1 |
-----BEGIN PGP SIGNED MESSAGE----- |
2 |
Hash: SHA1 |
3 |
|
4 |
On Thursday 05 February 2004 12:20, Marius Mauch wrote: |
5 |
> X-Mailer: Sylpheed version 0.9.8claws (GTK+ 1.2.10; i686-pc-linux-gnu) |
6 |
> Mime-Version: 1.0 |
7 |
> Content-Type: text/plain; charset=US-ASCII |
8 |
> Content-Transfer-Encoding: 7bit |
9 |
> |
10 |
> On Thu, 5 Feb 2004 11:20:42 +0100 |
11 |
> |
12 |
> Spider <spider@g.o> wrote: |
13 |
> > So my question was for the generic case of things like this, Can |
14 |
> > the binaries be made more reliable? |
15 |
> > |
16 |
> > Perhaps this sort of thing be worked around in portage-code? ( wrap |
17 |
> > a consistency check of linking before installing? demand that a |
18 |
> > system is"updated" ? ) |
19 |
> > |
20 |
> > Or should binaries have more metadata in them, perhaps a specific |
21 |
> > requirement for libraries? (sheesh, then its down to RPM again. |
22 |
> > that's bad. ) |
23 |
> |
24 |
> Why more metadata? Can't we just run a ldd check on the binaries, |
25 |
> check that the relevant libraries are present, if not do a lookup in a |
26 |
> (to-be-created) list of library<->package mappings and add the |
27 |
> relevant packages as PDEPEND? (or yell at the user to provide that |
28 |
> library ;) |
29 |
|
30 |
Partly because first of all whe then would unpack those libs and do some |
31 |
magic to take local libs into account. Further the mappings is exactly |
32 |
the thing that redhat does that we cannot do easilly (which we probably |
33 |
don't want either). |
34 |
|
35 |
Paul |
36 |
|
37 |
- -- |
38 |
Paul de Vrieze |
39 |
Gentoo Developer |
40 |
Mail: pauldv@g.o |
41 |
Homepage: http://www.devrieze.net |
42 |
-----BEGIN PGP SIGNATURE----- |
43 |
Version: GnuPG v1.2.4 (GNU/Linux) |
44 |
|
45 |
iD8DBQFAIiijbKx5DBjWFdsRAv7/AKCyu7NaqkCbJijnFyEGANtqLPQCFQCfZ8DZ |
46 |
s8PeQUx1nOkCU++R57bCyUE= |
47 |
=3la3 |
48 |
-----END PGP SIGNATURE----- |
49 |
|
50 |
-- |
51 |
gentoo-dev@g.o mailing list |