[00:53] In the following webpage there is a section entitled "bzr-builddeb" that tells me to install bzr-builddeb. I do not have that installed, but I do have bzr-debbuild. I just wanted ot make sure that was correct, and if so, see what to do about getting the wiki updated (I would be glad to do it, but need to know how): https://wiki.ubuntu.com/SimpleSbuild#bzr-debuild [00:53] darkxst sarnold ^ [00:54] (to see if bzr-debbuild is correct, not bzr-builddeb) [00:56] also, it does not say how to install schroots for different versions - I am assuming for vivid, for instance, it would be "mk-sbuild vivid"? [00:57] or, is are there other steps I need to do to make an schroot for other versions before that? [01:06] aeoril, bzr-builddeb is correct [01:06] easiest make to make schroots is with sbuild-launchpad-chroot [01:06] darkxst ok, i thought so - should I e-mail the owner of that page to inform them it is mis-spelled? [01:07] aeoril, no, that is something else [01:08] darkxst what is something else? [01:08] oh, youi said builddeb is correct, not debbuild ... sorry [01:09] aeoril, they are 2 different tools [01:09] by the looks of it [01:11] darkxst yes, I understand now - I guess I need to install bzr-builddeb? This is a fresh machine and I did not install any build or developer tools other than what were in the directions on that page. Apparently, I was supposed to know to install bzr-builddeb - are there other things I am missing on a fresh machine for development? is there a wiki to follow to get set up for building [01:11] properly? I am familiar with the BeginnerTeam wiki, but know that that team has not been active for quite some time [01:12] aeoril, there is this http://packaging.ubuntu.com/html/getting-set-up.html [01:13] but there is no definitive guide since they are multiple ways to do every step of packaging [01:13] darkxst ok, I will follow that - I guess once I get everything set up correctly, I need to remake my schroot to reflect the changes in my machine? [01:14] your schroot is isolated from your machine [01:15] darkxst where did it pull its packages from? I guess not my machine, but the "magic appropriate place on the web???" [01:15] if you used sbuild-launchpad-chroot, it pulls an image off launchpad [01:16] if you use mk-sbuild, I suspect it would use debbootstrap to make a minimal environment [01:16] anyway gtg [01:16] hmmmm ... on that wiki page, it just had me do 'mk-sbuild trusty' [01:16] ok, thanks [01:17] sbuild-launchpad-chroot, gives you an schroot almost identical to the archive builders [01:40] darkxst yes, I have used it now and will experiement - thanks! I think this will really help! === Guest33721 is now known as benonsoftware [05:33] Good morning [05:33] mitya57: it's already gone from private jenkins, the public one needs an RT to clean up [06:13] I have set up schroots with sbuild-launchpad-chroot on a trusty 14.04.1 LTS machine. After successfully running sbuild-launchpad-chroot create and and creating a local repo of ubiquity with "bzr branch lp:ubiquity" I ran the sbuild command to build for vivid from the ubiquity directory. Unfortunately, it did not complete successfully and failed as follows: http://paste.ubuntu.com/10367718 [06:13] / The isntructions had a final parameter "" I left off - not sure what to put there. Any ideas why this is failing? [06:13] darkxst sarnold ^ [06:20] Do I need to do the following first? "dch -i; update-maintainer; debuild -S" [06:22] aeoril, yes build the sauce package, then pass that .dsc file to sbuild [06:22] source even [06:26] darkxst debuild -S failed with the same error as sbuild as well as the following: "dpkg-buildpackage -rfakeroot -d -us -uc -S failed" [06:27] aeoril, do you have pyflakes installed? [06:27] I think ./test/run-pyflakes is failing [06:27] oh, maybe not - just a minute === benonsoftware is now known as Guest48867 === Guest48867 is now known as benonsoftware [06:39] darkxst ok, got past pyflakes and pep8 install and run, but failing