[19:30] cgoldberg: do you know if this bug is on the near roadmap? [19:30] https://bugs.launchpad.net/autopilot/+bug/1308762 [19:30] Ubuntu bug 1308762 in Autopilot "Capture a screenshot on errors and failures" [High,Triaged] [20:02] morning [20:42] morning all [21:54] barry: got a moment? [21:54] thomi: sure [21:55] barry: so, I'm sure you'll hear more about this from Max tomorrow, but we're attempting to convert the existing python-pyramid package to include a python3 binary package [21:55] barry: looking at it now, the version in debian is really old (1.4.5 vs 1.5.1) [21:55] is it worth, do you think, pushing up a newer package version to debian before we begin? [21:56] I should be able to do that with uscan, rebuild, and dput to mentors, right? [21:56] thomi: yes, i'd say so. then we'll have to merge any remaining ubuntu changes. [21:57] thomi: are you a member of dpmt? [21:57] barry: hell no :) [21:57] ah :) [21:57] will they be upset if I try and submit this via mentors.d.o? [21:58] thomi: here's what i'd suggest. branch the dpmt svn, and work out the changes. then file a bug on https://bugs.debian.org/cgi-bin/pkgreport.cgi?ordering=normal;archive=0;src=python-pyramid;dist=unstable;repeatmerged=0 and attach the diff [21:58] thomi: then ping me and i'll review, sponsor, upload as appropriate [21:59] barry: you're part of the dpmt? [21:59] thomi: but of course! and papt. and dd :) [22:00] barry: so it should be fairly fast to get the new version uploaded, right? I need to get the python 3 package done by Friday, is all :) [22:02] thomi: it won't be that fast if there's a new binary package. it has to be approved by the ftpmasters. of course, it won't be that fast either in ubuntu. all NEW packages must be approved (but you'll probably have better luck pinging ubuntu admins). i'd still prep the debian patch, but then you can -0ubuntu1 it in utopic for now. we'll have to manage a merge/sync later [22:02] barry: yeah... ok [22:03] now I just need to remember hwo to use svn :) [22:03] hah! [22:04] barry: how much longer till you EOD? [22:04] whenever the dinner bell rings :) but at least another 30-60m i think [22:04] ok [22:04] wow, this is a huge repo :-/ [22:27] barry: what is the '+dfsg' part of the package version? [22:28] barry: should I include it? If so, how do I tell 'uscan' to include that in the symlink it creates for me when it downloads the new upstream tarball into .. ? [22:28] thomi: it means that some part of upstream is deemed "not free software", so upstream tarball is probably repacked to eliminate those bits [22:28] dfsg == debian free software guidelines [22:29] it should ignore the +dfsg part in the symlink [22:31] barry: any ideas? http://pastebin.ubuntu.com/7620484/ [22:31] looks like dh is looking for a source package with the +dfsg in the filename [22:31] thomi: hmm, i would have thought uscan would dtrt [22:32] let me try to grab the vcs [22:36] thomi: take a look at d/README.source and you'll see the original reason for the +dfsg. might be good to see if that's still relevant [22:37] thomi: and fwiw, i always use svn-buildpackage to build the source package from svn checkout. it has lots of nice features, like --svn-download-orig :) [22:37] * thomi looks [22:37] barry: that's still the case. [22:38] k [22:38] barry: should I remove the docs and re-pack it under the new name>? [22:38] I'd be interested to know why CC by-nc-sa isn't considered good enough for the dfsg ? [22:39] barry: oh wait, the README.Sources says 'CC-BY-NC-3.0-US' [22:39] thomi: looks like d/rules has the command someone used to do that. might be better to add a get-orig-source target to d/rules to do that correctly [22:40] yeah, IANADL [22:40] but it's now 'by-nc-sa' [22:40] * barry looks [22:41] barry: https://wiki.debian.org/DFSGLicenses#Creative_Commons_Attribution_Share-Alike_.28CC-BY-SA.29_v3.0 [22:41] says it's compatible :) [22:41] nice! so all that crap can be dropped and upstream tarball used [22:41] sweet [22:42] barry: so the version should be '1.5.1-1' ? [22:43] thomi: yep [22:43] barry: what do you think of http://paste.ubuntu.com/7620524/ ? [22:44] thomi: probably add a note in the changelog about why the +dfsg was dropped [22:44] barry: new line, or same line? [22:45] thomi: don't sweat it too much, add it as a new line. i'll think about it when i review/sponsor [22:45] ok [22:50] :( [22:51] for some reason now it cannot find the BFG_HISTORY.txt file, so dh_installdocs fails, but it's there in the upstream tarball [22:51] hmmm [22:55] barry: is there any reason why 'dpkg-buildpackage' isn't the right thing to use? [22:57] thomi: not really. my personal dpmt workflow is all based on svn-buildpackage, just for the convenience [22:58] hmmm [22:58] pdebuild seems to work [22:58] thomi: yeah, lots of ways to do it. more perl-like in that regard than python :) [22:59] barry: OK, it seems to build a sensible package [22:59] I should file a bug against python-pyramid with the svn diff? [23:00] thomi: that would be great. paste me the bug #. i'm eod so probably won't get to it today, but should be able to tomorrow uos permitting. i can also do the -0ubuntu1 package for the short term [23:00] (but i think you can do that too) [23:00] barry: well, this is just bumping the upstream version number. What I need by friday is the python3-pyramid package [23:00] so I'll work on that now. [23:01] thomi: ah, right, yes