Re: [R-sig-Fedora] R 3.4.0 RPMS

[Tom Callaway]

Thanks. I've been traveling almost non stop for the last several months. I
appreciate your help here.

On May 12, 2017 10:25 AM, "José Abílio Matos"  wrote:

> On Thursday, 11 May 2017 17.37.41 WEST Martyn Plummer wrote:
> > Dear Tom,
> >
> > I see that RPMS for R 3.4.0 are not successfully built on Fedora, nor
> > on RHEL 7.
> >
> > https://koji.fedoraproject.org/koji/packageinfo?packageID=1230
> >
> > When I build with mock on Fedora 25 I also get a build failure. R is
> > not able to establish the time zone inside the chroot and this triggers
> > a regression tests which is new in R 3.4.0. Specifically, in file
> > tests/reg-tests-1d.R this test fails:
> >
> > ## PR#17186 - Sys.timezone() on some Debian-derived platforms
> > (S.t <- Sys.timezone())
> > if(is.na(S.t) || !nzchar(S.t)) stop("could not get timezone")
> > ## has been NA_character_  in Ubuntu 14.04.5 LTS
> >
> >
> > The work-around is to set the environment variable TZ to any non-empty
> > string before running make check, in the %check section of the SPEC
> > file, e.g.
> >
> > TZ="Europe/Paris" make check
> >
> > best regards
> > Martyn
>
> Thank you Martyn. :-)
>
> I am rebuilding R bumping the version since the previous version succeeded
> to
> build on el6.
>
> BTW Tom, it is probably time to get rid of the
> %defattr(-, root, root, -)
>
> after we are already in the Spring here in the North hemisphere. :-)
>
> Regards,
> --
> José Abílio
>
> ___
> R-SIG-Fedora mailing list
> R-SIG-Fedora@r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-fedora
>

[[alternative HTML version deleted]]

___
R-SIG-Fedora mailing list
R-SIG-Fedora@r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-fedora

Re: [R-sig-Fedora] R 3.4.0 RPMS

[José Abílio Matos]

On Thursday, 11 May 2017 17.37.41 WEST Martyn Plummer wrote:
> Dear Tom,
> 
> I see that RPMS for R 3.4.0 are not successfully built on Fedora, nor
> on RHEL 7.
> 
> https://koji.fedoraproject.org/koji/packageinfo?packageID=1230
> 
> When I build with mock on Fedora 25 I also get a build failure. R is
> not able to establish the time zone inside the chroot and this triggers
> a regression tests which is new in R 3.4.0. Specifically, in file
> tests/reg-tests-1d.R this test fails:
> 
> ## PR#17186 - Sys.timezone() on some Debian-derived platforms
> (S.t <- Sys.timezone())
> if(is.na(S.t) || !nzchar(S.t)) stop("could not get timezone")
> ## has been NA_character_  in Ubuntu 14.04.5 LTS
> 
> 
> The work-around is to set the environment variable TZ to any non-empty
> string before running make check, in the %check section of the SPEC
> file, e.g.
> 
> TZ="Europe/Paris" make check
> 
> best regards
> Martyn

Thank you Martyn. :-)

I am rebuilding R bumping the version since the previous version succeeded to 
build on el6.

BTW Tom, it is probably time to get rid of the
%defattr(-, root, root, -)

after we are already in the Spring here in the North hemisphere. :-)

Regards,
-- 
José Abílio

___
R-SIG-Fedora mailing list
R-SIG-Fedora@r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-fedora