/srv/irclogs.ubuntu.com/2008/06/29/#ubuntu-motu.txt

sistpotyaaaaahhh... all my mail seems to be gone00:07
sistpotydear kmail, please don't shock me like this, sincerely yours, stefan.00:08
cyberixIf I send a patch for Hardy, should I send the patch for Intrepid too?00:14
persiacyberix: The general rule is that nothing not fixed in intrepid will be uploaded to hardy.00:17
gesercyberix: for an SRU?00:17
persiaTo avoid confusion, I'd recommend putting both patches in the same bug.00:17
cyberixno, just a small patch00:18
cyberixActually there is no use to ship this for Hardy00:18
persiacyberix: Any change to hardy at this point is an SRU, even a just a 1-bit change00:18
cyberixI'm just trying to understand how the processes work00:18
cyberixk00:18
cyberixso it is stable instead of source :-)00:19
geserhas someone an idea what went wrong during this build? http://launchpadlibrarian.net/15562990/buildlog_ubuntu-intrepid-i386.libgnupg-interface-perl_0.36-1_FAILEDTOBUILD.txt.gz00:24
sistpotygeser: hm... "which gpg ... gpg (GnuPG) not found at Makefile.PL line 21." might give some clue?00:26
persiasistpoty: But build-essential depends on dpkg-dev, which depends on gnupg00:27
sistpotypersia: which doesn't mean that Makefile.PL is correct thought ;)00:27
persiasistpoty: True.00:27
gesersistpoty: it builds in my pbuilder00:28
sistpotygeser: hm... strange00:28
* persia tries in sbuild, to see if it is due to pbuilder-satisfybuilddepends00:28
sistpotypersia, geser: oh, dpkg-dev doesn't b-d on gnupg00:29
persia!00:30
sistpoty(at least not 1.14.20ubuntu1)00:30
persiaHow did that become Suggests: ?  That's rather unexpected.00:30
geserbut gnupg is installed (see the gpg signature check after the source is extracted)00:30
persiaNo, that's just me reading apt-cache output wrong.00:31
persiageser: There's something very odd: that buildlog also has "sh: gcc: not found"00:33
sistpotyhm... dear kmail, don't disconnect me from the outside world, by not letting me receive mails from a local mbox. thanks.00:33
persiaBuilds cleanly in my intrepid sbuild chroot.00:33
sistpotypersia: hm... however it has: "Setting up g++-4.3 (4.3.1-1ubuntu2) ..."00:34
persiasistpoty: Right, which is why I thought "sh: gcc: not found" was odd.00:34
persiaI find it especially odd right after "Toolchain package versions: ... gcc-4.3_4.3.1-1ubuntu2"00:35
geserif I understand the Makefile.PL correnctly the second "gpg" in the line "which gpg ... gpg" should come from the system() call, right?00:35
persiageser: Right.  I would have expected "/usr/bin/gpg"00:36
persiaI think that system($gpg_find) is returning nothing, and the second gpg is coming from 'die "gpg (GnuPG) not found";'00:37
geseryes, you are correct00:38
geserperl -e "system('which gpg')" returns here /usr/bin/gpg00:38
persiaLooks like this was the first build since edgy.  Maybe this is a case of someone hand-adjusting things.00:39
geserpersia: gcc-4.3 only provides /usr/bin/gcc-4.3, /usr/bin/gcc is in gcc but that's also installed00:39
persiageser: Still, it's odd to see a report that gcc is not found.00:39
gesertrue00:40
gesersomeone messing with $PATH?00:40
persiahttp://launchpadlibrarian.net/3087635/buildlog_ubuntu-edgy-i386.libgnupg-interface-perl_0.33-6_FULLYBUILT.txt.gz looks rather different: Makefile.PL likely changed a bit upstream.00:41
persiaCould be the common situation with arch:all that the Debian build wasn't run in a sanitised chroot...00:41
geserbut why does the same source package build in my pbuilder and your sbuild but not on the buildd?00:43
persiaDo you have gnupg installed in your chroot?  It seems to be in mine.00:43
bobbosistpoty: thanks for sponsoring my memaker upload :)00:44
sistpotybobbo: thanks for providing a debdiff in the fisrt place ;)00:44
bobbosistpoty: hehe, my pleasure00:44
geserpersia: I have, but how did dpkg-source (on the buildd) check the gpg signature without gnupg? (gpg: Signature made Tue Jun 24 07:31:19 2008 BST using DSA key ID 9725F63B)00:46
* sistpoty goes to bed now... gn8 everyone00:46
geserpersia: the "sh: gcc: not found00:49
geser" seems to be normal now00:49
geserit's also in successfully build logs, e.g. http://launchpadlibrarian.net/15590444/buildlog_ubuntu-intrepid-i386.apache2_2.2.9-2ubuntu1_FULLYBUILT.txt.gz00:50
persiaThat's very odd.  Maybe the "gcc" package isn't installed, or poorly configured, or something.00:51
geser"Setting up gcc (4:4.3.1-1ubuntu2) ..." seems ok to me00:53
persiageser: I have to agree that gpg is installed.  I don't know why "which gpg" wouldn't work.00:54
geserpersia: n8, I'll try to ask the buildd admins if they have an idea why libgnupg-interace-perl FTBFS  (and also about the "sh: gcc: not found" thing)00:54
LaserJockis anybody using sbuild+some apt cacher with both Debian and Ubuntu ?00:54
* persia checks the default /bin/sh in the sbuild schroot00:54
persiaNo, I seem to have dash, so that's not it.00:55
* persia is baffled00:55
geserpersia: a reason why gcc and gpg are not found could be that /usr/bin is not in the PATH for some reason on the buildd00:56
persiageser: Maybe, but why would it be able to call dpkg-source (also in /usr/bin)?00:57
persiaFurther, dpkg-source checks for the existence of /usr/bin/gpg, but then calls gpg without a prefix path.00:57
persiaIt's backticks rather than system(), but that oughtn't be sufficient to make it work or fail.00:58
geserdoes it get called as "dpkg-source" or "/usr/bin/dpkg-source"?01:01
geserhmm, dpkg-source calls gpg without the full path to do the signature check, so PATH seems to be ok01:02
persiaRight.01:02
geserwhat about giving it back to the buildds and hope that it gets tried on a different buildd?01:05
