2008/06/20/#ubuntu-bugs.txt

=== LucidFox is now known as LucidFox_KOTOR30
=== LucidFox_KOTOR30 is now known as LucidFox
[04:36] <Hew> Hey guys. How do I find which package installed a certain file? I'm sure there was some command that did this, but I can't remember it.
[04:39] <bimberi> Hew: dpkg -S /path/to/file
[04:39] <Hew> bimberi: excellent, thank you :-)
[04:40] <bimberi> yw :)
=== asac_ is now known as asac
=== thekorn_ is now known as thekorn
=== pedro___ is now known as pedro_
[14:06] <Rioting_pacifist> how do i find the backwards dependancies of a program (what depends on a library)?
[14:06] <seb128> apt-cache rdepends library
[14:10] <Rioting_pacifist> thx
[14:11] <Rioting_pacifist> a program is listed twice is that a problem or just the way to apt-cache works works?
[14:12] <seb128> you might have several sources providing a different version for this one
[14:13] <seb128> you can apt-cache rdepends library | sort | uniq too
[14:34] <seyed-mehdi> hi everyone
[14:35] <seyed-mehdi> I can't open any https pages. can anyone help please ?
[14:38] <seyed-mehdi> any suggestion ?
[14:41] <persia> seyed-mehdi: You might ask on #launchpad if there is any way around it, although be warned that there may be problems with some data.
[14:42] <seyed-mehdi> persia: thanks 4 suggestion :) which data may have problems ?
[14:43] <persia> seyed-mehdi: I'm not sure.  Personally, I'd not be surprised if e.g. private bugs were not available without https.
[14:45] <seyed-mehdi> persia: I donno, its better checking #launchpad :) thanks
[14:47] <seyed-mehdi> persia: where r u from ? ur ID is name of my country !
[14:49] <Rioting_pacifist> could somebody with kubuntu please confirm/deny #240472 for me please
[14:49] <persia> seyed-mehdi: Not since 1935, but the United States
[14:50] <persia> bug #240472 (not that I use kubuntu, but links are good)
[14:50] <ubottu> Launchpad bug 240472 in kubuntu-meta "firefox3 breaks in kubuntu when certain programs are installed" [Undecided,New] https://launchpad.net/bugs/240472
[14:52] <Hobbsee> persia: er, why is it a launchpad thing, if he can't open https?
[14:52] <persia> Hobbsee: Huh?  Maybe I have the context wrong.
[14:53] <Hobbsee> persia: the guy never mentioned launchpd at all - he just said he couldn't open https pages.
[14:53] <persia> seyed-mehdi: Are you here to talk about a bug with not being able to open https, or are you having trouble helping with bug triage because you are behind a firewall that prohibits https?
[14:53] <Hobbsee> persia: haven't you noticed a disturbing trend of people coming here for premier support, as #ubuntu is too busy?
[14:54] <persia> Hobbsee: Not recently.  I'll try to be more careful, with that in mind.
[14:54] <seyed-mehdi> persia: no Im not behind any firewall or proxy. I can have not any problem with opening such urls in Windows
=== Hobbsee changed the topic of #ubuntu-bugs to: Ubuntu BugSquad | http://wiki.ubuntu.com/BugSquad | Documentation: http://wiki.ubuntu.com/HelpingWithBugs | If you have been triaging bugs for a while, please apply to https://launchpad.net/~ubuntu-bugcontrol/ | Want to report a bug? Read https://help.ubuntu.com/community/ReportingBugs | User support (not related to triage) is in #ubuntu
[14:54] <Hobbsee> persia: that might make it clearer?
[14:54] <persia> seyed-mehdi: Did you already search for a bug that matches your issue?
[14:55] <persia> Hobbsee: Indeed.
[14:55] <seyed-mehdi> persia: yeah, but didnt find anything
[14:55] <seyed-mehdi> persia: there was problem with opening htts in OO or evolution or other progs, but my problem is not those
[14:56] <persia> seyed-mehdi: Maybe you have a new bug then.  Have you tried multiple browsers?
[14:57] <seyed-mehdi> persia: yeah, any known browser such as Firefox, opera, Konqueror and others
[14:58] <seyed-mehdi> persia: i hadnt this problem in other distro's of ubuntu
[14:58] <persia> seyed-mehdi: Hrm.  Not sure which package to file the bug against then.  You might start with a question, and after some more investigation file a bug.
[14:59] <seyed-mehdi> persia: the point is that I've reinstall ubuntu, either kubuntu but nothing changes, even in Live CD I cant browse those.
[15:00] <persia> seyed-mehdi: Right.  It sounds like a bug.  Unfortunately, as it seems to happen under multiple conditions, I don't know where to suggest you file it, which is why I recommended a question.
[15:00] <persia> You might also find support in #ubuntu helpful towards determining the nature of the bug so it can get filed and processed, and this doesn't happen to anyone else.
[15:01] <seyed-mehdi> persia: I was in that channel, no one have any answer for that
[15:02] <persia> seyed-mehdi: Then a question is your best bet.
[15:03] <seyed-mehdi> persia: it seems that it's last chance !
=== thekorn_ is now known as thekorn
[16:07] <snap-l> COuld I get a member of the bug squad to prioritize / triage a bug? bug 99445
[16:07] <ubottu> Launchpad bug 99445 in sun-java6 "Java 6 and KDE integration" [Undecided,Confirmed] https://launchpad.net/bugs/99445
[16:12] <yuriy> snap-l: there are also some bugs about java not working in konqueror and such
[16:12] <snap-l> This is a different issue... it's about appearance, not the plugin
[16:13] <yuriy> snap-l: I think the best thing to do would be to put together a spec documenting all the stuff broken with java in KDE and what needs to be done to fix it
[16:13] <james_w> snap-l: I think you've done a good job with the bug, is there anything else that you think needs doing to it
[16:13] <james_w> apart from fixing it of course :-)
[16:13] <snap-l> james_w: I can't think of anything at the moment
[16:14] <snap-l> yuriy: This is across KDE distros vs. GNOME. I checked it out with Kubuntu and Mandriva 11 last night / this morning using jbidwatcher
[16:14] <james_w> I'll set the Ubuntu status to triaged
[16:14] <snap-l> Thanks!
[16:14] <snap-l> yuriy: The differences between how it appears in GNOME / GTK vs. KDE / QT are pretty striking.
[16:15] <james_w> snap-l: are you not a member of bugcontrol yet?
[16:15] <snap-l> No, I am not.
[16:15] <james_w> have you considered applying?
[16:15] <snap-l> james_w: I've considered it, but would like to get more familar with the process before having more authority. :)
[16:16] <james_w> sure, no rush.
[16:16] <james_w> if java looks really bad under KDE then a couple of screenshots added to the bugs might help a bit.
[16:17] <snap-l> I'll attach some tonight.
[16:20] <yuriy> snap-l: since there's a script that supposedly fixes it, screenshots of how it should look would be good too
[16:22] <snap-l> Yeah, I'll put them all up
[16:23] <snap-l> Funy enough, I don't even use KDE, so I installed Kubuntu and Mandriva under Virtualbox. :)
=== mpih_ is now known as mpih
[16:34] <yuriy> heh, gj snap-l! thanks for your work!
[16:39] <snap-l> yuriy: Thank you.
[16:56] <dholbach> "How to run a Bug Jam" session in #ubuntu-meeting in 4 minutes
[18:07] <bdmurray> calc: bug 228604 has a test document now but it doesn't crash for me - it just looks really odd
[18:07] <ubottu> Launchpad bug 228604 in openoffice.org "Segmentation Fault with Excel sheet" [Undecided,Incomplete] https://launchpad.net/bugs/228604
[18:09] <calc> bdmurray: i just followed up to it, it doesn't crash for me either on hardy-updates. the user claimed it looked weird under excel as well but didn't crash there
[18:09] <calc> so the fact the document is bad isn't a problem afaict, just the old crash
[18:09] <calc> but that looks like it is fixed now
[18:10] <bdmurray> a screenshot of what it looks like w/ excel might be interesting
[18:10] <bdmurray> but yeah not crashing is progress!
[18:15] <Old_Soldier> it limd of defeats the purpose of a test file :) and creates a new / different bug
[18:15] <Old_Soldier> *kind
[18:15] <Old_Soldier> ack must have coffee!
