yum-utils-1.1.17 is now released.
Check the changelog [1] for changes.
tarball:
http://linux.duke.edu/projects/yum/download/yum-utils/yum-utils-1.1.17.tar.gz
SRPM:
http://linux.duke.edu/projects/yum/download/yum-utils/yum-utils-1.1.17-1.src.rpm
[1]:
Brian Long wrote:
On Thu, 2008-08-28 at 07:43 -0400, Seth Vidal wrote:
On Thu, 2008-08-28 at 07:46 +0200, Tim Lauridsen wrote:
Are there any plans to push yum-3.2.19 into Fedora 8, because
yum-utils-1.1.16 needs yum-3.2.19 [1], and it would be nice to be able
to get it into Fedora 8 too
yum-utils-1.1.16 is now released.
Check the changelog [1] for changes.
tarball:
http://linux.duke.edu/projects/yum/download/yum-utils/yum-utils-1.1.16.tar.gz
SRPM:
http://linux.duke.edu/projects/yum/download/yum-utils/yum-utils-1.1.16-1.src.rpm
[1]:
%setup -q
@@ -481,6 +492,10 @@ fi
%config(noreplace) %{_sysconfdir}/yum/pluginconf.d/keys.conf
/usr/lib/yum-plugins/keys.*
+%files -n yum-NetworkManager-dispatcher
+%defattr(-, root, root)
+/etc/NetworkManager/dispatcher.d/*
+
%changelog
* Thu Aug 7 2008 Tim Lauridsen [EMAIL PROTECTED
Seth Vidal wrote:
On Fri, 2008-08-22 at 09:53 +0200, Terje Røsten wrote:
python-iniparse faster to gain any more speed :).
Just saw this:
http://pyfig.alecwh.com/
Might be faster than iniparse.
unless I'm misreading it doesn't handle the config file format we use,
does it?
-sv
Simon Goodall wrote:
I've written a yum plugin to use Avahi to find proxy server's on a
local network (see the attachment). The idea is that I do not have to
configure a proxy server in the yum config, especially due to;
a) My laptop is used across many networks and I do not want to mess
with
yum-utils-1.1.15 released.
Changlog:
http://devel.linux.duke.edu/gitweb/?p=yum-utils.git;a=blob;f=ChangeLog;h=321e3f07a65ee7066c96c90dba7871ec71d49f68;hb=19196969cbd710251ce420e1351e8ec6c00b4732
tarball:
James Antill wrote:
So this is something I wrote during the Fed-9 freeze so that you could
easily see what was meant by a group being installed/not-installed.
Example output is (without verbose mode output is identical):
% yum grouplist -v
[...]
Installed Groups:
Group name
Kulbir Saini wrote:
Hi!
Since we now have more than 20 plugins for yum and more are popping up, it
would be a good idea to have a single configuration file for all the
plugins with a format similar to that of .repo files.
[plugin_name1]
enabled=1
parameter1=foo
parameter2=bar
Florian Festi wrote:
If editing all those files is too cumbersome someone could write an
plugin options dialog for pirut or any other graphical tool. That could
offer a simple interface for configuring and switching on/off plugins.
Sending patches the project chosen could help.
Florian
Florian Festi wrote:
Hi!
This is another API issue thta I already tried to address some time ago:
Transaction.add* do not report back what exactly they did. They
sometimes even return a txmbr that did not get added at all. Not to talk
about updated txmbrs and others that might also be the
Robin Bowes wrote:
James Antill wrote:
On Wed, 2008-05-28 at 15:49 +0100, Robin Bowes wrote:
I'm not sure what you mean here. What are group conditional packages ?
See man yum.conf for:
enable_group_conditionals
...it's kind of an extra type, where a package gets installed if
another
I have created a new yum-utils index page on the yum Wiki
http://wiki.linux.duke.edu/YumUtils
There is also subpages for each util and plugin.
The pages is populated with subjects and description from yum-utils.spec.
Feel free to add extra information :)
Tim
Florian Festi wrote:
Hi!
This is something that annoys me for quite a while. But it was clear to
me that it was not fixable in ther past. So now is time to speak up:
What about normalizing the PRCO data to a sane format. There is a wild
mixture of integers, comparator signs, two letter
Florian Festi wrote:
Hi!
