2008/02/28/#ubuntu-motu.txt

[00:02] <Arwen> though maybe that's gcc's fault now that I think of it..
=== Pricey is now known as PriceChild
[00:51] <LaserJock> mok0: ping
[00:51] <emgent> heya
[00:51] <mok0> LaserJock: pong
[00:52] <LaserJock> mok0: regarding rasmol
[00:52] <mok0> yes
[00:52] <LaserJock> are you gonna do a new debdiff?
[00:52] <mok0> LaserJock: eeerr, I kinda think my icons are prettier
[00:52] <LaserJock> I think perhaps linking to the 32x32 icon might be better than the 48x48
[00:53] <LaserJock> mok0: eeerr, I don't really care ;-)
[00:53] <LaserJock> you're creating a 300+KB diff because you think the icons are "prettier"?
[00:53] <mok0> LaserJock: Well, I spent time on the icon, and generated it in all the different sizes so I think that is the correct solution
[00:53] <LaserJock> no, it's not
[00:54] <LaserJock> we only need 1 icon, and we already have it
[00:54] <LaserJock> it's creating a lot more work for us
[00:54] <mok0> LaserJock: ?
[00:54] <LaserJock> there's  already and icon
[00:54] <mok0> LaserJock: I've already done the work
[00:54] <LaserJock> it works just fine
[00:55] <LaserJock> mok0: yeah, I'm sorry about that
[00:55] <mok0> LaserJock: attention to detail?
[00:55] <LaserJock> but there's just no way we should upload that
[00:55] <mok0> LaserJock: lots of packages populate the /usr/share/apps directory
[00:56] <LaserJock> *if* there currently isn't anything
[00:56] <LaserJock> we have an icon
[00:56] <LaserJock> and we don't need a 300KB diff
[00:56] <mok0> LaserJock: ok, well invalidate the bug then
[00:56] <LaserJock> well, there is a bug, in that we need to add a symlink
[00:56] <LaserJock> because the icon isn't showing up
[00:57] <mok0> LaserJock: right
[00:57] <LaserJock> but you probably should have looked at that before doing all the work to create all those new icons
[00:57] <LaserJock> you could also send them to the rasmol authors
[00:57] <LaserJock> although rasmol is basically a dead upstream
[00:58] <mok0> LaserJock: I have a script creating the icons, it's gimping the largest one that took the time.
[00:58] <mok0> I could perhaps replace the xpm
[00:58] <LaserJock> anyway, I'm not upset or anything, and I'm sorry about the work you've done
[00:59] <LaserJock> but when we've already got an icon it's a good idea to use it
[00:59] <mok0> LaserJock: I accept the decision, but I don't agree with it
[00:59] <mok0> When the icon gets enlarged it looks really bad
[00:59] <mok0> That
[01:00] <LaserJock> when should they get enlarged?
[01:00] <mok0>  that is why you need all those sizes
[01:00] <mok0> That is up to the user.
[01:00] <mok0> You can choose icon size in konqueror
[01:00] <LaserJock> yeah, but I doubt you're gonna get a menu icon over 48x48
[01:00] <mok0> right
[01:01] <mok0> There's a lot of talk about scalable icons on d-d
[01:02] <mok0> Well, I'll change it to a link. *sigh*
[01:03] <LaserJock> hmm
[01:03] <LaserJock> the icons are done by Debian I guess
[01:03] <LaserJock> perhaps send an icon to the Debian maintainer
[01:03] <mok0> LaserJock: Sure
[01:04] <mok0> LaserJock: they are talking about a transition towards SVG icons
[01:05] <LaserJock> sure
[01:05] <LaserJock> who knows how long that'll take ;-)
[01:05] <mok0> LaserJock: good question :)
[01:05] <LaserJock> do they want to create pngs at build time?
[01:06] <mok0> LaserJock: No, they want SVG icons that are scalable
[01:06] <mok0> real time scaling
[01:06] <LaserJock> yikes
[01:06] <LaserJock> that's a bit of a performance hit
[01:06] <LaserJock> hopefully we can do both
[01:06] <mok0> LaserJock: that was the discussion :)
[01:07] <LaserJock> we've been talking a bit about that for Edubuntu because there's a new icon theme that'd be great
[01:07] <LaserJock> but it's mostly svg
[01:07] <mok0> LaserJock: OS X does some kind of real time scaling of icons
[01:07] <LaserJock> and on thin clients or old hardware it seems like there might be a significant amount of performance issues
[01:08] <mok0> LaserJock: yeah
[01:08] <mok0> what's the new icon theme for edubuntu?
[01:09] <LaserJock> well, we haven't moved over yet or anything
[01:09] <LaserJock> it's called Ubuntoon
[01:09] <LaserJock> based on Gartoon
[01:09] <mok0> For applications etc?
[01:10] <LaserJock> the icons are in a bzr branch at https://edge.launchpad.net/ubuntoon/
[01:10] <LaserJock> yeah
[01:10] <LaserJock> it's a decently full theme now
[01:10] <mok0> Cool.
[01:11] <mok0> I'd like to see a Kedubuntu :-)
[01:11] <LaserJock> yes, we've had work on that
[01:11] <LaserJock> there's an edubuntu-desktop-kde package for that
[01:11] <mok0> It'd be cool to get away from the dull blue KDE themes
[01:11] <LaserJock> since like half of our edu apps are KDE anyway ;-)
[01:13] <mok0> LaserJock: I'll upload another debdiff for rasmol tomorrow. It's getting late in my timezone
[01:14] <mok0> LaserJock: Please unsubscribe u-u-s
[01:14] <LaserJock> I did
[01:14] <LaserJock> rasmol is in Main
[01:14] <LaserJock> :-)
[01:14] <mok0> Oh
[01:15] <LaserJock> feel free to ping me and I'll sponsor it for you
[01:15] <mok0> LaserJock: Thanks
[01:15] <mok0> There are several other apps that lack icons
[01:15] <mok0> Is there an effort to go through them?
