[00:05:54] <dr0ptp4kt>	 vibha: sorry, just saw the irc message. thanks for following up on gchat. gonna head out. have a good rest of the day
[00:07:57] <vibha>	 sent you an email also.
[00:08:15] <vibha>	 have a lovely ride home
[08:28:03] <joakino>	 mornint
[08:28:08] <joakino>	 morning
[14:01:48] <mdholloway>	 dbrant: does the main page now fail to reload for you when returning from a fragment accessed via the nav drawer?
[14:03:10] <dbrant>	 mdholloway: hmm, so it does :(
[14:03:41] <bearND>	 oops
[14:03:47] <mdholloway>	 rats.  could've sworn i'd done that a couple dozen times while testing. 
[14:04:09] <mdholloway>	 bearND: maybe a product of one of the most recent changes?
[14:04:27] <bearND>	 mdholloway: yeah, i'm pretty sure it is
[14:04:41] <bearND>	 just don't yet how to fix it
[14:21:04] <matanya>	 dbrant: editing is mostly not possible on the newest alpha, would triage my bug report as UBN, but i won't step on your feet :)
[14:21:32] <dbrant>	 matanya: we're on it! thanks for your diligence
[14:21:44] <matanya>	 thanks much
[14:46:13] <dbrant>	 bearND: looks like the problem lies in the fact that you recreate the pageLoadStrategy every time onActivityCreated is called, which has the effect of resetting the backstack.
[14:46:59] <bearND>	 dbrant: I see
[14:47:35] <bearND>	 thanks
[15:02:02] <bearND>	 dbrant: mdholloway : I think I got a fix for the backstack issue. But now have to deal a crash when trying to edit. I think the editHandler needs to be set at the right time.
[15:02:12] <dbrant>	 bearND: as for editing... it looks like we're simply not calling editHandler.setPage() in enough places.  It's only called when the page finishes loading from the network. If the page is loaded from cache, it's not set.
[15:02:19] <dbrant>	 bearND: lol beat me to it.
[15:02:46] <bearND>	 dbrant: i had a few instances commented out. Will see if I can bring them back
[15:05:26] <mdholloway>	 bearND: dbrant yeah, you know, when i was looking over the code yesterday i was wondering if pageLoadStrategy should be a proper singleton, but it didn't seem to be causing any issues so I let it go
[15:15:53] <bearND>	 dbrant: mdholloway : ok, I think I've got it: https://gerrit.wikimedia.org/r/215924
[15:16:10] <mdholloway>	 bearND: great, i'll take it for a spin
[15:17:15] <dbrant>	 bearND: ah, covering both issues! looking now...
[15:19:57] <bearND>	 I didn't like sharing the editHandler between the fragment and the pageloader. Oh well.
[15:30:01] <mdholloway>	 bearND: how specifically was editing broken?  i'm back in master checking it out and there's no one specific thing going on that I can see, though i did get one crash and a couple of potentially bogus editing permission issues
[15:30:41] <mdholloway>	 also, i get one of the bad old "you don't have a network connection" error messages when trying to edit my user page from in-app (even after that patch), but i'm pretty sure that's a separate issue
[15:32:25] <mdholloway>	 nm, i see the phab task.
[15:33:42] <bearND>	 dbrant: mdholloway: I'm going to make a really minor amend, to update the name of the method and add the Pahb task as well, wasn't aware of it earlier
[15:34:11] <bearND>	 mdholloway: s/update the name of the method/update commit message/
[15:35:02] <matanya>	 bearND: should i add you as a cc next time ?
[15:35:37] <bearND>	 matanya: no, that's not necessary. But thanks
[15:37:35] <bearND>	 dbrant: mdholloway : i just updated commit message
[15:38:57] <kristenlans>	 Good morning coreyfloyd. Would you be up for a hangout today to run through the Q1 brainstorm etherpad to try to break out the ideas per platform (at least the ones you knwo are only app specific)? 
[15:39:38] <coreyfloyd>	 kristenlans: yeah sure
[15:40:22] <coreyfloyd>	 kristenlans: I’m free until standup if you want to do it nowish
[15:41:38] <kristenlans>	 cool thanks coreyfloyd that will be helpful. I actually have a meeting with Toby in 20 minutes that I want to prep for. Are you plannign on attending the monthly metrics meeting in real time today (2 PM ET)?
