Bioequivalence and Bioavailability Forum • Some code

Some code [Software]

posted by Helmut – Vienna, Austria, 2019-03-22 16:58 – Posting: # 20068
Views: 1,770

Hi Dr Anonymous :-D

» I agree, patience is the key :), I use R studio predominantly and have a GitHub account (need to be more active though).

So you are an advanced RUser!

» » ....This is the number of seconds since midnight, January 1, 1960…
»
» Woah, thanks for the explanation.

Welcome. Didn’t know it until yesterday either.

» […] I believed ISO8601 was majorly the standard time format

In many countries. In Austria it is not just a norm but legally binding. Rarely used in daily life.

» Thanks for working it out, I tried replicating this and this looks promising for me to try it out further.

I was curious myself. My latest code:

library(lubridate) # Makes job easier but has large footprint. See also:

# https://blog.revolutionanalytics.com/2009/06/converting-time-zones.html

# https://www.gormanalysis.com/blog/dates-and-times-in-r-without-losing-your-sanity/

# https://stackoverflow.com/questions/51236962/how-to-format-a-difftime-object-to-a-string-with-hhmmss

# For a list of abbreviated locations / time zones: OlsonNames()

# or https://en.wikipedia.org/wiki/List_of_tz_database_time_zones

# I recommend to give the location; if applicable it will observe

# local daylight time switches. In case you are lost, retrieve the location

# R is running in: Sys.timezone()

lct       <- Sys.getlocale("LC_TIME") # Save original locale.

invisible(Sys.setlocale("LC_TIME", "C"))

my_tz     <- 'Europe/Vienna'

my_time   <- now(tzone=my_tz)

IBM_tz    <- 'US/Eastern'

IBM_time  <- with_tz(my_time, IBM_tz)

your_tz   <- 'Asia/Kolkata'

your_time <- with_tz(my_time, your_tz)

NZ_tz     <- 'Pacific/Auckland'

NZ_time   <- with_tz(my_time, NZ_tz)

tz        <- c(my_tz, IBM_tz, your_tz, NZ_tz)

dt        <- c(as.character(my_time), as.character(IBM_time),

               as.character(your_time), as.character(NZ_time))

UTC.dt    <- offset <- rep(NA, length(dt))

for (j in seq_along(dt)) { # Cannot be vectorized!

  dt[j]     <- format(as.POSIXct(dt[j], tz=tz[j]), tz=tz[j], usetz=TRUE)

  UTC.dt[j] <- format(as.POSIXct(dt[j], tz=tz[j]), tz='UTC', usetz=TRUE)

  x         <- as.numeric(difftime(dt[j], UTC.dt[j], units="hours"))

  offset[j] <- sprintf("%+02d:%02d", x %/% 1, (x - x %/% 1)*60)

}

df <- data.frame(location=tz, local.datetime=dt, offset, UTC.datetime=UTC.dt,

                 loc.SAS.datetime=format(as.POSIXct(strptime(dt,

                  format='%F %T')), format='%d%b%y:%H:%M:%S'),

                 UTC.SAS.datetime=format(as.POSIXct(strptime(UTC.dt,

                   format='%F %T')), format='%d%b%y:%H:%M:%S'),

                 row.names=NULL, stringsAsFactors=FALSE)

invisible(Sys.setlocale("LC_TIME", lct)) # Restore original locale.

print(df, row.names=FALSE)               # Here we are.

Which gave on my machine:

         location           local.datetime offset            UTC.datetime loc.SAS.datetime UTC.SAS.datetime

    Europe/Vienna  2019-03-22 18:32:57 CET  +1:00 2019-03-22 17:32:57 UTC 22Mar19:18:32:57 22Mar19:17:32:57

       US/Eastern  2019-03-22 13:32:57 EDT  -4:00 2019-03-22 17:32:57 UTC 22Mar19:13:32:57 22Mar19:17:32:57

     Asia/Kolkata  2019-03-22 23:02:57 IST  +5:30 2019-03-22 17:32:57 UTC 22Mar19:23:02:57 22Mar19:17:32:57

 Pacific/Auckland 2019-03-23 06:32:57 NZDT +13:00 2019-03-22 17:32:57 UTC 23Mar19:06:32:57 22Mar19:17:32:57

Note the different local times: Poughkeepsie changed EST ↑ EDT on March 10^th, Austria will change CET ↑ CEST on March 31^th, and New Zealand will change NZDT ↓ NZT on April 7^th.
See this funny story.

—
Cheers,
Helmut Schütz

The quality of responses received is directly proportional to the quality of the question asked. ☼
Science Quotes

Complete thread:

CDISC datasets or R. Acceptable ?? WhiteCoatWriter 2019-03-21 07:13 [Software]
- CDISC datasets from R. Why not? Helmut 2019-03-21 10:47
  - CDISC datasets from R. Why not? WhiteCoatWriter 2019-03-21 12:13
    - CDISC datasets from R. Why not? Helmut 2019-03-21 16:43
      - CDISC datasets from R. Why not? WhiteCoatWriter 2019-03-22 06:56
        
        Some code Helmut 2019-03-22 16:58
        
        Some code WhiteCoatWriter 2019-03-24 07:59
        
        SAS only? Helmut 2019-03-24 12:34
        
        SAS only? WhiteCoatWriter 2019-03-25 09:39
        
        Confusing… Helmut 2019-03-25 13:44
        
        Confusing… WhiteCoatWriter 2019-03-27 06:55