r - Lubridate as_date and. as_datetime differences in behavior

Question

I have a vector of numerics representing the number of milliseconds since January 1, 1970. I would like to convert these into a date time object using lubridate. A sample of data is below:

raw_times <- c(1139689917479, 1139667123031, 1140364113915, 1140364951003, 
               1139643685434, 1139677091970, 1139691963511, 1140339448413, 1140368308429, 
               1139686613641, 1139666081813, 1140351488730, 1140346617958, 1141933663183, 
               1141933207579, 1140360125149, 1140351845108, 1140365079103, 1141933549825, 
               1140365601476)

Knowing that the documentation for as_date and as_datetime indicate they take a numeric vector representing the number of days since January 1, 1970 , I tried the following:

library(lubridate)

as_date(raw_times / (1000 * 60 * 60 * 24))
"2006-02-11" "2006-02-11" "2006-02-19" "2006-02-19" "2006-02-11" 
"2006-02-11" "2006-02-11" "2006-02-19" "2006-02-19" "2006-02-11" 
"2006-02-11" "2006-02-19" "2006-02-19" "2006-03-09" "2006-03-09"
"2006-02-19" "2006-02-19" "2006-02-19" "2006-03-09" "2006-02-19"

(Obviously using the fact that there are 1000ms in a second, 60 seconds in a minute, 60 minutes in an hour, 24 hours in a day.)

When I run the same the code with as_datetime, I get the following:

as_datetime(raw_times / (1000 * 60 * 60 * 24))
"1970-01-01 03:39:50 UTC" "1970-01-01 03:39:50 UTC" "1970-01-01 03:39:58 UTC" "1970-01-01 03:39:58 UTC" "1970-01-01 03:39:50 UTC" "1970-01-01 03:39:50 UTC"
"1970-01-01 03:39:50 UTC" "1970-01-01 03:39:58 UTC" "1970-01-01 03:39:58 UTC" "1970-01-01 03:39:50 UTC" "1970-01-01 03:39:50 UTC" "1970-01-01 03:39:58 UTC"
"1970-01-01 03:39:58 UTC" "1970-01-01 03:40:16 UTC" "1970-01-01 03:40:16 UTC" "1970-01-01 03:39:58 UTC" "1970-01-01 03:39:58 UTC" "1970-01-01 03:39:58 UTC"
"1970-01-01 03:40:16 UTC" "1970-01-01 03:39:58 UTC"

The results are different. I would assume there is some other argument which I am missing, but I can't find anything in the documentation which would tell me what that would be.

Session information below:

> sessionInfo()
R version 3.3.2 (2016-10-31)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows 7 x64 (build 7601) Service Pack 1

locale:
[1] LC_COLLATE=English_United States.1252  LC_CTYPE=English_United States.1252    LC_MONETARY=English_United States.1252 LC_NUMERIC=C                          
[5] LC_TIME=English_United States.1252    

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
[1] lubridate_1.6.0

loaded via a namespace (and not attached):
[1] magrittr_1.5  tools_3.3.2   stringi_1.1.2 stringr_1.1.0

Answer 1

Not a (package name redacted) solution, but you can do this with base::.POSIXct:

R> options(digits.secs=3)
R> .POSIXct(raw_times/1000)
 [1] "2006-02-11 14:31:57.479 CST" "2006-02-11 08:12:03.030 CST"
 [3] "2006-02-19 09:48:33.914 CST" "2006-02-19 10:02:31.003 CST"
 [5] "2006-02-11 01:41:25.434 CST" "2006-02-11 10:58:11.970 CST"
 [7] "2006-02-11 15:06:03.510 CST" "2006-02-19 02:57:28.413 CST"
 [9] "2006-02-19 10:58:28.428 CST" "2006-02-11 13:36:53.641 CST"
[11] "2006-02-11 07:54:41.812 CST" "2006-02-19 06:18:08.730 CST"
[13] "2006-02-19 04:56:57.957 CST" "2006-03-09 13:47:43.183 CST"
[15] "2006-03-09 13:40:07.578 CST" "2006-02-19 08:42:05.148 CST"
[17] "2006-02-19 06:24:05.108 CST" "2006-02-19 10:04:39.102 CST"
[19] "2006-03-09 13:45:49.825 CST" "2006-02-19 10:13:21.476 CST"