Re: [gentoo-dev] Re: Python 3.1: Stabilization and news item

[William Hubbs]

On Thu, Mar 11, 2010 at 02:24:46AM +0100, Ben de Groot wrote:
> On 11 March 2010 01:25, William Hubbs <williamh@g.o> wrote:
> > ??If someone has a package that truly works with either python 2 or 3,
> > ??what is the harm in automatically pulling in python 3 and installing
> > ??the package for both python 2 and 3?
> >
> > ??As long as pulling in python-3 doesn't change the system's default
> > ??python interpretor I don't see a problem with having them both
> > ??installed.
 
> I've seen enough python-3 specific bugs to know it is not without
> problems. It's a waste of time and resources for something that is
> not ready to be used anyway. While it can be argued that that is
> what our testing branch is for, it is certainly not something that
> should be pushed to stable users.
 
 What does upstream say about python 3.1?  Are they calling it stable?
 Yes, it is incompatible with python-2, but, it is set up so both can be
 on a system at the same time.  I'm no expert on python, but I think
 even upstream has python deliberately set up that way.

> Even if it would be just "dead weight", it is not something we should
> wish for. It is bloat, it is unnecessary, and causes more problems
> than that it solves. Why should users have to compile multiple
> python versions, if they only use one anyway?
 
 If they are only using python-2 and all of the packages they use only work
 with python-2, then the dependencies of the packages should be fixed to
 reflect that.

Even if python-3 is stable and the dependencies of the
 packages they have say that they only support python-2
 python-3 will not be on their systems.

Someone compared pythohn to gcc earlier in this thread, but I'm not sure
that is a fair comparison.  AFAIK, gcc is not slotted by upstream, and
python is.  I think that makes a difference in how we handle it.

William