Looping through date in R loses format

Question

This has been frustrating me. Even with lubridate I can't get dates to maintain their type when I loop over them. For example:

require(lubridate)
yearrange = ymd(20110101) + years(seq(4))
yearrange
#[1] "2012-01-01 UTC" "2013-01-01 UTC" "2014-01-01 UTC" "2015-01-01 UTC"
class(yearrange)
#[1] "POSIXct" "POSIXt" 

However, if I try to loop through years (creating a separate plot for each year in my data set): I lose the formatting of the year, and would have to re-cast the data

for (yr in yearrange) { show(yr) }
#[1] 1325376000
#[1] 1356998400
#[1] 1388534400
#[1] 1420070400

If I loop though specifying indices, I get date objects back:

for (i in seq(length(yearrange))) { show(yearrange[i]) }
#[1] "2012-01-01 UTC"
#[1] "2013-01-01 UTC"
#[1] "2014-01-01 UTC"
#[1] "2015-01-01 UTC"

Is there an easy way to avoid the indexed option, without using foreach, or is that the only way?

Answer 1

Try this

for (yr in as.list(yearrange))  { show(yr) }

I think for (yr in yearrange) coerces yearrange into a vector and POSIXct is not one of the supported types that vector coerces into.

Answer 2

lapply doesn't seem to have the same problem, e.g.:

for (x in yearrange) plot(1, main=x)
#Main title = 1420070400
lapply( yearrange, function(x) plot(1, main=x) )
#Main title = 2015-01-01