[00:02:58] jdlrobson with https://github.com/wikimedia/WLMMobile/pull/137 i am getting monuments for the top-level admin levels rather than subdivision admin levels [00:03:13] that is, i click on 'United States', i get a list of monuments, rather than the next set of subdivisions [00:04:01] unless… MaxSem could that be related to your api changes ^ [00:04:02] ? [00:06:12] awjr: [00:06:13] http://wlm.wikimedia.org/api/api.php?action=adminlevels&uselang=en [00:06:19] http://wlm.wikimedia.org/api/api.php?action=adminlevels&uselang=en&admtree=us [00:06:24] http://wlm.wikimedia.org/api/api.php?action=adminlevels&uselang=en&admtree=us|us-mh [00:06:31] WFM [00:06:45] oh MaxSem i didn't update my admin_tree table after your changes [00:20:55] awjr: works for me [00:21:26] jdlrobson if you go us -> us-ca -> [[alameda county]] what happens? [00:21:57] awjr: i get monuments [00:22:15] ! [00:22:25] that's good [00:23:10] what is weird is that is not the behavior i am seeing from my checkout of that pull request [00:23:11] one sec [00:24:12] jdlrobson what is also weird is that you should not be getting monuments, you should be getting another list of subdivisions. [00:24:44] jdlrobson you should be getting: http://wlm.wikimedia.org/api/api.php?action=adminlevels&uselang=en&admtree=us|us-ca|%5B%5BAlameda%20County%2C%20California%5D%5D [00:25:02] awjr: are you using your api or wlm.wikimedia.org [00:25:02] ? [00:25:09] or toolserver? [00:25:12] jdlrobson i've tried with both [00:25:22] both wlm.wikimedia.org and my own [00:25:27] ok we'll look at it tomorrow [00:25:28] i [00:25:33] i need to prepare for presentation [00:25:39] ok [00:25:43] good luck! [00:48:21] awjr: i don't understand your comment btw on #137 [00:48:24] we should be encoding [00:48:45] jdlrobson indeed but it appears to be getting double encoded [00:48:50] urls with foreign characters for instance [00:49:05] mm it shouldnt [00:49:12] i know, that is the problem :) [00:49:21] the example you submit though is fine [00:49:22] it [00:49:25] it does the right thing [00:49:33] eh? [00:49:51] well %5B%5B is [[ encoded [00:50:14] maybe i'm misunderstanding [00:50:27] jdlrobson oh i see what you mean - but what i posted there is what the api sees AFTER the $_GET params have been decoded [00:50:44] * jdlrobson doesn't really understand and will talk about it tomorrow [00:50:45] which is why i think the values are getting double encoded [00:50:54] ok [00:50:58] code looked fine to me though [01:58:33] Why is utorrent tweeting pictures of Tomasz? [01:58:46] https://twitter.com/utorrent/status/235917274442252288 [01:59:19] Ahh, there's some meetup.. [02:23:17] Reedy: I didn't know uTurrent had a Twitter feed. [02:23:21] uTorrent, too. [02:23:32] I think I use rtorrent these days. [02:23:35] With ruTorrent. [17:07:13] jerith: ping [17:33:38] MaxSem can you take a look at r855 in the api? if it looks ok, can you get it pushed to yttrium? [17:34:05] *looking* [17:34:40] the problem it's attempting to address is adm tree names that have a "|" in them - the admintree api explodes on "|" so it was causing problems with admin tere values like [[Alameda County, California|Alameda]] [17:34:42] awjr, first of all - it breaks b/c [17:34:50] MaxSem: ? [17:35:10] how? [17:35:54] eh, if it's just for admin tree it's ok [17:36:01] it is :) [17:36:45] awjr, was there any discussion whether wikitext in DB makes any sense at all? [17:36:50] it requires a change in the app which i'm going to coordinate with jon to escape the "|" in admin tree values with a "\", so [[Alameda County, California|Alameda]] would become [[Alameda County, California\|Alameda]] [17:37:11] MaxSem: IRC, yes [17:37:21] olololo [17:37:27] just realised: [17:37:28] MaxSem: i believe the problem is that there is a lot of it. [17:38:18] how should we sort array( 'foo', '[[bar]]' )? [17:39:02] MaxSem: my inclination is to let it sort the way it would automagically and not worry about it [17:39:18] we have enough other problems to cope with :p [17:39:19] wikitext makes alphabetic sort a chimera [17:39:29] wikitext is, in short, evil. [17:39:42] it's not once you filter it out [17:39:50] kwadhwa: yo [17:40:06] MaxSem: you are saying it is not evil once you get rid of the evil bits? :p [17:40:12] tfinc: hey [17:41:12] awjr, s/\[\[(.