While I like the fact that the yum wiki get used a bit more I still
think some things need to go into the source tree. What about starting a
file that documents the API changes so 3rd party vendors (and our own
developers) can see what we changed in the API for
Florian Festi wrote:
Patch is a bit lengthly as I needed to touch more or less all test
cases. But most changes are just query/replace. May be someone should
have a brief look over the few manual changes.
Florian
Florian Festi wrote:
Hi!
As already announced on IRC I still have a pile of patches that were
waiting for the 4.0.x branch to open.
These two switch to using generators instead of lists for the
Depsolve.check* family. This allows us to restart the checking when ever
the transaction is
yum-utils-1.1.14 is now released.
Check the changelog [1] for changes.
tarball:
http://linux.duke.edu/projects/yum/download/yum-utils/yum-utils-1.1.14.tar.gz
SRPM:
http://linux.duke.edu/projects/yum/download/yum-utils/yum-utils-1.1.14-1.src.rpm
[1]:
James Antill wrote:
So atm. we have a couple of super core plugins which add repos.
and/or tweak network stuff and so really should be on all the time. So
I'd like to propose that we change the defaults to plugins=True for all
of the yum-utils commands that don't have it atm.
Can anyone think
seth vidal wrote:
For those of you who are familiar with the innerworkings of comps -
there are two identifiers users can use to get to a group: groupid and
group name
the groupid is a single word
the group name can be multiple words
when multiple comps files are available yum merges them
seth vidal wrote:
On Sat, 2008-03-29 at 16:08 -0400, James Antill wrote:
On Sat, 2008-03-29 at 09:11 -0400, seth vidal wrote:
On Sat, 2008-03-29 at 02:22 -0400, James Antill wrote:
Maybe this was more obvious to others, but:
yum --enablerepo=development upgrade
...takes about twice as long
Jeremy Katz wrote:
On Tue, 2008-03-25 at 07:11 -0400, Tim Lauridsen wrote:
@@ -67,14 +68,15 @@ class RPMBaseCallback:
Base class for a RPMTransaction display callback class
'''
def __init__(self):
-self.action = { TS_UPDATE : 'Updating',
-TS_ERASE
James Bowes wrote:
Add the following options for --leaves:
* --exclude-devel :: Any package whose name ends in devel is not a
leaf.
* --exclude-bin :: Any package that has a file with 'bin' in the
name is not a leaf.
* --leaf-regex:: A regular
yum-utils-1.1.12 is now released.
Check the changelog [1] for changes.
tarball:
http://linux.duke.edu/projects/yum/download/yum-utils/yum-utils-1.1.12.tar.gz
SRPM:
http://linux.duke.edu/projects/yum/download/yum-utils/yum-utils-1.1.12-1.src.rpm
[1]:
James Antill wrote:
So while I should have been asleep I did this, please don't look if
you are feeling queasy, and I'm pretty sure it still works the same.
Basic idea behind it is that we pre-load the entire requires DB into a
python hash+arrays, as we are losing out when we call
seth vidal wrote:
Hi folks,
If anyone wants to help some of us sift old yum bug reports and close
out some of them come by #yum on irc.freenode.net on thursday march 12,
2008. Pretty much any timezone you want. Someone will probably be up.
bug reports for yum and friends are here:
seth vidal wrote:
On Sat, 2008-03-01 at 17:58 -0500, James Antill wrote:
So as some of you know, I've been working on doing a yum-verify plugin
to add a verify command ... based on the pkg.verify() work Seth did.
I've finally finished it, and verify-rpm does the same thing as rpm -V
and
seth vidal wrote:
Just somethings a few of us have talked about in the past few days:
1. createrepo --noxml - generates ONLY the sqlite dbs, no xml at all
2. yum --break-up-transaction: Takes a resolved yum transaction and
attempts to break it up into as many discrete, dependency-closed sets
i have some problems with using YumBase.searchGenerator from the
PackageKit yum2 backend (dbus based backend).
searchlist = ['name', 'summary', 'description', 'group']
key = 'firmware'
res = self.yumbase.searchGenerator(searchlist, [key])
for
seth vidal wrote:
On Thu, 2008-02-21 at 11:10 +0100, Tim Lauridsen wrote:
i have some problems with using YumBase.searchGenerator from the
PackageKit yum2 backend (dbus based backend).
searchlist = ['name', 'summary', 'description', 'group']
key = 'firmware
Florian Festi wrote:
Hi!