[15:42:24] <coreyfloyd>	 kristenlans: no wasn’t planning
[15:44:07] <kristenlans>	 OK, cool, I'll pop out a meeting aroudn then if that's cool.
[15:45:49] <dbrant>	 kristenlans: did I hear app-specific ideas?
[15:46:12] <kristenlans>	 ah yes dbrant it woudl be good to have you there too!
[15:47:12] <kristenlans>	 dbrant I had coreyfloyd on the brain because he put forth the idea in email. I sent you an invite.
[15:48:58] <mdholloway>	 kristenlans: that's curious, i don't have a monthly metrics meeting invite on my calendar.  
[15:50:59] <dr0ptp4kt>	 dbrant: jdlrobson2 any cards that should be tagged with 'roadmap' for https://phabricator.wikimedia.org/tag/roadmap/ ? if so, would you please tag those release cards and drag them to the correct column on https://phabricator.wikimedia.org/tag/roadmap/ ?
[15:51:13] <dr0ptp4kt>	 bearND ^
[15:51:40] <mdholloway>	 kristenlans: i have one for 6/11, though...
[15:52:27] <coreyfloyd>	 kristenlans: yeah just send the invite
[15:57:01] <kristenlans>	 mdholloway|brb: ah i think my claendar is out of sync with the WMF staff calendar
[15:57:12] <kristenlans>	 mdholloway|brb: 6/11 is right AFAICT
[15:57:34] <dbrant>	 kristenlans: any chance we can do 1/2 hour earlier?
[15:58:20] <kristenlans>	 dbrant: yep, updated
[15:58:40] <dbrant>	 kristenlans: thx!!
[16:01:33] <matanya>	 dbrant: new crash. reporting in a moment :)
[16:04:20] <Krinkle>	 dbrant: Hi
[16:04:36] <dbrant>	 Krinkle: hello!
[16:04:55] <Krinkle>	 dbrant: I noticed that commits from mobile apps routinely disturb the continuous integration system. Anytime this repo merges a commit, the whole CI stack clusterwide is halted for 10-30 minutes
[16:05:20] <Krinkle>	 This is because (most likely inintentionally) you merge commits manually instead of post-test
[16:05:57] <Krinkle>	 The convention is to use CR+2 and then Jenkins merges it for you.
[16:06:19] <Krinkle>	 Gerrit has terrible UX for this and has a confusing button "Publish and submit" where submit=merge immediately
[16:06:38] <Krinkle>	 instead always use 'Publish comments'.
[16:06:42] <Krinkle>	 I can remove the button if you like. 
[16:06:55] <Krinkle>	 Many projects have done so already, but it's there by default.
[16:07:17] <dbrant>	 Krinkle: ahh, I never knew that
[16:07:21] <dbrant>	 bearND: ^
[16:07:30] <Krinkle>	 I'll put this in an e-mail but just wanted to cathc up here first
[16:07:32] <Krinkle>	 see https://integration.wikimedia.org/zuul/
[16:07:56] <Krinkle>	 Because the commit is already merged, when the test finishi, it fails to try and merge it in Gerrit. And tries again and again.
[16:08:28] <Krinkle>	 That is an upstream bug in Zuul, but nothig we can change about that. Either stop force-merging or remove Jenkins from the CR+2 pipeline.
[16:08:35] <dbrant>	 Krinkle: so, simply pressing the "publish comments" button will solve all of this?
[16:08:39] <Krinkle>	 Yep
[16:08:40] <bearND>	 Krinkle: which gerrit repos does this happen in? We have several.
[16:08:53] <Krinkle>	 I see it most often in apps/android/wikipedia and ios
[16:09:26] <Krinkle>	 If there is no particular reason to bypass Jenkins, then all is good and you can just switch to using the normal Publish comments button
[16:09:43] <Krinkle>	 If you do need to bypass Jenkins, we can work out a way that works well with the system, no problem.
[16:12:07] <dbrant>	 Krinkle: unless bearND knows of a reason otherwise, we should just remove the "publish and submit" button in this case.