*|)?([^|]*?)\]\]/$1/ [17:41:21] kwadhwa: how's india treating you ? [17:41:34] exhausting [17:41:43] I have to jump on a call right now [17:41:49] the sun's too shiney? [17:41:49] MaxSem to clean up the db? [17:41:51] we're putting in some long days [17:42:04] awjr, this'll have to be done in the bot [17:42:13] w/ Amit and Patrick [17:42:25] MaxSem we'll have to talk with maarten about that. last we chatted, i believe he stated the wikitext was a requirement. [17:42:55] preilly: Pong. [17:43:14] Blame kwadhwa for the amount of time supper took. :-) [17:43:16] MaxSem getting rid of the wikitext will only marginally solve some of the problems. there are still weird issues in the app with encoding [17:45:02] MaxSem: what i checked in will at least solve part of the problem, esp if we wind up needing to keep the wikitext. [17:45:14] don't believe jerith, he was chatting away w/ his partner in crime, dfoy [17:45:48] has anyone seen jon around? [17:46:13] tfinc, preilly ^ [17:46:59] awjr: at his desk [17:47:08] preilly can you tell hi to get on irc/ [17:49:05] just want to let people know I am in a bunch of product meetings today, but if there is anything you need from me, ping me on gchat [17:49:13] awjr: he's a coming [17:49:17] thanks! [17:49:19] there he is [17:49:33] hey sorry for some reason i didn't auto join this group [17:49:49] hi jdlrobson no problem [17:50:07] jdlrobson can we chat for a minute about the encoding/recursion problem? [17:50:13] sure [17:50:29] gimme a sec to get setup with where i got to yesterday [17:50:31] MaxSem: if that rev looks ok can you push it to yttrium? [17:50:41] jdlrobson cool sounds good. ring me when you're ready [17:50:47] oh, pushing [17:50:50] thanks [17:53:08] hey jdlrobson [17:53:10] how did the meetup go? [17:54:44] stabstabstab, git-svn + gerrit, stabstabstab [17:55:15] * yuvipanda smears little bit of wolfsbane on MaxSem's knife [17:56:33] awjr: > http://wlm.wikimedia.org/api/api.php?action=adminlevels&format=json&admtree=us%7Cus-ca%7C%255B%255BAlameda%2520County%252C%2520California%255D%255D [17:56:37] awjr: > admtree:us|us-ca|%5B%5BAlameda%20County%2C%20California%5D%5D [17:57:10] jdlrobson: http://wlm.wikimedia.org/api/api.php?action=adminlevels&format=json&admtree=us%7Cus-ca%7C%255B%255BAlameda%2520County%252C%2520California%255D%255D [17:57:15] jdlrobson: http://wlm.wikimedia.org/api/api.php?action=adminlevels&format=json&admtree=us|us-ca|%5B%5BAlameda%20County%2C%20California%5D%5D [17:58:52] jdlrobson: tree[ i ] = tree[ i ].replace( /\|/g, '\\|' ); [18:00:17] jdlrobson: http://dpaste.org/BtCuB/ [18:04:26] jdlrobson: im having networking problems with my isp - hence very slow speeds on my end [18:05:16] awjr, now someone's gotta approve https://gerrit.wikimedia.org/r/#/c/20050/ :) [18:05:36] MaxSem: looking [18:06:09] derp code comments in languages i don't know :( [18:06:58] MaxSem: merged [18:07:04] cool, thanks [18:08:01] awjr, deployed [18:08:33] w00t! [18:37:14] awjr: http://img2.ranker.com/list_img/40487/385342/full/the-very-best-of-the-success-kid-meme.jpg?version=1320517773000 [18:43:59] yuvipanda: ping [18:50:01] awjr: have fun https://github.com/wikimedia/WLMMobile/pull/137 [18:51:02] it's funny that jquery would urlencode components automatically but not decode them [18:51:34] jdlrobson: i'll take it for a spin momentarily [18:51:47] im just trying to sort out networking issues with my isp [18:51:58] awjr: it's not that [18:52:06] it's because i encode them in page history [18:52:10] oh! [18:52:23] it's a bizarre issue related to the fact that the api doesn't tell me when i'm not going to get any sub levels [18:52:45] what happens at the moment is we hit level 1, level 2, level 3.. level 4 returns an empty list so we then go back and then loads the results page [18:53:01] the go back step was causing level 3 url encoded to be loaded [18:53:06] ohhh [18:53:08] which was screwing with the history [18:53:10] makes sense [18:53:28] it's hacky.. that's why the api should be nicer :) [18:54:13] there are a lot of things the api should be :) and even more that the data should be! [18:57:41] Jarry1250: pong [18:58:05] yuvipanda: I noticed you mentioned push notifications [18:58:43] Jarry1250: indeed I did [18:58:49] I plan on having that on in a week's time [18:59:10] (over the weekend, basically) [18:59:17] (which starts tomorrow, actually!) [18:59:21] I was just going to say, on iOS I used http://urbanairship.com 's free package, made it nice and simple. Not sure how much easier they make Android pushes though (they do support them) [18:59:38] I say "package" as in "price point". [18:59:52] Jarry1250: https://developer.android.com/guide/google/gcm/index.html [18:59:55] free from Google :P [19:00:16] yuvipanda: Ah :) good shout. [19:01:49] yuvipanda: Not sure what system you've got going, but [19:02:05] Jarry1250: 'system' as in? [19:02:10] I run the bot that handles Signpost publication, going to be refitting it shortly [19:02:35] So if you want me to trigger whatever, don't mind adding that in. [19:02:39] Jarry1250: https://github.com/yuvipanda/WPSignpost/tree/master/api [19:02:59] Jarry1250: i've a nice trigger point already built in, but I think a trigger point in the bot would be nice too [19:03:17] Oh, what trigger have you got written in? [19:03:38] Jarry1250: there's a scraper written, and I have a nice 'commit' point [19:03:57] so I'm planning on sending out the notifications when a commit happens, which happens when a new issue is published [19:04:03] will also add a few more checks [19:04:15] But who triggers the commit? You? [19:04:31] Jarry1250: cron [19:04:39] MaxSem: awjr_lunch : anything need to be done on the client to support translations of campaign names? [19:04:44] so yeah, a webhook system of sorts would be nice [19:05:13] yuvipanda: Yup, I mean, you could cronwatch WP:.../issue hourly or something [19:05:27] Jarry1250: yeah, that's the idea [19:05:28] But if there's some easier way that I can give a nudge to trigger the GCM push [19:05:34] * MaxSem has no idea [19:05:45] Jarry1250: yeah, since that's a bot, it could easily do that - but has auth / security issues [19:05:54] will poke you if I think that's going to be a better solution than a hourly cron [19:06:06] yuvipanda: For urbanairship, they just required a username password to authenticate with [19:06:28] I don't know if GCM has a similarly easy API [19:06:39] Jarry1250: this runs on my server, I'm wondering more about how random people can hit a web URL and trigger notifications [19:06:44] and cronwatch is simple enough [19:07:29] Hit a web URL with the correct username/password :P Yeah, cronwatch is simple enough, I guess. [19:07:56] sigh. I've done a lot of performance engineering just to discover that sorting by name makes no sense due to wikitext [19:09:24] Jarry1250: so, can I get png renders of this? [19:09:26] http://harryburt.co.uk/files/placeholder-notext.svg that is [19:09:57] yuvipanda: Er, why do you need PNG renders? Does Android not support native SVGs? [19:09:58] MaxSem what is the scope of the problem? from just casually looking through the db, wikitext doesn't seem to be super common, and when it happens, it seems to be that all monuments for a given admin level use wikitext [19:11:20] awjr_lunch, http://dpaste.org/9W7Jc/ [19:12:25] :( [19:12:35] Jarry1250: not natively [19:12:51] but MaxSem those all seem to have different