persiaThat's why they put the button there :)01:06
gesernothing changed :( still FTBFS01:11
* geser goes to bed now01:12
persiageser: Sleep well.01:13
=== macd_ is now known as macd
=== rockstar` is now known as rockstar
RAOFHm.  Gnome-do-plugins in Ubuntu has a do-plugins as its source package, but gnome-do-plugins in Debian has gnome-do-plugins as its source package.  I'm rebasing the Ubuntu package on the Debian package; do I upload as gnome-do-plugins and ask for removal of do-plugins?07:56
wgrantRAOF: Please... do.08:11
RAOFwgrant: :P08:12
RAOFwgrant: Do you know if I need to ask for the removal of do-plugins source before uploading gnome-do-plugins?  Otherwise there'll be a binary package name conflict, and I'm not sure how those get dealt with.08:16
wgrantRAOF: As long as the version is greater, it will be fine.08:16
wgrantThey'll go through binary NEW, but then dominate the old binaries, so your new ones won't be removed with the old source just because they have the same name.\08:17
wgrantGrmph. Why is Full Circle Magazine recommending Webmin?08:24
wgrantAnd 'sudo su'?08:24
wgrantEhem, and they used a pre-Hardy interview...08:36
wgrantAnd they recommend using Debian packages. Ew ew ew.08:38
Flannelwgrant: Where?08:49
wgrantFlannel: To which bit?08:50
\shmoins08:51
Flannelwgrant: Um, all of them if theyre convienent08:51
jpdswgrant: I think you can find them lurking in #fullcirclemagazine ;-=08:51
jpdsmoin \sh08:51
wgrantjpds: Ah, perhaps.08:51
wgrantFlannel: I glanced through the latest edition (#14).08:51
wgrantThey're fairly easy to find with any search functionality.08:52
Flannelwgrant: Alright, thats all I needed to know08:52
FlannelDidn't want to start from 008:52
IulianG'morning09:03
sistpotyhi folks12:07
geserhi sistpoty12:11
sistpotyhi geser12:12
sistpotyhm... last nights dist-upgrade with many kde parts getting upgraded left me with a gnome session... epic fail *g*12:12
gesersistpoty: it's part of the hidden master plan to convert all kde users to gnome :)12:13
sistpotyhaha12:13
elkyi like the sound of this plan...12:14
RAOFHm.  I'm just trying Intrepid KDE.  Again.12:15
RAOFIs the display settings panel _meant_ to do anything, or just look (not) pretty? :)12:16
jussi01someone remind me of the build a .dsc command?12:16
RAOFsbuild -d intrepid foo.dsc ? :)12:17
RAOFOn the other hand: dear GTK: please implement something like the smooth animation of the output list.12:17
jussi01no, the dpkg one?12:17
RAOFIs there one?12:17
jussi01hrm... /me goes to google more12:18
RAOFYou dpkg-source -x it, then dpkg-buildpackage the extracted source.12:18
RAOF(or debian/rules binary, if you're confident it's policy-compliant :))12:18
jussi01RAOF: ahh, yeah thats the one12:18
jussi01RAOF: its just a personal thing, nothing for ubutnu atm.12:19
RAOFUm.  Why does Kubuntu default to playing a 'ding' every time I press a button?12:19
wgrantRAOF: Because it's kool.12:29
RAOFGah!  KDE is so easy to break!12:29
RAOFI seem to have created a panel covering my entire second monitor.12:30
wgrantNice!12:31
wgrantKDE4 or not?12:31
RAOFKwin now believes that maximise about 30x20 pixels in the top left corner, too.12:33
RAOFKDE412:33
RAOFHm.  It is apparently imperitive that I don't alt-tab, lest kwin consume 100% cpu and lock the desktop.12:35
persiaRAOF: Just use the panel on your second monitor to control which app you are using.  That way you won't need the keyboard at all :)12:36
RAOFThe panel on my second monitor is entirely empty, and unresponsive to mouse clicks :)12:37
RAOFMaybe it's not even a panel?  Maybe something just decided to fill my second monitor with all black.12:38
=== Kopfgeldjaeger2 is now known as Kopfgeldjaeger
RAOFOooh, maximising works again!12:39
RAOFI was going to spend some time using kde to file some bugs, but kubuntu appears to be composed entirely of bugs.  I must be an expert kde-breaker.12:42
StevenK"I'd file bugs, but maybe what I think are bugs are really features."12:42
persiaEverything in a feature, for the right use case...12:43
RAOFThe biggest "feature" seems to be semi-awareness of dual head.12:43
RAOFIt knows my laptop's running at 1680x1050, so the panel is 1680 wide.12:43
RAOFHowever, when it started up my two monitors were cloned.  Thus, the desktop is the size of the smaller of the two.12:44
RAOFSadly, this means that the left hand side of the panel, containing, among other things, the launcher, is offscreen.12:44
StevenKTeehee12:44
RAOFW.T.F?12:45
* RAOF blinks again.12:45
persiaRight-justification is always correct :)12:45
RAOFPressing the "configure this panel" button (not labelled as such in any way, of course) brings up, among other things, 3 arrows.12:46
RAOFIn some as yet undiscovered way, these three arrows influence the size of the panel.12:46
StevenKThat sounds intutive.12:47
RAOFAs well as some left/centre/right justification boxes.12:47
RAOFAlso intuitively: selecting the text which is left justified causes the panel to be right justified on my right hand monitor.12:47
RAOFCentre works as expected, but the right-justification button causes it to be right-justified on my left monitor.12:48
RAOFHm.  How do I make the clock show 12h time?  Hint: it doesn't involve right clicking on the clock, apparently.12:51
StevenKRAOF: You switch to Ubuntu?12:52
RAOFEntirely possible, yes.  I'd like to file at least _some_ bugs first, though.12:52
RAOFSuch as "has no concept of multi-head"12:52
persiaCould anyone explain why pycentral might report "pycentral: pycentral pkginstall: not overwriting local files", as in http://launchpadlibrarian.net/14327059/DpkgTerminalLog.txt?  Is this likely due to user customisation, or is it a packaging issue?13:00
geserpersia: looking into /usr/bin/pycentral this seems to be set from the option "overwrite-local" in /etc/python/debian_config13:05
geserI don't have it set in my /etc/python/debian_config13:06
wgrantpersia: I've seen the problem with a Dapper upgrade.13:06
geserman pycentral mentions this option in the FILES section13:07
wgrantBut it doesn't look like it could be the problem here.13:07
persiaSo, do I need to do anything to the package to fix it, or just leave it alone?  (I'm processing NBS, so like to minimize actual changes)13:07
wgrantIn my case, python-opengl didn't conflict with python2.4-opengl, so pycentral refused to overwrite the python2.4 opengl bits.13:07
wgrantAdding a conflicts against it was the solution there, but it could be a local installation in this case.13:08
persiaShould python-csound conflict with python2.5-csound, just to be sure?13:09
persiaIt currently Provides: it, which makes me think that such a package used to exist (or maybe python2.4-csound)13:10
geserthe provides are generated: Provides: ${python:Provides}13:12
emgentmorning13:13
persiageser: Right.  I'm just wondering if it needs a conflicts, or if bug #227651 is not really valid.13:15
ubottuLaunchpad bug 227651 in csound "package python-csound 1:5.08.0.dfsg2-1ubuntu3 failed to install/upgrade: subprocess post-installation script returned error exit status 1" [Undecided,New] https://launchpad.net/bugs/22765113:15
persiaGrumble.  Publisher appears out of date for glib2.0-dev on archive.ubuntu.com :(13:21
wgrantpersia: Hm?13:25
wgrantThere is no glib2.0-dev, is there?13:26
wgrantAnd all right libglib2.0-devs are there..13:27
persiaErr.  Right.  libglib2.0-dev.  1ubuntu2 is in packages for amd64, and 1ubuntu3 is in the pool (1ubuntu2 is gone).  I presume it will catch up soon.13:27
wgrantOops.13:27
wgrantThat is a bug.13:27
wgrantThough it may be a sysadmin bug.13:28
persiaYep.  The first package I ran worked, but the ~12:00 UTC run seems to have befuddled it.13:28
persiaIs the bug likely to be a Packages.gz construction issue, or just a timing issue?13:28
wgrantUm, glib2.0 has been a sync for three weeks now.13:29
wgrantNo ubuntuX.13:29
geserpersia: hardy or intrepid?13:31
wgrantOh, right, Hardy makes more sense.13:31
wgrantI forget that it still exists.13:31
persiageser: intrepid.  I'm tesing bare package install now, rather than looking at a build failure, just in case I did something wrong.13:32
wgrantWhat's the full version you want?13:32
wgrant'Version: 2.17.2-1'13:33
wgrantThat's in amd64 Packages.gz on a.u.c.13:33
persiaRight.  I'm not sure where I got the error.  I may have inadvertantly spelled intrepid h-a-r-d-y.13:34
wgrantHeh13:34
krzysz01A have a project readfile. i have made .debs for it and they are in my ppa (krzysdrewniak on launchpad) check it out.13:37
krzysz01what should i do now?13:37
krzysz01i made a .deb of wine 1.013:37
persiakrzysz01: What is your goal?13:37
krzysz01i'm trying readfile into the repos, but it needs some polishing still13:38
persiakrzysz01: Are you up for polishing it, or do you need review?13:39
krzysz01i am adding some new features before i submit it for review13:40
krzysz01i'm still working on it13:40
persiakrzysz01: Then the answer to the question "what should i do now?" is "add some new features before submitting it for review".13:41
krzysz01stay tuned it will probably be good when i get to 1.013:41
krzysz01see ya13:41
DktrKranzbah... packages.{ubuntu.com,debian.org} are not responding13:42
persiaDktrKranz: packages.qa.debian.org, launchpad.net/ubuntu/+source/$(pkgname)13:44
DktrKranzunluckily, I needed them for some tests, so I guess I have to wait13:45
devfilpersia: can you please take a look at https://bugs.edge.launchpad.net/ubuntu/+source/crystalspace/+bug/242961 ? I need 2 sponsors for it13:45
ubottuLaunchpad bug 242961 in crystalspace "Please update crystalspace to 1.2.1" [Undecided,Confirmed]13:45
persiadevfil: I really know nothing about crystalspace, except that it's a mess.  Anyway, why do you need two sponsors?13:46
devfilpersia: it is a new package13:46
DktrKranzNEW? mmh13:47
devfilpersia: a new huge work as wxwidgets2.813:47
devfilnot new package, new version, sorry13:47
DktrKranzprobably it deserves four eyes to look at it, but it's not NEW :)13:47
rzrthere are a few RC  in crystal space affecting debian's release13:49
devfilrzr: this is why I upgraded it to the latest stable version13:50
rzrgood job13:50
devfilrzr: I also asked Debian maintainer to let me work at it (NMU), but I didn't receive any reply for now13:53
rzri know him personnaly13:53
rzrhe's offline now13:53
rzrcan you NMU it ?13:53
devfilrzr: yes13:54
bliZZardzwhat is NMU?13:54
devfilbliZZardz: Non Maintainer Upload in Debian13:55
geserNon-Maintainer Upload (Debian)13:55
rzronly DD can NMU packages right?13:55
bliZZardzand does that mean that people who do not maintain it, contribute a patch/feature?13:55
devfilrzr: I don't know, if I do it, probably, Debian maintainer should sponsor it13:56
DktrKranzrzr, DD can upload packages, but everyone can propose a NMU if there's RC or release-goal bugs13:57
rzrdevfil:  well I would attach a debdiff to the bug first13:57
bliZZardzi was exploring Dbus yday, i need some mentor to help me out understand it better. Would this be the right forum?13:57
persiarzr: A debdiff of a new upstream?  Isn't that hard to read (and might it be impossible to generate given binary changes in the upstream tarball)?13:58
persiabliZZardz: For some very limited types of questions, yes.  For deeper questions, probably not.13:58
bliZZardzpersia : and that would be?13:59
bliZZardzpersia: you :) ?13:59
persiaNot I, no :)13:59
devfilrzr: why Debian maintainer of crystalspace is offline?14:05
rzrhe probally has a life :)14:05
DktrKranznot so bad, after all14:06
sistpotybah, rl is overrated *g+14:06
=== bliZZardz is now known as bliZZardz_
DktrKranzbliZZardz_,  I guess the most qualified person is pitti, but he's on holidays at the moment and he's a very busy developer14:07
bliZZardz_DktrKranz : presently i am doing some triaging, but i would like to get into Dev - hence am looking out for some mentors. I am comfy with Python.14:08
DktrKranzwhoa! I've just become a DM! \o/14:09
persiaDktrKranz: Congratulations!14:09
=== bliZZardz_ is now known as bliZZardz
DktrKranzpersia, thanks14:09
bliZZardzDM? (deb mentor?)14:10
DktrKranzbliZZardz, well... if you're comfortable with dbus (or utopia-related stuff), pitti has some mentoring slots available14:10
bliZZardzDktrKranz: just started with DBus..need some initial pointers14:11
DktrKranzso I guess triaging bugs should be good at the moment14:12
DktrKranzpersonally, I haven't any documentation available14:12
bliZZardzDktrKranz: what abt projects(in python) that have mentoring slots available?14:13
DktrKranzbliZZardz, for python related bugs, you may want to look at pythonistas bugs14:14
bliZZardzDktrKranz : yes - am there.14:15
DktrKranzcool :)14:15
bliZZardzjust getting a feel of the bugs ; looks like most of them need complete module knowledge14:16
DktrKranzI'm sure there are some entry-level bugs14:16
DktrKranzeventually, you can look at bugs marked as bitesize14:17
\shwe made it ! http://leonov.tv/content/leonov-launchpad-desktop-client-first-coming14:31
persia\sh: Congratulations.  Nice job.14:32
\shpersia: thx :)14:32
* \sh is sweating like hell14:32
devfil\sh: great job!14:33
persia\sh: The bugs will come.  Have no fear :)14:33
bliZZardzpersia : LOL ")14:33
\shpersia: seele was reviewing our work already...so I have a lot of them already :) and no, I don't fear any bugs :)14:33
sistpoty\sh: apt-cache showsrc leonov doesn't yield a result... please fix :P14:34
persiasistpoty: Please file a needs-packaging bug to request new software :p14:35
DktrKranzACK from me, please upload to NEW queue :D14:35
\shsistpoty: lol14:36
* sistpoty is now off again, enjoying the weekend... cya14:42
tacone\sh: congratulations :) when will we be able to download bug informations for offline usage? (ok ok.. jocking)14:43
\shtacone: please file a wishlist report (seriously) :)14:43
\shtacone: I thought about that already, tbh14:44
tacone\sh: I feared you would have asked that.14:44
taconeehehe14:44
tacone\sh: seems I can't mark it as wishlist myself https://bugs.launchpad.net/leonov/+bug/24396014:47
ubottuLaunchpad bug 243960 in leonov "Please store bug information for offline usage" [Undecided,New]14:47
\shtacone: done :)14:48
tacone\sh: nice14:48
\shah da nixternal14:48
\shphew...14:52
nixternalyay power14:52
persianixternal: You're one of those kubuntu folk right?  Are you guys chasing NBS yet?  There seems to be a few packages listed...14:53
\shnixternal: dude, you are also mentioned in our announcement :)14:54
nixternalwhat announcement?14:55
\shnixternal: check p.u.c14:55
nixternalpersia: dunno, ask Riddell...I have been working to much lately, so unfortunately I haven't been that close to what is going on14:55
nixternalhehe, \sh...that has come along quick man...I liked reading the posts seeing how much it changed almost daily14:58
nixternalgood job!14:58
geser\sh: p.u.c is ambiguous: people.u.c, packages.u.c, planet.u.c :)14:59
\shnixternal: that's only the beginning of something really bug^Wbig14:59
wgrantpopcon as well!14:59
wgrantI wonder if any other letter is as ambiguous.14:59
nixternalgeser: ya, I did packages.ubuntu.com at first :P14:59
\shdamn...b.u.c is missing...beer.ubuntu.com ;)14:59
nixternalthen I went, OH! Planet!14:59
\shpackages is pkg.u.c. ;)15:00
Riddellpersia: NBS?15:29
gesersiretart: Hi, as you seem to care about ffmpeg-free: should "ffmpeg-config --libs avformat" result in the same output as "pkg-config --libs libavformat"?15:38
geserI ask because I stumbled about a package which FTBFS (moc) because ffmpeg-config still lists "-la52"15:38
persiaRiddell: http://people.ubuntu.com/~ubuntu-archive/NBS/15:42
persiaLooks like either there's a lot of changes expected, or lots of rebuilds to process.15:42
Riddellchanges are expected, since we moved to KDE 415:45
persiaRiddell: OK.  I won't poke them then.  Thanks.15:47
siretartgeser: ffmpeg-config will go away in the next upload. please make the package use pkg-config instead15:50
gesersiretart: I've done that. I just wondered if the output of ffmpeg-config should match the output of pkg-config as long as ffmpeg-config still exists.15:55
siretartgeser: it probably should. that's one of the reasons why we want to get rid of ffmpeg-config16:04
persiaDoes anyone have a good ddeb recovery hook for sbuild?  I'm generating ddebs, but throwing them away, which doesn't seem ideal.16:20
=== wolfger__ is now known as wolfger
devfil_persia: wow, a lot of upload today16:26
persiadevfil_: Trying to clean up some of the outstanding transitions for universe.16:27
devfil_persia: good job16:27
AnAntHello16:27
devfil_hi AnAnt16:27
persiaBah.  Uploaded things out of order again :(  At least the retry button works.16:28
AnAntpersia: that brainstorm discussion is dead ?16:29
devfil_persia: nobody is perfect16:29
persiaAnAnt: I think so.16:29
AnAntpersia: ok, I know I'm asking for a big thing, I mean about glibc16:40
persia-ECONTEXT16:41
