Coerce all dates to the same day, preserving time of day.

Sets the date of every element in times to 2000-01-01 while keeping the time-of-day unchanged. This lets you overlay recordings from different calendar days on a shared x axis using ggplot2::scale_x_datetime and label_hour. The arbitrary date 2000-01-01 was chosen because it is far enough from DST transition dates to avoid edge cases in most time zones.

Usage

commontime(times, tz = Sys.timezone())

Arguments

times

A POSIXct vector.

tz

Time zone string (e.g. 'America/New_York'). Defaults to your system time zone. Should match the tz argument passed to label_hour.

Value

A POSIXct vector of the same length as times, with the date component set to 2000-01-01.

Details

Time zone tip: pass the same tz here as you pass to label_hour to ensure the hour labels match the data.

See also

label_hour for the matching x-axis label formatter, time_of_day for a numeric alternative suitable for modelling.

Examples

times <- as.POSIXct(
  c('2023-08-09 06:15:00', '2024-07-27 14:30:00'),
  tz = 'America/New_York'
)

# Both dates become 2000-01-01; times of day are preserved
commontime(times, tz = 'America/New_York')
#> [1] "2000-01-01 06:15:00 EST" "2000-01-01 14:30:00 EST"

# Typical ggplot2 usage (not run here):
# ggplot(df, aes(x = commontime(bin_datetime, tz = 'America/New_York'),
#               y = detectionrate_ins_buzz, color = flower)) +
#   geom_line() +
#   scale_x_datetime(labels = label_hour(tz = 'America/New_York'))