adm levels [19:13:18] they're the same for adm0...2 [19:13:33] MaxSem: sure but we won't return monuments until there are no more subdvisisions to return [19:13:45] yuvipanda: Really?! Weird. [19:13:48] okay [19:13:53] committing [19:14:04] so sorting would only really matter for the bottom-most adm (i guess except for in the case of nearby) [19:14:08] the solution is really insane [19:14:14] oh good [19:16:28] Jarry1250: yeah, so we actually need different sized images [19:16:37] also I'm waaaay too sleepy to do anything useful right now [19:16:43] so I'm just going to give in to my body and go sleep [19:16:48] sorry :( [19:16:55] Well, the 256px [19:17:05] 256px != 256px [19:17:06] version will be at s/.svg/.png/ [19:17:11] it's actually 256dp :P [19:17:22] which could be 256px, 512px, or something in between, depending on DPI [19:17:29] Jarry1250: so make me a 512px version too :D [19:17:33] Hmm. It's not a large file [19:17:36] so I can scale it down for the other versions [19:17:36] Just downscale? [19:17:56] philinje, I've made admin level lookups sort alphabetically [19:18:22] yuvipanda: Cool, it'll be a t s/.svg/.png/ [19:18:25] when you wake up. [19:18:30] thanks! [19:18:33] gnite for now :) [19:18:43] Jarry1250: and I'm still not fully convinced about crop vs fit :) [19:18:51] for tomorrow! [19:28:55] MaxSem: great news! [19:29:29] how is performance? [19:30:21] acceptable, otherwise it wouldn't have been done:) [19:32:44] anyone know if mf_mobileFormat is the cookie name used in productie as well for UseFormatCookie ? [19:36:36] thedj: on the wmf cluster, no [19:37:09] thedj on the cluster the only cookie that gets used is stopMobileRedirect - when that cookie is present, automattic redirection to the mobile site will not happen [19:38:15] hmmm. there is something wrong there [19:38:25] we have had many reports of the cookie failing to clear [19:39:02] thedj yeah the reason why stopMobileRedierct doesn't get unset is partially related to how caching works in production [19:39:02] hmm, we set the expiration with -3600... [19:39:20] ah caching was my other bet. [19:39:24] :) [19:39:47] anyways, -3600 is a bad value, always should be more than 25 hours, just to make sure you have no timezone trouble EVER. [19:40:23] awjr: is the recirect being cached or something ? [19:40:40] thedj there's been some good discussion on https://bugzilla.wikimedia.org/show_bug.cgi?id=38009 as well as a possible solution proposed - including a working interim solution [19:40:50] [WikipediaMobile] brion pushed 1 new commit to master: http://git.io/vgipbw [19:40:50] [WikipediaMobile/master] Update Windows 8 app from RC to final RTM version - Brion Vibber [19:41:17] Project WikipediaMobile - Nightly builds build #370: SUCCESS in 17 sec: https://integration.mediawiki.org/ci/job/WikipediaMobile%20-%20Nightly%20builds/370/ [19:41:18] Brion VIBBER: Update Windows 8 app from RC to final RTM version [19:42:17] thedj the solution is to unset the stopMobileRedirect cookie using javascript on the desktop version of the site rather than rely on the backend handling the cookie, and i've been thinking this is probably the solution that should be implemented in MobileFrontend anyway [19:42:45] ah, i put the common.js thing in. [19:43:01] the solution in common.js will not work; in fact it might even cause more problems [19:45:24] it doesn't set any extra cookie for me. [19:46:13] that is good - i looked at it quickly this morning, but i was concerned the cookie paths won't match between what's in MediaWiki:common.js and what gets set by MobileFrontend [19:46:21] thedj is it actually unsetting the cookie for you? [19:47:14] looks to work on safari. [19:47:29] on ios it seems my script pages are aggressively cached by ios though. [19:47:42] so i need to clear the browser cache, and then it starts to work. [19:48:08] how bout we do this: $.cookie( 