AnAnthuh ?16:41
AnAntwhat's that ?16:41
persiaContext error, meaning I have no idea about what you are writing16:41
AnAntpersia: the suggestion I mentioned on brainstorm post16:42
lagapersia: i prefer this: http://heresabunnywithapancakeonitshead.com/16:42
AnAntis packages.ubuntu.com working ?16:42
persiaAnAnt: Oh, right.  Alternate calendars.  I have no idea how that might work, but think it deserves it's own idea on brainstorm, rather than getting mixed in with that one.16:48
AnAntis there a problem with packages.ubuntu.com  ?16:49
i4xhi everyone!16:52
i4xI want to use a .deb package to install some files, but I'm not sure where to put them. I'm using the openweakhardy/PackagingA tutorial and the only problem I have is to install that files!16:55
i4xwhere do I put them? in ed-0.9/debian/ folder??16:56
i4xI created 'ed-0.9/debian/tmp/mp/' and 'ed-0.9/debian/tmp/database/' folders. made the 'debuild -us -uc' and it returned no errors!16:58
persiai4x: What kind of files?16:59
i4xbut when installing, the postinst can't find the folders and ....16:59
i4xpersia: I need to install: /tmp/database/mp.sql  and /tmp/mp/runner (script)17:00
bobboAnAnt: yeah package.ubuntu.com is down for me too17:01
bobbos/package/packages17:02
nhandlerI have a quick question. The package I am working on comes with .desktop files (not the debian directory). The package does not have a patch system in place. Should I add a patch system to the package in order to modify the .desktop files? If so, which type of patch system should I use? Or should I just modify the .desktop files in the debdiff?17:02
persiai4x: You really don't want to install anything to /tmp.  Why not install the sql into /usr/share/$(package)/$(package).sql, and embed the script in the postinst?17:02
gesernhandler: I'd only add a patch system for big changes17:03
nhandlergeser, So for this instance, it would be better to modify something outside of the /debian directory in the debdiff than to add a patch system?17:03
persianhandler: I'd recommend checking other packages by the same maintainer, and possibly introducing a patch system that matches their style.17:03
gesernhandler: especially if the package has already some changes (from Debian) adding a patch system isn't wise (for small changes) as you end with some changes as patches (Ubuntu) and some changes applied directly (Debian)17:04
* persia very strongly agrees with geser about packages with changes in diff.gz17:04
i4xpersia: the script will be installed to init.d to run at startup, and the database will only be used once by postinst that installs the database and then removes it.17:05
gesernhandler: depending how much you want to change, I wouldn't add a patch system just to change a few lines in a .desktop file17:05
persiai4x: For the init script, investigate dh_installinit.  For the sql, I still think it belongs in  /usr/share/$(package)/$(package).sql for later reference by the user.17:06
nhandlergeser, It looks like I'll just be adding Icon lines to a few .desktop files that are in the package. I might also be making a few minor changes to the .desktop files to make them validate.17:06
persiageser: I've had more success getting patches into Debian when I follow the maintainer's style, which is why I recommend that research.17:06
i4xpersia: I know there are some .ex files as init.d.ex .. but I don't understand so much to use them.  and, are you saying to install the sql to /usr/share/$(package)/$(package).sql ? if so, how?17:08
persiai4x: Put it in debian/ then add it to your $(package).install file.17:09
i4xpersia: hm, ok.. I'll search for the *.install file..17:10
persiai4x: read the manual pages for dh_install and dh_installinit17:10
nhandlerpersia and geser: It looks like the Debian maintainer only has 7 packages. I'll look at them and see if I can find a preferred patch system. If I can't, I think I'll just include the changes in the debdiff17:10
i4xpersia: thx, I'll do that! (y)17:11
nhandlerpersia, I just checked all 7 of the Maintainer's packages. None of them use a patch system. In this case, do you agree that I should just include the changes to the files that are outside of the debian directory in the debdiff?17:18
persianhandler: Very much so.17:18
nhandlergeser and persia: Thank you for your help. I'll go and make the changes now.17:18
porthosewarp10: ping17:32
alex-weeji'm looking to build the debian package for nautilus-open-terminal on Hardy17:38
alex-weejthe version for Debian Experimental requires a newer version of libpango (i guess because it was just built against it)17:38
alex-weejhow do i get the source from DebExp?17:38
ScottK-laptopIf it's in experimental, it's probably there for a good reason.17:39
alex-weejthere's nothing much experimental about nautilus-open-terminal17:39
alex-weejit gets about 3 commits a year17:39
ScottK-laptopOne was is to go to packages.qa.debian.org, find the package package page, and then dget -x the .dsc.17:39
alex-weejthanks, i see the DSCs17:40
ScottK-laptopIt might be building against the new libpango is the only reason it's there.17:40
alex-weejScottK-laptop: that got me a .orig.tar.gz and a .diff.gz17:41
ScottK-laptopDid you use -x?17:41
alex-weej(sorry, i'm used to apt-get source wiping my bottom for me)17:41
alex-weejyes17:42
ScottK-laptopIt should have unpacked the source too then.17:42
ScottK-laptopIf it didn't, you can dpkg-source -x filename.dsc to unpack it.17:42
alex-weejit bombed here" dscverify: can't find any Debian keyrings17:42
ScottK-laptopSo it couldn't verify the signature.17:43
ScottK-laptopIIRC, the package with the Debian Keyring in it is deceptively called debian-keyring.17:43
