I have fixed this in R-devel and will port it over to the R release branch in
due course.
The underlying issue is that the conversion from POSIXlt to POSIXct uses the
local time zone and not the CET time zone. I believe this is a bug, but I will
take up that discussion elsewhere.
Martyn
________________________________________
From: Patrick Connolly <p_conno...@slingshot.co.nz>
Sent: 19 May 2017 06:13
To: Martin Maechler
Cc: peter dalgaard; Martyn Plummer; Joris Meys; R-devel
Subject: Re: [Rd] [R] R-3.4.0 fails test
On Thu, 18-May-2017 at 05:46PM +0200, Martin Maechler wrote:
|>
.....
|>
|> Being pretty "stretched" time wise currently, I'm happy for
|> timezone-portable propositions to change the test.
Meantime, anyone who lives where DST happpens in December who wants to
get through the remaining tests can avoid this one by changing the line
> stopifnot(length(fd) == 10, identical(fd, format(dct <- as.POSIXct(dlt))))
to
> stopifnot(length(fd) == 10, identical(fd, format(dct <- as.POSIXlt(dlt))))
^^^^
(which effectively isn't testing anything much)
A less lazy way would be to comment out the relevant lines.
|>
|> Martin
--
~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.
___ Patrick Connolly
{~._.~} Great minds discuss ideas
_( Y )_ Average minds discuss events
(:_~*~_:) Small minds discuss people
(_)-(_) ..... Eleanor Roosevelt
~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.
-----------------------------------------------------------------------
This message and its attachments are strictly confidenti...{{dropped:8}}
______________________________________________
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel
