[00:24:51] sorry DarTar, that's some weird stuff [00:25:06] I'm not sure why it's redirecting to localhost [00:25:22] the good news is that it's still signing you in, so if you go back to the original page and refresh, you're ok [00:25:43] i'm adding the localhost redirect thing as a bug though [00:29:02] yep, it's definitely not a blocker just a little nuisance :) [12:48:22] hi [12:48:27] hi dr0ptp4kt [12:48:32] milimetric: in the airport in Iasi [12:48:37] milimetric: free wifi [12:48:44] surprisingly [13:14:32] free wifi! None of the airports here have free wifi. Only Amsterdam did [14:26:36] brb [15:20:06] milimetric: did you get the gerrit / gitweb issues worked out? [15:20:11] yes [15:20:14] chad pinged me [15:20:18] he said he messed something up [15:20:21] what was the fix? or did it fix itself [15:20:22] i see [15:20:31] i'm hot on the trail of making the namespace edits work [15:20:42] niiice [15:21:03] yeah, just give me like 20 minutes and then maybe we can hangout and talk documentation? [15:21:09] oh - it's still early for you [15:21:11] whenever [15:21:50] no i'm in the office [15:21:54] !log decommissioning analytics1020 from hadoop cluster in preparation for reinstall [15:21:56] I'm good to go whenever [15:27:19] hey erosen, is there a way to get ipython to re-load source once i've imported it? [15:27:37] sort of [15:27:53] rather, there is a config option, but I tried it once and didn't get it to work [15:28:00] k, cool [15:28:07] i'll start making temp scripts then [15:28:24] milimetric: http://ipython.org/ipython-doc/dev/config/extensions/autoreload.html [15:32:10] erosen: I've added a file called adhoc.py to tests, we can "from tests.adhoc import *" when we wanna mess with sqlalchemy and stuff [15:32:26] it's useful for me as I fumble around in the dark through the awesome query syntax [15:33:07] nice [15:37:15] ok erosen, I pushed [15:37:18] k [15:37:31] milimetric: pulled [15:37:39] hangout? [15:37:39] I'm gonna run out for some food, but take a look and we'll talk when I'm back? [15:37:41] k [15:37:43] sounds good [15:38:25] NamespaceEdits is not working quite yet because it doesn't account for when there are no revision records [15:38:34] but it issues the correct query from what I can tell [15:38:46] you're welcome to make the tests pass if you'd like :) [15:38:56] (those should at least be good) [16:04:48] alright erosen, batcaving [16:04:58] omw [16:17:18] hello [16:35:49] hello! [16:35:52] average! [16:35:54] :) [16:36:01] i want to help with dclass but we need to talk first I think [16:56:15] mmmmmmmah [17:00:52] ottomata: scrum [17:01:34] whoops! [18:19:12] milimetric: question about multiple engines per declarative base when you get a sec [18:19:35] yeah, that's a tough one [18:19:43] i'll look it up [18:19:45] similar but not useful: http://turbogears.org/2.1/docs/main/MultipleDatabases.html [18:21:44] ottomata: you around? [18:22:05] yup hiya [18:22:37] howdy, I need a quick permission/ownership fix on stat1 [18:23:14] yuppers, what'sup? [18:25:11] milimetric: very intersting stack overflow question: http://stackoverflow.com/questions/1308765/multiple-mappers-for-the-same-class-in-different-databases [18:25:21] erosen: I think you basically could just create a MediawikiSessionEN, MediawikiSessionAR, etc., probably stored in a dictionary [18:25:42] * DarTar rsyncing an extra half pint to brooklyn [18:25:47] and just use the in-memory sqlite one for the metadata [18:26:14] k, reading [18:26:29] erosen, milimetric: in full sqlalchemy swing? [18:26:33] milimetric: ooh, interesting, i didn't realize that declarative_base is independent of session [18:26:41] DarTar: yup [18:26:46] sweet [18:26:48] yeah Dario, it's fun [18:26:57] DarTar: yeah, pretty fun so far [18:27:07] we've almost got NamespaceEdits working (was the easiest one) [18:27:16] so evan's working on making it work against arbitrary connections now [18:27:52] great, I was talking to adrian yesterday: another one you guys may want to experiment with is threshold, it's been heavily tested on v1 and should be fairly straightforward to port [18:28:29] (good if we want to compare the output of v1 and v2) [18:28:34] just a thought [18:28:45] cool [18:29:05] sounds good, we were thinking of doing bytes_added next since Frank wants that out of the gate [18:29:17] yeah, I heard that, makes sense [18:29:20] but threshold will be next on our list if ye say it's easy :) [18:29:46] yeah, erosen it looks like the only times you bind the declarative base to the engine is when you do stuff, like "create_all" [18:29:49] so we should be ok [18:29:52] yu[ [18:29:59] and for that, it is fine to loop through the sessions [18:30:14] and create_all for all projects [18:30:22] alright, back to my dashboards (the stat1->stat1001 rsync folder is operational in case you guys didn't notice, ori and ottomata fixed it yesterday) [18:30:31] nice [18:30:52] I haven't spread the word yet [18:31:15] just migrating already public data from the TS [18:57:03] gotta run for a min, be back in a 30 mins [20:17:52] phewwweee back [21:46:48] erosen: just pushed working web stuff [21:46:58] niiiiice [21:46:58] copied the templates too [21:47:03] awesome [21:47:12] i'm gonna go update mingle and call it a day afterwards [21:47:19] i think i'm getting the hang of the sqlalchemy stuff [21:47:20] but i wanna make sure you're good before that [21:47:22] nice :) [21:47:26] it's so slick right?! [21:47:35] i love it, best ORM I've ever worked with [21:47:52] yea