Dne 22.1.2018 v 12:31 Miroslav Suchý napsal(a):
You took pretty basic example without context. So let me give you
different example.
There were attempts to get Ruby on Rails into EPEL. It is around 80
packages. Some packages were RHEL contidiontalized. But the effort to
get Ruby on Rails was never successful. Even if it was successful,
nobody maintained/updated it later, for example because more recent Ruby
on Rails, developers decided to drop support for older Ruby which are in
RHEL. That left some Fedora packages with some RHEL conditions and makes
every update of such package pain. These conditionals won't be ever used
again.
Also, from my experience maintaining Ruby in RHEL and RHSCL, in 90 %
percent of cases cherry-pick of the specific fix I want to get from
Fedora works without conflicts and if there are conflicts, they are just
in changelog.
Vít
Well I think maintainers should chose the ideal solution for their case. If use of macros
just complicates things too much, maybe use of branches might make more sense. On the
other hand does it really make sense to enforce maintaining different branches for a few
conditionals?
Also the issue is not just in maintaining branches. As it stands right now, one can grab
almost any spec file from a src.rpm and have it work properly everywhere.
Having said that, some specs does indeed look like a mess at a first glance, but if
maintainer chose this way as a best solution for his case, I respect that.
To prevent such cases become popular, I think this should be documented on guidelines so
that new maintainers can make their choice appropriately and not just copy what others
did.