R devel - Jun 2026 - Revisit c.POSIXlt , [<-.POSIXlt , [[<-.POSIXlt ?

>>>>> Martin Maechler 
>>>>>     on Mon, 15 Jun 2026 18:08:43 +0200 writes:
>>>>> Suharto Anggono Suharto Anggono 
>>>>>     on Sun, 14 Jun 2026 13:37:13 +0000 (UTC) writes:
    >> Could at least a minimalistic fix be put for the upcoming R 4.6.1?
For example:
    >> ln '[<-.POSIXlt' and '[[<-.POSIXlt', change
    >> ? ? else as.POSIXlt(as.POSIXct(value), tz = tz[1L]))
    >> to
    >> ? ? else as.POSIXlt(as.POSIXct(value), tz = if(is.null(tz))
"" else tz[1L]))

    >> In 'c.POSIXlt', apply 'floor' to the 'sec'
component before converting to POSIXct like in Bug 18989 Comment 9
(https://bugs.r-project.org/show_bug.cgi?id=18989#c9).


    > Yes, this *could* be done, last minute, i.e., fixing some bugs
    > in POSIXlt <--> POSIXct  handling.

    > I've tried in my local copy of  "R 4.6.1 beta".

    > and indeed, these minimal patches already fix the two problems
    > you mention below (and I've marked with  '1.'  and
'2.' ).

    > *However* your wish is rather late (we have code freeze in 1.5 days and
    > should only fix somewhat urgent problems with small fixes), 
    > *and* the two problematic examples below are not new bugs, they
    > do "not work" already in 4.4.x (I tried one) and 4.5.0, ..
4.5.3, etc

    > Whereas your proposed changes *are* nicely small and look
    > sensible, the bugs (below) have been in R for a while now *and*
    > are fixed in R-devel (already with svn r89926 I think).

    > So I don't know if there is anybody (but you ;-) who'd be very
    > happy for such a late bug fix.

    > Other opinions -- notably R core / Release Manager ?

    > Thank you in any case, for the suggestion, and the effort to
    > find such a minimal (informal) patch.

    > Best regards,
    > Martin


I've decided to commit these -- together with the
as.matrix(<POSIXlt>)  bug fix, ported from R-devel.

So these should make into R 4.6.1  next week.

Best regards,
Martin

Isn't the NEWS item too condensed?

On 'c.POSIXlt', the part from "Changes in R-devel" can be
taken over:
c(x) could become one second ahead of x in R 4.6.0

On '[<-.POSIXlt' and '[[<-.POSIXlt', the following can be
stated:
x[i] <- val and x[[i]] <- val now also works as intended for POSIXlt
object 'x' without "tzone" attribute.
Or:
For POSIXlt object 'x' without "tzone" attribute, x[i] <-
val and x[[i]] <- val now works and converts 'val' as intended.
Or:
In R 4.6.0, if 'x' is a POSIXlt object without "tzone"
attribute and 'val' is not a POSIXlt object without "tzone"
attribute, x[i] <- val and x[[i]] <- val failed or didn't convert
'va' to current time zone.

______
On Tuesday, 16 June 2026 at 06:25:42 am GMT+7, Martin Maechler <maechler at
stat.math.ethz.ch> wrote:

>>>>> Martin Maechler
>>>>>? ? on Mon, 15 Jun 2026 18:08:43 +0200 writes:
>>>>> Suharto Anggono Suharto Anggono
>>>>>? ? on Sun, 14 Jun 2026 13:37:13 +0000 (UTC) writes:
? ? >> Could at least a minimalistic fix be put for the upcoming R 4.6.1?
For example:
? ? >> ln '[<-.POSIXlt' and '[[<-.POSIXlt', change
? ? >> ? ? else as.POSIXlt(as.POSIXct(value), tz = tz[1L]))
? ? >> to
? ? >> ? ? else as.POSIXlt(as.POSIXct(value), tz = if(is.null(tz))
"" else tz[1L]))

? ? >> In 'c.POSIXlt', apply 'floor' to the 'sec'
component before converting to POSIXct like in Bug 18989 Comment 9
(https://bugs.r-project.org/show_bug.cgi?id=18989#c9).


? ? > Yes, this *could* be done, last minute, i.e., fixing some bugs
? ? > in POSIXlt <--> POSIXct? handling.

? ? > I've tried in my local copy of? "R 4.6.1 beta".

? ? > and indeed, these minimal patches already fix the two problems
? ? > you mention below (and I've marked with? '1.'? and
'2.' ).

? ? > *However* your wish is rather late (we have code freeze in 1.5 days and
? ? > should only fix somewhat urgent problems with small fixes),
? ? > *and* the two problematic examples below are not new bugs, they
? ? > do "not work" already in 4.4.x (I tried one) and 4.5.0, ..
4.5.3, etc

? ? > Whereas your proposed changes *are* nicely small and look
? ? > sensible, the bugs (below) have been in R for a while now *and*
? ? > are fixed in R-devel (already with svn r89926 I think).

? ? > So I don't know if there is anybody (but you ;-) who'd be very
? ? > happy for such a late bug fix.

? ? > Other opinions -- notably R core / Release Manager ?

? ? > Thank you in any case, for the suggestion, and the effort to
? ? > find such a minimal (informal) patch.

? ? > Best regards,
? ? > Martin


I've decided to commit these -- together with the
as.matrix(<POSIXlt>)? bug fix, ported from R-devel.

So these should make into R 4.6.1? next week.

Best regards,
Martin  
	[[alternative HTML version deleted]]