Dne 3.4.2013 14:12, Florian Festi napsal(a):
On 04/03/2013 12:58 PM, Vít Ondruch wrote:
The only thing you get wrong is that you take a look at Fedora packages
and do some statistics. You don't see the packages which could be in
Fedora if RPM/YUM would do better job.
Just as an example, I guess everybody would welcome Redmine [1] in
Fedora (you can substitute GitLab [2] or Aeolus [3] for Redmine if you
like). It was not possible to do so for several releases of Fedora,
since Redmine was using Ruby on Rails 2.3 where in Fedora, there was
Ruby on Rails 3.x. If we would like to move to Ruby on Rails 4 in Fedora
as soon as they'll be releases, we will have actually two options (1)
forget about the upgrade of Ruby on Rails in Fedora and wait for
upstream or possible become upstream, to help with migration of the
project (2) break Redmine and every application which is using Ruby on
Rails in Fedora. Neither of these options are good options. So the
easiest solution is to not have Redmine in Fedora at all.
So now, please, could you count also the cost of missed opportunities?
I have some difficulties believing that the only reason for this is that
the name "rubygem-rails" was already taken. May be you can elaborate a
bit more why getting Rails 2.3 into Fedora would have been fundamentally
easier if the name was still available?
Florian
Because it would be natural process.
Lets suppose we are in F14 timeframe and 2.3 are the newest Rails
available and lets say we have in Fedora several Rails applications
using Rails 2.3 API. They work just perfect to suit needs of their users.
Now, in F15 timeframe, Rails 3.0 were introduced. So we would like to
move the packages to Rails 3.0. But what to do with the old
applications? How to provide support for them? Note that even at this
point, although Rails 2.3 are not actively developed, they are supported
by upstream, so no need for upstreams to migrate your application.
Ok, if we don't want to break anything, we have two options
1) Just introduce new package rails30 and new applications can depend on it
2) Move the rails package to Rails 3.0 and reintroduce rails23
compatibility package
That looks quite simple, but you doesn't count that what is called Ruby
on Rails is collection of 40 packages (the number vary from version to
version, but tends to increase), which would need to be re-reviewed,
although they were just perfect moment ago.
Ok, so lets say we introduce the rails23 compatibility packages (which
is IMO the better option, since the nonversioned package should always
point to newest and greatest release), we do the reviews and we possibly
double the amount of packages. We even fix all the application
dependencies from rubygem-rails to rubygem-rails23. We might try to
migrate some applications to Rails 3.0 and propose patches to upstream,
but considering this, it is already too much work.
Now, there comes security fix for Rails. Typically it impacts every
stable Rails branch, i.e. it should be applied to two packages. But you
already denied me to use cherry-pick straight forward, since I am not
working with one package/repository but with two packages. So again more
work then it necessarily needed to be.
Yes, you might change the policy that re-reviews are not necessary, but
anyway, you'll end up with mess of packages such as rails23, rails30,
rails31, rails32, rails40 and you lost the meaning of version. Actually
then somebody comes and he things that he doesn't need just Rails 3.0,
but he needs specifically Rails 3.0.5, so he well do another new package
rails305. You cannot stop this explosion of various versioned modules.
So please, don't allow this dark future. Leave the name of the package
and the version of the package its meaning and give us tools to allow
parallel installation and easy migration between various versions of
packages.
I am pretty sure it will not be today, nor tomorrow, but I'd like to see
that you acknowledge that this is issue and one day, it will be solved.
Thank you.
Vít
--
devel mailing list
devel@xxxxxxxxxxxxxxxxxxxxxxx
https://admin.fedoraproject.org/mailman/listinfo/devel
