path: root/meeting-logs/20100902.txt

[21:14:55] <tampakrap> roll call
[21:15:02] <reavertm> 1
[21:15:05] <spatz> 2
[21:15:07] <dilfridge> 3
[21:15:18] <spatz> we lost abcd
[21:15:28] <ABCD> 4
[21:15:34] <tampakrap> and i am number five
[21:15:55] <tampakrap> i would like to have a small talk about amarok in the
end if jmbsvicetto manages to find us
[21:15:59] <tampakrap> either way:
[21:16:02] <ABCD> not even a quorum (for most definitions of "quorum")
[21:16:08] <tampakrap>
http://git.overlays.gentoo.org/gitweb/?p=proj/kde.git;a=blob;f=Documentation/maintainers/meetings/meeting-2010-09-02;h=e5a2764f178127e3f04e3311c12563b68a1b8dbe;hb=HEAD
[21:16:15] <tampakrap> here's the agenda
[21:16:23] <tampakrap> 1) KDE-4.5 status and plans to put it in Portage
[21:16:29] <tampakrap> reavertm: you can start
[21:16:39] <reavertm> ok, here it is:
[21:17:04] <reavertm> there are following issues I'd consider a blockers
[21:17:26] <reavertm> (unless we say - screw you users - upstream)
[21:17:53] <reavertm> https://bugs.kde.org/show_bug.cgi?id=230247
[21:18:47] <dilfridge> hmm here everything works fine _since_ the upgrade to
4.5.0
[21:19:02] <reavertm> result - you need to start any akonadi-client app twice
in order to use it (first start will make akonadi start bug won't detect that
it started)
[21:19:18] <dilfridge> but I'm not logging in/out very often
[21:19:38] <reavertm> one blocker - kwin FBO bug has been fixed (commit
reverted as I requested)
[21:19:56] <reavertm> remaining ugly bugs: multiple plasma glitches in system
tray
[21:20:20] <tampakrap> plasma glitches were highly reduced in 4.5.1
[21:20:31] <dilfridge> not seen for a while
[21:20:37] <reavertm> https://bugs.kde.org/show_bug.cgi?id=246931
[21:20:45] <tampakrap> if the akonadi bug is the only blocker i'd say let's
put it in tree, and let the people know about it
[21:20:46] <reavertm> no, they are still there
[21:21:02] <reavertm> tampakrap: and spam out bugzilla for no reason?
[21:21:20] <dilfridge> ah that yes that is still there... I got used to that
bug...
[21:21:27] <reavertm> https://bugs.kde.org/show_bug.cgi?id=247144 - another
plasma glitch (folderview)
[21:21:39] <reavertm> and most important - memory leaks in plasma-desktop
[21:21:49] <reavertm> (minor leaks, but they are there)
[21:22:19] <reavertm> dolphin tooltips issues as well as nepomuk related
crashes seen to be gone at least
[21:22:49] <tampakrap> i'd suggest let's put it in tree, but announce those
bugs and make it clear that it's not going to be stabilized
[21:22:50] <reavertm> oh, and one bug present in 4.4 alteady -
https://bugs.kde.org/show_bug.cgi?id=247204
[21:23:07] <tampakrap> i'm getting enough spam on IRC already, and i think it
is usuable enough
[21:23:21] <tampakrap> that's my opiniion, but it is your call mostly
[21:23:24] <reavertm> with that state I don't think *any* 4.5 patch release is
going to be stabilized, I won't allow it
[21:23:31] <spatz> there are regressions in every major release but that's why
we have testing. 4.5.1 seems good enough for tree
[21:23:50] <dilfridge> seconded (I personally can work with it fine)
[21:23:53] <reavertm> unless we clearly communicate - "we know it's broken -
we reported bugs and they keep being ignored"
[21:24:17] <tampakrap> reavertm: give me a list of the bugs, i can put them on
the meeting summary
[21:24:18] <spatz> 4.4 took a lot of time to be stabilized because of
regressions too, that's how kde works, apparently
[21:24:26] <tampakrap> and put the summary on my blog on planet gentoo and kde
[21:24:32] <reavertm> I won't wrange any 4.5 bugs in our bugzilla, I want to
make it clear :P
[21:25:09] <reavertm> spatz: plasma devs utilized 'Broken Development Model"
that's why
[21:25:31] <spatz> I'm not pointing fingers, just stating facts :)
[21:26:05] <tampakrap> ok, we made our points, reavertm: your call
[21:26:13] <reavertm> tampakrap: fine, but you'll blog about it :) (with bug
links there)
[21:26:14] <tampakrap> since you fixed a couple of upstream bugs
[21:26:23] <reavertm> aah, one more blocker
[21:26:32] <reavertm> but on our side - handbooks handling
[21:26:38] <dilfridge> yes that is true
[21:26:51] <dilfridge> something is seriously broken there
[21:27:01] <tampakrap> bug #?