[01:15] <LaserJock> I think it's a good idea
[01:16] <LaserJock> we need to make sure that there actually isn't an icon
[01:16] <LaserJock> sometimes they just don't get installed or stuff like that
[01:16] <mok0> Perhaps we should make a list on the wiki?
[01:16] <LaserJock> and then working well with upstream to get the icons included
[01:16] <LaserJock> since that's really not diff we want to maintain
[01:17] <mok0> I think the priority is getting them working in hardy, then we can contribute upstream once hardys window has closed
[01:17] <LaserJock> right
[01:18] <LaserJock> set up a wiki page if you want
[01:18] <mok0> OK, will do
[01:18] <LaserJock> I think it'd be worthwhile
[01:18] <mok0> Me too. It makes the ui look much better
[01:19] <mok0> It looks kludgy when a fourth of the icons are missing
[01:19] <mok0> sloppy
[01:19] <LaserJock> yeah
[01:19] <LaserJock> just make sure
[01:20] <mok0> yes?
[01:20] <LaserJock> that it actually needs an icon
[01:20] <LaserJock> rather than having an installation issue
[01:21] <LaserJock> or something like that
[01:22] <mok0> LaserJock: hmmm, I will make the list of the icons missing on my system, but I don't think I have time to check them all. The folks who fix the particular ones need to help with the checking
[01:23] <LaserJock> sure
[01:23] <LaserJock> just having a candidate list would be good
[01:23] <mok0> Right. I will make a note on the wiki re: your concerns
[01:25] <mok0> LaserJock: Does this work have to be finished before the artwork FF?
[01:26] <mok0> Or can it be treated as bugs?
[01:29] <LaserJock> mok0: I would think as a bug
[01:30] <LaserJock> depending on the package though a little
[01:30] <LaserJock> if it's gonna be in screenshots or something we might need to do something more
[01:30] <mok0> So, bugs can be fixed right up to the official release right?
[01:31] <LaserJock> well
[01:31] <LaserJock> you might need to do some notifications
[01:32] <LaserJock> most all the Main packages should have icons so it shouldn't matter too much
[01:32] <mok0> LaserJock: Fine. I think a couple of weeks to do this project is enough
[01:33] <LaserJock> yeah, if we have a nice list and hit it it should be fast
[01:33] <mok0> LaserJock: I'll get on these things tomorrow. Right now I am fading
[01:33] <LaserJock> yeah
[01:34] <mok0> See you later!
[01:34] <LaserJock> cya
[01:55] <bddebian> Heya gang
[01:57] <LaserJock> hi bddebian
[01:57] <bddebian> Heya LaserJock
[02:38]  * ScottK2 ponders: Does he upload the patch the shows the final U/I to meet the U/I freeze, but crashes all the time or hold off so the obsolete U/I is still there, but the program at least doesn't crash ...
[02:40] <slangasek> or get a waiver for uploading it after the UI freeze so you don't have to worry? :)
[02:40] <LaserJock> ScottK2: it's an either or?
[02:41] <LaserJock> you can't have a final U/I that works? :-)
[02:41] <slangasek> he probably can't have a final UI that works by tomorrow... :)
[02:43]  * ScottK2 is leaning towards the waiver.  He's building with debugging enabled now, so maybe he'll get lucky.
[02:44] <ScottK2> Upstream removed on-access scanning from Klamav, but then conveniently left the U/I in place but non-working to give end users and obvious targe for filing bugs.
[02:51] <ScottK2> slangasek: Does U/I freeze apply to Universe?  We don't have a process for that listed for Universe.
[02:52] <slangasek> ScottK2: logically, it would apply to anything the documentation team is writing documentation for, or the translators are translating for
[02:53] <ScottK2> Logically.
[02:53] <ScottK2> OTOH, I'm just subtracting, so there's no risk that they'll have to translate something extra.  Urgh.
[02:54]  * ScottK2 cheers for helpful error messages: KCrash: Application 'klamav' crashing...
[03:03] <bddebian> heh
[03:08] <effie_jayx> what does a dsl stand for in the package version
[03:14] <bddebian> dsl in a package version?
[03:53] <emgent> heya people
[03:56] <emgent> someone can open task for Gutsy, Feisty, edgy, Dapper in Bug #195949 ?
[03:56] <ubotu> Launchpad bug 195949 in vlc "VLC Arbitrary memory overwrite in the MP4 demuxer" [Medium,Fix released] https://launchpad.net/bugs/195949
[03:57] <emgent> i can only nominate, only >=MOTU can open task :P
[05:04] <LaserJock> hmm, any math/physics nerds about?
[05:04] <RAOF> Yo!
[05:05] <RAOF> Less of the physics, more of the pure maths, but let's see where this goes.
[05:06] <RAOF> LaserJock: ^^
[05:06] <emgent> LaserJock, heya
[05:06] <emgent> can you open task for Gutsy, Feisty, edgy, Dapper in Bug #195949 ?
[05:06] <ubotu> Launchpad bug 195949 in vlc "VLC Arbitrary memory overwrite in the MP4 demuxer" [Medium,Fix released] https://launchpad.net/bugs/195949
[05:06] <emgent> i can only nominate, only >= MOTU can open task
[05:07] <RAOF> LaserJock: I also prefer the term "geek", but whatever :)
[05:09]  * somerville32 wishes the meanings of geek and nerd were reversed.
[05:09] <LaserJock> sorry, was afk
[05:09] <LaserJock> RAOF: so do I
[05:09] <LaserJock> but I'm a nerd, plain and simple
[05:09] <emgent> lol
[05:09] <LaserJock> I've got an equation and it's got a : in it
[05:10] <LaserJock> something like a dot product, but not
[05:10] <LaserJock> or something
[05:10] <RAOF> An inner product?
[05:11] <RAOF> Some more context might be nice?  I can parse latex if necessary.
