Bug#793976: ruby-netcdf: FTBFS with netcdf in experimental / update to 0.7.1

Youhei SASAKI uwabami at gfd-dennou.org
Wed Jul 29 14:23:45 UTC 2015


Hi, thanks for your notice.

I'll prepare new upstream release ASAP.

2015/07/29 wrote:
>Package: ruby-netcdf
>Version: 0.6.6-2
>Severity: normal
>User: debian-gis at lists.debian.org
>Usertags: netcdf-split-c-f-cxx
>
>Hi Youhei,
>
>We're preparing for the upcoming netcdf transition which will update
>from
>NetCDF 4.1.3 to 4.4.0 and its new lanuague specific releases (NetCDF
>C++4 4.2.1, Netcdf C++ legacy 4.2 & NetCDF Fortran 4.4.2).
>
>ruby-netcdf (0.6.6-2) FTBFS with these new netcdf packages in
>experimental:
>
> checking for netcdf.h... yes
> checking for main() in -lnetcdf... no
> *** extconf.rb failed ***
>Could not create Makefile due to some reason, probably lack of
>necessary
> libraries and/or headers.  Check the mkmf.log file for more details.
> You may
> need configuration options.
>
>The full build log is attached.
>
>Just specifying the Multi-Arch library path with the following is not
>sufficient to resolve main() check failure:
>
> override_dh_auto_configure:
> 	dh_auto_configure -- \
> 		--with-netcdf-include=/usr/include \
> 		--with-netcdf-lib=/usr/lib/$(DEB_HOST_MULTIARCH)
>
>While looking into this build failure I noticed that your ruby-netcdf
>package limits checks for new upstream releases to the 0.6.x versions,
>which excludes the recent 0.7 & 0.7.1 upstream releases.
>
>The RubyNetCDF 0.7.x releases contain improvements for NetCDF-4 support
>which is very relevant for upcoming netcdf transition.
>
>Are there good reasons to stick to the 0.6 releases, and would you
>object to removing the version constraint in the watch file?
>
>I'd like to update the ruby-netcdf package to the latest upstream
>release and upload it to experimental along with the other new netcdf
>package.
>
>It may be that the new upstream release is not sufficient either, and
>that the new NetCDF built only with the HDF5 MPI variant cause the
>incompatibility with ruby-netcdf.
>
>Kind Regards,
>
>Bas



More information about the Pkg-grass-devel mailing list