[06:27] good morning [07:22] davidcalle, I have another fix I need to get into trunk, it's for https://bugs.launchpad.net/developer-ubuntu-com/+bug/1565639 [07:22] Launchpad bug 1565639 in Ubuntu Developer Portal "Two tests fail with postgres" [Undecided,New] [07:22] davidcalle, and I suggest we add a "test all" command or something, which runs the tests both on postgres and sqlite [07:43] dholbach: +1, and in the meeting with the webteam, there was mention of a Jenkins instance (they would setup) in the future :) [07:43] oh wow [07:43] that'd be awesome [07:43] new MP, code reviews, jenkins runs tests, deploy... or something :-P [07:44] Yes please :) [07:50] dholbach: btw, the db dump is on the server, I'll load it in th db in a few minutes [07:50] cool [07:50] I'll let you know once I fixed the issue(s) [10:17] davidcalle, this might look oddly familiar to you: https://daniel.holba.ch/temp/:'-%28.png [10:17] I'm trying to rebuild this in a test setting - the tests in postgres are already fixed [10:18] dholbach: is that a messed up tree? :( [10:19] I vote we should go for local testing using nothing but postgres :) [10:19] or it's what I should do [10:20] dholbach: I think we should change make dev to use postgre, indeed [10:20] for now I created https://bugs.launchpad.net/developer-ubuntu-com/+bug/1565676 [10:20] Launchpad bug 1565676 in Ubuntu Developer Portal "Add a test_all command" [High,Triaged] [10:20] opinions and ideas welcome [10:28] ok, the find_parent function is on crack [10:28] time to add some more tests for it === dholbach_ is now known as dholbach [14:08] davidcalle, https://code.launchpad.net/~dholbach/developer-ubuntu-com/1565639/+merge/290855 [14:08] problem solved AFAICS [14:09] dpm, ^ :-) [14:09] \o/ [14:09] dholbach ! [14:09] You rock [14:10] dholbach: looking at it as soon as I'm done with the krita thing [14:11] davidcalle, no worries [14:11] let me give you a test-case [14:12] dpm, that now leaves some time to look at the redirects thing :) [14:13] dholbach, excellent. Is the call later on too late for you? If not, we can talk about it there, or alternatively tomorrow morning [14:14] dpm, it's fine [14:14] dholbach: by the way, I've been having troubles with loading the dbdump on staging, it's not just the "db content" we have been passed, but the whole db declaration, which conflicts a little with the staging db (user names, etc.). For the sake of rapidity, I'll quickly reproduce a basic snappy page layout instead, to test an import. [14:15] sure [14:16] I added instructions for testing [14:17] thanks [14:52] dholbach: I've a few issues when running the importer on the site: snapcraft reference doesn't generate: missing python3-requests, python3-docopt, and ultimately, it fails with http://paste.ubuntu.com/15613951/ [14:53] what what [14:53] ouch [14:53] yeah, I have those installed locally :-( [14:53] dholbach: no worries, I'm more concerned about the pastebin [14:54] feel free to leave out the reference generation for now [14:54] I'll file a bug [14:54] Alright. I'm not trying with your branch yet [14:54] just remove the post_checkout_command for now [14:55] Yep. Also, I can confirm I'm reproducing your trees issues :) WIll test with your branch in a short moment ;) [15:18] dholbach: your branch is working on staging \o/ [15:18] <3 <3 <3 [15:18] looks like it was worth ignoring everything else today ;-) [15:18] Indeed :D [15:26] davidcalle, mhall119: I'm currently looking at implementing the small redirects app - and I'm struggling somewhat using the RedirectsView - maybe I'm doing the wrong thing? https://code.launchpad.net/~dholbach/developer-ubuntu-com/404-redirect/+merge/290887 (it's a mid-flight branch) [15:35] oh... I just found https://docs.djangoproject.com/en/dev/ref/contrib/redirects/ [15:35] I'll try that out instead :) [15:35] dholbach: woah, strange behaviour on staging. Run the importer once, it fills up. A second time, it empties everything (no more pages at all in snappy/). A third time, fills up again. [15:36] let me see if I can reproduce this [15:36] I actually added a couple of tests for running imports more than once [15:38] dholbach: wait, I'm wondering it the cache is not acting up... [15:38] best to check locally then - I'll do it with a fresh db [15:41] davidcalle, I ran it three times in a row, every single time the imported files were all there [15:42] dholbach: maybe I need to wait more between two runs [15:42] dpm: dholbach: have we let the folks supplying the chinese content know about the impending IA changes? [15:42] mhall119, not AFAIK [15:43] mhall119: not yet afaik [15:43] davidcalle, you're right :-(( [15:43] I reproduced it [15:45] dholbach: are you sure about adding the RedirectView to the urlpatterns? I thought we were going to override the 404 handler to do this [15:46] mhall119, I couldn't find a way to do this properly, and I just saw I need to fix something more important now :-/ [15:46] and maybe https://docs.djangoproject.com/en/dev/ref/contrib/redirects/ would be an option too [15:48] dholbach: from my point of view, the redirect is more urgent than the importer. I can look into redirects if you want [15:48] 'cause I haven't, yet :) [15:48] dholbach: it doesn't look like the redirects middleware supports regex or base-path redirects, only 1:1 path matches [15:49] but it might be easier to fork that, rather than start from scratch [15:50] davidcalle, ok, cool - I'll look into fixing this issue then [15:50] ok :) [15:51] dholbach: I'm merging your branch from today into trunk to make things easier, hold on [15:52] ok, thanks [15:52] dholbach: done [15:52] * dholbach hugs davidcalle [16:06] mhall119, davidcalle, dholbach, previous call running over [16:08] thought so ;-) [16:13] mhall119, joining the desktop docs call? [16:49] dpm: sorry, got carried away with lunch [16:49] mhall119, got time for a quick hangout? We can reuse the same url as the previous call [16:50] sure