[05:11] <LaserJock> hmm, let me see if I can put something together
[05:13] <LaserJock> something along the lines of: P = \chi^2 : E_vis E_IR
[05:13] <LaserJock> where they're all vectors
[05:13] <LaserJock> I've not seen just a " : " sitting in an equation before
[05:14] <RAOF> E_vis & E_IR are both vectors?  There doesn't seem to be an operator between them?
[05:14] <LaserJock> probably just a dot product
[05:15] <RAOF> Hm.  It doesn't ring bells, although it could be "proportion".  Which I use an awful lot.
[05:16] <RAOF> ":" is such a nice, easy to write character it's probably overloaded several times in different fields/papers to mean different things.
[05:16] <LaserJock> bah
[05:17] <RAOF> Could also be a probability distribution?
[05:17] <LaserJock> I've got a presentation tomorrow and I get stuck on what ":" means :-)
[05:17] <LaserJock> I don't think so
[05:18] <RAOF> I don't suppose they go so far as to _define_ the ":" operator? :)
[05:19] <LaserJock> of course not
[05:19] <LaserJock> that would be too easy
[05:19] <LaserJock> "he's getting his PhD, he can sweat for it"
[05:20] <RAOF> "Colon" is unlikely to be the easiest thing to google for, either.
[05:23] <somerville32> http://en.wikipedia.org/wiki/Table_of_mathematical_symbols
[05:23] <somerville32> :P
[05:23] <somerville32> "For matricies, the colon notation may be used."
[05:24] <LaserJock> ah ha!
[05:25] <LaserJock> idiots
[05:25] <RAOF> Right.  Inner product!
[05:25] <LaserJock> pfft, and the say math is the universal language ;-)
[05:25] <LaserJock> *they
[05:26] <RAOF> That doesn't make a whole lot of sense still.
[05:26] <LaserJock> well, the \chi is a tensor
[05:27] <LaserJock> but the E_* should be just vectors
[05:27] <LaserJock> so whatever, as long as I know it's an inner product I'm cool with it
[05:29] <LaserJock> thanks RAOF and somerville32
[05:29] <somerville32> no problem :)
[05:30]  * RAOF grumbles about physicists not understanding how simple tensors are, and fiddling with crazy indicies.
[05:31] <LaserJock> dude, I'm a chemist. I hear "tensor" and I run
[05:32] <RAOF> LaserJock: Bah!  It's just the cartesian product of a couple of rings modulo a fairly simple equivalence relation :P
[05:32] <LaserJock> mmmmhm
[05:33] <RAOF> True, tensors as used in, say, a GR course are crazy arcane objects involving a hojilion indicies.
[05:33] <RAOF> But that's because physicists tend to suck at maths.
[05:35]  * LaserJock raises his hand for the "suck at maths" part
[05:36] <StevenK> And LaserJock isn't even a physicist!
[05:37] <LaserJock> StevenK: yeah, I suck so bad at math they wouldn't let me in ;-)
[05:37] <StevenK> Haha
[05:39] <LaserJock> of *course* I had to pick the paper to present that has all the scary math
[05:39] <LaserJock> :/
[05:40] <ScottK2> Anything that doesn't kill you makes you stronger.
[05:41] <jml> or gives you a hangover
[05:41] <ScottK2> There is that.
[05:41] <RAOF> Or possibly cuts of your hands.
[05:42] <StevenK> "cuts of your hands" ...
[05:42] <RAOF> Oh, yeah.  Absolutely.
[05:42] <RAOF> That's clearly not in relation to the previous statements; it's merely meant to provide an existential statement from which conversation can naturally crystalise.
[05:43] <StevenK> Hah
[05:43]  * LaserJock wonders if he should stop his presentation while he still has hands :s
[05:43] <ScottK2> Conversation that doesn't require math.
[05:43] <StevenK> They're wheeling in the bone saw now
[05:43] <RAOF> Maybe you'll get choice cuts of someone else's hands?
[05:44] <LaserJock> ok ...
[05:44] <LaserJock> so we went from ":" to "physicists suck" to "I'm a cannibal"
[05:46] <RAOF> Via a spelling mistake, yes.
[05:52] <jml> mmmmm handrump
[06:33] <dholbach> good morning
[06:37] <nxvl_work> :D
[06:38] <Hobbsee> mornign dholbach!
[06:38] <dholbach> hiya Hobbsee
[06:38] <dholbach> hey nxvl_work
[06:39] <dholbach> nxvl_work: you're always '_work', right? :)
[06:39] <nxvl_work> heh
[06:39] <nxvl_work> nop, just when i'm at work
[06:39] <nxvl_work> :P
[06:39] <nxvl_work> here i use xchat, everywhere else irssi
[06:42] <nxvl> and i have always my irssi running logging everything :D
[07:01] <Hobbsee> nxvl_work: proxies ftw.
[07:01] <nixternal> looking for a RegEx junky :)
[07:01] <nixternal> convert %f "`echo %f | perl -pe 's/\.[^.]+$//'`.gif"
[07:01] <nxvl_work> Hobbsee: heh, i used to use it, but i'm fine this way for now :D
[07:02] <nixternal> if I have a file named    'No. 22.jpg'    then that RegEx obviously will not work..is there a way to modify that regex so it will work?
[07:02] <nixternal> or am I wasting time because someone filed a bug against a damn servicemenu and has retarded filenames?
[07:11] <dholbach> hi tmarble
[07:11] <dholbach> hi nixternal
[07:11] <tmarble> dholbach: hello
[07:11] <jml> nixternal: so, you want to remove extensions from filenames?
[07:12] <nixternal> hiya dholbach
[07:12] <nixternal> jml: don't want to remove the extension, what it does is replace the extension
[07:13] <nixternal> convert is an imagemagick program to convert from one image format to another, all the regex does is take the original name, replace the .jpg with .gif and rock on :)
[07:13] <jml> shouldn't in be this: convert %f "`echo %f | perl -pe 's/\.[^.]+$/.gif/'`
[07:13] <jml> err, maybe with a " at the very end.