here. I don't understand why it is doing this since that directory does not appear to exist: http://paste.ubuntu.com/10367924/ [06:45] aeoril, are you building from a bzr branch? use bzr builddeb -S if so [06:45] darkxst yes, I am building from a bzr branch thanks [06:51] darkxst well, it went a lot better, much more stuff happened, but I got an error that included the text "you may need to install the Debian::Debhelper::Sequence::d_i module" - I can give you the entire error if you need it [06:51] dh: unable to load addon d-i: Can't locate Debian/Debhelper/Sequence/d_i.pm in @INC (you may need to install the Debian::Debhelper::Sequence::d_i module) ... [06:51] aeoril, I am going to guess you are missing some debhelper tool [06:52] darkxst yes, I figured that, but I don't know which one ... [06:52] apt-get build-dep ubiquity (should hopefully get it) [06:52] or try apt-file find d_i.pm [06:53] darkxst hmmmm ... I thought the whole idea was to do this using sbuild in an schroot as close to the build environment on launchpad as possible? I am confused as to exactly what I am accomplishing right now ... [06:54] aeoril, some tools are required to build source packages (sbuild can't help there) [06:55] sbuild is for building the binaries [06:56] darkxst wouldn't it be better just to find the tools necessary to build the source packages, rather than getting *all* the build dependencies? Wouldn't getting build dependencies gunck up my regular machine? [06:56] (in other words, do your second suggestion?) [07:02] aeoril, do whichever way you prefer [07:04] darkxst I think everything built, but now I have a gpg error. I thought I set up GPG right - is there a way to test it?: http://paste.ubuntu.com/10368218/ [07:05] aeoril, you need to set a default key (or add your own changelog entry) [07:05] darkxst oh, in .bashrc? [07:06] aeoril, in ~/.gnupg/gpg.conf I think [07:07] good morning [07:07] darkxst this says to do it in .bashrc? https://help.ubuntu.com/community/GnuPrivacyGuardHowto (search for "default key") [07:07] dholbach good morning :) [07:07] hi aeoril [07:08] aeoril, I imagine either would work [08:05] so i've got a bluetooth mouse here that refuses to reconnect after going into powersave mode [08:05] how do i debug this? === kickinz1|afk is now known as kickinz1 === ttx2 is now known as ttx [08:18] hmm it keeps failing with bluetoothd[25955]: plugins/hciops.c:link_key_request() Matching key not found [08:26] darkxst thanks, everything is working now. Setting the default gpg id in both .bashrc and .gpg.conf did not work - I had to add a new changelog entry for my new version [09:03] Hi, I've been in Laos and I told people there that Ubuntu is very great because it tries to be available for everyone (disabled people, all languages, and so on). Then, I found out that Lao language is not correctly supported by the default-fonts in Ubuntu (for example Ubuntu or Sans). Shouldn't we/Ubuntu implement this to keep our/its promise? [09:04] Or ... Who shold I talk to to tackle this problem? === kickinz1 is now known as kickinz1|afk === kickinz1|afk is now known as kickinz1 === kickinz1 is now known as kickinz1|afk === kickinz1|afk is now known as kickinz1 === vrruiz_ is now known as rvr === _salem is now known as salem_ === chrisccoulson_ is now known as chrisccoulson [13:39] cjwatson: hi! I was wondering if you might be able to help me figure out / tell me what commands I might use on my phone to install a click package into the preinstalled db at /usr/share/click/preinstalled/ [13:44] pete-woods: Odd to do that on the phone rather than in a chroot elsewhere while preparing an image! [13:44] pete-woods: Should be something like "sudo click install --all-users --root=/usr/share/click/preinstalled foo.click" [13:45] cjwatson: yeah, just testing something out. thanks for the help! [13:46] np [14:10] Q: Am I wasting my time trying to fix FTBS 'à la' PlusOneMaintenance ? [14:18] caribou: why do you ask? [14:18] It's generally useful, assuming you send fixes upstream. [14:18] Though I did find it generally unending and thankless, and didn't feel it made that much of an impact when I did it - particularly for packages in universe. [14:18] rbasak: because there doesn't seem to be any formal activity with the PlusOne Maintenance team anymore [14:18] It did make my application for MOTU much more useful though :) [14:19] s/useful/credible/ perhaps [14:19] rbasak: it is also partly my intent : get exposure for all sort of packaging problems to gain experience [14:20] rbasak: for instance, the deja-dup FTBS is peculiar as it does build as PPA and locally and not in the builder [14:20] That certainly is useful for experience :) [14:20] I think that doing universe FTBFSes helped me more than I thought at the time. [14:23] rbasak: so what should I do with this one ? open a bug & attach the debdiff ? branch & MP ? [14:24] rbasak: since I cannot upload myself [14:24] maybe mterry will want to sponsor it === Guest81339 is now known as rcj === rcj is now known as Guest41495 [14:25] caribou, rbasak: a deja-dup ftbfs? fun. Point me at bug and I'll try to get to it [14:25] mterry: oh, I fixed it [14:25] caribou, but you need a sponsor, right? [14:25] mterry: ok, I'll create the bug & attach the debdiff. Or maybe you prefer a branch [14:26] mterry: yep [14:26] caribou, either is fine, but don't bother with a branch if you already have a debdiff [14:27] mterry: hmm, it'll have to be a debdiff, there's an in-flight source modification into -proposed [14:27] matter of fact, this is what introduced the FTBS [14:28] Noskcaj: you have the deja-dup uploaded into -proposed [14:30] mterry: ok, I'll create a bug & attach the debdiff against the source in -proposed [14:33] caribou: I'll leave you with mterry. Just FYI, when I was doing +1 and had a bunch of FTBFS fixes, rather than creating a branch I just stuck all the updated source packages online somewhere. Sponsors were happy to use that, rather than doing the busywork of creating a ton of bugs. [14:34] rbasak: good to kwow, thanks [14:34] caribou: I guess the MP/bug debdiff workflow works better for actually putting things in the sponsorship queue though [14:34] But if you have a sponsor lined up, they usually don't mind how they get the work. [14:34] rbasak: it always depend if the sponsor is already identified (as it is in the current case) [14:34] Right [14:35] rbasak: i was just surprized by the difference of build behavior b/w local & the archive builder [14:35] Ah. [14:35] That surprised me too when I started. [14:36] rbasak: I did have a previous situation where it was a parallelism issue involved [14:36] Apparently it's quite difficult to replicate the archive builder exactly, so most devs don't bother and it works in the majority of cases. [14:36] In the odd cases where it doesn't, comparing the build logs can often identify the issue (I use meld) [14:37] And when that still doesn't work, I get help here from the devs who are closer to the builders (generally archive admins, etc) [14:37] darkxst I found two equivalent .deb files created by the ubiquity build (same name but different version as oem-config_2.21.13_all.deb and oem-config-gtk_2.21.13_all.deb) in the Ubuntu .iso for the livecd in the following directory: pool/main/u/ubiquity Do I just replace those two package files with the ones I made with my compiled changes in the .iso then use that .iso to create a [14:37] bootable live cd to test the install? [14:37] caribou, yeah I can take debdiff however you like, if you don't want to bother with bug [14:37] or whoever knows [14:38] mterry: no worry, bug is done already, just doing a bit of description there [14:40] mterry: it's all there LP: #1424652 [14:41] Launchpad bug 1424652 in deja-dup (Ubuntu) "deja-dup 32.0-0ubuntu3 fails to build following change in Build-dep" [Low,Confirmed] https://launchpad.net/bugs/1424652 [14:48] pitti: i love the typo you made re:mtab vs mountinfo "Yes, it arses it just fine" [14:48] vs parses =) [14:48] xnox: that was a case of SCNR indeed :) [14:50] pitti: so given the new enough util-linux with that toggle enabled, can the /etc/mtab symlink be dropped? [14:50] pitti: or is there anything else that reads it? [14:50] xnox: no, there's still lots of other things which look at that [14:51] pitti: name one [14:51] or, let's rather say, we don't know [14:51] udev, systemd, util-linux no longer do..... [14:51] we could drop it and see what happens, but that's not a good idea post-FF [14:51] indeed, any well-behaved program/script ought to parse /proc/self/mounts or the output of "mount" [14:53] http://codesearch.debian.net/results/%2Fetc%2Fmtab/page_0 has 186 pages, ugh [14:55] xnox: the last page has a few [14:55] http://codesearch.debian.net/results/%2Fetc%2Fmtab/page_185 [14:56] xnox: so cleaning those up sounds like a nice vivid+1 project [15:06] where is KMTAB defined?! [15:15] nevermind non-linux codepath.... [15:16] infinity: FYI, a fix for percona-galera-3 !Intel FTBFS is in progress. [15:17] Once done, I'll poke you to review all the binNEWs. [15:17] And to process bug 1417328 [15:17] bug 1417328 in percona-xtradb-cluster-5.5 (Ubuntu) "Please remove 5.5 versioned MySQL and variants from vivid" [Undecided,New] https://launchpad.net/bugs/1417328 [15:21] pitti: glibc seems to be using /etc/mtab =( [15:26] second CfV sent to u-d-a@ if someone could moderate [15:28] Laney: done [15:28] cheers! [15:31] anyone else reported menus living in their windows title bar rather than the top unity panel? [15:32] that happened on purpose [15:33] oh. really [15:33] smoser: http://www.webupd8.org/2015/02/locally-integrated-menus-lim-set-as.html [15:33] http://www.webupd8.org/2015/02/locally-integrated-menus-lim-set-as.html [15:33] DAMN! [15:33] \o/ [15:33] read my link, it's better [15:33] coming from UK, so closer to you, obviously :) [15:35] Laney, thanks. [15:35] yw [16:14] caribou, uploaded, only change was adding bug number to changelog [16:15] mterry: oh right, I built the debdiff before opening the bug [16:15] mterry: thanks ! [16:15] caribou, I'm curious why it fails in archive but not locally [16:17] mterry: couldn't figure it out either since appdata-validate still works but throw the DEPRECATED message [16:17] mterry: I'll be attentive to the next build [16:23] mterry: ok, I know why it worked locally : appdata-validate was not the reason it was failing [16:23] caribou, yeah, we need to pass --nonet [16:23] caribou, just saw the failure myself [16:24] caribou, I can do that myself real quick [16:24] mterry: ok, but I'd like to know what you changed. It'll outline why it didn't work on the builder to me [16:27] caribou, http://bazaar.launchpad.net/~ubuntu-desktop/deja-dup/ubuntu/revision/42 [16:28] caribou, the validate tool was hitting network and couldn't, so it failed validation [16:28] mterry: ah, ok I get it [16:29] caribou, though I would have expected the PPA builds to fail then? [16:29] caribou, and I think that was a behavior difference in the tool too -- it didn't used to try to hit the network to test the screenshots in the file. But they added a feature that does that I guess [16:29] mterry: me too, that's why I built both .3 & .4 in my PPA prior to going for sponsor [16:29] caribou, well... maybe .5 will fail for some other reason ;) [16:37] caribou, .5 worked on i386... so that's good [16:38] mterry: good ! [16:41] mterry: builds fine on amd64 as well [16:42] caribou, OK. Chalk that up to mysterious network access in the PPA...? [16:43] caribou, it was hitting launchpad.net [16:43] caribou, maybe PPAs allow doing that because they sometimes pull from other PPAs [16:43] mterry: yes, most probably [16:44] Shouldn't be any difference. [16:44] mterry: & the previous behavior of appdata-validate changed as well [16:44] launchpad.net itself, or a subdomain? [16:45] cjwatson, https://launchpad.net [16:46] Other PPAs aren't on launchpad.net, neither the same physical host nor the same DNS name [16:46] So that's not the reason [16:47] And in any case the distinction is not likely to be PPA vs. Ubuntu, but non-virtual vs. virtual [16:47] And non-virtual builders build some specialised PPAs too [16:48] "url not found" hides some underlying error, so it's hard to say. [16:52] PSA: d-jenkins is unhappy again, so if you see a bunch of autopkgtest failurs which look weird, we're on it [16:52] Laney, Mirv, etc. ^ [16:53] 91.189.89.30 - "10.122.37.20, 127.0.0.1" "launchpad.net:443" [23/Feb/2015:16:19:21 +0000] "GET /deja-dup/32/32.0/+download/screenshot-2.png HTTP/1.0" 302 526 9 0.105017185211 594 102 "Anonymous" "LibraryFileAlias:+index" "" "libappstream-glib" [16:53] I wonder why that's getting a 302 [16:53] Oh, of course, 302 to the librarian [16:55] Then AFAICS the librarian never sees the redirected request. [16:56] But that's odd since buildds explicitly have librarian access. [16:57] infinity, cjwatson, wgrant: FYI, fisher03's apache hasn't responded in 2 days for ddeb retrieval; if that keeps being broken, we'll lose ddebs in 5 days [16:58] pitti: Needs infinity; wgrant and I can't help, but thanks for the heads-up [16:59] cjwatson: yeah, I figured; it was more an FYI for you [17:02] wgrant: Do you know if we have access to the specialised buildd DNS code, and if it's the same for non-virt buildds and scalingstack? Surely it should be, but I can't see why the deja-dup thing above would have broken on non-virt builders but apparently not on scalingstack. === kickinz1 is now known as kickinz1|afk [17:11] cjwatson: I'd be willing to bet scalingstack's DNS setup isn't the same, actually (and that should be fixed). [17:11] pitti: Looking. [17:12] [602822.116012] sd 0:0:0:0: rejecting I/O to offline device [17:12] ^-- qemu emulating real hardware a bit TOO well. [17:13] just online it :P [17:14] infinity: I'm a little surprised that anyone would have bothered to do it differently. [17:15] There's a nagios test defined in canonical-is-openstack-deploy to the effect that hp.com returns NXDOMAIN. [17:15] So there's definitely something similar in spirit in place. [17:15] Hrm. Curious. [17:16] Well, it's going to be a different view (or a different server entirely) because it's not the same subnet, or possibly even network, but yeah, you'd have expected it to be either a straight cargo-cult or wide open. [17:17] infinity, Do you have a sec? [17:17] It's the same server, according to puppet. [17:17] (salmonberry) [17:17] Oh, no, that's the puppeteer [17:18] flexiondotorg: Perhaps. [17:18] I think ... don't know this stuff very well [17:18] infinity, I'd just like to ask when it might be possible to try build Ubuntu MATE images? [17:19] flexiondotorg: Oh. This week. Maybe even today. Let me wake up a bit and confer with cyphermox on what he believes the status of things are. [17:19] flexiondotorg: We can kick off another attempt any time. [17:20] flexiondotorg: Or, if Colin has context, what he said. :P [17:20] I mean, if the necessary packages are in place. [17:20] cjwatson, Another attempt? [17:20] Has it been tried before? [17:20] flexiondotorg: Yes. [17:20] cjwatson, I didn't know that. [17:20] It failed, fairly obscurely but I think you were known to be missing some packages at the time. [17:21] Are there some error logs I can look at? If there is stuff to be fixed I'll get stuck in. [17:21] http://people.canonical.com/~ubuntu-archive/cd-build-logs/ubuntu-mate/vivid/daily-live-20150220.log [17:21] Thanks. [17:21] But https://launchpadlibrarian.net/198284153/buildlog_ubuntu_vivid_amd64_ubuntu-mate_BUILDING.txt.gz doesn't really have much to go on. [17:22] That's not the most informative failure... [17:23] cjwatson, infinity Reading both logs. [17:23] Yeah, the first log I can't do anything with. [17:23] The first just links (indirectly) to the second. [17:23] I meant the livefs log wasn't informative. :P [17:23] right, that failure tells me nothing as to what might have gone wrong [17:23] The cdimage log is totally informative, just brief. [17:23] infinity: It's a bit opaque ... [17:24] cjwatson: "a bit". [17:24] next step for me was to ask you cjwatson or infinity for help [17:24] cjwatson: I think we're log past due cranking up the verbosity on live-build. [17:24] Wait. Does the ubuntu-mate-core task actually exist? [17:24] cjwatson: As in, for the love of god, tell me what command you were trying to run. [17:25] Nobody ever changed ubuntu-archive-publishing to generate it ... [17:25] ah, that's something we didn't know about [17:25] * cjwatson fixes that [17:26] Thanks guys. [17:26] cyphermox, Hi 😃 [17:26] Should be better in a couple of publisher cycles. [17:27] cjwatson: was that something we could change in a code branch or did that require special access? [17:27] cyphermox, mate-tweak and mate-menu are still in the Upload Queue. [17:27] cyphermox: lp:ubuntu-archive-publishing [17:27] flexiondotorg: that won't help [17:27] cjwatson: thanks [17:27] Yeah, I didn't have time to look at them on Friday, I'll poke today. [17:27] Oh, wait. I looked a bit. [17:27] flexiondotorg: I had a copyright complaint about mate-menu. :P [17:27] https://code.launchpad.net/~acidburn0/+recipe/ubuntu-archive-publishing-daily WTF [17:27] Copyright: *No copyright* [17:27] License: GPL-2+ [17:28] ^-- That is literally impossible. [17:28] You can't have a copyright license and assert no copyright. And "no copyright" is almost certainly a lie anyway. [17:28] Given that all creative works (in most jurisdictions) have copyright attached to them just by being created. [17:29] pitti: The fishers should be marginally less grumpy. [17:29] infinity, OK. That copyright stuff is what was recommended by the DD I am working with. [17:30] infinity: that may have rather meant that the file itself didn't have a copyright notice but there is COPYING there [17:30] infinity: cheers! [17:30] I thought there was a "assuming blah blah" line just below or in Comment: or something [17:30] cyphermox: I know exactly what it means, but it also won't do. [17:31] cyphermox: "I don't know who owns the copyright and who supposedly gave me the license to distribute" isn't a helpful license. [17:31] infinity: fortunately flexiondotorg here is supposed to be the upstream and able to explain [17:31] cyphermox: They're asserting the license is probably the one in COPYING (a fair assumption), but no copyright. [17:31] flexiondotorg: So, uhm... Who owns the copyright on those files? :P [17:31] oh [17:32] the copyright *owner* [17:32] Yes. [17:32] I see now [17:32] infinity, I forked another project. They did not provide copyright headers in all the sources. [17:32] A copyright license without provenance isn't a license. [17:32] infinity: of course not [17:32] flexiondotorg: Headers in all the files aren't necessary, but a statement of copyright ownership on the copyrightable bits is important. [17:33] flexiondotorg: Certainly, some of this junk is just glue and not subject to copyright anyway, but I see logos and such under this non-copyright, which won't do at all. [17:33] infinity, OK. One sec then. [17:33] Copyright: 2007-2014, Clement Lefebvre [17:33] 2015, Martin Wimpress [17:34] That ^^^^^ is the suitable copyright. Given that it include the original author and myself (new upstream maintainer). [17:34] flexiondotorg: And that applies to, say, data/icons/mate-logo.svg ? [17:34] flexiondotorg: Clement is the person who designed the logos (for example) originally in the project you forked? [17:34] infinity, Do you want me to resubmit or can/will you make the required change? [17:35] flexiondotorg: If so, get someone to fix the copyright file to be sane in that regard (and perhaps also note correct copyrights in the upstream project too) [17:35] cyphermox, No, that particualr file was created by myself and Sam Hewitt. [17:35] infinity, Updating the upstream project won't happen. I've tried that already. [17:35] Not upstream [17:35] flexiondotorg: I mean your upstream project. ;) [17:36] flexiondotorg: Not the one you derived