On 2012-11-27 19:33, Jan Pokorný wrote:
> Hi review experts,
>
> I think the check from $SUBJ does not make much sense nowadays
> (perhaps it did in the past). At least I am unable to find
> any clause about it in Fedora Packaging Guidelines.
>
> If I am wrong, please make obvious what is meant with this check,
> otherwise I would suggest dropping this check so as to avoid
> confusion in the package review process [*].
>
> [*] As an anecdotic reference, I found a review [1] where
> the packager understood the check title as to literally use
> "%{name}" macro as a prefix (something not exactly a subject
> of the check, doesn't hurt, though).
>
> [1]
https://bugzilla.redhat.com/show_bug.cgi?id=825593#c2
>
> Thanks,
> Jan
> _______________________________________________
> fedorareview mailing list
> fedorareview(a)lists.fedorahosted.org
>
https://lists.fedorahosted.org/mailman/listinfo/fedorareview
Review experts... I don't know who you refer to, but it's certainly
not me :)
That said, the link in the subject test is [1]. This is not (and never
has) describing any conditions like this on SourceX/PatchX. So, simply
speaking we have a bug.
Furthermore, at a second look the check on SourceX is just silly.
Sources are named by upstream, and as packagers we will live with
whatever name they use.
OTOH, patches is another issue. As maintainers we often handle more
than one patch, they are sent via email etc. In this context, it does
make sense to prefix the patches with the package name IMHO.
We *do* have a place for tests which are not backed up by the
guidelines, the EXTRA section. One way to go would be to skip the test
on SourceX, and place the PatchX check under EXTRA. Another way is of
course to just skip the check.
Or perhaps we are both wrong, and there is a GL mandating this kind of
naming?!
Thoughs?
--alec
1:
https://fedoraproject.org/wiki/Packaging:SourceURL?rd=Packaging/SourceURL.
Found an url [2] which actually promotes prefixing patches with the
package name. My gut feeling is that even if this is not mandated by
the GL, it's sort of best practise.
Opened a bug : [3]
2: