[R] Preserving the class of POSIXt objects

From: Kenn Konstabel <lebatsnok_at_gmail.com>
Date: Fri, 25 Mar 2011 21:14:50 +0000


Dear all,

I am working with a list of objects each of which contains two POSIXct objects (say, $Start and $End) and a number of different data in addition to that. Now an easy way to extract Start times of all object could be sapply(x, "[", "Start") but this converts them all to numeric, and so does sapply(x, "[[", "Start"). lapply preserves the class but is obviously a list rather than an atomic vector but using unlist is a way of getting rid of the class again. I know the vector can be converted back using as.POSIXct and that one can set the class attribute back to "POSIXct" but is there an easier trick that I can't just figure out? Thanks in advance for any suggestions.

A reproducible example:

foo <- as.list(Sys.time()+1:10)
sapply(foo, "[", 1)

I've also noticed that trunc.POSIXt does something unexpected (to me) in some cases:

foo <- as.POSIXct("2011-03-25 21:03:20")+1:10 length(foo) ## obviously 10
length(trunc(foo, "day")) # 1!

This makes it impossible to use [ and length() with trunc'ed POSIXct vectors (and it is a little unexpected too that the result of using trunc with a POSIX_ct_ object is a POSIX_lt_ object) but the cure seems easy: in the code of [.POSIXt,

##               old           ##  new
            x$sec <- 0      ##  x$sec[] <-  0
            x$min <- 0L    ##   x$min[] <- 0L
            x$hour <- 0L   ##   x$hour[] <- 0L

After these changes, [ and length would work "correctly" (in my possibly naive understanding), and length(trunc(foo, "day")) would return 10 in the example above.

Best regards,

Kenn

Kenn Konstabel
National Institute for Health Development Hiiu 42
Tallinn
Estonia



R-help_at_r-project.org mailing list
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. Received on Fri 25 Mar 2011 - 21:18:01 GMT

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.2.0, at Fri 25 Mar 2011 - 23:50:24 GMT.

Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help. Please read the posting guide before posting to the list.

list of date sections of archive