This is a revisited patch to finally add a set of unresolved transaction
members to tsInfo that is maintained by the tsInfo object itself. That
way it cannot get outdated when plugins or other code pieces add or
remove pkgs form the transaction behind the back of the
Florian Festi wrote:
Hi!
Reimplementing .processConflict() is on the todo list for too long now.
The first patch removes pkgs from updates (self.up) to avoid errors when
trying to update to an already skipped package (which is no longer found
in the .pkgSack). The second patch is the new
James Bowes wrote:
On Tue, Feb 12, 2008 at 12:56:44PM +0100, Tim Lauridsen wrote:
James Antill wrote:
No because a txmbr isn't created and nothing is added to the pkgdict,
but the bigger picture here is that IMO this shouldn't be part of the
Transaction Data class.
Although it'd be nice
of
the Depsolve class?
I agree the skipped packages don't fit nicely into tsInfo.
I have created a patch to do it in another way.
Tim
From 93d6e50123372ef0dd11897f79d330f5a2a52451 Mon Sep 17 00:00:00 2001
From: Tim Lauridsen [EMAIL PROTECTED]
Date: Tue, 12 Feb 2008 12:54:52 +0100
Subject: [PATCH
Hans-Peter Jansen wrote:
Am Freitag, 8. Februar 2008 schrieb Dr. Peter Poeml:
On Fri, Feb 08, 2008 at 11:22:27AM +0100, Hans-Peter Jansen wrote:
In spite of the fact that yum installed the key in a correct way, it
failed to use it in the same session, and succeeds in subsequent
sessions only.
Neal Becker wrote:
If you run yum-cron, and have yum-merge-conf, then yum will go into an
infinite loop generating an infinite email.
Previously this was really annoying. Today if filled my disk to 100%.
I guess what's needed is for yum-merge-conf to detect whether the tty is
actually a
Jeremy Katz wrote:
On Thu, 2008-02-07 at 04:59 -0500, Tim Lauridsen wrote:
Make the _() translation wrapper make sure that the translated string is
Unicode, to solve problem with ISO-8859-1 translation files
It's probably easier to just make sure the po/mo files have UTF-8
instead. See
seth vidal wrote:
How do all the committers feel about a 3.2.11 release soon ish b/c of
all the new stuff we've added/fixed since 3.2.10?
-sv
___
Yum-devel mailing list
Yum-devel@linux.duke.edu
https://lists.dulug.duke.edu/mailman/listinfo/yum-devel
python console.
Anybody have any objections to adding this patch.
Tim
From 724690d2e6adb88dc8c7240c35588a1607c0a85b Mon Sep 17 00:00:00 2001
From: Tim Lauridsen [EMAIL PROTECTED]
Date: Sun, 3 Feb 2008 14:11:23 +0100
Subject: [PATCH] * make package object show some useful info when printed
James Antill wrote:
On Sun, 2008-02-03 at 14:19 +0100, Tim Lauridsen wrote:
Today if you run the attached test script, you will get this output.
[EMAIL PROTECTED] yum]$ sudo python ~/yum-repr-test.py
[yum.packages.YumInstalledPackage object at 0xb73c56c
yum-utils-1.1.11 is now released.
Check the changelog [1] for changes.
tarball:
http://linux.duke.edu/projects/yum/download/yum-utils/yum-utils-1.1.11.tar.gz
SRPM:
http://linux.duke.edu/projects/yum/download/yum-utils/yum-utils-1.1.11-1.src.rpm
[1]:
Arjun Roy wrote:
Howdy - my name's Arjun Roy, and this is my first time posting on this
list.
I'm an undergraduate cs student, and am interested in yum (among various
other things).