ScottK-laptopI've got a little X windows problem here, so I'll be back (hopefully) in a little bit.17:44
ScottK-laptopHopefully that will be better.17:48
alex-weejOK, nautilus-open-terminal package in Debian Experimental builds fine on Hardy17:48
alex-weejand works as expected17:48
alex-weejwhat do i need to do to get it pushed into Hardy Universe?17:48
ScottK-laptopIt needs to get into Intrepid.17:48
alex-weejcan it then go via hardy-backports->hardy-updates?17:49
ScottK-laptopNot usally.17:49
ScottK-laptopWhat bug are you fixing to get it updated?17:49
alex-weejthe on-desktop behaviour drops you in $HOME previously17:49
ScottK-laptopIs there a bug?17:49
alex-weejit was fixed, with the old behaviour being in a gconf option17:49
alex-weeja long time ago17:49
alex-weejand 0.9 was released on 28 February 2008, so the debian packages have just been lagging17:50
ScottK-laptopIntrepid already has 0.917:50
alex-weejoh... hmm... i probably should have just used that17:51
alex-weejhaha, oh well17:51
alex-weejyay, now to just drop the habit of typing cd ~/Desktop every time i open a terminal17:52
ScottK-laptopalex-weej: I don't think this qualifies for -updates.  It's not a regression, data loss, or crash issue.17:53
ScottK-laptopYou could get the new version in -backports.17:53
alex-weejok, is it a lot of effort?17:54
ScottK-laptopNo.17:54
alex-weejwhat do i do?17:55
ScottK-laptopYou've already done virtually all the work (testing).17:55
ScottK-laptop!backports | alex-weej17:55
ubottualex-weej: If new updated Ubuntu packages are built for an application, then they may go into Ubuntu Backports. See https://help.ubuntu.com/community/UbuntuBackports - See also !packaging17:55
ScottK-laptopYou file a bug against the hardy-backports project as described above.17:55
alex-weejgot it, thanks17:55
ScottK-laptopalex-weej: Ping me when it's done and I'll approve it.17:56
alex-weejwell... i haven't really tested it properly... :S17:56
alex-weeji mean, it works17:56
alex-weejmaybe i should use it for a bit first17:56
ScottK-laptopFor backports, the standard is "Builds, installs, runs" - You pass that.17:57
ScottK-laptopMore testing is welcome, but you've done the minimum.17:57
ScottK-laptopNote that backports is not enabled by default for what are generally considered good reasons ...17:57
ScottK-laptopActually we have very few problems.17:57
alex-weejyeah of course17:57
alex-weejok just a mo17:58
alex-weejScottK: you beat me to the original bug :P18:03
ScottK-laptop;-)18:03
alex-weeji will link it in the backport request18:03
=== _neversfelde is now known as neversfelde
alex-weejScottK: https://bugs.launchpad.net/hardy-backports/+bug/24401718:06
ubottuLaunchpad bug 244017 in hardy-backports "Backport nautilus-open-terminal 0.9 from Intrepid to Hardy" [Undecided,New]18:06
ScottK-laptopalex-weej: Mark it confirmed to show you've tested it.18:07
alex-weejdone18:08
ScottK-laptopalex-weej: Approved and passed to the archive-admins.  It should get backported the next time the process backports.18:10
alex-weejcheers18:10
i4xppl, I'm trying to understand how to use 'dh_install', but I'm confused.. anyone knows how to use it and gives me an example? I just need to understand how the package recognizes the folders/files to install and moves them!18:13
ScottK-laptopi4x: Did you read man dh_intall?18:14
i4xScottK-laptop: yes, I did!18:16
ScottK-laptopi4x: Did you see the bit about "debian/package.install"?18:17
i4xScottK-laptop: I've no '($package).install' file..18:19
ScottK-laptopMake one that lists the files you want installed.18:19
i4xScottK-laptop: do I have to modify anything in rules?18:20
i4xok, but where do I create it? debian/ ?18:20
ScottK-laptopYes.18:21
i4xand what should I put there?18:21
ScottK-laptopA list of the files to install.18:21
i4x'/folder/to/create/file' ?? only?18:22
i4xand will the 'package.install' be recognized automatic while debuilding? or should I run 'dh_install' something?18:25
ScottK-laptopIs your debian/rules cdbs or debhelper?18:25
ScottK-laptopIf it's CDBS, it's automatic magic.  If it's debhelper, dh_install needs to be called, but it's in the standard template.18:26
i4xScottK-laptop: "# Sample debian/rules that uses debhelper."18:31
i4x'standard template'?18:31
ScottK-laptopYes.18:32
i4xScottK-laptop: what do you mean with 'standard template'?18:33
ScottK-laptopWhen you run dh_make to start making a debhelper based package, it makes a standard rules file.  That's what I mean.18:34
i4xScottK-laptop: "dh_install needs to be called, but it's in the standard template." it means that the rules (standard template) calls 'dh_install' ?18:37
i4xif so, my problem is resolved. if not, how when and how do I call dh_install?18:37
ScottK-laptopi4x: Please pastebin your debian/rules.18:38
i4xScottK-laptop:  http://pastebin.com/d400d5b5d18:41
ScottK-laptopLooking18:42
ScottK-laptopi4x: Look at line 74.  It's commeted out.  Remove the "#" and you should be good.18:43
i4xScottK-laptop: ok, I'll do that! another 2 questions: since the package has the name 'ed-0.9', should I create 'debian/ed-0.9.install' ? and to use 'dh_installinit' is the same thing as 'dh_install', isn't it?18:50
ScottK-laptopNo and No.18:51
i4xScottK-laptop: so, what name I call the 'package.install' file?18:53
ScottK-laptoped.install18:54
ScottK-laptop-0.9 is the version.18:54
ScottK-laptopjdong: Just because it's summer vacation for you doesn't mean you're off duty on backports.19:09
* ScottK-laptop just ack'ed ~20 for Hardy.19:09
nxvlsiretart: around?19:16
