Re: lmodern fonts and sarge
frank@kuesterei.ch (Frank Küster) wrote:
> Thank you. But to be really sure, I would suggest some additional tests.
[...]
> First of all, it might be lucky that bar is before foo in the
> alphabet. Perhaps try it the other way round... And, more important, let
> the packages change files. That is, in your setup (without swapped
> alphabetical order, your filenames and versions), bar_1.0 would contain
> the file /usr/share/baz, but bar-1.1 would not, whereas now foo-1.1
> would contain it.
Works like a charm.
- installed the following (dummy) packages:
Package: bar
Version: 1.0-1
Does not contain /usr/share/evil/evil-file
and
Package: foo
Version: 1.0-1
Contains /usr/share/evil/evil-file
- "dselect update" with the following new packages available:
Package: bar
Version: 1.1-1
Conflicts: foo (<< 1.1)
Contains /usr/share/evil/evil-file
Package: foo
Version: 1.1-1
Does not contain /usr/share/evil/evil-file
- "dselect select" -> nothing particular. The new versions are shown as
available, but no dependency/conflict resolution screen appears.
- "dselect install":
Reading Package Lists... Done
Building Dependency Tree... Done
The following packages will be upgraded:
bar foo
2 upgraded, 0 newly installed, 0 to remove and 0 not upgraded.
Need to get 3620B of archives.
After unpacking 0B of additional disk space will be used.
Do you want to continue? [Y/n]
Get:1 copy: sid/binary-all/ foo 1.1-1 [1756B]
Get:2 copy: sid/binary-all/ bar 1.1-1 [1864B]
Fetched 3620B in 0s (0B/s)
(Reading database ... 79773 files and directories currently installed.)
Preparing to replace foo 1.0-1 (using .../apt/archives/foo_1.1-1_all.deb) ...
Unpacking replacement foo ...
Preparing to replace bar 1.0-1 (using .../apt/archives/bar_1.1-1_all.deb) ...
Unpacking replacement bar ...
Setting up foo (1.1-1) ...
Setting up bar (1.1-1) ...
I've updated:
deb http://people.via.ecp.fr/~flo/debian-experiments sid/binary-all/
deb-src http://people.via.ecp.fr/~flo/debian-experiments sid/source/
accordingly so you can try this at home.
--
Florent
Reply to: