[00:57] <micahg> DktrKranz: hi, pkg-multimedia would like a new version of scons, but I'm wondering what impact that has on existing packages, do all the rdepends need a rebuild?
[04:24] <arand> The authour section in the manpage, does that refere to the manpage or the application?
[04:27] <lifeless> manpage I believe
[04:28] <lifeless> probably best to be explicit though
[04:29] <arand> ok, cheers
[06:01] <Mase_wk> hi guys, i am trying to package an application and I would like it to add it's directory to the PATH on installation. What is the preferred way to do this ?
[06:01] <micahg> Mase_wk: to not do this at all...why can't you install in the appropriate existing paths
[06:04] <Mase_wk> good question. I guess there is no real reason at all other than the install guides telling me to set it up this way
[06:06] <Mase_wk> micahg: this is a small python utility which dynamically looks at the path of the executable to see where it should load the rest of it's components from
[06:06] <Mase_wk> i think i would have to change that behavior to jam it in /usr/bin
[06:06] <Mase_wk> is there a recommended way of doing that or should i just hard code it ?
[06:07] <Mase_wk> it has a bunch of xslt, xml files as well as other python libs
[06:08] <micahg> Mase_wk: I'm not so familiar with python utilities
[06:08] <RAOF> The xslt, xml stuff sounds like it wants to go in /usr/share/$PACKAGENAME, and the other python bits treated differently.
[06:09] <ScottK> The actual executable should still go in /usr/bin even though it's written in Python.
[06:09] <Mase_wk> RAOF: right, so i'm not really sure what the best way to modify this application to make that happen is. Do i go through and hardcode it , or should i try an make it a config option and upstream it ?
[06:10] <Mase_wk> at the moment it's reasonably dynamic, you just run the binary and it figures out where stuff is.
[06:10] <RAOF> Yeah, but it's wrong :)
[06:10] <RAOF> Or, at least, that's not *NIX-y.
[06:10] <Mase_wk> yeh i understand that, i am just trying to work out how best to modify it
[06:11] <RAOF> If you can see a clean way to do it upstream, that's better.
[06:11] <RAOF> If you can't, it's ok to distro-patch.
[06:11] <Mase_wk> hmm ok thanks
[07:48] <arand> How can I run a lintian check on a plain manpage without having to go through the build procedure?
[08:04] <dholbach> good morning
[11:36] <X3lectric> hi guys, im trying to uplaod to my ppa, but lately idk why the orig.tar.gz gets stuck at 93622k out of 93623k
[11:36] <X3lectric> irrespective of file size
[11:43] <geser> X3lectric: I've heard about this "last 1k" problem, but don't remember what the issue is or if it's fixed or can be worked-around, try asking in #launchpad
[11:43] <geser> I see you're already there
[11:45] <Laney> it's a dput bug
[11:45] <Laney> or a bug which is exposed by dput
[11:45] <Laney> bug 193848
[11:46] <Laney> I think you can work around it by using sftp
[11:46] <Laney> X3lectric: ^^^^
[11:50] <Bachstelze> that's weird
[11:50] <Bachstelze> it's old, but I never had this issue
[11:50] <X3lectric> Laney: thx
[11:51] <Laney> it depends on how your router handles passive FTP I think
[12:03] <X3lectric> Laney: my router has no problems with passive ftp, its not a home network error
[12:04]  * X3lectric knows networking backwards
[12:59] <Laney> good for you
[13:12] <X3lectric> Laney: this is a actual bug that ppa upload ned to bypass ftp and use sftp but the config for sftp on dput is proving a chanllenge
[13:13]  * X3lectric is not ppa/launchpad jedi
[13:14]  * X3lectric is not even bzr experienced
[13:15] <Laney> X3lectric: https://help.launchpad.net/Packaging/PPA/Uploading#SFTP%20and%20older%20versions%20of%20Ubuntu
[13:16] <X3lectric> ya read that
[13:16] <X3lectric> Permission denied (publickey)
[13:16] <X3lectric> Unable to connect to SSH host ppa.launchpad.net; EOF during negotiation
[13:17] <Laney> you need to add your ssh key into launchpad
[13:17] <X3lectric> google is not helping either
[13:30] <X3lectric> Laney: it is added man
[13:31] <X3lectric> https://launchpad.net/~x3lectric
[13:35] <c2tarun> I was trying to install the build dependencies for a package and I got this error http://paste.kde.org/6685/, what is meaning of this error?
[13:37] <X3lectric> c2tarun: try running a-t-get install -f
[13:37] <X3lectric> er
[13:37] <X3lectric> apt-get install -f
[13:38] <c2tarun> X3lectric: here is what I got 0 upgraded, 0 newly installed, 0 to remove and 13 not upgraded.
[13:39] <X3lectric> try installing the deps again
[13:39] <X3lectric> you my wanna check wahts not getting upgraded
[13:40] <X3lectric> idk kde
[13:41] <X3lectric> apt-get install -f is supposed to fix any broken packages
[13:42] <c2tarun> X3lectric: here are the build dependencies reported by dpkg-checkbuilddeps r-base-dev r-cran-xtable r-cran-spc r-cran-maptools r-cran-vcd r-cran-msm r-cran-vr r-cran-colorspace r-cran-matrix   while installing them I am getting the same error.
[13:42] <X3lectric> dpkg --configure -a
[13:43] <X3lectric> you may want to remove the respective packages from /var/cache/apt/archives/
[13:43] <c2tarun> X3lectric: nothing happened, no output. What is the problem exactly?
[13:43] <X3lectric> broken packages
[13:44] <X3lectric> http://www.google.co.uk/search?q=Broken+packages&ie=utf-8&oe=utf-8&aq=t&rls=org.mozilla:en-GB:official&client=firefox-a
[14:01] <c2tarun> does dpkg-checkbuilddeps checks dependencies mentioned in debian/control file?
[14:22] <c2tarun> I got this error while building a package http://paste.kde.org/6690/, when looked into file I found this line creating the problem http://paste.kde.org/6689/ I need help with the second argument passed in the function, what does it mean? is it typecasting?
[16:49] <DktrKranz> micahg: which version? I pushed 2.0.1 soon after squeeze, isn't that enough?
[16:50] <micahg> DktrKranz: yeah, but natty only has 2.0.0, so I'm wondering about an FFe, I just want to know how bad it affects everything
[16:51] <DktrKranz> micahg: scons upstream is *very* smart about that, so minor versions are all about bugfixes
[16:52] <DktrKranz> there could be some new features, but backward compatibility is almost always a guarantee
[16:52] <micahg> DktrKranz: ok, what about rebuilds, do we need them for all build-depends or just stuff that is experiencing issues
[16:52] <DktrKranz> (e.g. 1.X -> 2.X was a no-op for most of the rdeps)
[16:52]  * micahg thought he saw a new feature or 2, which would require an FFe
[16:52] <DktrKranz> no need to rebuild
[16:53] <DktrKranz> I don't think there will be any differences
[16:53] <DktrKranz> except failing packages which become buildable :)
[16:53] <micahg> DktrKranz: ah, ok, that's great news, thanks
[16:54] <DktrKranz> if you want, I could setup a test framework to rebuild all rdeps, just to make sure it works
[16:55] <DktrKranz> (long time I didn't manage that, though)
[16:55] <micahg> well, I just want to make the release team happy
[16:57] <DktrKranz> also, some rebuilds have been managed by lucas in debian, no major issues reported
[19:54] <delimiter> anyone can mentor me? I'm having trouble getting a package built for redis
[19:57] <Bachstelze> delimiter: what's the problem?
[19:59] <delimiter> I'm trying to get redis 2.2.2 built. when I do the pbuilder build step it bombs out. See http://pastebin.com/mAjUVzuw
[20:20] <LLStarks> what do i apport against if i have a needs-packaging
[20:22] <kklimonda> you don't
[20:23] <kklimonda> you just open a new bug by hand
[20:33] <delimiter> Bachstelze: presumably debian/tmp/redis-server should be auto created since "Package: redis-server" is in debian/control
[20:41] <Bachstelze> delimiter: not 100% sure but I think so, yes
[20:54] <LLStarks> kklimonda, but file against ubuntu, right?
[20:58] <kklimonda> LLStarks: yes
[21:15] <psusi> so is there some sort of magic bridge that connects gtk signals to dbus signals?
[21:24] <broder> psusi: yes, but it's not pure magic. you ahve to tell glib about the dbus signals you care about
[21:25] <broder> hmm...actually i'm not sure if that's true with gdbus. it was with dbus-glib
[21:26] <psusi> ok.  now things are starting to make sense... now if I could just find this magic bridge...
[21:26] <broder> psusi: are you using gdbus or dbus-glib?
[21:27] <psusi> dunno, I'm looking at gnome-power-manager and UPower
[21:27] <broder> dbus-glib, then
[21:27] <broder> at least for g-p-m
[21:27] <psusi> and don't really know anything about gtk or dbus before today ;)
[21:27] <broder> (happened to have the code in front of me already)
[21:27] <psusi> ohh?
[21:28] <broder> so i don't know how things work if you're generating proxy classes from dbus introspection data
[21:28] <broder> but if you're just using DBusGProxy, you use dbus_g_proxy_add_signal to tell the object to listen for the signal
[21:28] <psusi> I've been studying it trying to understand how the gpm gconf keys to disable locking the screen and starting the screen saver got broke... at least when you choose to suspend from the menu
[21:28] <broder> and dbus_g_proxy_connect_signal to setup a callback
[21:30] <psusi> then I think the SIGNAL_SUSPENDING in gpm is vestigial now and should be removed... it is no longer part of the dbus interface and I can't find any code connecting to it within gpm...
[21:31] <psusi> hrm... if you add an argument to a dbus signal, will that break existing clients, or will they just ignore the new argument?
[21:44] <broder> it depends on the framework, but it will likely break existing clients
[21:55] <psusi> broder: damn... so adding an argument to a well known interface like org.freedesktop.UPower.Sleeping is a no go?  Have to add a whole new signal or a property to get the desired info instead?
[21:55] <broder> yeah
[21:56] <broder> i'm fairly positive that python-dbus, at least, would blow up. i think gdbus and dbus-glib will both depend on implementational details, but you'd have to be pretty lucky
[21:56] <broder> networkmanager has had to do this before - c.f. /org/freedesktop/NetworkManager org.freedesktop.NetworkManager.{sleep,Sleep}
[21:57] <broder> well, those are methods not signals, i guess
[21:57] <lucidfox> Oh crap, I seem to have flooded Planet Ubuntu with old posts
[21:57] <lucidfox> any way to undo that?
[21:58] <micahg> slingshot around the sun?
[21:58] <psusi> broder: so something like add a SuspendingEx() signal that has the argument, and emit both?
[21:58] <psusi> err, SleepingEx() rather
[21:59] <broder> psusi: yeah, i think so. or expose the extra piece of information as a property, and sufficiently intelligent clients can query for it when they get the Sleeping signal
[21:59] <psusi> micahg: even doing that, you'd never get to warp factor 10 with current technology ;)
[21:59] <micahg> psusi: that didn't seem to be a requirement before...
[22:00] <warp10> psusi: do you need some warp10?
[22:01] <psusi> micahg: that was the purpose of slingshotting around the sun... to get the kingon bucket of bolts to warp factor 10 for timewarp ;)
[22:01] <micahg> psusi: warp 10 wasn't infinite in TOS, only in TNG and on, and this is getting too offtopic, happy to continue discussion in PM
[22:18] <lucidfox> Okay, I've fixed the RSS so that it outputs the correct dates (it previously formatted month names in Russian...)
[22:19] <lucidfox> will the posts with wrong dates disappear from the front page automatically upon next update?
[22:20] <soren> lucidfox: I think so.
[22:24] <lucidfox> Yay, they went away
[22:24] <lucidfox> foosh
[22:43] <ari-tczew> hellooooo