how to filter a timestamp by time of day?

[von_olaf]

Hello there,

This is an old issue, but surprisingly the last version of lubridate does not seem to handle this very well.
Consider this simple example

dataframe <- tibble(gmt_time = c('2016-07-08 04:30:10.690'), value = c(1))

library(hms)
library(lubridate)

dataframe %>% 
  mutate(gmt_time = ymd_hms(gmt_time),
         est_time = with_tz(gmt_time, 'America/New_York'),
         myhour1 = as.hms(est_time),
         myhour2 = as.hms(gmt_time))

# A tibble: 1 x 5
  gmt_time                value est_time                myhour1       myhour2      
  <dttm>                  <dbl> <dttm>                  <time>        <time>       
1 2016-07-08 04:30:10.690     1 2016-07-08 00:30:10.690 30'10.690000" 30'10.690000"

What is going on here? I would like to extract the time component of the timestamp, so that I can run filters like time > 04:30:00

Thanks!

[andresrcs]

as.hms() is not a lubridate function but a hms function, when you use as.hms() this way, you have to specify the time zone, like this.

library(dplyr)
library(hms)
library(lubridate)

dataframe <- tibble(gmt_time = c('2016-07-08 04:30:10.690'), value = c(1))

dataframe %>% 
    mutate(gmt_time = ymd_hms(gmt_time),
           est_time = with_tz(gmt_time, 'America/New_York'),
           myhour1 = as.hms(est_time, tz = 'UTC'),
           myhour2 = as.hms(gmt_time, tz = 'America/New_York'))
#> # A tibble: 1 x 5
#>   gmt_time            value est_time            myhour1      myhour2      
#>   <dttm>              <dbl> <dttm>              <time>       <time>       
#> 1 2016-07-08 04:30:10     1 2016-07-08 00:30:10 04:30.690000 30'10.690000"

^{Created on 2019-01-18 by the reprex package (v0.2.1)}

[von_olaf]

Thanks! that is very nice. Its a lot of tz repetition but that seems to be working correctly

time_str = c('2013-01-03 21:59:21.549', '2013-01-04 22:00:21.549', '2013-01-05 22:01:21.222', '2013-01-06 22:06:23.559' )
value = c(1,2,3,4)

data <- tibble(time_str, value)
data %>%  mutate(timestamp_utc = ymd_hms(time_str, tz = 'UTC'),
                 timestamp_est = with_tz(timestamp_utc, 'America/New_York'),
                        time_est = as.hms(timestamp_est, tz = 'America/New_York')) %>% 
  filter(time_est >= hms::as.hms('16:59:00', tz = 'America/New_York'),
         time_est <= hms::as.hms('17:01:00', tz = 'America/New_York'))

gives

# A tibble: 2 x 5
  time_str                value timestamp_utc           timestamp_est           time_est 
  <chr>                   <dbl> <dttm>                  <dttm>                  <time>   
1 2013-01-03 21:59:21.549     1 2013-01-03 21:59:21.549 2013-01-03 16:59:21.549 16:59.549
2 2013-01-04 22:00:21.549     2 2013-01-04 22:00:21.549 2013-01-04 17:00:21.549 17:00.549

[andresrcs]

You can set the defaul time zone with this pkgconfig::set_config("hms::default_tz", "UTC")