[07:14] <nixternal> I will give it a shot
[07:15] <jml> nixternal: because, afaict that regex should work for 'No. 22.jpg', shell quoting issues aside
[07:15] <nixternal> jml: that worked
[07:15] <nixternal> you da man
[07:16] <jml> nixternal: you have no idea how much soul I had to lose in order to gain that knowledge
[07:16] <nixternal> I am afraid to learn it :)
[07:16] <nixternal> I have a great book by O'Reilly, I just need to go through it
[07:16] <jml> nixternal: it's a great read.
[07:17] <jml> nixternal: the first two chapters are enough to give you kick-ass regex skills
[07:17] <jml> and then it keeps going
[07:17] <nixternal> ya, I just split the book in half and looked, and fainted
=== greeneggsnospam is now known as jsgotangco
[09:42] <liri> how often do developers check newly added packages or changed packages in the revu archive?
[09:43] <dholbach> liri: at the moment not that much - we're in Feature Freeze (http://wiki.ubuntu.com/HardyReleaseSchedule) so NEW packages are only allowed with a special exception
[09:43] <geser> liri: currently not very often as new packages aren't added to hardy anymore
[09:51] <liri> well I uploaded my first package to revu at the beginning of January (2008)  but I only got around to fix the reported errors recently and uploaded the changed package yesterday...
[09:52] <persia> liri: The issue is that Hardy is frozen for new packages.  If the package is in good shape, it will likely get reviewed early, and included in the next release.
[09:56] <liri> what does it mean in terms of time table? :)
[09:59] <persia> liri: Roughly speaking, late May or early June.
[09:59] <liri> persia: so only by that time will the package get more reports on if it's ok or would it need more changes? kinda long time to wait :)
[10:01] <persia> liri: Most likely.  There are exceptions, but Beta Freeze is in about a week, so from then exceptions are unlikely.  Also, the REVU Coordinator for the next release may start activities as early as late April, although this is sooner than has been done in the past.
[10:02] <liri> persia: bummer... I thought I could get some feedback much before that
[10:02] <persia> While it's a bit of a wait, that is in part because it is just two weeks past the freeze deadline.  While there is considerably more complexity involved, release schedules often consist of about three months for new stuff (new versions, new applications, new functions, etc.) and three months of testing and polish,
[10:02] <dholbach> liri: that's the crux of time-based releases: at some stage you have to focus your efforts on fixing the stuff that is in the archive alaready
[10:03] <liri> alright well, I guess I'll just have to wait
[10:04] <persia> liri: You may be able to get feedback, but the vast majority of developers are concentrating on other things.  If you really want to work on that package rather than distribution polish, you might consider working to get it into Debian: I think their freeze cycle doesn't start until next month, and Ubuntu will automatically sync in April or May.
[10:04] <liri> I want to continue with Ubuntu
[10:04] <liri> this is just about getting it into the motu repository, isn't it?
[10:05] <persia> Well, there's not really a "MOTU" repository, but getting it into the universe component of the Ubuntu repository, yes.
[11:57] <liri> persia: thanks
[12:04] <\sh> hmm..tomcat5.5 seems uninstallable
[12:05] <\sh> grmpf...
[12:06] <\sh> tomcat5.5 needs a dependency on a java jdk
[12:07] <\sh> or icedtea-java7-jdk needs a provides: java-virtual-machine
[12:35] <Iulian> Hi
[13:11] <ScottK> So apparently the launchpad approach to bug triage for the developer to edit your bug into something different and then file a new bug that's what your original bug is about and then tell you you did it wrong.
[13:11] <ScottK> https://bugs.launchpad.net/malone/+bug/194601
[13:11]  * ScottK is done with them.
[13:11] <ubotu> Launchpad bug 194601 in malone "Inline bug attachments aren't attached" [High,Confirmed]
=== StevenK_ is now known as StevenK
[13:54] <RainCT> Hi
=== neversfelde_ is now known as neversfelde
=== _czessi is now known as Czessi
=== asac_ is now known as asac
=== warp10_ is now known as warp10
[15:00] <emgent> heya
[15:21] <bddebian> Heya gang
[15:21] <emgent> gang ? :O
[15:21] <bddebian> Sure :)
[15:21] <emgent> lol
[15:27] <tbf> RainCT: duh, seems i overwrote your advocation for gnome-lirc-properties, by uploading a bugfixed version?
[15:28] <tbf> (note to myself: don't put symlinks to your working copy into /usr/lib/python2.5, and remember that pylint doesn't check automake files)
=== santiago-z is now known as santiago-ve
=== santiago-ve is now known as santiago-vea
=== santiago-vea is now known as santiago-ve
[16:30] <bobbo> lutin: Did jpatrick speak to you about Bug #195806 yesterday?
[16:30] <ubotu> Launchpad bug 195806 in purrr "please update purrr to version 0.8 [FFe-granted]" [Undecided,In progress] https://launchpad.net/bugs/195806
[16:31] <Lutin> ah, yes, maybe
[16:32] <bobbo> Lutin: he wanted a second opinion on whether it was good to upload (Im the person working on it)
[16:33] <Lutin> bobbo: as I said in the bug report, I'm wary given the upstream tarball name. I'd check with him whether it's sure or not (as it seems to me that he rather forgot to update the link, or something like that)
[16:33]  * rainct_ would apreciate if someone could check if he is missing anything on bug #195913
[16:33] <ubotu> Launchpad bug 195913 in dialog "Please sync dialog 1.1-20071028-3 from Debian(Unstable)" [Undecided,New] https://launchpad.net/bugs/195913
[16:33] <bobbo> Lutin: ok will talk to the upstream author then get back to both of you
[16:34] <Lutin> bobbo: thanks
[16:36] <jeromeg> hello
[16:37] <jeromeg> jdong: I think we can process the pidgin backport now, it has been tested a lot
[16:37] <jeromeg> without any issues
=== rainct_ is now known as RainCT
[17:01] <RainCT> bobbo: why is bug 186121 set to "In Progress"?