[18:51] <afflux> Is there a way to let E*Botu only announce new bugs which are not sync/merge requests? :)
[18:52] <james_w> it can be useful to see them.
[18:53] <afflux> probably
[18:53] <afflux> I felt bored and scrolled through my scrollback in -bugs-announce, and I had the impression that 99% were sync requests :P
[18:54] <james_w> that's good! :-)
[18:54]  * thekorn did not know that -ananounce is active and usefull again
[18:55] <afflux> for some weeks now, iirc
[18:57] <bdmurray> james_w: hi there
[18:58] <thekorn> bdmurray, hi!
[18:58] <bdmurray> thekorn: hello, how are things with you?
[18:59] <james_w> hi bdmurray
[18:59] <james_w> hi thekorn
[18:59] <thekorn> bdmurray, good, lot's of soccer here in europe :)
[18:59] <afflux> wuuuh, yeah, we won yesterday, yeah!!1111 :P
[18:59] <thekorn> hi james_w
[19:00] <bdmurray> I've updated my greasemonkey patches script - http://people.ubuntu.com/~brian/greasemonkey/lp_patches.user.js
[19:00] <bdmurray> It now identifies patches in bug comments not just in the attachments portlet
[19:00] <afflux> school was fun today... holidays started today, so it was the last day, we got our reports and so on, and most people were boozed :)
[19:01] <thekorn> bdmurray, I recently experimented with javascript and greasemonkey, I started a script which integrats the activitilog in the main bugpage,
[19:01] <thekorn> so you see each log item in a comment box
[19:02] <thekorn> like in other bug tracker
[19:02] <thekorn> +s
[19:02] <bdmurray> that might be neat, especially seeing who modified the description
[19:03] <thekorn> yes, exactly and you can see the reaction of comments in a chronical order
[19:03] <thekorn> but its not mature yet
[19:04] <thekorn> I've also done some work on py-lp-bugs
[19:04] <afflux> bdmurray: btw.: I intend to go to the EMEA board for ubuntu membership. It would be very nice if you could add some testimonials to my wikipage https://wiki.ubuntu.com/KjellBraden
[19:05] <afflux> bdmurray: I'm leaving for holidays on monday or tuesday, so no hurry ;)
[19:05] <bdmurray> afflux: okay, I don't see a specific section for testimonials
[19:05] <thekorn> bdmurray, my plan for  the weekend is to reduce the number of unmerged branches of py-lp-bugs. and I'm not sure about the procedure:
[19:05] <afflux> hum, will create that in 20 mins..
[19:06] <bdmurray> afflux: okay great, ping me when its ready for me
[19:06] <thekorn> should I merge them into a copy of the .main branch and then merge this copy back to .main
[19:06] <thekorn> or should I merge the changes directly to .main
[19:07] <bdmurray> thekorn: I think the bzr log is easier to parse if they are committed directly but james_w might know best
[19:07] <thekorn> bdmurray, note: some of this changes a really big and some of them are changeing the API
[19:07] <james_w> yep bdmurray is right
[19:07] <thekorn> or removing obviously unused features
[19:08] <james_w> however, if you want to set up a devel branch where you stage all of these big things for some integration testing then that works as well.
[19:08] <thekorn> bdmurray, some functions are so buggy, nobody seems to use them, else we would get lots of bugreports
[19:09] <bdmurray> If they are that large I think having a separate branch would be best as it could negatively effect apport
[19:09] <bdmurray> or the retracing service rather
[19:09] <thekorn> otherwise: if nobody uses this branch, we won't get real-life test results
[19:10] <bdmurray> I'm happy to test it, I'm just concerned about killing the retracer
[19:10] <thekorn> ok,
[19:11] <thekorn> so I will create a dev brach,
[19:11] <thekorn> merge changes,
[19:11] <thekorn> announce it,
[19:11] <thekorn> and after a week of testing merge it to .main
[19:12] <thekorn> btw, leonov is a new tool using py-lp-bugs!  https://edge.launchpad.net/leonov
[19:12] <bdmurray> I've seen some stuff on planet about it.  It looks neat
[19:13] <bdmurray> oh, I'd really like to see bug 233948 merged it main
[19:13] <ubottu> Launchpad bug 233948 in python-launchpad-bugs "add username to http-request header" [Undecided,In progress] https://launchpad.net/bugs/233948
[19:14] <thekorn> bdmurray, damn, I forgot about this bug
[19:15] <bdmurray> thekorn: no problem, I understand
[19:17] <thekorn> bdmurray, there is an (optional) file which contains the name of the user, I named this file ~/.python-launchpad-bugs-cookie in this patch,
[19:17] <thekorn> but it is actually not a cookie,
[19:17] <thekorn> do you have a naming suggestion
[19:19] <bdmurray> maybe '.rc' or '.conf'
[19:19] <thekorn> ~/.python-launchpad-bugs.conf
[19:20] <bdmurray> that works for me
[19:22] <thekorn> james_w, is it know that  'bzr patch' is broken in hardy?
