, x = t, y = y, type = 'scatter')
p

# Now zoom. Needs to be the number of milliseconds since 01/01/1970. 
# I'm deliberately using lubridate functions.
min_Date <- ymd_hms("2010-03-14 12:00:00")
min_Date_ms <- interval("1970-01-01 00:00:00", min_Date) / dmilliseconds(1)
max_Date <- ymd_hms("2012-03-14 12:00:00")
max_Date_ms <- interval("1970-01-01 00:00:00", max_Date) / dmilliseconds(1)

p %>% layout(xaxis = list(range = c(min_Date_ms, max_Date_ms)))

There's probably a more elegant way of doing this but it should work.

0
votes

So for range, you should set it to a vector length 2, ie: c(min value,max value).