Michael Schroeder [EMAIL PROTECTED] writes:
So A requires B and B has a requires(post) to A? This isn't installable
at all, thus it's a packaging bug.
i don't agree.
not knowing exactly what rpmlib does, here is how i understand the pb:
Requires is same as Requires(postrans), ie if A
Michael Schroeder [EMAIL PROTECTED] writes:
Package A contains /foo
Package B has Conflicts: /foo
Since repository metadata can't possibly contain this [*],
Mandriva has chosen to get rid of file requires/conflicts, unless
explicitly provided. eg:
% rpm -q --provides bash | grep '^/'
/bin/sh
there is a bug in rpmbuild triggered by hardlink on softlink.
ie things like:
lrwxrwxrwx 2 pixel pixel 1 Sep 10 13:28 b - a
lrwxrwxrwx 2 pixel pixel 1 Sep 10 13:28 c - a
the cpio generated by rpmbuild is broken.
suggested fix:
replace
!S_ISDIR(st-st_mode) st-st_nlink 1
(inspired by 90ca5e5989ec289a51d2e1c7c8caa59063a6fb70)
check chroot() result
- bail out early if it fails instead of blindly continuing and potentially
messing in real root (chroot can fail for priviledged user too)
diff -p -up rpm-4.4.2.3/lib/psm.c.pix rpm-4.4.2.3/lib/psm.c
---
Jindrich Novy [EMAIL PROTECTED] writes:
rpmlib(PayloadIsLzma) = 4.4.6-1 is needed by glibc-2.8-1.i686
rpmlib(PayloadIsLzma) = 4.4.6-1 is needed by glibc-common-2.8-1.i686
rpmlib(PayloadIsLzma) = 4.4.6-1 is needed by glibc-devel-2.8-1.i686
rpmlib(PayloadIsLzma)
Tom Brown [EMAIL PROTECTED] writes:
Hi
I have an rpm called say 1 that had dependencies on rpm's 2, 3 and 4 -
so when i install 1 then 2, 3 and 4 also get installed which is the desired
result.
Is there anyway that i can have rpm 1 install first out of the 4 as there are
pieces in rpm 1
Thomas Fitzsimmons [EMAIL PROTECTED] writes:
How are rebuilddbs handled during deployment? If the process is cumbersome or
error-prone then maybe a slower rpmdbCountPackagesArch is a worthwhile
trade-off for avoiding an RPM database rebuild.
i've never had to handle this. I don't remember
Thomas Fitzsimmons [EMAIL PROTECTED] writes:
[...]
In the typical case, rpmtsRun() is already doing 2 iterations on
Name doing headerLoad. And there are a few more.
IMO if we look at optimisations, there are many things that should be
done before this. For example, it would be far more
Thomas Fitzsimmons [EMAIL PROTECTED] writes:
I'm working on a way to pass per-architecture package counts to
scriptlets [1]
for the record, mandriva is not allowing to install 2 pkgs with
different arch. So any change on this subject won't hurt/help us :)
To achieve biarch support Mandriva is
Thomas Fitzsimmons [EMAIL PROTECTED] writes:
I'm working on a way to pass per-architecture package counts to
scriptlets [1]. Jeff Johnson suggested creating a new database index
for this purpose [2].
Attached are two patches against Fedora 9's rpm. The first adds a
Namearch index to the
Seth Vidal [EMAIL PROTECTED] writes:
On Mon, 16 Jun 2008, Pixel wrote:
However I can imagine an one-time unsubscription counterpart (something,
which will evaluate list of all files going to be removed or overwritten
in the forthcoming transaction and then issue a command with a such
list
Stanislav Brabec [EMAIL PROTECTED] writes:
[...]
Mandriva patch defines patterns evaluated by RPM runtime.
[...]
* no, install-info is not a good candidate fot this technique,
see http://wiki.rpm.org/Problems_of_Scriptlets (this is an
implementation of Database rebuild, install info
Stanislav Brabec [EMAIL PROTECTED] writes:
[...]
Using %preun is acceptable only for uninstallation, not for upgrade.
But this is yet another problem mentioned in the document, not related
to one-time-scriptlets.
oops, right. mandriva never went that far :)
[...]