[19:23] <thekorn> or, better: do you know about it?
[19:23] <james_w> I didn't
[19:23] <james_w> what's the error?
[19:24] <thekorn> bzr: ERROR: bzrlib.plugins.bzrtools.errors.PatchFailed:
[19:24] <thekorn> will file a bugreport with all output
[19:24] <james_w> it's actually a bzrtools command, so can you file it against that package please?
[19:25] <thekorn> yes, ok, but then there is another bug: it says: Please report a bug at https://bugs.launchpad.net/bzr/+filebug
[19:25] <thekorn> :)
[19:26] <james_w> we can always reassign it, but if I can get you to do the work then it's easier for me :-)
[19:29] <thekorn> james_w, ok, it is bug 241738
[19:29] <ubottu> Launchpad bug 241738 in bzrtools "bzr: ERROR: bzrlib.plugins.bzrtools.errors.PatchFailed: " [Undecided,New] https://launchpad.net/bugs/241738
[19:29] <james_w> ah, ok, it is a bug
[19:29] <james_w> but the command won't apply a patch that won't apply cleanly
[19:30] <james_w> use --force to get the .rej etc.
[19:31] <thekorn> james_w, hmm, sorry, bzr patch --force <file> ?
[19:31] <james_w> yeah, I think so
[19:31] <thekorn> there is no --force for the patch command for me
[19:35] <james_w> there's something weird going on here
[19:35] <james_w> you don't happen to have bzr aliased to bzr -Derror do you?
[19:36] <james_w> ah no, the arguments are there
[19:36] <thekorn> no
[19:37] <james_w> do you have the .rej files?
[19:39] <james_w> I don't know why it is coming up with a backtrace, but it seems like this is intended behaviour
[19:41] <thekorn> james_w, I have .rej and the result looks also ok
[19:41] <thekorn> everything seems to be patched correct
[19:41] <thekorn> but it does not look userfreindly
[19:42] <james_w> yeah, you shouldn't have the backtrace.
[19:42] <james_w> I can't see why you do from looking at the code. I'm sure Aaron will know what's going on, so I'll leave it for him to look at.
[19:43] <thekorn> james_w, thanks a lot
[19:43] <james_w> no problem, thanks for the bug report.
=== Toadstoo1 is now known as Toadstool
[19:54] <afflux> bdmurray: added a section on the wikipage
[20:09] <bdmurray> afflux: are you editing it? your lock has timed out
[20:11] <afflux> bdmurray: hm, probably screwed up with tabs.. pressing "abort" in the edit page should release the lock, right?
[20:11]  * bdmurray shrugs
[20:29] <bdmurray> afflux: I've added my testimonial
[20:29] <afflux> thanks!
[20:30] <bdmurray> My pleasure, good luck!
[20:30] <afflux> huh, that was detailed.  :)
[20:47] <bdmurray> thekorn: is there anyway to test bug 233948?
[20:47] <ubottu> Launchpad bug 233948 in python-launchpad-bugs "add username to http-request header" [Undecided,Fix committed] https://launchpad.net/bugs/233948
[20:48] <bdmurray> oh, a packet trace!