Alright, with that out of the way, I'd like to submit a small plugin for
yum. It's a tiny
history plugin - it
seth vidal wrote:
Something we talked about before. This is a simple fs-as-a-db backended
additional database. It fits in around the rpmsack and the
YumInstalledPackage objects to let you have fairly seamless access to
this information from any given installed package object
Things I'm thinking
Hi
I a moment of insanity, i have started the process of translation
support to yum.
* I have moved i18n.py to yum/i18n.py to make it possible to translate
yum base.
* i have add po/Makefile with the magic pieces to
generate/update/install yum.pot, *.po etc.
* i have added to stuff to
Look like we have forgot to add 'Obsoletes: yum-skip-broken' to
yum.spec, else it will fail for uses there have yum-skip-broken installed.
I will release a new yum-utils release where
yum-skip-broken conflict with yum 3.2.8, but the obsoletes is needed
before i do that.
Tim
James Antill wrote:
This does the yum list vendors feature from the TODO list[1], as well
as anything else useful I could think of in a generic package.
Any objects to commiting? Any requests for other fields?
I'd also like to do a similar plugin which does filtering, Ie. exclude
all
James Bowes wrote:
On Thu, Jan 17, 2008 at 07:38:34PM -0500, seth vidal wrote:
On Thu, 2008-01-17 at 19:37 -0500, James Bowes wrote:
Hey all:
Attached is a possible fix to our one failing unit test (where 'yum
install kernel' will bring in the oldest uninstalled kernel, and
installonlyn
Michael E Brown wrote:
On Mon, Jan 14, 2008 at 04:19:36PM +0100, Tim Lauridsen wrote:
seth vidal wrote:
On Mon, 2008-01-14 at 09:32 -0500, Jeremy Katz wrote:
On Mon, 2008-01-14 at 09:18 -0500, seth vidal wrote:
Talking to Tim about plugins on the irc channel it occurred to me, to
keep
Look like there is a problem with pup/pirut and the aliases plugin.
https://www.redhat.com/archives/fedora-test-list/2008-January/msg00154.html
I have answered the user to open a bz report the problem.
The problem is the
from i18n import _
this only works when running from the cli, bacause
James Bowes wrote:
On Fri, Jan 11, 2008 at 12:50:21PM +0100, Tim Lauridsen wrote:
Tim Lauridsen wrote:
Hi
I have created a patch to make the depsolver error messages show more
information.
I objections to this ?
Tim
seth vidal wrote:
repodiff:
http://skvidal.fedorapeople.org/misc/repodiff.py
- take 2 repos on the cli
- return the added, removed and modified pkgs
do we already have something like this?
I don't think we have something like this, could be useful
+1 for adding it to yum-utils.
Tim
Hi
I have created a patch to make the depsolver error messages show more
information.
I objections to this ?
Tim
From f1ed8565bd595a6fcf0a2068d3693fa256c86732 Mon Sep 17 00:00:00 2001
From: Tim Lauridsen [EMAIL PROTECTED]
Date: Fri, 11 Jan 2008 12:20:05 +0100
Subject: [PATCH] make
Question of the day.
What problem does this test case illustrate.
def testProvidesAndDepsGetRemoved(self):
po1 = self.repoPackage('Spaceman')
po1.addProvides('money')
po2 = self.repoPackage('GutlessGibbon')
po2.addRequires('money')
Tim Lauridsen wrote:
begin captured stdout -
Checking deps for GutlessGibbon.noarch 0-1-0 - u
looking for ('money', None, (None, None, None)) as a requirement of
GutlessGibbon.noarch 0-1-0 - u
looking for ('nice', None, (None, None, None
Florian Festi wrote:
Again?
Yes, again...
Sorry, I was trying to install F8 Everything '*'
and the result of the skip-broken plugin is not very useful.
The attached patch goes back to remove just the actually problematic
packages. Result looks good on first sight.
Florian
It looks like
Florian Festi wrote:
seth vidal wrote:
I had a silly thought as the year ticked over last night. - Would it
make sense for us to put in an update nag for yum in yum with a hard
date in it. So, for example, if 3.2.12 is released in march 2008, we
have a check in the startup of the code that says
Florian Festi wrote:
Hi!
