[Mageia-dev] [changelog] [RPM] cauldron core/release kdelibs4-4.9.0-2.mga3
Colin Guthrie
mageia at colin.guthr.ie
Wed Aug 1 21:31:15 CEST 2012
'Twas brillig, and Shlomi Fish at 01/08/12 11:00 did gyre and gimble:
> On Wed, 01 Aug 2012 03:44:32 +0200
> Nicolas Lécureuil <nicolas.lecureuil at free.fr> wrote:
>
>> Le mardi 31 juillet 2012 22:20:45 Balcaen John a écrit :
>>> Le mardi 31 juillet 2012 18:42:00 tv a écrit :
>>>> Name : kdelibs4 Relocations: (not relocatable)
>>>> Version : 4.9.0 Vendor: Mageia.Org
>>>> Release : 2.mga3 Build Date: Tue Jul 31
>>>> 18:15:54
>>>> 2012 Install Date: (not installed) Build Host:
>>>> ecosse.mageia.org Group : Graphical desktop/KDE Source RPM:
>>>> (none)
>>>> Size : 11884847 License: GPLv2 and LGPLv2
>>>> and GFDL and BSD and QPL and Artistic Signature : (none)
>>>> Packager : tv <tv>
>>>> URL : http://www.kde.org
>>>> Summary : K Desktop Environment - Libraries
>>>> Description :
>>>> Libraries for the K Desktop Environment.
>>>>
>>>> tv <tv> 2:4.9.0-2.mga3:
>>>> + Revision: 276808
>>>> - rebuild for new libattica
>>>>
>>>> + mikala <mikala>
>>>>
>>>> - Silent: Fix Source URL
>>>> - Update tarball to KDE 4.9.0
>>>
>>> Oh damn that !!!
>>> kde 4.9.0 is not supposed to be released & now because a padawan without a
>>> commit right send a patch, his mentor feels obligate to push the package
>>> *without* checking the side effect of this change (yeah it's a update &
>>> we're late let's push it ) of course we could have stay only with kdelibs
>>> 4.9.0 but now i noticed that fwang is pushing the whole kde stack
>>> especially when we're not sure that some kde package are not going to be
>>> respun ....
>>> damn that !
>>>
>>> I guess i'll won't commit anything more if it's just to see some stuff like
>>> this .
>>
>> please don't, we need you :(
>>
>> Rindolf we can't loose packagers like mikala, you really need to follow rules.
>>
>> If you need more infos about rules please ask we are here to help but don't
>> touch mikala's packages w/o his authorisation.
>>
>> just do a mgarepo maintdb get $package and see who is the maintainer and
>> ask before commiting.
>
> OK, I'll keep it in mind. I thought the change was harmless because it didn't
> bump the library's libmajor, but apparently the KDE packages require an exact
> version 0.3 of attica's libraries which seems broken to me.
I have to agree here that something is "funny" in the libattica package
which ultimately helped to contribute to this issue.
e.g. on my system before update (tho' with similar results after):
[colin at jimmy ~]$ rpm -q --provides lib64attica0
libattica.so.0.3()(64bit)
lib64attica0 = 0.3.0-1.mga2
lib64attica0(x86-64) = 0.3.0-1.mga2
[colin at jimmy ~]$ rpm -ql lib64attica0
/usr/lib64/libattica.so.0.3
/usr/lib64/libattica.so.0.3.0
IMO this *should* be:
[colin at jimmy ~]$ rpm -q --provides lib64attica0
libattica.so.0()(64bit)
lib64attica0 = 0.3.0-1.mga2
lib64attica0(x86-64) = 0.3.0-1.mga2
[colin at jimmy ~]$ rpm -ql lib64attica0
/usr/lib64/libattica.so.0
/usr/lib64/libattica.so.0.3.0
So I can see how this mistake was made and TBH I could have made the
same mistake myself (with the caveat that I likely would not have bumped
the version of someone else's package with out confirming first and that
it should have been obvious from testing and installing the build)
But either way this seems like an issue to fix properly (possibly with
an upstream fix or some modification to the library policy when the
minor version is "presented" like this).
