[04:20] Laney: opencv finished, I'm testing wallch with the changes [04:25] lenios: so, asm3 doesn't even build in oneiric [04:40] Laney: new opencv uploaded, so in ~5-6 hours you should be able to retry your build [04:40] Laney: I mean no-change-rebuild opencv, not new :) [06:44] good morning === dholbach_ is now known as dholbach [08:15] micahg: oh, cool. wallch worked then [08:16] ? [08:16] Laney: hmm, failed in LP worked locally, maybe it didn't publish, trying again [08:19] http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=641670 [08:19] Debian bug 641670 in docky "docky: Spelling Error" [Minor,Open] [08:19] * Laney giggles [08:31] Laney: wallch finished, so go ahead and retry yours [08:31] okey dokes [08:33] https://launchpad.net/ubuntu/+source/libdecodeqr/0.9.3-5ubuntu1/+build/2784201 [08:33] keep an eagle eye out [08:34] we [08:34] err, we'll see how much longer I'm awake, but will check in the morning [08:45] build of asm3 lacking mh_clean... [08:50] micahg: yeah, looks like it worked [08:50] good call [08:50] awesome [08:55] i can get the latest asm3 build, but i need to remove the mh_clean added in debian/rules (which is said to clean up after the maven helpers) [09:00] looks like maven-repo-helper >= 1.5 is needed to use mh_clean [09:00] and oneiric only have 1.2 [09:09] so, basically eclipse needs to go in a PPA for oneiric [09:09] wait, maven-repo-helper is actually 1.5.2 in oneiric [09:11] i think i can put it in my ppa [09:14] what i don't know is why asm3 building is not finding maven-repo-helper even if it's in Build-Depends-Indep. I do have to install the package on my machine [09:29] how are you building it? [09:31] i install maven-repo-helper (i don't know why it doesn't work without it), and then use pbuilder and pdebuild [09:34] you sometimes need extra packages to run a clean target [09:45] and how do i specify extra packages when uploading to ppa? [09:46] by "run a clean target" I meant in order to build the source package on your machine [09:46] dpkg-buildpackage -S or similar [09:47] if you need to add extra packages then put them in build-depends(-indep) [09:50] lenios: I just built asm3 and it worked just fine [09:51] try pbuilder build asm3_3.3.2-1.dsc [09:51] there's no problem with 3.3.2? [09:51] you mean 3.2? [09:52] try http://ftp.de.debian.org/debian/pool/main/a/asm3/asm3_3.3.1-1.dsc [09:52] i don't have any issues with latest ubuntu (3.2) but they added an mh_clean in debian/rules in the 3.3.1 [09:53] what release are you trying to build on? [09:53] i dget the debian dsc and try to build on oneiric [09:53] it works... only if i install maven-repo-helper before using pbuilder [09:54] what /exact/ commands are you using? [09:54] 3.3.1 works fine too. [09:54] (I am using sbuild as opposed to pbuilder though) [09:55] using pdebuild with http://pastebin.com/wqz3buBy [09:58] never used it, but I suspect that the problem is you need maven-repo-helper and/or ant installed to build the source package. This happens /outside/ of pbuilder. [09:58] would it work on ppa? [09:58] that's why I suggested just doing pbuilder build ...dsc [09:59] should do [10:03] pbuilder build ...dsc works for me too [10:03] great [10:03] so what happens is that when you build a source package, the clean target from debian/rules is invoked, and sometimes this needs extra packages [10:04] i see === kentb-afk is now known as kentb === bulldog98_ is now known as bulldog98 === almaisan-away is now known as al-maisan === al-maisan is now known as almaisan-away [16:49] hm how does one do a bzr-merge-upstream when the patches are applied in the branch? [16:50] as the new upstream may conflict with the patches [16:51] why does bzr do this, it only causes me problems and I don't see any advantage .. [16:59] do what? applied patches? [16:59] yes [17:00] because then bzr can manage them. Yes it's currentyl very not-ideal [17:00] eventually they should be managed as a patch queue === almaisan-away is now known as al-maisan [17:12] gna, what should be a 2 minute operation turns into 30 minutes, screw bzr I'll make a debdiff === al-maisan is now known as almaisan-away [17:13] :) [17:13] because of .pc conflicts? [17:15] jtaylor: quilt pop -a; bzr merge --force; while quilt push; do quilt refresh; done [17:15] yes but verifing the patch is unecessarilty painful [17:15] WFM [17:23] btw force does nto seem to work with merge-upstream [17:49] jtaylor: no - that's a bug I think [19:26] jtaylor: heh, Debian just uploaded a