[17:01] <ubotu> Launchpad bug 186121 in ed "Typo in man page of 'ed'" [Wishlist,In progress] https://launchpad.net/bugs/186121
[17:03] <bobbo> RainCT: because ive submitted a debdiff, is that not the right thing to do?
[17:04] <RainCT> bobbo: I think it should be confirmed when it's ready for review ('in progress' would mean that you are working on it)
[17:05] <bobbo> RainCT: ah, ok didnt know that :)
[17:05] <bobbo> RainCT: want me to change it now?
[17:05] <RainCT> bobbo: nah, it's not necessary :)
[17:05] <bobbo> RainCT: ok thanks
[17:06] <RainCT> bobbo: just wanted to check if it should be reviewed or not
[17:06] <bobbo> RainCT: yeah its ready, just a tiny fix though..
[17:10] <RainCT> bobbo: ok.. the version is wrong
[17:10] <RainCT> bobbo: should be 0.7-1ubuntu1, and you've to modify the Maintainer field
[17:11] <bobbo> RainCT: ok thanks :)
[17:15] <LaserJock> RainCT: ping
[17:17] <RainCT> LaserJock: pong
[17:18] <LaserJock> RainCT: regarding rasmol, it's in Main, btw
[17:20] <RainCT> LaserJock: oh, well, my comment applies anyway heh
[17:20] <LaserJock> yeah, except sub'ing u-u-s
[17:20] <LaserJock> it doesn't work in Gnome either, btw
[17:20] <RainCT> it's the second package I see today that has u-u-s but is in main then :/
[17:21] <RainCT> LaserJock: ah right. ok, thanks
[17:21] <LaserJock> they changed the icon name and forgot to change the .desktop ;-)
[17:23] <RainCT> LaserJock: commented about this :)
=== \sh is now known as \sh_away
[17:37] <bobbo> RainCT: new debdiff up for Bug #186121 if you aren't busy
[17:37] <ubotu> Launchpad bug 186121 in ed "Typo in man page of 'ed'" [Wishlist,In progress] https://launchpad.net/bugs/186121
=== dholbach_ is now known as dholbach
[17:41]  * RainCT looks
[17:43] <RainCT> bobbo: ok, now the distro is wrong :P (gutsy -> hardy)
[17:43] <RainCT> (sorry for not noticing that before)
[17:44] <bobbo> RainCT: no problem :) will go and make *another* debdiff
[17:44] <RainCT> bobbo: well, I'll just change that myself if you don't want to create another one :)
[17:45] <bobbo> ive got nothing better to do
[17:45] <RainCT> heh
[17:45] <bobbo> might aswell save you the effort :)
[17:46] <RainCT> well, isn't really any effort at all.. I'd need the same time to download a new debdiff again :P
[17:47] <bobbo> RainCT: did it anyway, new debdiff: http://launchpadlibrarian.net/12277409/ed_0.7-1ubuntu1.debdiff
[17:47] <jpatrick> pochu: hola
[17:48] <pochu> buenas jpatrick
[17:50] <bobbo> jpatrick: Spoke to Lutin about the purrr update (the debdiff i showed you yesterday) and he told me to email upstream
[17:50] <RainCT> bobbo: well, if you have nothing better to do.. want to create yet another one? :P
[17:50] <jpatrick> bobbo: I saw
[17:50] <bobbo> RainCT: Might as well, what have i done wrong this time?
[17:51] <RainCT> bobbo: well, the package is using a patch system (dpatch), so it would be best to use a patch to fix that
[17:51]  * bobbo goes to dig out log of that patch session pitti did last week
[17:53] <bobbo> RainCT: theres already an ed.1 spelling mistake dpatch, *embarrased*
[17:53] <bobbo> doesnt fix the current bug though (obviously)
[17:53] <RainCT> bobbo: yes, but it fixes another problem
[17:53] <RainCT> right..
[17:53] <hellboy195> LucidFox: around?
[17:53] <LucidFox> yes
[17:53] <RainCT> bobbo: so better you create a new one.. (08_ubuntu-doc-typo-fixed.dpatch or whatever)
[17:54] <bobbo> RainCT: ok
[17:54] <RainCT> bobbo: (because a) so it's easier to use again if Debian isn't responsive, and b) the existing patch refers only to ed.1, not the other files)
[17:54] <hellboy195> LucidFox: you remember the problem with starplot? pitti built it again on LP for me but it fails with the same strange error :\
[17:55] <RainCT> bobbo: to create the patch just run "cdbs-edit-patch 08_ubuntu-doc-typo-fixed.dpatch", do the changes, and then "exit 230"
[17:55] <bobbo> RainCT: thanks :D
[17:55] <LucidFox> I have no idea about it, sorry
[17:56] <hellboy195> RainCT: what's for ACK'ing dialog. in the evening I would have written the same. so thx so far :)
[17:56] <RainCT> hellboy195: no problem :)
[17:59]  * RainCT notes that if someone is bored it would be great if he could test open-invaders 0.3 from Debian unstable (especially the sound)
[18:04] <CarlFK> I am having trouble building on hardy - hoping you build experts can help
[18:04] <CarlFK> build error that others don't get on gentoo:  http://dpaste.com/37229/  "expected specifier-qualifier-list before 'size_t'"  Who's problem is this? :)
=== luka74 is now known as Lure
[18:10] <bobbo> RainCT: ping
[18:11] <tsmithe> hi - could a motu upload the new version of mscore as detailed at bug 195170
[18:11] <ubotu> Launchpad bug 195170 in nspluginwrapper "npviewer.bin crashed with SIGSEGV in g_slice_free_chain_with_offset()" [Undecided,New] https://launchpad.net/bugs/195170
[18:11] <tsmithe> oops
[18:11] <tsmithe> bug 195179
[18:11] <ubotu> Launchpad bug 195179 in mscore "Please update mscore to 0.9.1d+dfsg-0ubuntu2 (debdiff attached)" [Undecided,New] https://launchpad.net/bugs/195179
[18:12] <tsmithe> persia, ^^?
