=== vorian is now known as v === v is now known as vorian === tumbleweed_ is now known as tumbleweed === broder__ is now known as broder === broder is now known as broder___ === broder___ is now known as broder [04:50] ScottK: so, if a reverse build dependency is broke, does that disqualify a backport? all the other tests were fine (bug 1074765) [04:50] bug 1074765 in Quantal Backports "Please backport ipython 0.13.1-1 (universe) from raring" [Undecided,New] https://launchpad.net/bugs/1074765 [04:50] I'm tempted to push it through [04:52] I guess I'd want to understand the "should be suggests" in there. [04:52] For the stuff that was already broken, no problem at all. [04:52] Fix the one that has an easy fix and then keep testing. [04:53] Given that there's some regression, I think the testing needs to be finished. [04:53] he actually ended up finishing it save for one [04:53] It doesn't disqualify it, it just means fix the other package. [04:53] OK. [04:53] Is the issue yade then? [04:54] If it's got an easy fix, backport the easy fix and make sure the backported ipython has breaks on the older yade. [04:54] hrm, what happened, I thought I just commented to him... [04:55] jtaylor: re ipython backport, it seems that nipype wasn't tested (reverse build dependency) [04:57] ScottK: so, would that fix be a changeful patch or a new backport? [04:57] (yade) [04:58] Whichever makes more sense. If a new backport does it, that's probably easier. [05:00] jtaylor: also, there's -2 now, do you want that instead? [05:00] Once I when I was trying to get a clamav backport done on Dapper, I had one rdepend that was totally fubar. I ended up making a franken package using the Dapper packaging, most of the package from Feisty (in order to avoid needing too new libs) and one file from (IIRC) Hardy that had the interface for clamav in it to work with the newest clamav. [05:00] Amazingly it all worked fine. [05:01] ok, I just wanted to make sure I stay within the rules of backports [05:02] * micahg has a nice backlog to get through === LoganCloud_ is now known as LoganCloud === LoganCloud is now known as Guest35627 === Guest35627 is now known as LoganCloud [07:59] good morning === almaisan-away is now known as al-maisan === yofel_ is now known as yofel === Ursinha is now known as Ursinha-afk === Ursinha-afk is now known as Ursinha === al-maisan is now known as almaisan-away [11:06] micahg: nobody is going to build yade, it needs like 6GB or ram ._. [11:07] jtaylor: Why is this a blocker? Just call for a volunteer: there's lots of folk who have that somewhere, or someone could use a PPA) [11:07] I don't understand why rbuild depends are even a concern [11:07] if its broken revert to normal package, build, and upgrade again [11:09] Because it prevents backport of the rbuilddep (either official or local), and may cause other odd unexpected behaviour with co-install. [11:09] With the test, only one person has to have the 6GB RAM, rather than everyone who wants to use the potentially broken rbuilddep [11:09] backports depending on backports doesn't work anyway [11:10] Why not? [11:10] because lp can't handle it [11:10] also a backport of yade is fixed anyway [11:11] How can't lp handle it? [11:11] don't know [11:11] Not, specifically what code path is broken, but in what manner is this non-handling? [11:11] let me look for the bug [11:12] bug 888665 [11:12] bug 888665 in Launchpad itself "Backports can't build-depend on other backports" [High,Triaged] https://launchpad.net/bugs/888665 [11:14] Ah, excellent. Hrm. That makes it annoying. [11:15] compared to micahg's favourate packages, 6GB of RAM is nothing [11:16] Anyway, so it becomes annoying, but there may still be folk who want to build, so we should do our best to ensue that the rbuilddeps can build with *either* version safely. [11:16] (this is why backporting libraries almost never happens: there are rbuilddeps that tend to make everything painful) [11:17] so basically anything with nontricial rdepends is not backportable? [11:17] because of non-automatic even backporting yade will not help [11:17] I'm not a backporter, but that's how I've generally understood the backport policy. [11:17] as people will not by default build the backported yade but the offficial one [11:18] ok one could add versioned depends into the backport === Tonio_aw is now known as Tonio_ === Quintasan_ is now known as Quintasan [13:40] tumbleweed, Laney: do you think we could get another "working with Debian/Upstream for fun and profit" session at UDW together? :) [13:40] https://wiki.ubuntu.com/UbuntuDeveloperWeek/Timetable still has a bit of free space === Tonio_ is now known as Tonio_aw === Tonio_aw is now known as Tonio_ === lfaraone_ is now known as lfaraone === dupondje_ is now known as dupondje [18:06] Hello guys [18:06] :D [18:06] I've a question :D [18:07] How can I get my application in ubuntu software center? === Tonio_ is now known as Tonio_aw === almaisan-away is now known as al-maisan === al-maisan is now known as almaisan-away === Logan_ is now known as Guest17481 === Logan__ is now known as Logan_ [20:27] jtaylor: do you need someone to test build yade (I have 6GB RAM) [20:30] infinity: poke on the backports depending on backports bug (feel free to decline over vacation) [20:35] micahg: I've been looking at mangling sbuild to satisfy everyone's wishes on that bug, but even if I fix it locally, we certainly won't roll it out over the holidays. [20:35] infinity: ok, sure, no problem, you just asked to be reminded when you weren't busy :)