R devel - Apr 2002 - strptime mysteriously adds a day - 0S-specific: Linux and (PR#1467)

On Thu, 18 Apr 2002, Martin Maechler wrote:
> >>>>> "Jason" == Jason Turner
<jasont@indigoindustrial.co.nz> writes:
>
>     Jason> strptime() mysteriously adds a day to a date, unless the year
>     Jason> is specified.  Tested on:
>     Jason> Linux (RedHat 6.0) - R version 1.4.1 and R-devel.
>     Jason> Windows - R version 1.4.1
>
>     Jason> Bug isn't found on OpenBSD, R version 1.4.1.
>
> I confirm the bug (R-devel, Redhat 7.1)
> To reproduce, it's sufficient to work with
> "%d %B" or "%d %b" .
>
> What happens is that January and February work okay, and all
> months from March produce the problem,
>
> e.g.
>
> > strptime("28 Feb",format="%d %b")# ok
> [1] "2002-02-28"
> > strptime("19 March",format="%d %B")# wrong
> [1] "2002-03-20"
The issue is that an unspecified year is assumed *by the IS* to be 1900, a
leap year. I'm not sure why glibc does this when the tm_year field is not
being set, but it confidently says this is day 101 of the year, and that's
the result you get.

The only portable workaround is to give the year: there is (at least in
R-devel) a warning in the help file about this.

-- 
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


-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To:
r-devel-request@stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

On Thu, 18 Apr 2002 ripley@stats.ox.ac.uk wrote:
> On Thu, 18 Apr 2002, Martin Maechler wrote:
> 
> > >>>>> "Jason" == Jason Turner
<jasont@indigoindustrial.co.nz> writes:
> >
> >     Jason> strptime() mysteriously adds a day to a date, unless the
year
> >     Jason> is specified.  Tested on:
> >     Jason> Linux (RedHat 6.0) - R version 1.4.1 and R-devel.
> >     Jason> Windows - R version 1.4.1
> >
> >     Jason> Bug isn't found on OpenBSD, R version 1.4.1.
> >
> > I confirm the bug (R-devel, Redhat 7.1)
> > To reproduce, it's sufficient to work with
> > "%d %B" or "%d %b" .
> >
> > What happens is that January and February work okay, and all
> > months from March produce the problem,
> >
> > e.g.
> >
> > > strptime("28 Feb",format="%d %b")# ok
> > [1] "2002-02-28"
> > > strptime("19 March",format="%d %B")# wrong
> > [1] "2002-03-20"
> 
> The issue is that an unspecified year is assumed *by the IS* to be 1900, a
> leap year. 
Sorry, 1900 is not a leap year. Maybe glibc doesn't know that? :-)
However, 'cal 1900' (on Linux) gives a correct answer.
 
Göran 
> I'm not sure why glibc does this when the tm_year field is not
> being set, but it confidently says this is day 101 of the year, and
that's
> the result you get.
> 
> The only portable workaround is to give the year: there is (at least in
> R-devel) a warning in the help file about this.
> 
> 
-- 
 Göran Broström                      tel: +46 90 786 5223
 professor                           fax: +46 90 786 6614
 Department of Statistics            http://www.stat.umu.se/egna/gb/
 Umeå University
 SE-90187 Umeå, Sweden             e-mail: gb@stat.umu.se

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To:
r-devel-request@stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._