[18:20] <james_w> CarlFK: I think that means that one of the types is unknown.
[18:20] <james_w> using cpp on the file might give a clue
[18:22] <bobbo> RainCT: Dpatch debdiff for ed bug: http://launchpadlibrarian.net/12278136/ed_0.7-1ubuntu1.debdiff
[18:32] <CarlFK> james_w: this? $ gcc ../lib/v4l2_driver.h
[18:32] <CarlFK> ../lib/v4l2_driver.h:26: error: expected specifier-qualifier-list before 'size_t'
[18:32] <CarlFK> t$ gcc  driver-test.c = same thing
[18:40] <RainCT> bobbo: great, uploading. please forward it to Debian
[18:44] <bobbo> RainCT: will do :D
[18:47] <james_w> CarlFK: "cpp file | less" to see if the types are resolved.
[18:47] <CarlFK> james_w: which file?  .c or .h
[18:49] <RainCT> bobbo: ehm.. ed is in main.
[18:49]  * bobbo slaps head
[18:50] <bobbo> RainCT: guessing its going to need another review from a core-dev or something?
[18:50] <RainCT> yes, I'm subscribing ubuntu-main-sponsors
[18:50] <james_w> CarlFK: the one that gives you the compile error, so h
[18:51] <bobbo> RainCT: thanks :D
[18:51] <CarlFK> james_w: same error
[18:51] <CarlFK> http://dpaste.com/37236/  (all 3)
[18:51] <bobbo> sorry for that RainCT
[18:54] <RainCT> bobbo: it's no problem at all :    *
[18:54]  * RainCT wonders how he send that if he didn't press enter.. :P
[18:54] <blueyed> How is a package like virtualbox-ose-modules-2.6.24-8-generic to be upgraded to virtualbox-ose-modules-2.6.24-10-generic automatically? I'm asking, because it does not appear to happen (any more?)..
[18:56] <RainCT> blueyed: with a transitional package, or if it's installed as a dependency for another package just change the dependency and I think the package manager should remove the old one as it's unused
[18:57] <blueyed> RainCT: so it needs a meta package then? Changing the dependency in virtualox-ose does not make sense IMHO. Probably it should get added to linux-ubuntu-modules-2.6.24?!
[18:59] <RainCT> blueyed: I can't really answer to that as I've no idea how virtualbox/kernel packaging works.. :(
[19:00] <blueyed> RainCT: I think I'll provide a "real" package virtualbox-ose-modules, which depends on the current variants.. instead of the current variants only providing the virtual virtualbox-ose-modules package.
[19:02] <james_w> CarlFK: I think you need to #include <stddef.h>
[19:04] <CarlFK> james_w: that got me much further
[19:04] <CarlFK> in fact, I think I can ignore what errored
[19:05] <CarlFK> james_w: bam!  thank you.
[19:07] <james_w> CarlFK: no problem. It's a very cryptic error, but I find it usually means "unknown type".
[19:14] <CarlFK> james_w: you a Python coder?
[19:15] <james_w> CarlFK: yes.
[19:16] <CarlFK> I may have 50gig of PyCon videos to send you on a stack of dvd's
[19:16] <CarlFK> pm me your email
=== \sh_away is now known as \sh
[19:23] <\sh> re
[19:25] <blueyed> Hi \sh. Thanks for mentioning me in your blog.. have you fixed my name? :)
[19:48] <\sh> blueyed: hmmm...no?! ;) I'll do it when you fix the vim xdebug plugin ;)
[19:52] <\sh> blueyed: fixing it now
[19:59] <\sh> blueyed: done
[20:19] <hellboy195> geser: around?
[20:19] <geser> hellboy195: yes
[20:20] <hellboy195> geser: you already know if this clean-la patch is still required?
[20:21] <geser> you mean for libnxml?
[20:21] <emgent> heya geser
[20:21] <geser> Hi emgent
[20:21] <hellboy195> geser: yep
[20:24] <geser> hellboy195: is it the only remaining change?
[20:24] <hellboy195> geser: nope. I will introduce a new one
[20:24] <hellboy195> geser: but except that it's the only one, yes
[20:25] <geser> hellboy195: I'd suggest to test-build it without clean-la.mk and check then the libnxml.la file if it contains references of other .la files
[20:26] <hellboy195> geser: yeah dholbach already told me. the only diff is that the debian la has the line "build-deps" filled with the depencies
[20:27] <hellboy195> geser: what I mean. Debian has dependency_libs=' /usr/lib/libcurl.la /usr/lib/libidn.la -lldap -L/usr/lib -lgssapi_krb5 -lkrb5 -lk5crypto -lcom_err -lz /usr/lib/libgnutls.la'
[20:27] <hellboy195> geser: and ubuntu instead only dependency_libs=''" . That's the difference
[20:27] <geser> yes, that's exactly the reason why clean-la.mk was uses
[20:28] <hellboy195> geser: so it's still necessary I suppose?
[20:28] <geser> yes
[20:28] <hellboy195> geser: great. thank you :)
[20:28] <geser> you might be lucky that it not needed currently and libmrss builds
[20:29] <hellboy195> geser: ??
[20:29] <geser> the problem in the past was that libgnutls.la moved from /lib/ to /usr/lib/ but the libnxml.la file had still the old location which lead to a FTBFS because it couldn't find the libgnutls.la file
[20:30] <hellboy195> geser: ah. k
[20:30] <hellboy195> geser: btw, I'm afraid. May you could help me with another problem?
