c(a, b) for POSIXct objects with tzone attributes?
Hi, Prof. Ripley, Bert:
Thanks for the comments.
Might there be a function similar to "c" but retains more
attributes than just names?
Before I write such, I felt a need to ask if anyone knows where
it may already have been done -- and if people have suggestions for how
they think it should function with particular classes of objects? For
example, if create an S3 generic "c2" for this, "c2.default" might
retains all the attributes that make sense from the first argument and
ignore attributes from later arguments if it makes sense to do so. This
would then retain the tzone attribute from the first of multiple POSIXct
arguments.
Thanks,
Spencer
On 7/9/2012 6:24 AM, Prof Brian Ripley wrote:
On 09/07/2012 07:12, Bert Gunter wrote:
Spencer: Just for fun, may I hazard a guess: It would be messy to retain time zones if your were concatenating objects with more than one time zone among them.
Actually, impossible as the design allows for only one timezone for each object.
"Normalizing" everything to a single zone probably also makes subsequent operations on the results much easier.
We considered having c() retain the timezone if it was common to all
the objects, but the main issue was that c() was documented to remove
attributes:
?c? is sometimes used for its side effect of removing attributes
except names, for example to turn an array into a vector.
?as.vector? is a more intuitive way to do this, but also drops
names. Note too that methods other than the default are not
required to do this (and they will almost certainly preserve a
class attribute).
So, sometimes removing and sometimes retaining attributes was going to
be confusing.
But in any case, the documentation (?c.POSIXct) is clear:
Using ?c? on ?"POSIXlt"? objects converts them to the current time
zone, and on ?"POSIXct"? objects drops any ?"tzone"? attributes
(even if they are all marked with the same time zone).
So the recommended way is to add a "tzone" attribute if you know what
you want it to be. POSIXct objects are absolute times: the timezone
merely affects how they are converted (including to character for
printing).
Not that it couldn't be done -- and maybe already has been in some package. If so, I would bet it's done via S4 classes, which would seem to be a natural way to go. Hopefully, you'll get a more informed and authoritative explanation, though. ... I'm just speculating and and may be all wet. Cheers, Bert On Sun, Jul 8, 2012 at 9:46 PM, Spencer Graves < spencer.graves at structuremonitoring.com> wrote:
Hello:
What is the recommended method for retaining the tzone
attributes
when concatonating POSIXct objects?
(d1 <- ISOdate(1970,1,1)) # Sets the tzone attribute = GMT
[1] "1970-01-01 12:00:00 GMT"
(d1.2 <- c(d1, d1)) # c(..) strips the tzone attribute, displays in
the time zone of the operating system [1] "1970-01-01 04:00:00 PST" "1970-01-01 04:00:00 PST"
attr(d1.2, 'tzone') <- 'GMT' d1.2 # tzone attribute manually restored
[1] "1970-01-01 12:00:00 GMT" "1970-01-01 12:00:00 GMT"
"c" is a generic function with a method defined for objects
of class
POSIXct, so the results here were what is returned by c.POSIXct [as
discussed with help('c.POSIXct')].
Is there some other function like "c2", say, that tries to
retain
attributes where "c" strips all but names?
Thanks,
Spencer
______________________________**________________ R-help at r-project.org mailing list https://stat.ethz.ch/mailman/**listinfo/r-help<https://stat.ethz.ch/mailman/listinfo/r-help> PLEASE do read the posting guide http://www.R-project.org/** posting-guide.html <http://www.R-project.org/posting-guide.html> and provide commented, minimal, self-contained, reproducible code.