However I can imagine an
Tom \spot\ Callaway [EMAIL PROTECTED] writes:
The only reason we use mktemp in there is because we couldn't make rpm
code changes to use the native glibc functions. As to rpm
--short-circuit, well, I honestly think we should think long and hard
about whether we want to keep it around.
well,
Stanislav Brabec [EMAIL PROTECTED] writes:
Particular problems there may have different severity and different
complexity. The worst one seems to be Problems of Scriptlets / Database
rebuild.
mandriva is currently experimenting something on this subject:
Florian Festi [EMAIL PROTECTED] writes:
Finger printing actually doesn't work as soon as there are some symlinks
involved that are not installed yet. I attached 3 spec files that show that.
Install the README rpm that then install FOO and FOO-DOC at once. As you can
see
Panu Matilainen [EMAIL PROTECTED] writes:
[...]
The ldconfig hackery in rpm is going away sooner or later, what I want to see
is some more generic mechanism for packages to queue actions to happen at end
of the transaction. Things like gtk-update-icon-cache only need running once,
so
Bill Nottingham [EMAIL PROTECTED] writes:
Pixel ([EMAIL PROTECTED]) said:
as for the cache, it is needed if you use /etc/ld.so.conf.d/xxx.conf,
which alas is the case for mandriva's qt4. It seems fedora also uses
the ld cache to handle its qt3?
Sure, but that's very few packages; the other
Any comment?
without this patch: rpm will skip running ldconfig for unordered packages
once it's done once. unordered packages seems to mean the package is not
Required(post) by another package (resp. Required(pre)).
The result is that if needed for a %post the ldconfig will be correctly done,
Michal Marek [EMAIL PROTECTED] writes:
I'm wondering - doesn't RPM support creating anyarch + noarch
subpackages in one build simply because it has never had the feature in
first place and no-one implemented it so far (for backwards
compatibility reasons perhaps)? Or are there good reasons
Panu Matilainen [EMAIL PROTECTED] writes:
Before anybody asks: I actually do think that ultimately rpm should be able to
support reliably rolling back transactions. It's just that the current
repackage+rollback combo fails to deliver it, as there's no way to undo script
actions.
for info,
Panu Matilainen [EMAIL PROTECTED] writes:
The questionable part is the size of the thing, it's not exactly trivial:
[EMAIL PROTECTED] ~]$ ls -l /lib64/libglib-2.0.so.0.1400.5
-rwxr-xr-x 1 root root 823936 2008-01-08 05:36 /lib64/libglib-2.0.so.0.1400.5
it's ok for mandriva:
- rpm is already
fixes http://qa.mandriva.com/show_bug.cgi?id=36672
$ rpm -ba something
[...]
Provides: firefo-devel testi5 jefae
Requires(interp): /bin/sh /bin/sh /bin/sh /bin/sh
Requires(rpmlib): rpmlib(PayloadFilesHavePrefix) = 4.0-1
rpmlib(CompressedFileNames) = 3.0.4-1
Requires(pre): /bin/sh
Requires(post):
Mark Rosenstand [EMAIL PROTECTED] writes:
Actually you can completely drop the path. So nss.h should be enough if
pkg-config works fine for NSS on Mandriva. Of course the defaults for
the compilation without pkg-config are wrong for your case.
But since upstream NSS doesn't include a .pc
Ralf Corsepius [EMAIL PROTECTED] writes:
- Does any application need this feature?
- How are applications supposed to use this feature?
ATM, I am not really convinced it is _really_ useful.
IMO it's really useful. In Mandriva we've wanted this feature for
years, even if we didn't know
Hi, Mandriva is currently using rpm 4.4.8. We are investigating the
cost to switch to 4.4.2.2
Since August 2007, Mandriva uses Suggests, which is implemented in
rpm 4.4.3 (?). In rpm 4.4.2.x, there is already partial support for it
since RPMSENSE_MISSINGOK flag is already there.
I've crafted a
Pixel [EMAIL PROTECTED] writes:
Hi, Mandriva is currently using rpm 4.4.8. We are investigating the
cost to switch to 4.4.2.2
Since August 2007, Mandriva uses Suggests, which is implemented in
rpm 4.4.3 (?). In rpm 4.4.2.x, there is already partial support for it
since RPMSENSE_MISSINGOK
Michael Schroeder [EMAIL PROTECTED] writes:
Hmm, IIRC rpm-4.4.8 puts suggests/enhances into the normal rpm
requires, which is incompatible with rpm versions that don't
understand the RPMSENSE_MISSINGOK flag.
I'd prefer the way we (SUSE) do it by putting the suggests/enhances
in the already
28 matches
Mail list logo