FWIW there are some other examples I can find:
This one just looks "odd" generally:
[colin at jimmy ~]$ rpm -qf /usr/lib64/libxenctrl.so.4.0
lib64xen3.0-4.1.2-4.mga2
[colin at jimmy ~]$ rpm -q --provides lib64xen3.0
libblktap.so.3.0()(64bit)
libblktapctl.so.1.0()(64bit)
libflask.so.1.0()(64bit)
libfsimage.so.1.0()(64bit)
libvhd.so.1.0()(64bit)
libxenctrl.so.4.0()(64bit)
libxenguest.so.4.0()(64bit)
libxenlight.so.1.0()(64bit)
libxenstore.so.3.0()(64bit)
libxlutil.so.1.0()(64bit)
lib64xen3.0 = 4.1.2-4.mga2
lib64xen3.0(x86-64) = 4.1.2-4.mga2
This one is even more strange as the "major" appears to be the
concatenation of the major and minor:
[colin at jimmy ~]$ rpm -q --provides lib64ice33
libFreeze.so.33()(64bit)
libGlacier2.so.33()(64bit)
libIce.so.33()(64bit)
libIceBox.so.33()(64bit)
libIceGrid.so.33()(64bit)
libIcePatch2.so.33()(64bit)
libIceSSL.so.33()(64bit)
libIceStorm.so.33()(64bit)
libIceStormService.so.33()(64bit)
libIceUtil.so.33()(64bit)
libIceXML.so.33()(64bit)
libSlice.so.33()(64bit)
lib64ice33 = 3.3.1-5.mga2
lib64ice33(x86-64) = 3.3.1-5.mga2
[colin at jimmy ~]$ rpm -ql lib64ice33
/usr/lib64/libFreeze.so.3.3.1
/usr/lib64/libFreeze.so.33
/usr/lib64/libGlacier2.so.3.3.1
/usr/lib64/libGlacier2.so.33
/usr/lib64/libIce.so.3.3.1
/usr/lib64/libIce.so.33
/usr/lib64/libIceBox.so.3.3.1
/usr/lib64/libIceBox.so.33
/usr/lib64/libIceGrid.so.3.3.1
/usr/lib64/libIceGrid.so.33
/usr/lib64/libIcePatch2.so.3.3.1
/usr/lib64/libIcePatch2.so.33
/usr/lib64/libIceSSL.so.3.3.1
/usr/lib64/libIceSSL.so.33
/usr/lib64/libIceStorm.so.3.3.1
/usr/lib64/libIceStorm.so.33
/usr/lib64/libIceStormService.so.3.3.1
/usr/lib64/libIceStormService.so.33
/usr/lib64/libIceUtil.so.3.3.1
/usr/lib64/libIceUtil.so.33
/usr/lib64/libIceXML.so.3.3.1
/usr/lib64/libIceXML.so.33
/usr/lib64/libSlice.so.3.3.1
/usr/lib64/libSlice.so.33
And finally here is one that provides symlinks at both levels of major:
[colin at jimmy ~]$ rpm -q --provides lib64qtxml4
qtxmllib = 4:4.8.2
lib64qtxml4 = 4:4.8.2-4.mga3
lib64qtxml4(x86-64) = 4:4.8.2-4.mga3
libQtXml.so.4()(64bit)
[colin at jimmy ~]$ rpm -ql lib64qtxml4
/usr/lib64/libQtXml.so.4
/usr/lib64/libQtXml.so.4.8
/usr/lib64/libQtXml.so.4.8.2
Not an exhaustive list, just some random looks at my /usr/lib64 folder.
So can someone familiar with the project have a bit of a think about it
and maybe see if they can fix it properly in a future proof way?
> BTW, you should not call me "Rindolf" outside IRC. My name is Shlomi.
Provided it doesn't offend you too much, most of us tend to chop and
change between real names and IRC nicks even on mailing lists. I'll
quite often refer to ennael, tmb, blino or tv etc. In some cases it's
because I can't remember the exact spelling of names and sometimes it's
pure laziness! While I'll try to remember your preference I hope you
don't take active office at it (otherwise it's a particularly bad choice
of IRC nicks :p).
FWIW I still get surprised when people call me "Col" rather than "Colin"
even tho' that's how I sign pretty much all my emails - same deal there
I suppose :)
Col(in)[g]
:p
--
Colin Guthrie
colin(at)mageia.org
http://colin.guthr.ie/
Day Job:
Tribalogic Limited http://www.tribalogic.net/
Open Source:
Mageia Contributor http://www.mageia.org/
PulseAudio Hacker http://www.pulseaudio.org/
Trac Hacker http://trac.edgewall.org/
More information about the Mageia-dev
mailing list