fixed version of tahoe-lafs [19:26] yes talk about bad timing ._. [19:26] I'll take care of syncing it next cycle [19:32] is there a diff left? [19:32] nothing worth syncing [19:33] and I don't want to check the package again for screwups ._. [19:36] well I might also bug someone from sr to fix it in natty ._. [19:36] it never ends with that package [19:36] jtaylor: I was wondering if you fixed that with the security upload [19:37] I was thinking about doing it [19:37] hmm I should ahve added the missing mock dep ... [19:38] right :) [19:38] that only fixes 50% but saves another upload [19:38] jtaylor: what do you mean 50%? [19:38] there are two issues [19:38] mock is the easy one [19:39] well the one where upstream does not disagree with me [19:40] for solving the pycryptopp issue there are two possibilities [19:40] the screw some users which have some broken corner case non package managed installation [19:40] or update a package in an sru [21:13] * ScottK votes SRU [21:14] Security updates should be a minimal change, so two uploads is appropriate in any case. === almaisan-away is now known as al-maisan === al-maisan is now known as almaisan-away [22:37] * davidsarah catches up [22:55] so that's all fixed in http://packages.debian.org/sid/tahoe-lafs 1.8.3-2, right? [22:58] * davidsarah checks the Ubuntu packages === yofel_ is now known as yofel [23:06] hmm. I have tahoe-lafs 1.7.1 installed on Maverick. I run Update Manager and install everything it prompts about. shouldn't that apply the security fix? [23:06] it doesn't [23:06] I'm not prompted to install any new tahoe-lafs package [23:09] davidsarah: pastebin the output of 'apt-cache policy tahoe-lafs' - launchpad says it was published 3 hours ago [23:10] * ajmitch can see the package on security.ubuntu.com, too [23:21] http://codepad.org/PnsARp1P [23:21] * davidsarah looks on security.ubuntu.com [23:21] that is the version that was uploaded a few hours ago by jtaylor [23:21] really? [23:21] https://launchpad.net/ubuntu/+source/tahoe-lafs/1.7.1-0ubuntu1.1 [23:22] I grepped for the fix [23:22] * davidsarah checks they didn't make a mistake doing that [23:22] diff between -0ubuntu1 & -0ubuntu1.1 is on the page if you want to check that [23:24] oh, it is updated [23:24] why didn't my original grep find that? [23:25] * ajmitch shrugs [23:25] as long as it is really fixed now? :) [23:26] still checking [23:27] yes, it is fixed [23:28] great [23:28] but why didn't I see a summary of the update in Update Manager? [23:28] I don't know what parts are involved in update manager listing changelogs [23:30] it didn't mention that tahoe-lafs was being updated at all [23:30] (but it was, because I ran the same 'grep -Rn fills-holes /usr/share/pyshared/allmydata' command before and after running Update Manager, so it had not already been updated) [23:30] don't know, sorry, I don't tend to use update-manager [23:31] I used it because I figured that is what a non-developer/techie user might use, and I wanted to see what they would see [23:31] it could depend on when it had checked for updates, if you'd checked automatically, and if security updates are set to install automatically [23:32] I didn't explicitly turn on any automatic updating, and I tend to turn off options like that if I see them [23:33] changelogs are fetched from a separate server. there at least used to be a delay between archive publishing and changelog generation. i don't know if there still is [23:34] broder: the package should still appear in u-m even if there's no changelog [23:34] yes, that's what I thought [23:34] summary of the update> i thought you were referring to the description section [23:34] "Install security updates without confirmation" is an option in software sources, I don't know if it's the default [23:35] I was referring to the top pane [23:35] * davidsarah looks whether that is set [23:35] you...did click the "check" button, right? [23:35] yes [23:35] might just have not gotten to your mirror yet [23:36] right, but it is installed now, somehow :) [23:36] yes, but I'm trying to understand why the package... what ajmitch said [23:36] ...oh, i see. that's what i get for not paying attention [23:37] so I was trying to figure out if it was silently installed, but I don't know update-manager's behaviour [23:37] you could go look in /var/log/apt/ [23:39] oh, /var/log/apt/history.log does include: tahoe-lafs:amd64 (1.7.1-0ubuntu1, 1.7.1-0ubuntu1.1) [23:39] maybe I just missed it in the list of updates