Formatting a duration as HH:mm:ss in R

Question

Forgive me if this is a duplicate question but I can't seem to type the perfect search to get what I need. I would like to convert seconds(as a numeric) into a duration in the format HH:mm:ss.

Below is a data frame with the original value and the expected value.

df <- data.frame(Duration = c(19, 250, 3, 3600), Expected = c("00:00:19", "00:04:10", "00:00:03", "01:00:00"))

[Convert seconds to days: hours:minutes:seconds](https://stackoverflow.com/questions/27312292/convert-seconds-to-days-hoursminutesseconds) seems relevant. — Henrik, Jan 14 '22 at 08:44
The above question is definitely relevant but I seem to like the updated answers here as well so I will keep this question up. Thanks! — Melissa Salazar, Jan 14 '22 at 20:17

Andre Wildberg · Answer 1 · 2022-01-14T01:38:52.493

This builds a date object with as.POSIXct and adds the duration, then truncates the output to Time with strftime. Resets to 00:00:00 if it counts higher than 24h. For higher numbers see the approach at the bottom.

df
  Duration
1       19
2      250
3        3
4     3600

df$Expected <- strftime(as.POSIXct("00:00:00", format="%H:%M:%S") + 
                        df$Duration, format="%H:%M:%S")

df
  Duration Expected
1       19 00:00:19
2      250 00:04:10
3        3 00:00:03
4     3600 01:00:00

In case you need to count higher than a day use this

df
  Duration
1       19
2      250
3        3
4     3600
5   431170

df$Expected <- paste0(sprintf("%02.f",floor(df$Duration/3600)),":",
                      sprintf("%02.f",(df$Duration/60)%%60),":",
                      sprintf("%02.f",df$Duration%%60))

df
  Duration  Expected
1       19  00:00:19
2      250  00:04:10
3        3  00:00:03
4     3600  01:00:00
5   431170 119:46:10

@BrianMontgomery Yep, thats correct, thx! Added an alternative approach for the OP to choose from! — Andre Wildberg, Jan 14 '22 at 01:05

score 1 · Accepted Answer · answered Jan 14 '22 at 01:18

And another answer using lubridate and hms:

library(lubridate)
library(hms)

df <- data.frame(Duration = c(19, 250, 3, 3600, 86401))
df$Expected <- hms::hms(seconds_to_period(df$Duration))

df
  Duration Expected
1       19 00:00:19
2      250 00:04:10
3        3 00:00:03
4     3600 01:00:00
5    86401 24:00:01

score 1 · Answer 3 · answered Jan 14 '22 at 02:35

1

We can consider duration column as seconds since epoch and use format to extract the time component from it.

A base R approach -

format(as.POSIXct(df$Duration, origin = '1970-01-01', tz = 'UTC'), '%T')
#[1] "00:00:19" "00:04:10" "00:00:03" "01:00:00"

answered Jan 14 '22 at 02:35

Ronak Shah

377,200
20
156
213

This will start over at 86400 seconds = 24 hours – Brian Montgomery Jan 14 '22 at 03:59

Brian Montgomery · Answer 4 · 2022-01-14T22:39:21.303

0

I wrote a little function for this:

hms <- function(s) {
  s <- as.integer(s)
  sprintf('%02i:%02i:%02i', s %/% 3600, s %/% 60 %% 60, s %% 60)
}

edited Jan 14 '22 at 22:39

answered Jan 14 '22 at 04:04

Brian Montgomery

2,244
5
15

Formatting a duration as HH:mm:ss in R

4 Answers4