I am going to apply this patch tomorrow as it drops creating this lists
from 200s to 0.5s on a 9000pkg transaction.
have fun
Florian
___
Yum-devel
Jeremy Katz wrote:
There seems to be a bit of a theme of plugins classifying themselves as
non-interactive (TYPE_CORE) but still depending on some of the pieces
presented by the cli, especially with regards to option parsing. This
leads to pretty regular tracebacks for pirut and other non-cli
seth vidal wrote:
On Tue, 2007-12-18 at 18:00 +0100, Florian Festi wrote:
I changed Package.hash to ._hash to avoid API clutter. Responsiveness in
file-resolutions shouldn't be affected by the ._files change. Filelists
aren't loaded anyway. Or are you talking about the searchFiles
Tim Lauridsen wrote:
Tim Lauridsen wrote:
Florian Festi wrote:
Hi!
The attached patch adds profiling to yummain.py. It's not that
beautiful as it replaces all kind of system.exit() calls by returns.
I don't like all these exit calls. But if anyone really likes them we
can just catch
James Antill wrote:
So currently the yum.conf documentation says:
[repository] OPTIONS
The repository section(s) take the following form:
Example:
[repositoryid]
name=Some name for this repository
baseurl=url://path/to/repository/
seth vidal wrote:
On Fri, 2007-12-14 at 05:10 -0500, Florian Festi wrote:
yummain.py | 110 +
1 file changed, 67 insertions(+), 43 deletions(-)
New commits:
commit 60b388ac8d47ae69bc7d60b23354272e8d70b4d1
Author: Florian Festi [EMAIL
Tim Lauridsen wrote:
Tim Lauridsen wrote:
Florian Festi wrote:
Hi!
The attached patch adds profiling to yummain.py. It's not that
beautiful as it replaces all kind of system.exit() calls by returns.
I don't like all these exit calls. But if anyone really likes them we
can just catch
Tim Lauridsen wrote:
Florian Festi wrote:
Hi!
The attached patch adds profiling to yummain.py. It's not that
beautiful as it replaces all kind of system.exit() calls by returns. I
don't like all these exit calls. But if anyone really likes them we
can just catch the SystemExit exceptions
seth vidal wrote:
fedora 8 - i686 box:
yum list kernel
Installed Packages
kernel.i686 2.6.23.1-49.fc8installed
Available Packages
kernel.i586 2.6.23.8-63.fc8updates
kernel.i686 2.6.23.8-63.fc8updates
seth vidal wrote:
It's either too late or night or I'm losing my mind:
For cases where matching case insensitively nor matching against a
subset of packages from any given sack are not at issue why are we using
packages.parsePackages() vs just using sackobject.matchPackageNames()
Specifically
Florian Festi wrote:
Shouldn't
depTree = self._buildDepTree()
be done in the while loop as the tsInfo changes with each loop?
No need to build the depTree again, it is just used to find the relation
between packages original in the transaction and which packages they
drag in as dependcies.
Florian Festi wrote:
Tim Lauridsen wrote:
Florian Festi wrote:
Shouldn't
depTree = self._buildDepTree()
be done in the while loop as the tsInfo changes with each loop?
No need to build the depTree again, it is just used to find the relation
between packages original in the transaction
How do we handle the conflict with current yum HEAD including a
'--skip-broken' option in yum-cli with the yum-skip-broken plugin in the
best way.
1. Add a conflict yum-skip-broken to yum.spec
2. Add a conflict yum 3.2.8 in yum-skip-broken-plugin spec
3. Add s obsoletes yum-skip-broken in
Florian Festi wrote:
Sorry. It looks like the skip broken code works exactly the opposite way
I thought: It removes the packages that leads to the problem.
Yes it does.
I've commited two test cases that give an idea why this might not be the
best strategy (hmm, may be I should add some doc
seth vidal wrote:
On Tue, 2007-12-11 at 14:45 +0100, Florian Festi wrote:
Sorry. It looks like the skip broken code works exactly the opposite way I
thought: It removes the packages that leads to the problem.
I've commited two test cases that give an idea why this might not be the
best
seth vidal wrote:
On Tue, 2007-12-11 at 15:12 +0100, Florian Festi wrote:
Florian Festi wrote:
Sorry. It looks like the skip broken code works exactly the opposite way
I thought: It removes the packages that leads to the problem.
