Re: gnome-pilot testing
On Sun, Jan 20, 2002 at 06:03:18AM +0900, Akira TAGOH wrote:
> >>>>> On Tue, 15 Jan 2002 13:20:41 -0600,
> >>>>> "CW" == Colin Watson <cjwatson@debian.org> wrote:
>
> CW> Which translates to saying that upgrading gnome-pilot from 0.1.54-2 to
> CW> 0.1.64-1 in testing makes the versions of gnome-pilot-conduits and
> CW> gnome-pim-conduits in testing uninstallable on alpha, and possibly on
> CW> other architectures alphabetically after alpha.
>
> CW> It looks to me as if forcing all of gnome-pilot, gnome-pilot-conduits,
> CW> and gnome-pim to be upgraded in testing simultaneously will sort this
> CW> out, as the packages are otherwise fine. Anthony?
>
> Well, update_output.txt has too little information.
update_excuses.html has too little information, but update_output.txt in
conjunction with madison and/or a handy mirror works just fine.
> for example, why isn't glib1.3 installed into testing?
> it seems that it also get same messages though.
skipped: glib1.3 (0) (126+11)
got: 42+0: a-42
* alpha: libatk-dbg, libatk-dev, libatk7, libfribidi-dev,
libfribidi0, libgtk1.3-11, libgtk1.3-common, libgtk1.3-dbg,
libgtk1.3-dev, liblinc-dev, liblinc0, liborbit2, liborbit2-dbg,
liborbit2-dev, libpango-common, libpango-dbg, libpango-dev,
libpango0.22, libwnck-1, libwnck-dev, libzvt2-0, libzvt2-dev,
orbit2
In other words, on the first architecture the testing script looked at
it found that upgrading glib1.3 would render the listed packages
uninstallable, so it didn't upgrade it. A brief glance at the
dependencies of the versions of each of these packages in testing (for
instance, libatk7 depends on libglib1.3-11) suggests to me that they all
have to be upgraded simultaneously, which means they all have to be
candidates for testing simultaneously, etc.
I thought GNOME 2 wasn't supposed to go into released distributions yet,
anyway?
--
Colin Watson [cjwatson@flatline.org.uk]
Reply to: