R devel - Jan 2004 - strange behaviour when converting from char to POSIX (PR#6422)

Full_Name: Christoph Schmutz, MeteoSchweiz, Switzerland
Version: R1.7.1, R1.8.1
OS: windows2000, solaris sunOS 5.8
Submission from: (NULL) (141.249.133.6)



I'm not sure if I don't get the clue, but please consider
this:
> strptime("19930870150","%Y%j%H%M")
[1] "1993-03-28 01:50:00"
> strptime("19930870250","%Y%j%H%M")
[1] "1993-03-28 01:50:00"
> strptime("19930870350","%Y%j%H%M")
[1] "1993-03-28 03:50:00"

The behaviour is the same with R1.7.1 on Window2000 as on a R1.8.1 on sunOS
5.8.

Thanks for the reply, Christoph

What exactly is strange, apart from posting a question as a bug report?
Please do read the R FAQ and don't misuse R-bugs to ask questions.

Hint: you most likely specified a non-existent time.  When did your locale
switch to DST in 1993?  I don't know, of course (you haven't told us
your
locale) but I bet it was that day.  That means the time you specified does
not exist if your locale switched at the same time as the UK (and I bet it
did).

Here's my evidence (WinXP)
> as.POSIXct(strptime("19930870001","%Y%j%H%M")) ->
foo
> foo
[1] "1993-03-28 00:01:00 GMT Standard Time"
> foo+3480
[1] "1993-03-28 00:59:00 GMT Standard Time"
> foo+3540
[1] "1993-03-28 02:00:00 GMT Daylight Time"


On Fri, 9 Jan 2004 christoph.schmutz@meteoschweiz.ch wrote:
> Full_Name: Christoph Schmutz, MeteoSchweiz, Switzerland
> Version: R1.7.1, R1.8.1
> OS: windows2000, solaris sunOS 5.8
> Submission from: (NULL) (141.249.133.6)
>
>
>
> I'm not sure if I don't get the clue, but please consider this:
> > strptime("19930870150","%Y%j%H%M")
> [1] "1993-03-28 01:50:00"
> > strptime("19930870250","%Y%j%H%M")
> [1] "1993-03-28 01:50:00"
> > strptime("19930870350","%Y%j%H%M")
> [1] "1993-03-28 03:50:00"
>
> The behaviour is the same with R1.7.1 on Window2000 as on a R1.8.1 on sunOS
> 5.8.
So it is probably intended, then!

-- 
Brian D. Ripley,                  ripley@stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272860 (secr)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595

On Fri, Jan 09, 2004 at 06:01:27PM +0100, christoph.schmutz@meteoschweiz.ch
wrote:
> Full_Name: Christoph Schmutz, MeteoSchweiz, Switzerland
> Version: R1.7.1, R1.8.1
> OS: windows2000, solaris sunOS 5.8
> Submission from: (NULL) (141.249.133.6)
> 
> 
> 
> I'm not sure if I don't get the clue, but please consider this:
> > strptime("19930870150","%Y%j%H%M")
> [1] "1993-03-28 01:50:00"
> > strptime("19930870250","%Y%j%H%M")
> [1] "1993-03-28 01:50:00"
> > strptime("19930870350","%Y%j%H%M")
> [1] "1993-03-28 03:50:00"
You are presumably hitting the switch from dayligh-saving to regular time in
that year.

I was just mucking about with that, but R (1.9.0 as of Jan 8, 2004, on
Debian unstable) still crashes reliably even when I explicitly set the TZ
variable (and it also crashed for TZ=GMT):
> Sys.getenv("TZ")
TZ
""
> Sys.putenv("TZ"="CDT6CST")
> Sys.getenv("TZ")
       TZ
"CDT6CST"
> format(strptime("199308070150","%Y%m%d%H%M"),
"%Y-%m-%d %H:%M:%S",
tz="GMT", usetz=TRUE)
[1] "1993-08-07 01:50:00"
> format(strptime("199308070150","%Y%m%d%H%M"),
"%Y-%m-%d %H:%M:%S %Z",
tz="GMT", usetz=TRUE)
Segmentation fault


Not nice. I wrote some code to transfer time stamps from one time zone to
another, but that sits at work right now and I can't get to it...

Dirk
> 
> The behaviour is the same with R1.7.1 on Window2000 as on a R1.8.1 on sunOS
> 5.8.
> 
> Thanks for the reply, Christoph
> 
> ______________________________________________
> R-devel@stat.math.ethz.ch mailing list
> https://www.stat.math.ethz.ch/mailman/listinfo/r-devel
> 
-- 
Bush behaved in Cabinet meetings "like a blind man in roomful of deaf
people".
  -- former Treasury secretary Paul O'Neill, quoted by Bloomberg on 9 Jan
2004