On 06/18/2014 05:23 PM, Vít Ondruch wrote:
Better to not go this way ...
Do you mean this whole approach is wrong?
I guess the cleanest option would be to have separate branches.
This is wrong:
ruby(abi) >= %{rubyabi}
it should be:
ruby(abi) = %{rubyabi}
Vít
Dne 18.6.2014 16:08, Achilleas Pipinellis napsal(a):
> For starters, the title says **most of**, because I'm not including
> EPEL5 :p
>
> To the point.
> Now that we have some common ground between the packaging guidelines,
> I'm trying to get together a template for packaging: fc19, fc20, fc21,
> el6, el7.
>
> I haven't yet packaged anything for EPEL6, so I searched ruby-sig and
> found some answers. So far I have written down the R/BR, so please
> correct me I am wrong. Here goes:
>
> ```
> %if 0%{?fc19} || 0%{?fc20} || 0%{?el7}
> Requires: ruby(release)
> Requires: ruby(rubygems)
> %endif
>
> %if 0%{?el6}
> Requires: ruby(abi) >= %{rubyabi}
> Requires: ruby(rubygems)
> BuildRequires: ruby(abi) >= %{rubyabi}
> %else
> BuildRequires: ruby(release)
> %endif
>
> BuildRequires: rubygems-devel
>
> %if 0%{?fc19} || 0%{?fc20} || 0%{?el7} || 0%{?el6}
> Provides: rubygem(%{gem_name}) = %{version}
> %endif
> ```
>
> Next parts are:
>
> 1) if gem has native extensions: different %install macro
> 2) if gem has minitest: build minitest5 for el7 [0]
>
> If I left something out please advise.
>
> Thanks!
>
>
> [0]
https://admin.fedoraproject.org/pkgdb/package/rubygem-minitest/
>
_______________________________________________
ruby-sig mailing list
ruby-sig(a)lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/ruby-sig
--
FAS : axilleas
GPG : 0xABF99BE5
Blog:
http://axilleas.me