[12:01] ok [12:01] or do you want to test it with all possible ubuntu grub setups ? [12:01] (thats a package that can really fuck your sytem if not working properly) [12:01] yeah maybe it will work with one of those 632.346 installs [12:03] heh === dredg [niall@malkovich.office.blacknight.ie] has joined #ubuntu-motu [12:04] hey Dr_Aevil [12:04] dredg :-) [12:05] ogra: it has gone... :-) [12:05] yeah [12:06] dredg: do you know anything about the status of your PythonTransition packages? [12:07] dholbach: all should be taken care of i think [12:07] ok... there are still some in "being transitioned" [12:07] oh, hmm [12:08] mailping, toursst, jaxml are fixed either by me or someone else [12:09] blender i had listed as problematic [12:09] hrm... hard to tell from apt-cache showsrc blender [12:10] wow, who killed planet.u.c with the stick of ugly? [12:11] dholbach: right, it requires python to build as scons is basically python [12:11] it builds in pbuilder [12:12] but on the buildd it fails because of ccache [12:12] ccache requires $HOME to be set. scons nukes the environment === dholbach can feel the pain [12:12] chicken. egg. chicken. egg. divide by chicken error. [12:12] or something === zerokarmaleft [~zerokarma@ip68-12-45-133.ok.ok.cox.net] has joined #ubuntu-motu [12:13] :-) [12:17] dredg: hah. look at gcc-2.95. [12:17] dredg: it depends on cpp-2.95 [12:17] dredg: cpp-2.95 depends on, you've guessed it, gcc-2.95 [12:17] diamond: that's allowed [12:17] dredg: ok, but how? [12:18] diamond: with enough crack anything is possible [12:18] please apply the above egg/chicken formula. [12:18] compiler-bootstrapping [12:18] diamond: nah, i know it's possible. i just dunno how :) === dholbach has some ... german ... lecture notes on that :-) [12:19] dholbach: should i just google? [12:19] i could probably read german [12:19] i think. it's been some time [12:19] if you want to know how it works [12:19] i'll have a look [12:20] not especially :) [12:20] mvo can tell... he knows everything about compilers [12:20] uhm, what are you wondering about? [12:21] how to get gcc-2.95 to build? [12:21] german info: http://www.inf.fh-dortmund.de/concute/contents/personen/dozenten/buechter/folien/foliensatz.zip :-) [12:22] Mithrandir: yeah [12:23] "don't care" not being an option? === dredg laughs [12:23] Mithrandir: depending -) [12:23] gcc-2.95 is dead upstream and programs should just be ported if they're broken with newer gccs. [12:23] Mithrandir: i'm inclined to agree, still tho [12:24] dholbach: hmm, seems that it's been some time that i've had to read german :) perhaps it will come back to me some other time, when my brain is feeling less mushy and spongey [12:24] gcc can be built with more or less any c compiler, it's just a bit of hassle. [12:24] :-))) [12:24] Mithrandir: aye, but the issue is that it build-deps upon itself. i don't understand why. === T-Bone [varenet@T-Bone.developer.debian] has joined #ubuntu-motu [12:25] me again [12:25] http://buildd.slashdirt.org/logs/multipath-tools_0.4.1-1_20050407-1343 <- fails in 'clean' target :P [12:26] diamond: probably because it's broken if you try to build it with any other compiler, 'cause it's broken by itself. [12:29] ajmitch: Apparently nscd is selinux aware. [12:30] jbailey: ok.. I'll look that up [12:31] ajmitch_: I just happened to notice it fly by when running configure. =) [12:31] bbl === jani [~jani@iv.ro] has left #ubuntu-motu [] [12:40] http://buildd.slashdirt.org/logs/torch-examples_2-3_20050407-1218 <- invalid build-dep, expecting libtorch-dev (>= 2.3). Versionned build dep on meta package, bad. Should be libtorch3-dev [12:40] out for a walk.. brb [12:40] T-Bone: torch-examples was removed [12:40] dholbach: because of the ftbfs maybe? :) === lamont_r [~lamont@phantom.acmeps.com] has joined #ubuntu-motu [12:41] T-Bone: because of ETOOMUCHPAIN [12:41] brb [12:42] oic ;) [12:45] ENOMARYJ [12:45] judging from your hackergotchi [12:45] http://buildd.slashdirt.org/logs/wnn6-sdk_1.0.0-12_20050407-1249 <- that one might be looking for libwnn6.so.1.0.0 (at the end of the build) === jani [~jani@iv.ro] has joined #ubuntu-motu [12:48] http://buildd.slashdirt.org/logs/sprng_2.0a-2_20050407-1215 <- doesn't stop where it should [12:49] tseng: LOL [12:50] dholbach: thanks, removed my and Riddell's tasks which are done in PythonTransitionTODO [12:51] haggai: ROCK [12:51] are you guys interested in -fPIC issues? [12:51] yes [12:51] oh [12:51] you're gonna love me then :) [12:51] http://buildd.slashdirt.org/logs/soya_0.9.1-2ubuntu1_20050407-1159 <- -fPIC === thoreauputic_ [~debianarc@wolax6-137.dialup.optusnet.com.au] has joined #ubuntu-motu [12:56] -fPIC affects amd64 [12:56] so [12:56] fix it :) [12:56] oh sure i will [12:57] if you grant me clearance to upload fixed packages ;) [12:57] actually it affects amd64, ia64 and hppa, afaik [12:57] http://buildd.slashdirt.org/logs/sagasu_2.0.5-1_20050407-1027 <- want old gnome? [01:00] http://buildd.slashdirt.org/logs/ngspice_15-2_20050407-0844 <- broken clean target maybe? binaries left over the place (long log) [01:01] i have also a bunch of XFree -> Xorg transition bug but i suppose you guys already know them [01:02] T-Bone: which ones? how old is your snapshot? [01:02] i don't have a snapshot [01:03] i'm up-to-date with current source tree, update every 3 hours [01:03] (from a.u.c) [01:03] ok [01:03] fire away, which ones [01:03] right now i'm stumbeling on iris and iterm, as it seems [01:04] dholbach: there are some packages in the archive that haven't changed since they were built for warty... which means they've only been tried in hoary-test, and that was only partial, etc. [01:04] http://buildd.slashdirt.org/logs/iris_0.12-1_20050407-0238 [01:04] http://buildd.slashdirt.org/logs/iterm_0.5-3.1_20050407-0241 [01:04] lamont_r: yes... can imagine [01:04] dholbach: those will be the ones that t-bone is hitting that you don't already have [01:05] take care of those [01:05] http://buildd.slashdirt.org/logs/sppc_1.0.1-8.1_20050407-1212 [01:06] sppc seems not so trivial [01:06] dholbach: in a general way, if you look at http://buildd.slashdirt.org/FAILED.html ignoring all KDE and mozilla stuff, you should find mostly stuff of interest to you guys [01:06] nite folks. ye guys are doing an amazing job [01:06] there are a bunch of hppa-centric issues, i'll give them some love ;) [01:06] diamond: you do too! [01:06] diamond: sleep tight === diamond waves [01:07] dholbach: most of the errors there are X, -fPIC, or general rules borkage; mostly [01:08] cool [01:08] and per Lathiat's request, mod-gzip should (if i have set it up properly) compress the logs so that you don't have to dl multi-megs files ;) [01:10] hping3 needs syncing btw [01:10] iris uploaded [01:10] hping3 should have been synced [01:11] http://buildd.slashdirt.org/logs/hping3_3.a2-1ubuntu1_20050407-0232 <- FTBFS (endianness issue, doesn't show on debian) [01:12] iterm uploaded [01:13] http://buildd.slashdirt.org/logs/mkhppa3/xfld-welcome_0.2-3_20050407-1421 <- looks like missing b-dep [01:14] http://buildd.slashdirt.org/logs/mkhppa3/muscle_3.51-1_20050407-1420 <- utmost sweetness "-mpentiumpro", yum... [01:14] crimsun, jani: ^-- xfce stuff [01:14] yes? [01:14] xfld-welcome [01:15] sppc uploaded [01:15] I think it's ok we don;t need that package it's not part of xfce core [01:16] jani: shall it be removed? [01:16] yes [01:16] i thought we were shipping all of debian main in universe? === T-Bone needs to update himself :} === xuzo [~xuzo@81-203-41-93.user.ono.com] has joined #ubuntu-motu === tritium [~rimbert@ee213-dhcp-9.ecn.purdue.edu] has joined #ubuntu-motu [01:24] wiki/UniverseUnmetDeps still has some nice stuff :-) [01:26] http://buildd.slashdirt.org/logs/mkhppa3/tightvnc_1.2.9-5_20050407-1211 <- most likely a rule issue. says : "To build Xvnc, do "cd Xvnc", followed by "./configure" then "make"", does "(cd Xvnc; /usr/bin/make World)" (lengthy log, issue at the end of course) [01:27] http://buildd.slashdirt.org/logs/mkhppa3/xmms-alarm-bmp1_0.3.6-0.1_20050407-1224 <- busted source? [01:29] http://buildd.slashdirt.org/logs/mkhppa3/blacs-mpi_1.1-23_20050407-1241 <- needs libmpich1.0-dev? (long log) === T-Bone notices the clock, needs to go to bed. You guys can give a look at http://buildd.slashdirt.org/FAILED.html - don't bother too much with stuff not obvious enough, hppa has some weird issues sometimes ;) [01:31] xmms-alarm-bmp1 was removed afaik [01:31] dholbach: ok. Of course, when a package is removed, the log remains, i have to purge them by hand :) [01:31] :-) [01:31] i see you guys uploaded some stuff, autobuilders has just started :) [01:31] s/has/have/ [01:32] FAILED.hmtl is uploaded every 1/2h, and the 'needs-build' list every 3h, fyi [01:32] have fun and keep up the good work, i'm off to bed ;) [01:32] night T-Bone [01:32] thx === lamont_r [~lamont@phantom.acmeps.com] has joined #ubuntu-motu [01:36] crimsun it looks like it was an -Wl,-E after all :) [01:36] I think it'll work :) [01:37] actually two, one for xfprint4 one for xfprint-manager [01:37] hiya everyone... how's progress on teh universe [01:37] hey schweeb [01:37] final job interview is tomorrow [01:37] yay [01:37] rocking! :-) === ajmitch_ wishes he could install ubuntu on this work machine now :( [01:39] why can't you? [01:39] because it's a work machine, not one of mine === T-None is running Warty at work ;) === T-None ducks and goes to bed for good ;) [01:40] heh [01:40] I have my hoary workstation [01:40] s/workstation/laptop [01:41] and I have my warty workstation [01:41] and like 12 Debian servers [01:41] but soon, I'll be on a Solaris WS ;_; === lamont_r had sid on his work machine [01:47] when the boss told me I should leave windows on it (2 weeks after it arrived), I told him it was too late. [01:47] any MOTU please upload http://sourcecontrol.net/~jmonoses/hoary/xfprint4/ [01:47] jani: will do [01:47] fixes xfce print manager hurray! [01:47] I can go home and sleep now :) [01:48] ROCK! :-) [01:48] thanks dholbach, you're an angel (or something ;) [01:48] jani: 300th upload! :-) [01:48] with all my love to the XFCE crew :-) [01:48] congrats to the most prolific MOTU the universe has seen! [01:48] :-))))) [01:49] lamont_r: doing solaris admin, I kinda need a solaris workstation, sadly enough [01:49] dholbach, WOOHOOO [01:49] \O/ [01:49] dholbach: man, weren't you up to 200 just like a week and a half ago? [01:49] schweeb: yeah, well.. === dholbach pipes innocently [01:50] jani: done [01:50] schweeb, solaris ? [01:50] how sad [01:50] crimsun, when you read this: thanks for fixing that 'quit' symbol problem :) I don't think I wouldn't have thought to this fix without remembering yours ;) [01:50] thanks daniel [01:50] goodnight all [01:50] any time :-) [01:50] ogra: it'll be a challenge, I'll have fun [01:50] sleep tight [01:50] night jani [01:50] I'll have to come to work in the morning === jani waves === jani [~jani@iv.ro] has left #ubuntu-motu [] === ogra wonders about fun and solaris.... [01:53] abakus is uploaded? [01:53] yop [01:53] i'll look at gaphor [01:54] woo [01:54] thanks ogra [01:55] dholbach: nooo! gaphor sucks, Umbrello rules [01:55] Riddell: what do you think? for how many of the Universe packages do i have REAL LOVE? [01:56] dholbach: could you just make gaphor a meta package which depends on Umbrello? pretty please? [01:56] Riddell: only if you fix wiki/UniverseUnmetDeps [01:57] Riddell, btw, how is the kubuntu testing coming along ? [01:57] (are we there yet ?) [01:58] Riddell: i could have slapped umbrello in the face six times a day, when i wrote about an uni project :-) [01:58] ogra: I think we still need to test the amd CDs [01:58] and not having an AMD64 machine that's kinday hard to do [01:58] dholbach, you didnt write pure latex in vim ? o_O [01:59] ogra: the uml diagrams by umbrello were a bit on crack [01:59] Riddell, i have no media anymore....and a broken dvd rom that only boots from dvd... [01:59] dholbach: well I never noticed any bugzilla entries from you [01:59] Riddell: that was like 9 months ago [02:00] or even longer [02:00] dia was even more broken :-) [02:00] dia is pretty good, it's just ment for different things [02:00] dia still is broken [02:01] and dia is a work of perfection compated to Kivio :) [02:01] compared [02:01] i used autodia, which was... i don't find appropriate words for it [02:02] my feelings towards it were, getting a huge club from somewhere === tritium has searched and searched for Hvels Original Bitterbier, but cannot find any in Indiana [02:03] Hvels rocks [02:04] it was your recommendation that initiated my search [02:04] :-) [02:04] yumm ... hvels [02:05] :-) [02:06] gaphor uploaded [02:09] anybody wants to join me fix the missing bits on UniversePythonTransitionTODO ? [02:09] i have a look at scalemail [02:19] scalemail fixed === arzajac [~arzajac@modemcable201.119-200-24.mc.videotron.ca] has joined #ubuntu-motu [02:48] anyone an idea why http://people.ubuntu.com/~lamont/buildLogs/g/gaphor/0.5.1-2ubuntu1/gaphor_0.5.1-2ubuntu1_20050408-0139-i386-failed didnt work? [02:49] played nicely with my pbuilder [02:49] oh wow, so it was -Wl,-E ? great. [02:49] (thanks dholbach) [02:49] crimsun: hm? [02:49] for uploading? [02:50] dholbach: yep [02:50] np :-) === herzi [~herzi@c149008.adsl.hansenet.de] has joined #ubuntu-motu [02:51] dholbach: nfc [02:51] that's what i thought [02:51] dholbach: "because it hateses you" === dholbach cries desperately === yfir [~yfir@dsl-129-139.aei.ca] has joined #ubuntu-motu [02:52] oh, herzi === arzajac [~arzajac@modemcable201.119-200-24.mc.videotron.ca] has left #ubuntu-motu ["Leaving"] [02:52] dholbach: how'd it do on the other architectures? [02:52] lamont_r: it's an all package [02:52] hi herzi [02:53] herzi, i just uploaded hula, didnt expect you to come back tonight (you forgot to change the distribution from unstable to hoary) [02:58] ogra: i'm on my way to bed [02:58] bad alcohol... [02:58] oh [02:58] so you had a release pre-party already :-) [02:59] kindof [02:59] weekly bar evening in the students doritory === herzi plays the barkeeper and goes after the last guest [03:00] good night [03:00] good night herzi [03:02] any low-hanging fruits in universe? [03:03] Have a good night, everyone. [03:03] night