from. [17:36] infinity, Understood (eventually) 😃 [17:36] OK, I will update the copyright now... [17:36] flexiondotorg: As in, you just said that data/icons/mate-logo.svg comes from you, not from your upstream, so "I don't know who owns the copyright" isn't even a factual answer. :) [17:36] flexiondotorg: So, yeah. Please sort. Make sure things are sanely attributed. [17:37] infinity, Like I say. That copyright file was prepared my the DD I'm working with. [17:37] flexiondotorg: I'll reject that one from the queue now, and expect a shinier version. And I'll go review the other one. [17:37] flexiondotorg: Check. Pass on my ire to him about the senselessness of "no copyright" then. :) [17:40] Riddell: can you have someone look at these PEP8 failures? http://pastebin.ubuntu.com/10375047/ [17:41] bdmurray: hmm? I was able to build and upload it from my laptop was I not? [17:42] Riddell: did you use bzr-buildpackage or debuild? [17:42] infinity, Would you accept this debian/copyright? [17:42] http://paste.ubuntu.com/10375062/ [17:45] flexiondotorg: If it's factually correct. [17:46] flexiondotorg: I see no mention of Sam Hewitt in there, who you highlighted above, for instance. [17:46] I've just added that 😃 [17:46] Files: data/icons/mate-logo.svg [17:46] Copyright: 2014, Sam Hewitt [17:46] License: GPL-3+ [17:46] bdmurray: I think I used bzr-buildpackage, let me try again [17:47] flexiondotorg: 3+, not 2+ like everything else? Whee. [17:47] infinity, Sam releases everything as GPL 3+. [17:47] I've include the GPL3+ license bioler plate too. [17:48] flexiondotorg: Kay. So, that also means the PNG produced from it is 3+ [17:48] flexiondotorg: But yes, that looks better. :) [17:48] infinity, Yes, the original artwork is GPL3+ licensed. [17:48] infinity, http://paste.ubuntu.com/10375144/ [17:49] flexiondotorg: mate-tweak has the same problematic "no copyright" clause. [17:49] infinity, On it.. [17:50] flexiondotorg: Oh, also, the copyright on COPYING isn't you (and, really, you shouldn't bother including it in the list of files at all). [17:50] Unless you'd like to pretend you wrote the GPL. ;) [17:55] bdmurray: ok I can recreate it, do you know what all the problems are with all the "'from PyQt5.QtCore import *' used; unable to detect undefined names" messages? does it just not like wildcard imports? [17:55] flexiondotorg: Oh, also, for future maintainability, it's MUCH simpler to collapse all those Clement/Martin copyrights down to "Files: *" and then just highlight the ones that differ. [17:55] flexiondotorg: Since having to adust the copyright file for every new added bit is painful and error-prone. [17:56] Riddell: yeah, it'd prefer from PyQt4 import QtCore instead of from PyQt5.QtCore import * [17:56] s/Qt4/Qt5/ [18:02] https://www.python.org/dev/peps/pep-0008/ has reasoning on why wildcard imports should be avoided [18:13] infinity, Shall I update the original packaging request bugs and direct you to them? [18:24] bdmurray: commited some fixes, let me know if that works or breaks [18:28] Riddell: looks good, thanks! [18:33] cyphermox, I've updated mate-menu and mate-tweak based on the feedback from infinity. [18:33] cyphermox, Can you upload them again? [18:49] flexiondotorg: sure [18:50] cyphermox, Thanks 😃 === guest11233454645 is now known as Streamstormer === guest123456 is now known as Streamstormer [19:46] aeoril, easiest way to test ubiquity is probably to boot a live session, update ubiquity with your packages, then run the installer [19:46] darkxst what do you mean by "live session"? [19:47] just boot from livecd? [19:47] yes [19:47] "try ubuntu" [19:50] I do not have a thumb drive to boot with - can I somehow do it with just the .iso from cdimage.ubuntu.com? [19:50] darkxst ^ [19:50] (I can always buy a small thumb drive for a few dollars though) [19:51] aeoril, use a VM [19:52] darkxst how do I get my ubizuity .debs into the live session though? [19:53] I could always just add them directly onto the livecd with iso master ... [19:53] ubiquity* [19:53] aeoril, scp, or network share [19:53] or ppa [19:53] there is no need to remaster ISO to test [19:55] darkxst actually, maybe adding to the iso would be easiest - it is just a copy/paste, save the iso then boot from it, I think [19:55] the iso from cdimage.ubuntu.com [19:56] iso master has a simple gui [19:57] aeoril, you can't just copy package files into the ISO [19:58] pitti: FYI, I've had someone report that the ureadahead upload broke the upgrade on their box as ureadahead would exit immediately on their box, leading to "start ureadahead" to fail [19:58] not sure if that's a common/normal case for ureadahead (I thought it might be for ssds and other case where ureadahead can't work) [19:58] darkxst I meant just to have them available after bootup, rather than scp or network share or ppa - don't I have access to the mounted .iso image from the live session? I could use dpkg to install them into my live session from the .iso image (I assume it will be mounted as a cd?) [19:59] darkxst but is scp is available, that would be very easy i guess - I guess the scp client is available default in the live session? [20:00] it should be but if not, you can just install it [20:01] darkxst I was just thinking I might have to set up ssh keys and such [20:04] ssh can use passwords too (unless you have disabled them) [20:04] ok, cool - that seems simplest then. I will try this now. Thanks. [20:09] darkxst thanks for all the help - watching the sbuild go to work was amazing yesterday. But, I had to make my own entry in changelog though to get around the gpg failure [20:16] cjwatson, What is the correct way to instruct the image build process to "--no-install-recommends"? [20:19] darkxst there are a bunch of .deb files created from the build process - how do I know which one to use to install the package? [20:19] install all except -dbg and -dev [20:20] darkxst ok, that would be all of them then ... [20:21] flexiondotorg: You'd want to make your tasks not follow recommends in that case, though it's better if you can avoid having to do that. [20:21] flexiondotorg: What problem is recommends causing you? [20:23] darkxst as an fyi, I found several places where icon names needed to be fixed - there was also a warning icon that needed to be renamed, not just the "finished" icon [20:24] well, "info" [20:27] darkxst I vaguely remember maybe that using the .dsc file will install them all appropriately? Is that just my own fantasy? [20:28] darkxst or is that just for source? [20:29] darkxst nm, looked it up myself [20:44] darkxst if I make a ppa, will "sudo apt-get install ..." automatically install all the .deb files for me that sbuild created? [20:44] darkxst also, would that be a good way to attach the fix to the bug? [20:45] (just reference the ppa?) [20:45] aeoril, yes, but you don't use sbuild in that case [20:45] darkxst would I build it with a recipe then? [20:45] aeoril, no, create a branch on your launchpad page and link to the bug [20:46] you can then create a merge proposal [20:46] darkxst how would I build/test it though? [20:46] you just did didnt you? [20:46] no, not yet - apparently, I am slow ... [20:48] darkxst you said I would not use sbuild ... so I thought there was some other way to build/test if i made a ppa - you mentioned installing it from a ppa to test from a live session [20:48] earlier [20:48] bdmurray, hi, please test-install your uploads ;) [20:50] darkxst so, now I am confused [20:50] darkxst I was thinking I could do the whole process using a ppa or something [20:52] darkxst you mentioned I could install my compiled ubiquity from a ppa - what exactly did you mean by that please? [20:52] aeoril, you can use a ppa to test, but not to submit your changes for review [20:53] aeoril, dput you package to a ppa, boot a livecd, install the ppa and update [20:53] then run the installer [20:54] darkxst ok, cool - what exactly do you mean by "install the ppa and update" please? [20:54] apt-add-repository ppa: [20:55]