[20:30] <geser> I can try :)
[20:31] <hellboy195> geser: https://edge.launchpad.net/ubuntu/+source/starplot/0.95.4-3  :)
[20:31] <\sh> ScottK: claws-mail-extra-plugins is building...so it seems the FFe report will hit you and the others tomorrow morning (my time)
[20:32] <hellboy195> \sh: poor man. I suppose wine is still a bad boy?
[20:32] <geser> hellboy195: my guess is that one of those *FLAGS made it fail, check config.log (from your pbuilder) for the exact reason
[20:32] <hellboy195> geser: but it builds in my pbuilder
[20:33] <geser> hellboy195: hmm
[20:33] <hellboy195> geser: that's the strange thing
[20:33] <\sh> hellboy195: yes
[20:34] <hellboy195> \sh: damn :)
[20:36] <geser> hellboy195: my pbuilder fails
[20:36] <hellboy195> geser: same error?
[20:36] <pochu> this brainstorm page is cool!...
=== santiago-php is now known as santiago-ve
[20:36] <pochu> ...but we don't really need 7 posts in the planet about it :)
[20:36] <geser> hellboy195: yes
[20:37] <hellboy195> pochu: 5-bugs-a-day got many more posts ^^
[20:37] <geser> hellboy195: from config.log:
[20:37] <geser> configure:3037: gcc "-Wall -pedantic -O2 -g"  -Wl,-Bsymbolic-functions conftest.c  >&5
[20:37] <geser> gcc: "-Wall: No such file or directory
[20:37] <hellboy195> geser: really really strange
[20:37] <pochu> hellboy195: yeah that too
[20:38] <geser> hellboy195: but easy to fix: removed the " from CFLAGS in debian/rules
[20:39] <hellboy195> geser: omg. wth? wired
[20:39] <geser> forget it, wrong fix
[20:39] <hellboy195> ^^
[20:40] <geser> partly correct, configure runs now but make is missing the " around the CFLAGS
[20:40] <hellboy195> geser: that means. configure hates " and make loves it
[20:41] <geser> hellboy195: no, it depends who sets the variable in the environment
[20:41] <\sh> ok...end of business
=== \sh is now known as \sh_away
[20:42] <hellboy195> geser: ah, k
[20:43] <geser> hellboy195: remove the " from line 9 and 11 and add them to line 42
[20:43] <hellboy195> geser: wait. I have to fetch the .dsc first. DaD already removed it ^^
[20:46] <hellboy195> geser: ah, sry. where to add it in line 42? CFLAGS="$ ?
[20:47] <geser> yes
[20:48] <hellboy195> geser: so I prepare the merge/debdiff now. I hope this fixes it. I can't test it because it worked in my pbuilder ..
[20:49] <geser> hellboy195: you can use your PPA
[20:49] <hellboy195> geser: true. hmm nvm. we have time :)
[20:55] <hellboy195> geser: that problem is that I always have problem with signing so I avoid using my PPA ^^
[20:56] <hellboy195> geser: uhh nice. a unset GPG_AGENT_INFO
[20:56] <hellboy195>  fixed it. how can I make that static?
[20:57] <geser> hellboy195: no idea, you could try to add no-use-agent to your ~/.gnupg/gpg.conf
[21:01] <hellboy195> geser: k, thx
[21:04] <hellboy195> geser: but if it builds fine in my PPA the first thing I will do, after I did the debdiff, is to report that back to debian :)
[21:07] <geser> emgent: have you seen http://packages.qa.debian.org/v/vlc/news/20080228T160212Z.html already?
[21:07] <emgent> nope, i go to see
[21:07] <hellboy195> geser: same error O_o
[21:07] <emgent> thanks geser
[21:08] <emgent> uhm other security fix..
[21:09] <hellboy195> geser: ah wait. maybe a mistake from me
[21:10] <hellboy195> geser: ah stupid question. Where does the " has to end in $(MAKE) all CFLAGS="$(CFLAGS) INSTALLPGM=$(INSTALLPGM) \
[21:10] <hellboy195> 		docdir=/usr/share/doc mandir=/usr/share/man sysconfdir=/etc ?
[21:10] <geser> hellboy195: CFLAGS="$(CFLAGS)"
[21:11] <hellboy195> geser: thank you very much. :)
[21:11] <geser> so the value from CFLAGS (line 9 or 11) gets passed as CFLAGS to make and not only the first part of it
[21:16] <hellboy195> geser: in some minutes we will know
[21:22] <hellboy195> anyone seen norsetto recently?
[21:36] <hellboy195> geser: PPA is hating me. md5sum sucks ^^
[21:39] <hellboy195> ah testing it tomorrow :) gn8 @ all
=== cprov is now known as cprov-out
[22:26] <blueyed> Any ideas why apt pulls in the -386 kernel if a package depends on "-386|-generic|..."? the -generic variant could be used without pulling another kernel image in.. http://pastebin.com/m363876fa
[22:28] <pochu> blueyed: do you have the -generic already installed? I think apt will see if there's any installed and if not, install the first one
[22:29] <pochu> blueyed: so change the order to be -generic | -386 should do the trick I think
[22:31] <RainCT> pochu: Are you sure it looks at the order? I though dpkg would sort them anyway how it wants.. (might be wrong though)
[22:32] <blueyed> pochu: changing the order appears to be a workaround only for -generic. The "end" package for -generic depends on linux-image-2.6.24-10-generic, which is installed.. but apt does not seem to resolve this properly to detect it being installed.. virtualbox-ose-modules-generic is not yet installed, but should get preferred, because it depends on the -generic kernel image (which is installed)
[22:32] <pochu> RainCT: nope, it won't sort OR'ed dependencies
[22:32] <RainCT> pochu: oh, cool :)
[22:32] <blueyed> that's right.. this way you can put a preferred one in front of a more generic one.