i4xScottK-laptop: I'm having this problem: "cp: cannot stat `debian/tmp//tmp/mp/runner': No such file or directory" after 'dh_install --sourcedir=debian/tmp'19:23
ScottK-laptopi4x: Try it without the /tmp19:24
i4x* "cp: cannot stat `debian//tmp/mp/runner': No such file or directory"19:25
ScottK-laptopI suspect you also need $(curdir)/debian (that's from memory so the syntax may not be perfect.19:25
i4xthe problem is in the both '/' in the middle!19:31
i4x'rules' has 'dh_install --sourcedir=debian'19:31
i4xand the 'ed.install' has 'tmp/mp/runner' .19:31
i4xbut it puts 2 '/' in the middle!19:31
i4xbut ScottK-laptop, do you know why it puts the // between 'debian' and 'tmp/mp/runner' ?19:36
nxvlScottK-laptop: is there any doc somewhere on how to test you .debs?19:38
ScottK-laptopi4x: No, but I don't think it matters.19:39
ScottK-laptopnxvl: Context please?19:39
nxvlScottK-laptop: i still have an rpath issue with augeas, but i don't know how to test the .debs to see if they still have it19:40
ScottK-laptopHow do you know you still have it?19:40
nxvlScottK-laptop: because siretart say i still have it (after reviewing it)19:41
ScottK-laptopAh.19:41
ScottK-laptopDoes lintian run against the .deb complain about it?19:41
nxvland i don'y want to try to fix the problem, upload it, and wait for someone to tell me it's already fixed or not19:41
nxvli have run it19:41
nxvlbut nothing shows up19:41
ScottK-laptopnxvl: Have you seen http://wiki.debian.org/RpathIssue19:44
nxvlScottK-laptop: yep19:45
ScottK-laptopnxvl: What happens when you grep aclocal.m4, configure.in or configure.ac for hardcode_into_libs19:47
i4xScottK-laptop: it makes the 'cp' return: "cp: cannot stat `debian/tmp//tmp/mp/runner': No such file or directory". So I think that the problem is there!19:47
i4x* "cp: cannot stat `debian/tmp//mp/runner': No such file or directory"19:47
ScottK-laptopi4x: At this point I usually login to my pbuilder chroot, use debuild to build the package and then see what dirs/files are present when it dies.19:48
nxvlScottK-laptop: a lot of "yes" show up19:49
ScottK-laptopnxvl: Then based on that wiki page, I'd say that's how you know.19:50
nxvlScottK-laptop: yes, but the problem is: should i use a patch system? or should i run sed on the debian/rules (this hasn't worked early)19:51
ScottK-laptopnxvl: I'm not sure.  I'd tend to go with patch system.19:52
nxvlScottK-laptop: but patch is before configure19:53
warp10porthose: pong19:58
=== fta_ is now known as fta
porthosewarp10: can I send you a pm :)20:03
warp10porthose: for sure! BTW, I'm just reading your mail :)20:03
porthosecool that's what it's about20:04
porthosewarp10:  email me with what you decide please, have to get ready for work (sigh)20:07
warp10porthose: no problem at all about a new mentee for me. BTW, once big Cesare expressed some concerns about people mentoring people speaking the same mother-tongue20:08
norsettowarp10: I'm not big, I'm just weight-challenged20:09
warp10norsetto: heh... speaking about the devil...20:09
norsettowarp10: yes, its part of the guidelines we discussed, try to avoid mentor-mentee of the same mother-tongue20:11
porthosenorsetto: np have another I can send warp10 after I do the work  thing20:13
porthosewarp10: would that be Ok20:13
warp10norsetto: I definitely agree. BTW, no problem for me on taking a new mentee (also because David is a little busy at now)20:13
warp10porthose: sounds good!20:14
norsettoporthose: sure, how are things going? having fun :-) ?20:14
porthosenorsetto: getting organized and trying to make my first match20:15
porthosenorsetto:  but alas duty (work) calls,  bye20:17
norsettoporthose: bye20:17
warp10bye porthose20:18
norsettowarp10: say the truth, you are also looking forward to working on Sunday, or Christmas Eve20:19
warp10norsetto: I like to work on Sunday, but not on Christmas Eve, indeed!20:19
bobbowarp10: hey Andrea!20:21
warp10Ciao bobbo! How do you do?20:21
bobbowarp10: un-banned from the computer :D20:22
bobbounfortunately im on holiday for a week tomorrow, then have a job all summer, but at least i can check my emails20:22
warp10bobbo: unbanned? great! Intrepid really needs your help to be a rocking distro!20:23
bobbowarp10: hehe, 30 uploads so far in Intrepid, itll be rocking already :P20:24
* norsetto didn't even know there was a way to see how many uploads one did20:25
warp10bobbo: 30, really? sounds great!20:26
bobbonorsetto: i wrote my own PHP+MySQL app to log my uploads, i add the details into the db then it shows a load of statistics etc about it20:26
bobbowarp10: hehe, i think its around 28 at the moment, but i have some to add in when my server gets back up20:27
warp10norsetto: Dktrkranz wrote a small python script to get the ranking20:27
* warp10 runs the script20:28
norsettobobbo, warp10: interesting, how do these scripts work? bobbo one I guess is local, but the dktrkranz one I guess is operating on the LP db?20:28
bobbonorsetto: yeah mine is local, Dktrkranzes sounds interesting20:29
warp10norsetto: nope, it gets data from <distro>changes@ mailing list20:30
norsettowarp10: ah, I didn't know the uploader was in those emails20:30
bobbowarp10: have you got a link to that script?20:30
warp10norsetto, bobbo: http://paste.beeseek.org/3620:31
bobbowarp10: thanks :D20:31
warp10bobbo: ;)20:32
norsettowarp10: ah, ok, thats not the uploader then20:32
warp10norsetto: nope, it's just the "changed by" field.20:33
warp10bobbo: you are 23rd in the list, with 26 packages20:33
bobbo26 packages, almost 30 (ish) :P20:33
norsettowarp10: I guess one could also see the uploader by checking the gpg key though20:33
bobbo1Jonathan Riddell (266 packages) o.020:34