'stopMobileRedirect', false, { [19:48:08] domain: '.wikipedia.org', [19:48:08] path: '/', [19:48:08] expires: -2 [19:48:08] } ); [19:51:30] thedj i think that should work - i've never used just an integeter in $.cookie for expires before - what does it signify? seconds? minutes? hours? days? eons? [19:52:07] days [19:52:19] thedj i tried this quickly this morning and seemed to work for me this morning but probably needs some more thorough testing: http://en.wikipedia.org/wiki/User:Awjrichards/common.js [19:52:46] thedj cool then i think that should work - give it a shot [19:58:21] hmm, i wonder.... [20:00:17] thedj this works for me in ios5.1 and android 4.04: http://en.wikipedia.org/w/index.php?title=User:Awjrichards/common.js&pe=1& [20:00:31] ah [20:00:45] awjr: have it, there is a bug in jquery.cookie [20:00:51] var days = options.expires, t = options.expires = new Date(); t.setDate(t.getDate() + days); [20:01:03] options.expires ? '; expires=' + options.expires.toUTCString() [20:01:07] it doesn't use t [20:01:13] so you need a string expires [20:01:35] i like the explicitness of the string anyway [20:01:51] anyway, thedj i've got to get back to other stuff - let me know how it goes! [20:07:00] bingo, that was it. [20:07:20] :D [20:08:22] jdlrobson your updates to pull req 137 look good - the only problem is, we're still not getting monuments [20:08:34] but at least the infinite recursion appears to have stopped :) [20:17:02] awjr: jdlrobson is out of the office [20:17:12] thanks preilly [20:17:39] argh [20:17:43] pesky wikitext [20:18:14] jdlrobson when you get back.. i think the reason we're not getting monuments is a bug in the api rather than the app [20:35:20] awjr / MaxSem : Sorry, bit short on time so answering emails takes a bit long...... [20:46:58] i must say that im enjoying using the nexus7 so far [20:49:47] preilly: kul is interested in us doing the Wikipedia app on FireFox OS for launch. i'm going to spin up a project page for it. will cc on some mails later today [20:50:33] MaxSem: when you have a chance can you look at r857 in the api? if it looks good enough can you also push it to yttrium? [20:52:45] awjr, unlike admintree, action=search is older and should have existing callers(?) [20:53:09] MaxSem: my change should be b/c - that is, existing callers are just as broken now as they were before :) [20:54:10] Platonides, what do you think about r857? [20:56:40] awjr, also - I'm thinking about filtering out all wikitext just on yttrium - what do you think about it? [20:57:07] MaxSem i do not think it is a good idea [20:57:26] we should try to keep yttrium and toolserver identical for failover [20:57:38] MaxSem: while the wikitext is a totaly pita, we can work around it [20:58:16] unless it's causing any other major complications im not aware of? [21:00:33] just sorting - if you say that it's ok let it be [21:02:35] MaxSem for now let's not worry about it. i have a feeling we'll burn a lot of time trying to clean it up AND make necessary changes elsewhere in the stack to accommodate it - i think it will be like opening a can of worms. the wikitext makes sorting a little weird, but i do not think that is a show stopper or that big of a deal for now [21:02:51] we should get everything else done first [21:08:07] awjr, deployed [21:08:15] :D thanks MaxSem [21:09:02] * MaxSem starts worrying that his repo for doing updates is not quite reproduceable:) [21:15:09] ruhroh [21:15:21] MaxSem: {"errors":["SQL error 1146: Table 'wlm.monuments_all' doesn't exist\nLast query was: SELECT `country`, `lang`, `id`, `country`, `lang`, `id`, `adm0`, `adm1`, `adm2`, `adm3`, `adm4`, `name`, `address`, `municipality`, `lat`, `lon`, `image`, `source`, `monument_article`, `registrant_url`, `changed` FROM `monuments_all` FORCE INDEX( `admin_levels2` ) WHERE (`adm0`='us') AND (`adm1`='us-ca') AND (`adm2`='[[Alameda County, California]]') A [21:15:38] I'm upgrading