[20:48] <thekorn> bdmurray, I recently thought about the same question
[20:49] <thekorn> oh, yeah, you are right
[20:49] <bdmurray> I could ask for logs too but a packet trace seems easier
[21:32] <thekorn> can someone please give me an opinion on which of these bugreports should be the master: bug 45516, bug 84098 and bug 205325
[21:32] <ubottu> Launchpad bug 45516 in vlc "Always On Top doesn't work" [Low,Confirmed] https://launchpad.net/bugs/45516
[21:32] <ubottu> Launchpad bug 84098 in vlc ""always on top" option in VLC not working" [Undecided,Confirmed] https://launchpad.net/bugs/84098
[21:32] <ubottu> Launchpad bug 205325 in vlc "vlc being compiled with wxwidgets 2.6 .. should use 2.8 to fix "allway on top" in Hardy " [Low,Confirmed] https://launchpad.net/bugs/205325
[21:33] <thekorn> the last one seems to have a solution, but the others have more comments (are older)
[21:33] <thekorn> I don't even know if this is a bug in vlc or compiz
[21:40] <bdmurray> thekorn: it looks like some of my changes got dropeed in python-launchpad-bugs
[21:41] <thekorn> oh
[21:41] <bdmurray> in basebuglistfilter.py
[21:42] <bdmurray> oh, it's just my comment
[21:44] <thekorn> but your comment between reporter and comment in the OPTION_DICT is still there
[21:44] <bdmurray> hmmm
[21:44] <bdmurray> might just be me then
[21:45] <thekorn> maybe a toggle full-line comments option in your $EDITOR
[21:46] <bdmurray> is there a way to force p-lp-b to use http instead of https?
[21:51] <thekorn> bdmurray, no, I think there is no way to use launchpad via http://
[21:51] <thekorn> http:// is always redirected to https://
[21:52] <bdmurray> hrm, packet traces won't be that helpful then. ;)
[21:55] <thekorn> bdmurray, you can packet trace this:
[21:56] <thekorn> python -c "from launchpadbugs.http_connection import HTTPConnection; c = HTTPConnection(); c._HTTPConnection__opener.open('http://www.google.de')"
[21:58] <bdmurray> thekorn: I've got confirmation it works
[21:58] <thekorn> super
[22:29] <sbeattie> thekorn|bdmurray: does http://www.nxnw.org/~steve/warez/python-launchpad-bugs-add_component_to_task.patch look sensible for exporting the component out of the task in launchpad?
[22:31] <thekorn> sbeattie, looking
[22:31] <sbeattie> thanks
[22:31] <thekorn> +    user = LateBindingProperty(get_component)
[22:32] <thekorn> has to be component = ..
[22:32] <sbeattie> doh!
[22:32] <sbeattie> monkey-see-monkey-do failed me.
[22:33] <sbeattie> thekorn: patch updated.
[22:34]  * sbeattie thinks "No wonder I was reduced to call get_component() when I tried to make use of my patch:
[22:35] <thekorn> ok, looks good so far,
[22:36] <thekorn> but what do you think
[22:36] <thekorn> about changing the default value of "component" to None instead of "unknown"
[22:37] <thekorn> so you can do something like     if not task.component: ....
[22:37] <thekorn> instead of   if task.component == "unknown"
[22:38] <sbeattie> sure, happy to do that, that was my area of concern actually, of how to handle the case when the task doesn't have a component listed.
[22:39] <thekorn> sbeattie, if you change  self._component = ....   to   self._component = value_dict.get("component", None)  task.component will be none if there is no component listed
[22:40] <sbeattie> yep, just testing that out now.
[22:41] <sbeattie> thekorn: fixed and updated.
[22:45] <thekorn> sbeattie, ok, cool, looks good, then add yourself to AUTHORS and I will commit this patch :)
[22:45] <sbeattie> thekorn: sure thing, thanks!
[22:46] <thekorn> thank you!
[22:47] <thekorn> sbeattie, if you are using bzr, can you please run   bzr diff > mypatch.diff   so committing is easier for me ;)
[22:50] <sbeattie> thekorn: done and pushed to http://www.nxnw.org/~steve/warez/python-launchpad-bugs-add_component_to_task.patch
[22:51] <sbeattie> (I was using quilt rather than bzr diff to generate the patch, just out of familiarity. I have it set to generate -p1 patches by default)
[22:53] <thekorn> sbeattie, ok, thanks again, I will run some tests now, and than commit it
[22:53] <sbeattie> thekorn: great, thanks again!
[22:59] <thekorn> sbeattie, pushed