warp10norsetto: well, I would like to modify that script using gpg keys to make a sponsors ranking, indeed20:34
warp10norsetto: it shouldn't be that difficult20:35
norsettowarp10: the difficult bit would be to collect all the gpg keys20:35
warp10norsetto: indeed. That would probably require some work with Launchpad20:36
* norsetto wonders who root was20:38
warp10norsetto: the real mysterious one is 141??????? (1 packages)20:39
norsettowarp10: also, i was surprised to see that siretart had only 1 upload, until I saw that he is uploading with his real name too :-)20:40
warp10:D20:40
=== bmk789__ is now known as bmk789
i4xnorsetto: are you there?21:37
norsettoi4x: yes sir21:38
=== pgquiles_ is now known as pgquiles
i4xnorsetto: already there? I'm having problems building a package.. because it puts 2 '/' between 'debian' and 'tmp/mp/database.sql', returning error.22:23
i4x*still there?22:24
norsettoi4x: that shouldn't be a problem; anyhow, check dir names in debian/rules (and perhaps also debian/*.install)22:25
i4xI checked!22:27
i4xnorsetto: debian/rules has 'dh_install --sourcedir=$(CURDIR)/debian'  and debian/ed.install has 'tmp/mp/database.sql'22:29
i4xI can't understand why appears '//' between  'debian' and 'tmp/...'22:30
i4xnorsetto:22:33
i4xdh_install --sourcedir=/home/i4/Desktop/final/package2/ed-0.9/debian22:33
i4xcp: cannot stat `/home/i4/Desktop/final/package2/ed-0.9/debian//tmp/mp/temporario': No such file or directory22:33
i4xany idea?22:34
norsettoi4x: the path seem wrong, you should use relative paths only unless you want things to fails in chroot, pbuilder, etc.22:36
warp10Any REVU admin around?22:37
jpdswarp10: key sync?22:43
geseri4x: and that path is really there?22:44
warp10jpds: no, REVU still sees me as a contributor, I would like to fix that22:44
jpdswarp10: One moment please.22:45
i4xnorsetto and geser: this is the path: '/home/i4/Desktop/final/package2/ed-0.9/debian/tmp/mp/'22:46
i4xbut I just need one '/', not 2.22:47
jpdswarp10: "Altering warp10@ubuntu.com to level reviewer"-22:48
geseri4x: what gives "ls -l /home/i4/Desktop/final/package2/ed-0.9/debian//tmp/mp" ?22:50
warp10jpds: mmm... my login is warp10@libero.it, but if you can change it to warp10@ubuntu.com I would prefer it. If I try to login with @ubuntu.com, it says "There is no REVU account for warp10@ubuntu.com, yet."22:51
i4xgeser: that returns: ls: cannot access /home/i4/Desktop/final/package2/ed-0.9/debian//tmp/mp: No such file or directory22:51
jpdswarp10: You'll have to upload a package with that email.22:51
jpdswarp10: Unless you'd prefer to use the old login.22:51
warp10jpds: I am seen as a contributor with the old login. BTW, I should upload a package within days, that should fix everything22:52
jpdswarp10: OK, ping me when you've uploaded with the new email and I'll bump the level.22:53
* jpds goes to bed. Night all.22:54
nxvlwhy do we install recommends by default22:54
nxvlis a mess22:54
nxvl:S22:54
geseri4x: the extra / should hurt, does "ls -l /home/i4/Desktop/final/package2/ed-0.9/debian//tmp" work?22:54
warp10jpds: great. Thank you so much!22:54
norsettoi4x: the problem is surely not the doubl /22:54
gesernxvl: :) just file bugs about wrong recommends22:54
i4xnorsetto and geser: it seems that 'debuild' removes the folder debian/tmp !!22:55
i4xand then, when needed, it can't find it!22:56
geseryes, the clean target gets called before build22:56
i4xgeser: what should I do?22:57
i4x:S22:57
nxvlgeser: all?22:57
nxvl:D22:57
gesernxvl: if all are wrong then yes22:57
nxvlyes22:57
nxvlthey ALL are wrong22:57
nxvl:P22:57
geseri4x: can you pastebin your debian/rules?22:58
gesernxvl: now you have a good reason to fix it now :)22:58
i4xsure!22:58
i4xgeser: ..a moment, please!22:58
i4xgeser: http://pastebin.com/d312c383d23:00
nxvlgeser: i mean, the recommends are not strictly necesary and they affect the choose option23:01
nxvlit's just wrong23:01
i4xgeser: should I move 'dh_install --sourcedir=$(CURDIR)/debian' before 'dh_clean' ?23:01
geserno, your rules file looks ok23:04
i4xgeser: do you have any idea?23:05
gesernot yet23:08
albert23i4x: do you have a file debian/compat? If so, which value is in it?23:08
i4xalbert23: 523:08
geseri4x: can you make the whole source package available? so I can look at it23:09
i4xiupiiiiiii!!23:10
i4xit seems it worked!!23:10
i4xgeser and albert23: I changed debian/rules with 'dh_install --sourcedir=/home/i4/Desktop/final/package2'23:11
i4xand moved the folder 'tmp' from debian/ to /home/i4/Desktop/final/package2/23:12
i4xnow the tmp is available to dh_install !23:13
geserhow did you create the files in debian/tmp before?23:14
geseri4x: I guess you don't need the --sourcedir anymore which has also the advantage that other people can build the package too23:15
i4xgeser: before, I moved the files to debian/tmp..23:16
geseri4x: which is bad as dh_clean removes the debian/tmp dir as that's where usually make install temporarily installs the build source and dh_install moves the files from there to the specified locations23:19
i4xand this is only a practice exercise. I'm making a mini-project to Sistem Programming (school).. but as I ever wanted to know how the packages are made, I decided to deliver de mini-project in one package! so I don't need to burn virtual-machines nor a big manual to the teacher evaluate! now, the teacher just installs the package!23:20
i4xgeser and albert23: thank you a lot for the help and time spent! See you someday.. I hope have some time to start building some decent packages! and learn how to fix some bugs! :D23:26
nxvlsiretart: around?23:30

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!