[22:33] <pochu> so if you have "foo, bar | foobar" it will become "bar | foobar, foo"
[22:33] <RainCT> I see :)
[22:33] <RainCT> pochu: thanks
[22:33] <pochu> RainCT: but that sorting will likely be reverted in dpkg
[22:34] <pochu> as a developer may have a reason to sort if in some special way
[22:34] <pochu> RainCT: there's a thread in debian-devel about it btw :)
[22:34] <pochu> that's why I know that... I'm no dpkg wizard ;)
[22:34] <RainCT> heh
[22:34] <RainCT> is there much traffic on debian-devel?
[22:34] <pochu> yeah
[22:35] <RainCT> :/
[22:35] <pochu> 1297 in January
[22:35] <pochu> 1216 in February as of today
[22:36]  * RainCT decides that he won't subscribe :P
[22:36] <pochu> heh
[22:37] <blueyed> how is apt-cache dotty supposed to work? "apt-cache dotty virtualbox-ose-modules | dotty  -" ?
[22:37] <RainCT> btw, that pypanel thing messed up my gnome-panel.. is this a feature or a bug? :P
[22:38] <tbf> RainCT: i had to upload another revision of gnome-lirc-properties, since a stupid symlink in /usr/lib/python2.5 covered a refactoring problem for me. can you re-attach your advocating flag, please?
[22:38] <tbf> RainCT: or isn't that needed?
[22:38] <RainCT> (messed up = the time is configured correctly but the clock applet shows a wrong time, some icons disapeared from the tray but their menu appears on right click on the void space, etc.)
[22:38] <RainCT> tbf: sure, one moment :)
[22:39] <tbf> RainCT: awesome :-D
[22:40] <tbf> RainCT: your gnome panel is from 2.22?
[22:40] <tbf> RainCT: maybe some ugly side effects from federico's intlclock merge?
[22:40] <RainCT> no, 1:2.20.1-0ubuntu1 (Gutsy)
[22:41] <RainCT> I tried that pypanel thing and since then gnome-panel is mad :S
[22:41] <tbf> RainCT: that sucks.
[22:42] <tbf> RainCT: i'd try to reset the /apps/panel tree in gconf-editor...
[22:43] <RainCT> yeh, specially the clock thing..
[22:43] <tbf> RainCT: or course that's kind of drastic....
[22:43] <tbf> ....but maybe the pythingy left bad entries there?
[22:43] <blueyed> RainCT: here's the dependency graph for virtualbox-ose-modules: http://codeprobe.de/tmp/temp.svg
[22:44] <bobbo> n/whois jpatrick
[22:44] <bobbo> sorry
[22:44] <jeromeg> RainCT: have you tried to restart your gnome-panel ?
[22:44] <RainCT> tbf: is there anything more beside /apps/gnome-settings/gnome-panel (which hasn't really much)?
[22:44] <RainCT> jeromeg: yes, the computer was off overnight
[22:45] <jeromeg> weird
[22:45] <jeromeg> no idea then :)
[22:45] <jeromeg> going to be now
[22:46] <jeromeg> good night
[22:46] <RainCT> jeromeg: good night :)
[22:46] <jeromeg> thank you
[22:46] <RainCT> blueyed: does this image fit on your screen? :P
[22:47] <blueyed> RainCT: not quite.. ;) (1920x1200 though)
[22:47] <blueyed> RainCT: the interesting part is the top only though. apt only looks at the first level apparently.
[22:48] <RainCT> +gnome-lirc-properties (0.2.1-1) hardy; urgency=low
[22:49] <RainCT> tbf: what happened to the "0ubuntu"? :P
[22:53] <RainCT> tbf: ping me once you fixed the version number
[22:53] <RainCT> but tomorrow, me goes to bed now :)
[22:53] <RainCT> good night all
[22:55] <mok0> Any news on qa.ubuntuwire.com?
[22:59] <tbf> STUPID uupdate tool!
=== lando_ is now known as lando
[23:17] <pochu> The server at brainstorm.ubuntu.com is taking too long to respond.
[23:17] <pochu> too many people visiting it?
[23:17] <jdong> slashdot.
[23:20] <tbf> is that normal, that uupdate eats "0ubuntu" tags?
[23:20] <jdong> uupdate by default does pump out -1, as if it were for debian
[23:20] <jdong> currently I just mangle the changelog by hand afterwards
[23:21] <azeem> huh, you'd think that'd be adapted to Ubuntu by now
[23:22] <tbf> jdong: oh... the "u" prefix stands for "update"
[23:22] <tbf> jdong, azeem: what!? "update-update"? what kind of name is that?
[23:22] <tbf> i read the tool's name as "ubuntu update"
[23:22] <jdong> tbf: I think Upstream Update
[23:23] <tbf> jdong: hmm. also possible.
[23:23] <jdong> azeem: I'm sure it's like a one-line patch
[23:23] <azeem> tbf: it predates Ubuntu by a couple of years
[23:23] <StevenK> Upstream Update would be my guess
[23:25] <tbf> azeem: i am a total packaging n00b
[23:26] <tbf> azeem: don't know such details.
[23:27] <azeem> details.
[23:30] <tbf> azeem: well, sense for details seems very important for packagers....
[23:30] <tbf> ...as i learned last week
[23:30] <azeem> did Murray send you to boot camp or what?
[23:31] <tbf> azeem: no, he just figured it's easier and more fun for him, if he delegates the neccessary packaging
[23:31] <azeem> heh
[23:31] <tbf> azeem: ...like for that lirc properties thing
[23:32] <tbf> azeem: hmm.... and i was too stupid to stand up, when i found that word in my contract ;-)
[23:33] <azeem> details ;)
[23:34] <tbf> well, should sleep now. have to master a 700km trip after wake up.
[23:34] <tbf> ciao
[23:34] <azeem> I hope he's flying
[23:59] <jdong> whee, pbuilder's so much less painful now that I hacked the backend to use lzop rather than gzip