I've commited two test cases that give an idea why this might
Terje Rosten wrote:
* Tim Lauridsen
|
| I think my work on adding support for skipping packages causing
| depsolve problems in yumbase is ready for inclusion in yum git
| master. the changes in the depsolve is minor, nothing there should
| break anything.
rawhide is currently a excellent
Jeremy Katz wrote:
On Fri, 2007-12-07 at 17:09 -0500, seth vidal wrote:
So my question is - where should the calling of this go? Into a
yum-util? like package-cleanup or something of its own? Into yum proper
with a subcommand or maybe an automatic event (I see you have an
aborted
yum-utils-1.1.9 is now released.
Check the changelog [1] for changes.
tarball:
http://linux.duke.edu/projects/yum/download/yum-utils/yum-utils-1.1.9.tar.gz
SRPM:
http://linux.duke.edu/projects/yum/download/yum-utils/yum-utils-1.1.9-1.src.rpm
[1]:
Tim Lauridsen wrote:
Tim Lauridsen wrote:
Tim Lauridsen wrote:
Florian Festi wrote:
Hi Tim!
Some more skip broken code:
Fix the endless loop (sorry, test case is still broken). Code is not
that beautiful - feel free to adjust your personal taste ;)=
Tim Lauridsen wrote:
Added some patch
James Bowes wrote:
Hi all:
I just commited some changes to the test dir that replace ffesti's test
discovery code with nose[1] (python-nose in fedora). You'll need to
install that for 'make test' to work.
Beyond test discovery, nose also has a pile of plugins for fun things
like profiling and
Florian Festi wrote:
Florian Festi wrote:
Florian Festi wrote:
Hi!
Attached patch completes the behavior change Jeremy introduced with
5badc2b67b9885d2993b6f77423bc632b8595b01 - when an obsolete is
available for a package, only install most matching arch rather than
all available arches
Florian Festi wrote:
Tim Lauridsen wrote:
Tim Lauridsen wrote:
James Bowes wrote:
While fiddling with some test cases for package removal, I ran into
some
problems with transactioninfo. It looks at the class of a package to
determine if its from the rpmdb, from a repo, or a local rpm
Tim Lauridsen wrote:
Florian Festi wrote:
Hi Tim!
Some more skip broken code:
Fix the endless loop (sorry, test case is still broken). Code is not
that beautiful - feel free to adjust your personal taste ;)=
Tim Lauridsen wrote:
Added some patch to fix issues with skip-broken and updates
seth vidal wrote:
On Wed, 2007-11-28 at 09:30 -0500, Jesse Keating wrote:
What I'd like to propose instead is that yum's default failover setting
of roundrobin be changed to priority. This would allow any yum tool
using Fedora's Mirror Manager to return mirror lists work without
adjusting
Tim Lauridsen wrote:
Tim Lauridsen wrote:
Florian Festi wrote:
Hi Tim!
Some more skip broken code:
Fix the endless loop (sorry, test case is still broken). Code is not
that beautiful - feel free to adjust your personal taste ;)=
Tim Lauridsen wrote:
Added some patch to fix issues
Florian Festi wrote:
Hi Tim!
Some more skip broken code:
Fix the endless loop (sorry, test case is still broken). Code is not
that beautiful - feel free to adjust your personal taste ;)=
Tim Lauridsen wrote:
Added some patch to fix issues with skip-broken and updates.
If we remove
Tim Lauridsen wrote:
James Bowes wrote:
While fiddling with some test cases for package removal, I ran into some
problems with transactioninfo. It looks at the class of a package to
determine if its from the rpmdb, from a repo, or a local rpm. This makes
testing with fake packages hard
seth vidal wrote:
The attached patch would allow callers to define copy_local at the
repository level so they could make file:// repos copy the packages to
the cache path. The use case for this is really just pungi but it seems
to make sense to me. The reason I'm worried about it is that I want
Florian Festi wrote:
Hi!