schema:P [21:15:43] phew! [21:15:53] MaxSem: can you ping me when it's done? [21:17:52] awjr, ping [21:17:58] MaxSem: pong [21:18:09] oh - it's done? [21:19:39] [WLMMobile] awjrichards pushed 4 new commits to master: http://git.io/hoFX4Q [21:19:39] [WLMMobile/master] avoid upstream scoping issues - jrobson [21:19:39] [WLMMobile/master] fix issue related to back button - jrobson [21:19:39] [WLMMobile/master] escape pipe symbols in campaign names - jrobson [21:19:54] Project WLMMobile - Nightly builds build #58: SUCCESS in 9.6 sec: https://integration.mediawiki.org/ci/job/WLMMobile%20-%20Nightly%20builds/58/ [21:19:54] * jrobson: avoid upstream scoping issues [21:19:55] * jrobson: fix issue related to back button [21:19:55] * jrobson: escape pipe symbols in campaign names [21:20:08] jdlrobson: i merged 137. with api fixes we just pushed out, things appear to be working correctly now :D [21:31:18] awjr: we need to pick a time to do our last m. moves. http://www.mediawiki.org/wiki/Mobile_default_for_sister_projects . when would you like to schedule them ? [21:31:55] tfinc: so… commons, wikitech, mediawiki.org? [21:32:11] i'm thinking of punting on wikitech for now [21:32:14] jdlrobson: is it [21:32:15] so just commons and mw [21:32:33] wmf.org too [21:33:10] awjr: the timelines is up to us. id just like to close out the project [21:33:45] tfinc: totally. where are we with all the work philinje wanted to do around commons before we enable there? [21:34:47] awjr: were not going to invest time into it. were simply going to enable the redirect. [21:35:03] tfinc is the same thing true for the other outstanding domains? [21:35:25] awjr: yes [21:35:29] great [21:35:32] preilly: can we close out this project http://www.mediawiki.org/wiki/Improved_Mobile_Device_Detection ? [21:36:17] tfinc i think we should make some improvements to redirector.c before we do the next batch [21:36:27] which i imagine we'll have some bandwidth to do after the all-staff [21:36:38] so let's schedule the switchovers for the end of sept [21:36:51] tfinc: yes [21:42:53] New patchset: Jdlrobson; "revert damage from 331df963" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/20226 [21:42:58] ^ preilly [21:43:29] Change merged: preilly; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/20226 [21:45:31] awjr: do we have a backlog post wlm on mingle ? [21:45:44] tfinc not that i am aware of [21:46:15] preilly: add a last update to the project and i'll close it out. [21:47:02] tfinc we'll need to change how we're using mingle post-wlm anyway since right now 'wlm' is a unique project on mingle. we'll probably want to repurpose the 'mobile' mingle project and set it up to track work for multiple products [21:47:15] awjr: id like to start prepping work for post wlm. when could we get that setup in mingle ? and yes i know that we might choose not to use mingle after but its easier to stay within one tool for now [21:47:33] awjr: tfinc brion just a reminder i'm out tomorrow [21:47:48] so if you want anything from me before the end of the week today is the time to bug me [21:47:57] jdlrobson: what time are you leaving todya? [21:48:16] jdlrobson: i know you had some concerns about getting the beta out tomorrow, so please make sure to sync up with phil about that before you leave [21:48:17] whenever needed [21:48:20] also could someone review https://github.com/wikimedia/WLMMobile/pull/135/files [21:48:32] it would be good to get all the remaining pull requests merged so we can see the state of the app [21:48:54] tfinc let me noodle on it for a bit. probably the thing to do is get the 'mobile' project in mingle cleaned up and start using that for building a backlog [21:49:27] New review: Siebrand; "The commit ID doesn't link. Wheres the offender?" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/20226 [21:49:32] currently reviewing https://github.com/wikimedia/WLMMobile/pull/132/files [21:49:34] tfinc i am pretty swamped with dev stuff for the beta through tomorrow, so it would be better for me to start messing around with it next week [21:50:19] awjr: thats fine. i'll keep it on our mw.org roadmap for now [21:50:34] tfinc cool. next week i'll see about cleaning up the mobile proejct on mingle [21:50:38] awjr: i wanted to flag it as early as possible for you so that we would have plenty of time to noodle on it [21:50:43] thanks :) [21:54:07] tfinc: updated status [21:54:16] preilly: thanks [21:56:29] Change merged: preilly; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/19395 [21:59:21] jdlrobson: I'm currently updating LocalisationCache for 1.20wmf9... [21:59:33] jdlrobson: the fixed key should be live soon'ish [21:59:59] thx preilly [22:00:09] jdlrobson: np [22:00:17] jdlrobson: thanks, for noticing and pointing it out [22:03:21] jdlrobson: i bet brion will trade you reviews for those pull requests. i have my hands full right now and can't really look [22:03:32] oh but brion is mia [22:04:10] i think he's in tech days planning - he should b eone at 4 i think [22:04:13] er [22:04:21] s/b eone/be done [22:04:35] if preilly has any spare cycles he might be able to help out with those pull requests too [22:05:16] do you guys need me to hunt down brion ? [22:05:18] let me know [22:05:18] awjr: I'm busy with VUMI right now [22:05:35] tfinc according to his calendar he's in a mtg til 4 [22:05:46] preilly will you have some spare time after? [22:05:52] let me txt him [22:06:18] tfinc i dont think it's ultra urgent, it can wait til he's done [22:06:36] k, let me know [22:06:40] thanks :) [22:12:59] jdlrobson: the message key should be fixed now [22:18:01] awjr: are there any remaining data questions? [22:18:16] or rather, language and data questions? [22:18:32] philinje i think we're ok for now given the info that we currently have [22:55:52] [WLMMobile] jdlrobson pushed 4 new commits to master: http://git.io/Tdg28Q [22:55:52] [WLMMobile/master] Check if monument articles exist and show a red/broken link if not. - Brion Vibber [22:55:52] [WLMMobile/master] use class for broken links - Brion Vibber [22:55:52] [WLMMobile/master] cleanup on article does not exist message - Brion Vibber [22:56:01] Project WLMMobile - Nightly builds build #59: SUCCESS in 7.4 sec: https://integration.mediawiki.org/ci/job/WLMMobile%20-%20Nightly%20builds/59/ [22:56:01] * Brion VIBBER: Check if monument articles exist and show a red/broken link if not. [22:56:02] * Brion VIBBER: use class for broken links [22:56:02] * Brion VIBBER: cleanup on article does not exist message [23:08:21] * tfinc watches chromebox update [23:12:31] * jdlrobson cries as he doesn't understand why tests have broken [23:12:50] did someone say chromebox? [23:13:02] * awjr hands jdlrobson a tissue [23:13:54] * jdlrobson discovers the cause [23:18:41] * tfinc realizes one critcal flaw in his chromebox strategy [23:19:10] awjr: tree[ i ] = tree[ i ].replace( /\|/g, '\\\|' ); broke my tests :( [23:19:34] jdlrobson: one sec [23:20:10] oh. nm. it *does* have built in speakers [23:20:12] woot [23:23:27] awjr: join http://bit.ly/N77SXr for a quick test at my desk [23:23:38] tfinc sec [23:24:06] i'll need an HDMI cable before i can setup it up in a conf room [23:24:13] set it up* [23:24:47] ok tfinc here i come [23:25:37] awjr: this guy may have no internal mic [23:26:13] jdlrobson i doubt this is the reason why but there's actually a problem in that line too [23:26:28] jdlrobson: i think that should be tree[ i ] = tree[ i ].replace( /\|/g, '\\|' ); [23:26:33] jdlrobson what broke exactly? [23:26:51] don't worry i'm looking into it [23:27:11] awjr: it's not picking up the yeti [23:27:18] let me tinker with it some more and then we can try it later [23:27:25] how is the video feed on its own though? [23:27:31] tfinc sounds good [23:27:35] good luck! [23:27:46]