R help - Mar 2017 - lubridate inserting unwelcome 9:21

Dear friends - I have a series of times on successive days and would 
like to convert them into a successive common time for each person (ID) 
. Using lubridate and adding days(1) does as expected apart from 
changing time zone to LMT from UTC and suddenly adding 9:21 (H:M) to all 
times. Individual parts of the instructions seem OK. I'm sorry for the 
clumsy demonstration - but the error comes through.

R version 3.3.2 (2016-10-31) - Windows.

All best wishes
Troels

library(lubridate)

SSS <- structure(list(ID = c(1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 3L,
3L, 3L, 3L, 3L, 3L, 3L), Time = c(-1L, 0L, 1L, 2L, 3L, 4L, 5L,
6L, 7L, 8L, 9L, 10L, 11L, -1L, 0L, 1L, 2L, 3L, 4L, 5L, 6L, 7L,
8L, 9L, 10L, -1L, 0L, 1L, 2L, 3L, 4L, 5L), VT = structure(c(3L,
5L, 11L, 11L, 8L, 10L, 8L, 10L, 8L, 7L, 11L, 5L, 5L, 12L, 5L,
7L, 7L, 5L, 5L, 8L, 10L, 8L, 7L, 7L, 7L, 2L, 8L, 7L, 8L, 7L,
10L, 8L), .Label = c("02:00", "03:00", "04:00",
"05:00", "06:00",
"09:00", "10:00", "11:00", "11:30",
"12:00", "13:00", "14:00",
"15:00", "17:00", "18:00", "21:30",
"23:00"), class = "factor")), .Names
= c("ID",
"Time", "VT"), row.names = c(NA, 32L), class =
"data.frame")

SSS$VT  <- parse_date_time(SSS$VT,"HM")
str(SSS$VT)

TT <- list()
for(i in 1:3) {
#i <- 1
BS <- subset(SSS,ID==i)
TT[[i]] <- c(BS$VT + (1:length(BS[,1])-1)*days(1))
}
BS$VT
(1:length(BS[,1])-1)*days(1)

#these appear as expected but

TT

#appears disturbed 9:21 inserted - LMT time zone - how comes?

Analyzing time data as POSIXct without dates is IMO unwise (likely to give you
trouble). My approach is to always keep the date and time together or to use
numeric hours-after-midnight. Others have invented packages like chron to deal
with such data.
-- 
Sent from my phone. Please excuse my brevity.

On March 28, 2017 8:28:30 AM PDT, Troels Ring <tring at gvdnet.dk>
wrote:
>Dear friends - I have a series of times on successive days and would 
>like to convert them into a successive common time for each person (ID)
>
>. Using lubridate and adding days(1) does as expected apart from 
>changing time zone to LMT from UTC and suddenly adding 9:21 (H:M) to
>all 
>times. Individual parts of the instructions seem OK. I'm sorry for the 
>clumsy demonstration - but the error comes through.
>
>R version 3.3.2 (2016-10-31) - Windows.
>
>All best wishes
>Troels
>
>library(lubridate)
>
>SSS <- structure(list(ID = c(1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
>1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 3L,
>3L, 3L, 3L, 3L, 3L, 3L), Time = c(-1L, 0L, 1L, 2L, 3L, 4L, 5L,
>6L, 7L, 8L, 9L, 10L, 11L, -1L, 0L, 1L, 2L, 3L, 4L, 5L, 6L, 7L,
>8L, 9L, 10L, -1L, 0L, 1L, 2L, 3L, 4L, 5L), VT = structure(c(3L,
>5L, 11L, 11L, 8L, 10L, 8L, 10L, 8L, 7L, 11L, 5L, 5L, 12L, 5L,
>7L, 7L, 5L, 5L, 8L, 10L, 8L, 7L, 7L, 7L, 2L, 8L, 7L, 8L, 7L,
>10L, 8L), .Label = c("02:00", "03:00",
"04:00", "05:00", "06:00",
>"09:00", "10:00", "11:00", "11:30",
"12:00", "13:00", "14:00",
>"15:00", "17:00", "18:00", "21:30",
"23:00"), class = "factor")),
>.Names 
>= c("ID",
>"Time", "VT"), row.names = c(NA, 32L), class =
"data.frame")
>
>SSS$VT  <- parse_date_time(SSS$VT,"HM")
>str(SSS$VT)
>
>TT <- list()
>for(i in 1:3) {
>#i <- 1
>BS <- subset(SSS,ID==i)
>TT[[i]] <- c(BS$VT + (1:length(BS[,1])-1)*days(1))
>}
>BS$VT
>(1:length(BS[,1])-1)*days(1)
>
>#these appear as expected but
>
>TT
>
>#appears disturbed 9:21 inserted - LMT time zone - how comes?
>
>______________________________________________
>R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
>https://stat.ethz.ch/mailman/listinfo/r-help
>PLEASE do read the posting guide
>http://www.R-project.org/posting-guide.html
>and provide commented, minimal, self-contained, reproducible code.

Thanks a lot - I got interested to make it work since just formatting  
in lubricate as a time  automatically puts the same complete date, which 
was OK for me - and also provided a utility for adding days - which 
apparently then needs tweaking -

Best wishes

Troels


Den 28-03-2017 kl. 20:38 skrev Jeff Newmiller:
> Analyzing time data as POSIXct without dates is IMO unwise (likely to give
you trouble). My approach is to always keep the date and time together or to use
numeric hours-after-midnight. Others have invented packages like chron to deal
with such data.