Re: Bug#221523: Causes build failures
On Thu, 2003-11-20 at 09:16, Junichi Uekawa wrote:
> > > The configure script is checking for 'python' script.
> > >
> > > What was the original reason for the 'recommends' instead of
> > > 'depends'? It's not really clear to me.
> >
> > some people wanted to be able to keep the default python version at
> > version 2.2, but install packages from unstable, when the default
> > versions were different in testing and unstable.
>
> I do doubt such special needs should be forced upon all of
> us.
The original reason was the policy in general was designed around using
"python" refer to the default version of python without having any hard
dependencies on exactly what version it was, and pythonX.Y for referring
to to a particular version of python without having any dependencies on
what the current default version is.
The reason for this philosophy was you start getting all sorts of weird
side affects once you start violating this clean separation. The above
complication is just one manifestation. The big one for me is it
complicates transitioning between versions... this dependency blocked
any simple packages that only depended on python2.3 from transitioning
into testing, because python2.3 couldn't get into testing until python
(2.3) did, which couldn't happen until _every_ package had transitioned
to python (2.3). Without this dependency, all the python2.3 packages
could have transitioned to testing _before_ migrating the default python
to python (2.3), giving us a more incremental and less traumatic
transition.
Unfortunately, the policy fails to take into account all the
source-package and build-depends issues. This means the clean separation
of python and pythonX.Y is not actually that clean... some more work is
needed on the policy to resolve these issues.
--
Donovan Baarda <abo@minkirri.apana.org.au>
http://minkirri.apana.org.au/~abo/
Reply to: