[02:03] <maxb> TheMuso: ok, will do - thanks
[07:31] <ebroder> The mit-scheme package is currently complete broken - uninstallable. Do I have a chance of getting a freeze exception to essentially re-merge with Debian?
[07:32] <ebroder> (Actually coming up with something like a minimal path would be more work than I'm interested in doing, but I think I mostly have the merge finished)
[07:37] <Hobbsee> quite likely
[07:38] <Hobbsee> is that even a new upstream version?  You could probably just do it, without the exception
[07:38] <ebroder> It's...not technically an upstream version, but it seems that the "releases" for a while have just been dated snapshots
[07:40] <ebroder> Cool - I have one more build to try. If it works I'll go ahead and open a bug
[07:47] <ebroder> Hmm...if I'm putting multiple LP closers in a changelog entry, is that "(LP: #123456, #654321)"?
[07:48] <Mithrandir> iirc we made that work, yes.
[17:04] <bluefoxicy> rhythmbox is eating freaking lots of CPU
[17:04] <bluefoxicy> 87% to play this MP3
[17:13] <xtknight> what configuration file is written to when the primary mixer track is changed under Sound Preferences?
[17:13] <xtknight> i'm having a problem where my system forgets the device i select upon next reboot.  trying to debug it
[17:27] <xtknight> also where is the sound preferences applet stored since there is no capplets/sound-properties in capplets-data
[18:34] <calc> cjwatson: i was reading on tytso blog that for non dual boot systems that are being freshly installed using GPT may be a better idea than using msdos partition format, has there been any discussion on doing this in ubuntu?
[18:34] <ITechJunkie> Does anyone here know why my terminals don't seem to be refreshing right? I have to press enter twice just to see the text output of the next line...
[18:34] <RainCT> ITechJunkie: Same here, please tell me if you find out how to fix it :P
[18:35] <ITechJunkie> gnome-terminal is the worst, xterm isn't quite as bad
[18:35] <RainCT> Well, it only happens some times here with irssi and such, not always
[18:35] <ITechJunkie> RainCT: I'm wondering if it was an update...
[18:35] <ITechJunkie> Yeah, i'm using irssi now and it's being janky...
[18:36] <RainCT> ITechJunkie: as a workaround, pressing Ctrl+L will force it to refresh and after doing that a few times it may work again
[18:36] <ITechJunkie> is that for irssi or gnome-term?
[18:37] <RainCT> ITechJunkie: for the terminal, I think
[18:37] <ITechJunkie> Is there a log where I can see the updates I have installed recently?
[18:38] <Laney> :q
[18:38] <Laney> whoops
[18:39] <geser> ITechJunkie: /var/log/dpkg.log
[18:40] <ITechJunkie> geser: thanks
[18:41] <ITechJunkie> I really hope it wasn't that kernel update that came out...
[18:42] <ITechJunkie> but I bet it was, i've had several problems since then including nvidia drivers
[18:53] <zyga> hello, I'd like to ask if pydoc3 is broken by design or is something invalid with py3 in jaunty? compare: 'pydoc os' vs 'pydoc3 os'?
[18:57] <ebroder> Do I need to argue for a freeze exception for bug #339618?
[19:03] <RainCT> kirkland: Uhm.. Where did the key binding to close windows in screen-profiles go?
[19:09] <JanC> zyga: there is a thread on c.l.py about that, I think
[19:58] <Laney> Is it possible to sync from lenny instead of sid or am I better off just uploading myself?
[19:58] <slangasek> it's possible, yes
[20:03] <slangasek> ArneGoetje: can you help me understand why we have -base vs. non-base langpacks?  the latest langpack uploads with duplication of the xulrunner translations have contributed to kicking us back oversize on CDs - I'd like to keep the CDs packed as tight as possible during development so we have early warning when something's getting bigger than it should, but that doesn't work so well when I get a full langpack's-worth of fluctuation from the langpack
[20:03] <slangasek> ... themselves :)
[20:21] <johanbr> tkamppeter: I recently tried printing 40 pages out of a 350 page pdf document with total size 4.5 megs. Those 40 pages were run through gs to produce a 100 meg ps file. Cups then converted this to pdf again to mess with it for a bit, giving a 166 meg temporary pdf file. Finally this was piped to pdf2ps to send to the printer.
[20:21] <johanbr> Is this really the way things are supposed to work? This is with Jaunty, by the way.
[20:36] <tkamppeter> johanbr, the plans are that no PostScript at all is involved in that process.
[20:37] <tkamppeter> The first conversion to PS is probably caused by the application from which you are printing. Which app did you use?
[20:38] <johanbr> evince
[20:38] <johanbr> which uses the gtk printing libraries
[20:40] <tkamppeter> CUPS is PDF-centric from Intrepid on. This means it turns everything to PDF and if it gets PDF it leaves it as PDF. Then it does page management (N-up, selecting pages, ...) on a PDF data stream which is more reliable than on PS. Now it converts PDF directly to the printer's native language (the native language of some printers is PostScript, here we have to convert to PS).
[20:41] <slangasek> but no one told our primary pdf viewing software about this?
[20:42] <tkamppeter> If your printer is not a PostScript printer make sure that you have a PPD file of Intrepid or later (with two or more "cupsFilter" lines), if not, re-create your print queue with system-config-printer.
[20:44] <johanbr> It is a postscript printer (HP Laserjet). And the print queue was created on a fresh install with a Jaunty daily build from a month or so ago.
[20:46] <tkamppeter> GTK has standard functionality for generating print data, opening a print dialog, ..., but unfortunately there are many ways to call these functions, and this makes a simple patch like in bug #258421 bot switching all apps to PDF. Some apps need to be taken care of individually. Patches welcome ...
[20:47] <johanbr> from bug: "Other programs, like evince and Firefox still send PostScript. Here there are probably similar patches needed in the applications themselves."
[20:48] <johanbr> I see. Thanks for the explanation.
[20:48] <tkamppeter> slangasek, perhaps the patch needed for evince is as simple as the patch of bug 258421.
[20:50] <calc> anyone able to spot what is wrong with my compiler args here: http://paste.ubuntu.com/128477/  ?
[20:51] <slangasek> tkamppeter: I just don't understand why it wasn't a priority to make sure the PDF viewer was doing the right thing as part of the cups transition...
[20:52] <calc> it compiles the code it just refuses to link it for some reason
[21:01] <johanbr> tkamppeter: looks like a simple patch to line 1016 of ev-print-operation.c would work.
[21:02] <johanbr> http://svn.gnome.org/viewvc/evince/trunk/shell/ev-print-operation.c?view=markup
[21:12] <RainCT> Arr.. If you move a file which Totem is playing it stops.. This didn't happen before :(
[21:13] <tkamppeter> johanbr, so I will add an evince task to bug 258421, please add the link to the bug report.
[21:14] <johanbr> tkamppeter: Sure. I'm now compiling evince with the change I mentioned. If it works, I'll attach the patch.
[21:17] <tkamppeter> johanbr, it looks like that the line 1016 gets the desired output format (PS or PDF) from some centralized configuration. Perhaps we should change this configurationm to get more apps to print in PDF.
[21:23] <TheMuso> dtchen: Will we be able to push a new rev of pulseaudio before Tuesday?
[21:25] <johanbr> tkamppeter: Right. Looking at that now.
[21:43] <UnixOne> hi
[21:43] <UnixOne> I've had some problems with the package comerr-dev. Kind of dependancy problems.
[21:43] <UnixOne> Is that a known error?
[21:43] <ebroder> What exactly are you running into?
[21:45] <UnixOne> ebroder: I tried compiling the latest kernel that requires installation of several -dev packages. which was never a problem for me. But my apt got stuck with the package comerr-dev which has some dependancies that were defect also because of it.
[21:45] <UnixOne> ebroder: It's a clean system no foreign repositories except the backports.
[21:45] <ebroder> You're still not being specific enough
[21:46] <UnixOne> ebroder: the un/install routine of comerr-dev is defect
[21:47] <ebroder> What version of Ubuntu is this on?
[21:48] <UnixOne> Intrepid
[21:49] <jdong> could you pastebin the exact error message?
[21:49] <UnixOne> only checking the backports maybe even the proposed updates. causes that huge error.
[21:49] <UnixOne> jdong: I've manuall uninstalled it by hand.. I don't like to run into that trouble again.
[21:49] <jdong> well then the short answer is no, this is not a known problem.
[21:50] <UnixOne> jdong_: I removed the installed files and reset the status of dpkg
[21:50] <ebroder> The group that I work with uses a lot of software that depends on comerr, and we haven't had any problems with it
[21:50] <jdong> *cringe*
[21:50] <jdong> you hand edited *WHAT*?
[21:50] <ebroder> (Well, I mean, we have, but those are related to comerr being worthless, not the packaging being wrong :-P)
[21:50] <UnixOne> ebroder: did you enable backporst and proposed updates?
[21:50] <jdong> I use both here on all my Ubuntu machines.
[21:51] <UnixOne> jdong: are you running kubuntu?
[21:51] <ebroder> I use -backports on virtually all of my machines, although not -proposed
[21:51] <ebroder> Shouldn't matter
[21:51] <jdong> backports doesn't do anything related to comerr
[21:51] <UnixOne> maybe proposed-updates is the key factor
[21:51] <jdong> and kubuntu vs ubuntu use the same repositories.
[21:51] <UnixOne> jdong_: indeed
[21:51] <jdong> I'm looking at this month's and last month's propsoed uploads
[21:51] <jdong> see nothing comerr related.
[21:52] <jdong> next time, please capture a precise error message -- without it diagnosing something like this is next to impossible
[21:52] <UnixOne> jdong how can I find the cause of that error?
[21:52] <ebroder> That's impossible to say without knowing what the error was
[21:52] <UnixOne> I mean how do I reversely look into that dependancy problem
[21:53] <jdong> I don't even know what error you are referring to.
[21:53] <ebroder> There is no dependency problem. The package installs cleanly
[21:53] <UnixOne> ebroder: ok, I'll install it again
[21:53] <jdong> and "some package has a defect" is not really helpful
[21:55]  * jdong notes there is some irony about errors in libcomerr :)
[21:55] <ebroder> You know what the best thing about the common error library is? There are so damn many of them
[21:56] <UnixOne> jdong: a question before I do install it. may it affect the system when I used backports+proposed to upgrade my system, when I remove proposed now?
[21:56] <jdong> not really?
[21:56] <UnixOne> k
[21:56] <slangasek> ebroder: only one in Ubuntu :-P
[21:56] <ebroder> slangasek: Incorrect
[21:56] <UnixOne> I'll uncheck proposed updates then
[21:57] <ebroder> AFS ships its own com_err :)
[21:57] <slangasek> eh, really?
[21:57] <jdong> well AFS is weird in all its own ways :)
[21:57] <slangasek> krb5 uses the common one
[21:57] <ebroder> Right
[21:58] <slangasek> I would've thought someone would send up the AFS stuff by now
[21:58] <ebroder> There's a lot of muttering about how someone really should, but I don't think anyone's actually gone and unified the libraries
[21:58] <ebroder> Actually...I bet Heimdal has a separate com_err from MIT Kerberos
[21:58] <ebroder> So that might be 3
[21:58] <slangasek> heimdal may not use it at all
[21:58] <UnixOne> comerr-dev is an error for me :P but it's installing.. will pastebin the logs
[21:59] <jdong> libopenafs-dev: /usr/include/afs/com_err.h
[21:59] <jdong> AFS definitely has its own :)
[21:59] <UnixOne> yay I've an errror with comerr
[21:59] <jdong> so we've heard the first time
[21:59] <ebroder> Yeah - the default API calls have afs_ prepended onto them, but there's a #define you can set to make it collide
[22:02] <UnixOne> ehrm how do I copy the log out of the terminal window in synpatic? STRG+C or STRG+SHIFT+C don't do it. Synaptic asks I really want to terminate. I say no. And I can't copy it.
[22:02] <UnixOne> but I was able to copy the popup that appeared before
[22:02] <jdong> ctrl-shift-c will do it
[22:03] <jdong> otherwise use mouse paste
[22:03] <jdong> or use a screenshot
[22:03] <jdong> I think /var/log/apt/term.log or something stores transcripts too
[22:06] <UnixOne> jdong: yes, I got it. It was related to klipper a clipboard tool default in kubuntu. It just gave me the earlier content of the clipboard. I deleted the clipboard history have done another ctrl+shift+c and voila.
[22:07] <UnixOne> short version(popup) http://paste.nn-d.de/839 very very long version -> (terminal) http://paste.nn-d.de/840
[22:08] <ebroder> Well, English is definitely my only language, but I'm pretty sure that the problem is with gettext, not comerr-dev
[22:09] <slangasek> both give the error
[22:09] <ebroder> Hmm...good point
[22:09] <UnixOne> I tried installing THIS http://paste.nn-d.de/841 in short build-essential kernel-package qt4-dev-tools libqt4-dev
[22:09] <slangasek> 'subprocess post-installation script returned 1'
[22:09] <jdong> well yay.
[22:10] <slangasek> UnixOne: 'info' is broken on your system
[22:10] <UnixOne> info ?
[22:10] <UnixOne> :D
[22:10] <UnixOne> lol
[22:10] <slangasek> specifically, the install-info script provided by dpkg to register info documentation
[22:10] <UnixOne> first the error was broken, now it's info
[22:10] <jdong> install-info --quiet --section "Development" "Development" \ --description="" /usr/share/info/gettext.info
[22:10] <jdong> that is what gettext attempts to do
[22:10] <slangasek> it's failing in the postinsts of any program that calls it
[22:11] <ajmitch> but which install-info is being used here?
[22:11] <UnixOne> I there's a part that needs translation or is important I could translate it
[22:13] <UnixOne> it is  info_4.11.dfsg.1-4_i386
[22:13] <UnixOne> the default one I guess
[22:13] <slangasek> UnixOne: what's the output of 'debsums -s dpkg' on your system?
[22:14]  * ajmitch was meaning more along the lines of 'whereis install-info', in case you have one in /usr/local/bin or otherwise
[22:14] <UnixOne> that package isn't installed currently
[22:15] <UnixOne> slangasek: debsums: checksum mismatch dpkg file /sbin/start-stop-daemon
[22:15] <slangasek> well, that's interesting
[22:15] <slangasek> but not directly relevant
[22:15] <UnixOne> yes, has nothing to with comerr-dev.. somehow. strange
[22:16] <slangasek> UnixOne: does /usr/local/bin/install-info exist, as ajmitch suggests?
[22:16] <jdong> which install-info
[22:16] <UnixOne> slangasek jdong yes /usr/local/bin/install-info
[22:16] <slangasek> yeah... you'll need to get rid of that
[22:17] <jdong> uhh.
[22:17] <jdong> yeah...
[22:17] <jdong> that could be problem #1...
[22:17] <jdong> though maybe dpkg shouldn't be operating with /usr/local in the PATH?
[22:17] <jdong> or is there actually a use-case for that
[22:17] <slangasek> jdong: the general principle is that dpkg shouldn't behave differently than the rest of the system wrt PATH
[22:18]  * jdong nods
[22:18] <ebroder> I could imagine wanting it for testing purposes
[22:18] <jdong> indeed the rest of the system is no different about succumbing to /usr/local poison
[22:19] <ScottK-desktop> That's by design.
[22:20] <ebroder> If the system didn't respect /usr/local, then you'd have people putting a bunch of stuff in /usr, and that just doesn't end well
[22:20] <ScottK-desktop> If an admin installs something, it should take precendence because presumably they know what they are doing.
[22:35] <andersk> python3-minimal is currently getting pulled into all Jaunty installs. Can someone look at my debdiff in bug 330613?
[22:40] <UnixOne> ebroder jdong can you help me?
[22:43] <UnixOne> I've sent you all logs. do you know what I can do to fix that error?
[22:46] <ebroder> UnixOne: remove the install-info in /usr/local
[22:48] <UnixOne> ebroder: sudo mv /usr/local/bin/install-info  /usr/local/bin/install-info.backup ?
[22:48] <ebroder> Sure
[22:48] <UnixOne> ok
[22:48] <UnixOne> and then reinstall those packages, right?
[22:48] <ebroder> Should work
[22:48] <UnixOne> ok
[22:48] <ebroder> If you've got anything else lying around in /usr/local/bin or /usr/local/sbin, you should think long and hard about whether they should be there
[22:49] <UnixOne> ebroder: why? I've just texlive2008 installed
[22:49] <UnixOne> in the default location
[22:50] <UnixOne> that's /usr/local/bin
[22:52] <UnixOne> ebroder: it worked wow
[22:52] <UnixOne> genius!
[22:54] <TheMuso> dtchen: With the work you are doing on pulse, will things be in a state such that we can push a change for alpha 6, due this week?
[22:54] <UnixOne> ebroder: why did install-info in /usr/local cause an error?
[23:09] <dtchen> TheMuso: yes
[23:09] <TheMuso> dtchen: ok great.
[23:10] <ebroder> Anyone willing to sponsor an upload for bug #339449?
[23:15] <rbrito> Hi there.
[23:15] <rbrito> I am an experienced linux user and I would like to help with the PowerPC port of Ubuntu.
[23:16] <rbrito> I have 5 packages on the Debian archive and I use PowerPC since 2000.
[23:16] <rbrito> I have also contributed some patches to the Linux kernel.
[23:17] <rbrito> What would it be necessary to be a contributor so that PowerPC gets in a better shape?
[23:22] <rbrito> For the task of helping with the PowerPC port, I can test it under an OldWorld computer (a PowerMac 9500/180MP upgraded to a G3/400MHz), a NewWorld computer (an iBook 12" G3 600MHz) and a Kurobox (an embedded device using a Freescale processor).
[23:22] <rbrito> Of course, the easier task would be to get things running with the NewWorld computers.
[23:23] <TheMuso> rbrito: please head over to #ubuntu-powerpc and we can chat there.
[23:23] <rbrito> But I would like to know how the ports of Ubuntu are maintained, so that I can take the proper measures.
[23:23] <rbrito> TheMuso, Thank you very much.
[23:25] <johanbr> tkamppeter: Forcing evince to export pdf doesn't seem to work very well. The pdftopdf filter runs out of memory on large documents, and on small ones I get black borders and similar artifacts.
[23:27] <tkamppeter> johanbr, can you direct the PDF printing output of evince into files to investigate the integrity, for example to see of acroread or gv handle this PDF correctly?
[23:27] <tkamppeter> johanbr, what is your Launchpad user name?
[23:36] <johanbr> tkamppeter: Sorry, I partially take that back. I think the small pdf document I started with was not very good.
[23:36] <johanbr> and I just subscribed myself to the bug
[23:37] <johanbr> For the "large" pdf document, pdftopdf grew to 1.6 gig resident size and the computer started swapping like crazy. Even though the original size is less than five megs.
[23:45] <johanbr> tkamppeter: Which options does cups use when it calls pdftopdf? I tried running it by hand, and it worked fine (even if the resulting pdf file is a bit large, about 100 megs).
[23:47] <tkamppeter> johanbr, can you attach the large PDF file which causes problems for you to the bug report?
[23:47] <tkamppeter> To test CUPS filter chains there is a utility named "cupsfilter". See also its man page.
[23:50]  * calc didn't realize he was going to end up learning how to write gnome code to fix bugs in OOo, heh
[23:50] <tkamppeter> In general a CUPS filter (or backend) is called with 5 or 6 options. See "man 7 filter".
[23:51]  * calc hopes he learns fast enough to fix the gvfs/gio bugs in OOo by beta
[23:54] <tkamppeter> johanbr, you can call CUPS filters simply with "<filter> 1 1 1 1 '' <input file>" and get the output file on stdout.
[23:55] <LaserJock> calc: OOo uses gcfs/gio directly?
[23:55] <LaserJock> *gvfs
[23:56] <johanbr> tkamppeter: Unfortunately I can't attach the document to the bug, but I could send it to you privately.
[23:57] <tkamppeter> johanbr, then please do so: till dot kamppeter at gmail dot com
[23:57] <johanbr> Interestingly enough, calling pdftopdf directly produces a large file (~ 100 megs), but calling cupsfilter produces a file of the same size as the original (~5 megs).
[23:58] <tkamppeter> johanbr, is the file too big to attach or is it of confidential content?
[23:59] <johanbr> The latter. Well, semi-confidential.