Re: pseudo package for upgrades from hamm
Robert Woodcock <rcw@debian.org> writes:
> We need to add a new field - call it anything you want - I called it
> "Was-Part-Of:" in an earlier post, but I'm sure there's a better name than
> that - "Previously:" maybe.
>
> Anyway, say slink contains a package 'foobar', version 1.2-3. The
> maintainer decides to split it into 'foo' and 'bar' for potato.
>
> In the control files for the foo and bar packages, the maintainer slips in
> that aforementioned field:
>
> Package: foo
> Version: 1.2-4
> Previously: foobar (<< 1.2-3)
>
> ... and does the same thing for the bar package. dselect and apt check that
> field, check the current version of foobar, and based on that, automagically
> select the new packages.
<RANT MODE>
Many, *many* people has proposed this idea before. So many, that you
would be tempted to consider it a simple, natural, and straightforward
idea. Nonetheless, it seems that this far, it has been impossible to
make it part of dpkg, or even to start working on the necessary
modifications for that purpose.
The list of bugs against dpkg grows almost daily, while the very few
people who are blessed to touch the source code continue to be too
buzzy to work on it. Opening the development model for dpkg would be
a great way to overcome this sad situation, but it seems that us, poor
mortal Debian maintainers, are not considered good enough for taking
care of the central and most important tool in our project.
</RANT MODE>
As a developer who have contributed to the project much less than I
would have wanted, but who appreciates it deeply, I would like to ask
the Debian leadership to seriously reconsider the dpkg development
model. It is now clear that our lack of capacity as a project to
adequately manage the evolution of dpkg, is causing us serious
problems, and that if things continue to be like this, we will face
much more serious difficulties in the near future.
Hoping for the best,
M. S.
------------
Martin A. Soto J. Profesor
Departamento de Ingenieria de Sistemas y Computacion
Universidad de los Andes masoto@uniandes.edu.co
Reply to: