[Rd] ok to segfault with POSIXlt zone=NULL zone=""?

frederik at ofb.net frederik at ofb.net
Tue Dec 6 18:27:43 CET 2016


Hi all,

Here's a more minimal version of my earlier bug report (thanks, Joshua
Ulrich):

d=as.POSIXlt(Sys.time()); d$zone=NULL; d$zone=""; d

I got some helpful, if glib, feedback from Joshua that the segfault
may be caused by the changing of the order of the list elements in 'd'
(representing the "internal structure" of the POSIXlt object).

He seems to think that it's OK for R to segfault - I was wondering if
someone else could lend a second opinion. My understanding is that we
should try to avoid segfaulting as a way of handling errors, if only
because they become much more difficult to debug when the R session is
forced to quit.

I don't know exactly which line is causing the bug, but looking at the
code for do_formatPOSIXlt in "src/main/datetime.c", it seems that
there would not be a huge performance penalty to add an extra sanity
check to prevent this from occurring.

Thank you,

Frederick

On Tue, Dec 06, 2016 at 04:37:20AM -0800, frederik at ofb.net wrote:
> Hi all,
> 
> I ran into a segfault while playing with dates.
> 
>     $ R --no-init-file
>     ...
>     > library(lubridate); d=as.POSIXlt(floor_date(Sys.time(),"year")); d$zone=NULL; d$zone=""; d
> 
>     Attaching package: ‘lubridate’
> 
>     The following object is masked from ‘package:base’:
> 
>         date
> 
>     Warning message:
>     package ‘lubridate’ was built under R version 3.4.0 
> 
>      *** caught segfault ***
>     address (nil), cause 'unknown'
> 
>     Traceback:
>      1: format.POSIXlt(x, usetz = TRUE)
>      2: format(x, usetz = TRUE)
>      3: print(format(x, usetz = TRUE), ...)
>      4: print.POSIXlt(x)
>      5: function (x, ...) UseMethod("print")(x)
> 
>     Possible actions:
>     ...
> 
> Hope I'm not doing something illegal...
> 
> Thanks,
> 
> Frederick
> 
> ______________________________________________
> R-devel at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-devel



More information about the R-devel mailing list