While playing around with the skipbroken code I run into some problems
when transaction members are removed and the packages is readded later
on. Reason for this is that __cmp__ and __hash__ methods are
overloaded in the TransactionMember class. This means that
James Bowes wrote:
While fiddling with some test cases for package removal, I ran into some
problems with transactioninfo. It looks at the class of a package to
determine if its from the rpmdb, from a repo, or a local rpm. This makes
testing with fake packages hard.
The attached patch switches
e0d67275a269ee8164cbda61c18ccb0bd462cd55 Mon Sep 17 00:00:00 2001
From: Tim Lauridsen [EMAIL PROTECTED]
Date: Mon, 26 Nov 2007 15:04:48 +0100
Subject: [PATCH] Fixed skip-broken on updates
---
yum/__init__.py |6 ++
1 files changed, 6 insertions(+), 0 deletions(-)
diff --git a/yum/__init__.py b/yum/__init__.py
index
Florian Festi wrote:
Hi Tim!
I applied your patch to my gold branch to have a fixed base as needed
for the test cases. I also cleaned up some whitespace and other quirks.
The main change is adding two test cases that really go wrong. One
removes a previously installed package (by cleaning
Michael E Brown wrote:
I was thinking about a couple enhancements to mock, and one of the
things that came up was an 'offline' mode for mock, ie. the ability to
build packages when disconnected from the internet.
I believe that everything is now in place in mock 0.8.x to allow this.
Mock now
seth vidal wrote:
On Wed, 2007-11-21 at 08:04 +0100, Tim Lauridsen wrote:
seth vidal wrote:
On Fri, 2007-11-16 at 21:36 -0500, James Bowes wrote:
On Thu, Nov 15, 2007 at 11:22:42AM +0100, Tim Lauridsen wrote:
I have created at patch to add skip-broken
seth vidal wrote:
On Wed, 2007-11-21 at 08:04 +0100, Tim Lauridsen wrote:
seth vidal wrote:
On Fri, 2007-11-16 at 21:36 -0500, James Bowes wrote:
On Thu, Nov 15, 2007 at 11:22:42AM +0100, Tim Lauridsen wrote:
I have created at patch to add skip-broken
Florian Festi wrote:
Hi Tim!
Having a first look over you patches the following comes to my mind:
Tim Lauridsen wrote:
I have created at patch to add skip-broken functionality to YumBase.
I works the following way:
1. I have added some code to the depsolver to store packages with
problems
Adel Gadllah wrote:
Looks good to me, but i have some problems apply the patch.
please do a 'git diff origin basearchonly.patch' and send the file.
ok done (git diff master) .. file is attached.
Adel Gadllah wrote:
2007/11/18, Jeremy Katz [EMAIL PROTECTED]:
Really, looking at the plugin code, it should be of type
TYPE_INTERACTIVE as opposed to TYPE_CORE given that you're depending on
having things in the command line.
Which would also avoid the system-install-packages crash as
There is some kind of problem in current Yum git HEAD
i have added a little test program the Traceback that is failing.
it is working in yum-3.2.7-1.fc8
Tim
import yum
yb = yum.YumBase()
yb.doConfigSetup()
txmbr = yb.install(name='BackupPC')
if txmbr:
print Adding %s to transaction %
There is a proposal in this bugzilla report[1] about changing the name of the
plugins in yum-utils
from yum-pluginname to yum-plugin-pluginname.
I just want to hear what YOU are feeling about this change.
Tim
[1] http://devel.linux.duke.edu/bugzilla/show_bug.cgi?id=848
Terje Røsten wrote:
Jeremy Katz
Having a test suite only means that we know it doesn't break anything
which we have a test for. That's no substitute for reviewing changes
and thinking about the intent and implications
Of course, my posting was not a reply to your review, I just saw that
James Bowes wrote:
On Tue, Nov 06, 2007 at 03:53:47PM +0100, Florian Festi wrote:
Tim Lauridsen wrote:
Florian Festi wrote:
Tim Lauridsen wrote:
Have some problems applying the patch to yum master, it is
complaining about a missing file.
Ehm, true
I have run the yum test cases on a x86_64 system and on a i686 system.
On the x86_64 all tests run without problems on the i686 system 15 test
fails.
The test need to be grouped some how, so the multilib test not is
performed on a non multi lib systems.
Tim
1 - 100 of 201 matches
Mail list logo