tritium [03:04] bye tritium [03:04] bye crimsun, dholbach :) === tritium waves === tritium [~rimbert@ee213-dhcp-9.ecn.purdue.edu] has left #ubuntu-motu ["Leaving"] === StoneTable [~stone@c-67-184-135-68.hsd1.il.comcast.net] has joined #ubuntu-motu [03:12] hey StoneTable [03:12] hey === haggai [~halls@i-83-67-20-196.freedom2surf.net] has joined #ubuntu-motu === kiko [~kiko@200-171-140-32.dsl.telesp.net.br] has joined #ubuntu-motu [03:26] hey dholbach? [03:29] duderino.. [03:30] kiko, he probably fell asleep in front of the keyboard.... :) [03:30] hey kiko [03:31] wooo [03:31] mdz tells me you have issues with gaphor [03:31] still true? [03:31] yes [03:31] http://people.ubuntu.com/~lamont/buildLogs/g/gaphor/0.5.1-2ubuntu1/gaphor_0.5.1-2ubuntu1_20050408-0139-i386-failed [03:31] okay [03:31] it's a broken pofile [03:31] and i don't have no clue at all [03:31] were translations for it recently updated? [03:31] nope [03:31] it only happens with boxes with lots of memory :) [03:31] just grabbed the sources from last build and changed some python2.3 -> python2.4 stuff [03:32] erm [03:32] can I see a diff? [03:32] just a sec [03:32] dholbach, is there only one pofile there? nl.po? [03:33] yes [03:33] strange [03:33] --- gaphor-0.5.1/debian/control [03:33] +++ gaphor-0.5.1/debian/control [03:33] @@ -2,7 +2,7 @@ [03:33] Section: devel [03:33] Priority: optional [03:33] Maintainer: Cdric Delfosse [03:33] -Build-Depends-Indep: debhelper (>= 4.0.0), python-dev, python-gtk2, python-diacanvas2, xvfb, xbase-clients, xfonts-base, xsltproc, docbook-xsl+Build-Depends-Indep: debhelper (>= 4.2.28), python-dev (>= 2.4), python-gtk2, python-diacanvas2, xvfb, xbase-clients, xfonts-base, xsltproc, docbook-xsl [03:33] that's not really much, hrm? :-) [03:33] that's it?! [03:33] yes [03:34] and a changelog entry [03:34] hmmm [03:34] so it's breaking because a very large int is being supplied to struct.pack [03:34] this large int is the number of messages in the pofile [03:34] I was thinking this broke because of language packs [03:34] no? [03:35] unlikely [03:35] language packs haven't affected the process yet at that point [03:35] they dont pull strings from universe [03:35] heh, that too [03:35] (or do they ?? ) [03:35] pushing strings is more relevant here :) [03:35] ogra: they do not, but even if they did, they only act on the build tree at the end [03:35] ah [03:36] can I see the pofile in the tree where the crash happened? [03:37] lamont_r might be able to arrange that [03:38] http://moz.gotdns.org/ubuntu/nl.po [03:38] if this is due to some regression in python 2.4.1, there will be hell to pay ;-) [03:38] dholbach: I thought it didn't fail for you locally? [03:38] mdz: build dirs all purge automatically - I can go force the issue, but would prefer to drive home first, now that my dvd is done downloading [03:38] mdz: no it didnt, just wanted to supply the pofile in question [03:38] I think kiko wanted to be sure it wasn't modified somehow during the build [03:38] lamont_r: I'm sure dholbach has other things to keep him busy ;-) [03:39] heh [03:39] mdz, right. just sanity check that the pofile is fine -- if it is, then something in the msgfmt.py code is broken, but that's standard python.. [03:39] running home now, back online shortly [03:39] so I need to roll home too [03:39] kiko, there is no single utf8 char in the po file [03:39] ogra, and it's pretty short. [03:39] yop [03:39] mdz: i collected all the low-hanging fruits in universe it seems :-) [03:39] ogra, dholbach: I need to run off unfortunately; let me give you a pointer in how to track this down [03:40] I suspect the pofile may be corrupt. [03:40] if it's not, msgfmt is just being crazy and parsing it wrong [03:40] hey [03:40] I'm able to reproduce the failure here [03:40] I'll do some debugging [03:40] it may be that python2.4's included msgfmt.py works and gaphor's included copy is broken [03:40] mdz, and it's a pristine pofile? [03:41] only 10 strings ? must be a small program.... [03:41] here's the code in msgfmt.py: [03:41] output = struct.pack("iiiiiii", [03:41] 0x950412de, # Magic [03:41] [...] [03:42] >>> import struct [03:42] >>> struct.pack("i", 0x950412de) [03:42] Traceback (most recent call last): [03:42] File "", line 1, in ? [03:42] OverflowError: long int too large to convert to int [03:42] yeah, same as: http://cvs.sourceforge.net/viewcvs.py/gaphor/gaphor/po/nl.po?rev=1.5&view=auto [03:42] hmm. [03:42] I don't think the pofile enters into it at all [03:42] why doesn't that happen for me? [03:42] that value can't be packed as "i" [03:42] err, sorry [03:42] err wait [03:43] there's an L missing there [03:43] oh-oh. [03:43] I get the same behaviour everywhere I try [03:43] that's the problem :) [03:43] output = struct.pack("Iiiiiii", [03:43] 0x950412deL, # Magic [03:43] 0, # Version [03:43] len(keys), # # of entries [03:43] 7*4, # start of key index [03:43] 7*4+len(keys)*8, # start of value index [03:43] 0, 0) # size and offset of hash tabl [03:43] that s the code in SVN: [03:43] http://cvs.sourceforge.net/viewcvs.py/gaphor/gaphor/utils/msgfmt.py?rev=1.3&view=auto [03:43] can someone find out where that L went? [03:43] why is this code part of gaphor anyway? [03:43] isn't there a 'pygettext' or such? [03:44] hoho [03:44] confirm that adding the L fixes the build [03:44] http://cvs.sourceforge.net/viewcvs.py/gaphor/gaphor/utils/msgfmt.py?r1=1.2&r2=1.3 [03:44] that's why: we're packaging an old version of gaphor. [03:44] the more recent version includes a fix [03:44] interesting that nobody else caught that one. [03:44] oh, well -- thought it was a more exciting issue. :) [03:45] so this probably broke with python 2.3->2.4? [03:45] i'll try packaging a new one [03:45] hmm [03:45] dholbach: would be simpler to apply the patch from CVS [03:45] aha [03:45] hmm, doesnt fail here on amd64... [03:45] it used to be a futurewarning: [03:45] >>> import struct [03:45] >>> struct.pack("i", 0x950412de) [03:45] :1: FutureWarning: hex/oct constants > sys.maxint will return positive values in Python 2.4 and up [03:45] '\xde\x12\x04\x95' [03:45] that's 2.3. [03:45] yep [03:45] mdz, so yes. [03:46] mdz: we want NEW crack :) [03:46] heh [03:46] mdz, it used to turn into a negative value, which was in itself pretty bad eh? [03:46] dholbach: what kind of machine did you do your build on? [03:46] amd64 [03:46] hehe [03:46] hoho [03:46] we're the blessed ones :) [03:46] that still doesn't make much sense [03:46] amd64's ints are long ints, aren't they? [03:46] int on amd64 is still 32 bits [03:47] hmmm [03:47] maybe python2.4 is rigged to use long ints everywhere on amd64? [03:47] at least in C [03:47] don't you hate all these different codepaths? [03:48] anyway, dholbach, mdz: can I help with anything else? [03:48] kiko: you are released, thanks for the help [03:48] thanks a lot [03:48] hopefully you'll be released shortly too [03:49] sleep tight kiko [03:49] building new version :-) [03:51] does that mean you don't need the file? [03:53] ok... gaphor-0.7.0.1 makes problems, i'll get the patch from cvs [03:54] hmm, jani has put nvu on NEW packages and reviewed it himself ?? [03:54]