[21:27:28] <dilfridge> bug 296345
[21:27:29] <reavertm> since 4.5 - docbooks are no longer bundled - so for
every +handbook we need to pull docbook stuff like for kdelibs
[21:27:30] <willikins> dilfridge: https://bugs.gentoo.org/296345 "kde 4.4.4:
Compilation error when creating help search index - bad xslt pattern"; Gentoo
Linux, KDE; NEW; ashl1future@gmail.com:kde@g.o
[21:27:49] <dilfridge> ah different problem
[21:27:52] <reavertm> err, i meant different handbook issue
[21:28:40] <tampakrap> if there is no bug #, let's open one, fix that first
and then put 4.5.1 to tree
[21:29:12] <reavertm> I can fix it anyway (my handbook isse) - for instance by
introducing KDE_HANDBOOK eclass variable instead of +handbook in IUSE (that
could hold additional handbook dirs as well)
[21:29:41] <tampakrap> ok
[21:29:49] <tampakrap> anything else on this?
[21:30:10] <reavertm> I think this is it wrt kde 4.5
[21:30:29] <tampakrap> i have one more thing to say about it
[21:30:36] <tampakrap> not so relevant
[21:31:11] <reavertm> ?
[21:31:12] <tampakrap> http://bugs.gentoo.org/show_bug.cgi?id=335123
[21:31:43] <reavertm> hmm, works for me...
[21:31:45] <tampakrap> these patches don't apply to kdepimlibs 4.5.0 and i
don't know if by applying them to 4.4.5 only will break 4.5 as well
[21:31:59] <tampakrap> i can reproduce here and the patches worked on my 4.4.5
laptop
[21:32:23] <tampakrap> so, any ideas?
[21:32:36] <reavertm> does it affect only kde 4.5 + kdepim 4.4?
[21:33:00] <reavertm> or it's some general bug?
[21:33:01] <tampakrap> the patches are meant for 4.4.5 only, haven't tested to
4.5
[21:33:32] <tampakrap> but i don't want to introduce a new bug by applying the
patches to kmail 4.4.5 and not to kdepimlibs 4.5.x
[21:34:08] <reavertm> I'd prefer them to appear in svn..
[21:34:13] <dilfridge> never had this problem here
[21:34:51] <tampakrap> ah they are not taken from svn
[21:34:54] <tampakrap> nice point
[21:35:04] <tampakrap> ok thanks
[21:35:10] <tampakrap> next issue i guess
[21:35:20] <tampakrap> 2) KOffice 2.2 status
[21:35:35] -*- reavertm shuts up
[21:35:46] <tampakrap> i was away because of my thesis and vacations for too
long, anyone knows the blockers of keeping this away from tree?
[21:35:48] <dilfridge> needs bump to 2.2.2 (I had no time yet but that should
not be difficult)
[21:36:10] <dilfridge> there is a "workaround"
[21:36:13] -*- reavertm disagrees, he saw multiple cmake changes and file/dir
additions/removals
[21:36:24] <dilfridge> ok have not checked yet
[21:36:58] <dilfridge> considering kchart
[21:37:10] <reavertm> I see when do svn update in koffice occassionally, it
needs someone to closely follow up - and in our case when it's been abandoned
for a while - full review from scratch likely
[21:37:27] <dilfridge> The libraries do not link if kchart is not built at the
same time, so what I did was
[21:37:48] <dilfridge> do make a dummy ebuild for kchart for the meantime.
[21:37:51] <reavertm> KMCOMPILECONLY can be used
[21:38:06] <dilfridge> well it is done in koffice-libs
[21:38:17] <dilfridge> but I wanted to keep something named kchart in portage
[21:38:26] <dilfridge> so later upgrade remains painless
[21:38:42] <reavertm> is it standalone app?
[21:38:55] <dilfridge> so, now koffice-libs also installs kchart, and kchart
does, well, nothing
[21:39:13] <dilfridge> no I think only library
[21:39:37] <tampakrap> i see
[21:39:43] <dilfridge> apart from that nobody gave me negative feedback on
2.2.1 yet
[21:39:50] <tampakrap> i'd appreciate if there were more comments on the bug
[21:39:56] <tampakrap> http://bugs.gentoo.org/show_bug.cgi?id=322147
[21:39:59] <reavertm> then should be in koffice-libs imho - I'd like to avoid
creating excessive libraries like used to be in kdepim
[21:40:00] --> pesa
(~Pesa@host169-125-dynamic.52-82-r.retail.telecomitalia.it) has joined
#gentoo-meetings
[21:40:02] <tampakrap> and i could help you with it
[21:40:20] <tampakrap> i agree with that
[21:40:48] <dilfridge> fine with me... but the plan is that kchart can be
built separately again with 2.3
[21:41:33] <tampakrap> is it already done in the live ebuilds?
[21:41:47] <dilfridge> did not check yet
[21:41:51] <tampakrap> ok
[21:41:51] <reavertm> maybe they'll just add some host for that kpart, then
kchart ebuilds will be justified, (but libkchart should still be in
koffice-libs imho)
[21:42:28] <reavertm> it can be embedded in any koffice app after all
[21:43:07] <tampakrap> ok, i'm done with it, dilfridge any further comments?
[21:43:15] <dilfridge> I'm kind-of-away from tomorrow for a week, so if
anything should be done quickly I'm out
[21:43:30] <dilfridge> but I guess its not that urgent
[21:43:34] -*- reavertm needs to remember how it was decided - koffice-libs
contains full functionality and kword for instance only word processor
application (but kpart is provided by koffice-libs) - or installing kword
makes kpart available
[21:43:35] <tampakrap> indeed
[21:44:08] <tampakrap> if the kpart is used by other ebuilds too it should be
in koffice-libs
[21:44:15] <tampakrap> else the split makes no sense
[21:44:35] <dilfridge> we can find that out with the dependencies / linking
[21:44:47] <dilfridge> since now all apps are for sure compiled after kchart
[21:44:52] <reavertm> no, kparts are runtime-only things mostly
[21:45:03] <dilfridge> ok true
[21:45:20] <reavertm> (well, buildtime+runtime, but can be missing at runtime)
[21:45:45] <dilfridge> i'd say we could give 2.2.1 a try
[21:45:59] -*- dilfridge keeps fingers crossed
[21:46:11] <tampakrap> ok
[21:46:24] <tampakrap> next?
[21:46:27] <reavertm> fine, but I'd suggest full review of them (for instance
to check whether all koffice subdirs are packaged in ebuilds)
[21:46:54] <dilfridge> reavertm: teach me how to do that, please! :) (in two
weeks)
[21:47:35] <tampakrap> ok next
[21:47:49] <tampakrap> for next issue i would like alexxy but he seems not to
be here
[21:47:57] <reavertm> simple - take a look at koffice/ and check subdirs one
by one against koffice ebuilds (KMEXTRA, KMCOMPILEONLY, KMEXTACTONLY)
[21:48:25] <dilfridge> ok
[21:48:48] <tampakrap> 3) KDEPIM beta packages are released
[21:48:51] <tampakrap> i'll skip it
[21:49:00] <tampakrap> and add a gentoo-desktop thread instead
[21:49:05] <reavertm> we just want whole koffice covered (or some parts
*intentionally* skipped)
[21:49:14] <dilfridge> ok
[21:49:27] <reavertm> skip it please
[21:49:36] <reavertm> I don't want kdepim betas in overlay :P
[21:49:47] <reavertm> (there are live ebuilds already)
[21:50:11] <tampakrap> sebas stated in his blog that upgrade and downgrade
would be easy since configs don't affect each other
[21:50:50] <tampakrap> either way, we'll discuss it in gentoo-desktop
[21:50:55] <tampakrap> 4) open floor
[21:51:20] <dilfridge> about digikam
[21:51:41] <dilfridge> the patches are cleaned up a lot, but it's not
completely finished yet
[21:52:02] <-- spatz (~spatz@gentoo/developer/spatz) has quit (Ping timeout:
276 seconds)
[21:52:23] <dilfridge> reavertm: if you want to have a look at it go ahead,
but dont send anything upstream yet.
[21:52:46] <dilfridge> at least 1.4.0 should now build fine
[21:53:04] <dilfridge> I'll talk with the sci guys about clapack.
[21:53:11] <reavertm> I'll wait
[21:53:21] <tampakrap> i'm also going to unmask knerworkmanager very soon,
please test
[21:53:34] <tampakrap> i'm a bit worried about the versioning though
[21:54:47] <reavertm> ah, as of open floor - I encourage people working with
live and tagged ebuilds to use kde-misc/kde-overlay-servicemenus
[21:55:07] <dilfridge> what does it do?
[21:56:00] <reavertm> I've added simple Compare/Merge service menu - in
dolphin select two files one by one to have them open in kompare (where you
can interactively sync them)
[21:56:15] <reavertm> to avoid manual copy paste
[21:56:17] <tampakrap> w00t
[21:56:19] <dilfridge> nice
[21:56:43] <tampakrap> last issue, i'm going to remove a few inactive people
from the team
[21:57:14] <tampakrap> and if there is not anything else, i guess i should
close the meeting
[21:57:33] <reavertm> thanks
[21:58:08] <tampakrap> meeting is over