[16:12:36] <mdholloway>	 dbrant: bearND: Krinkle: ^ i agree.
[16:13:00] <bearND>	 Krinkle: hmm, strange. I wouldn't have expected that being done in those repos. Yes, please remove the button.
[16:13:28] <thedj>	 https://twitter.com/jaffathecake/status/595637850286206976
[16:13:38] <thedj>	 "Somewhat alarming that Wikipedia's mobile site gets to first paint later than the main one http://www."
[16:13:46] <bearND>	 Krinkle: We do have a few other repos where we actually have to do it (I think apps/android/java-mwapi, mediawiki/services/mobileapps, mediawiki/services/mobileapps/deploy) to get things merged. It would be great if we didn't.
[16:16:27] <Krinkle>	 bearND: hm.. can you elaborate?
[16:17:30] <Krinkle>	 Ah, I see.
[16:17:41] <bearND>	 Krinkle: I'm not sure why that is but on the aforementioned projects a simple CR+2 is not enough to get a patch merged.
[16:18:22] <Krinkle>	 OK. I'm gonna mess with the settings. If at any point later today things look wrong (e.g. no longer able to merge) let me know.
[16:18:33] <Krinkle>	 It's hard to test since I'm gerrit admin, so it always works.
[16:19:02] <Krinkle>	 bearND: Is there any repo under apps/* that doesn't have a Jenkins pipeline?
[16:19:18] <Krinkle>	 e.g. granting Jenkins merge rights and hiding that button would be fine to do for apps/*
[16:19:34] <Krinkle>	 By default on +2 jenkins will re-run the test and if passing, merge the commit.
[16:20:48] <bearND>	 Krinkle: well, apps/android/java-mwapi, I guess, unless you just changed it
[16:21:02] <Krinkle>	 But does that repo trigger Jenkins?
[16:21:38] <Krinkle>	 I'm trying to avoid a situation where you have a repo with no Jenkins tests and as such this would leave you unable to change the repo
[16:24:06] <bearND>	 Krinkle: there's no Jenkins job triggered from there, but I think it should. That project uses Maven and has test that could run on any JVM
[16:24:26] <Krinkle>	 I'll start with android app only for now then
[16:24:54] <Krinkle>	 Done
[16:24:54] <bearND>	 Krinkle: ok, and for iOS app it's best to check with bgerstle 
[16:25:12] <Krinkle>	 Verified>0 and "Submit" are now hidden
[16:25:35] <bearND>	 Krinkle: I think some of the iOS guys use the "New style" of Gerrit interface. It has different buttons for CR+2
[16:25:36] <bgerstle>	 what what?
[16:25:47] <bgerstle>	 i was using new style, but i think i switched to old? i don't know anymore
[16:25:52] <bgerstle>	 all gerrit styles are old
[16:25:57] <Krinkle>	 True that
[16:26:00] * bgerstle rimshot
[16:26:19] <bgerstle>	 Krinkle: if you're hooking stuff up to gerrit & jenkins, i'd love to talk to you
[16:26:29] <bgerstle>	 i.e. nerd snipe you to set some stuff up for us, potentially
[16:26:30] * Krinkle runs out of the building
[16:26:42] <Krinkle>	 I'm not that guy anymore.
[16:26:57] <Krinkle>	 Talk with RelEng (Greg/Antoine)
[16:27:09] <bgerstle>	 yeah, i planned to, just thought i'd ask :-P
[16:27:19] <Krinkle>	 I'm fixing this out of self interest because "invalid" merges break Zuul and slow everyone else down (incl. me)
[16:27:41] <Krinkle>	 it takes up to 30 minutes to realise the commit was alreayd merged and then it moves on
[16:28:05] <Krinkle>	 upstream wontfixed because one isn't supposed to use a gate with a way around it
[16:28:06] <bearND>	 bgerstle: the tl;dr is to not use the V+2 and the submit patch buttons in gerrit anymore. Krinkle is hiding the submit button
[16:28:19] <Krinkle>	 oh wait bearND != bgerstle
[16:28:22] <Krinkle>	 You are here
[16:28:42] <bearND>	 Krinkle: yep: bearND = Android, bgerstle = ios
[16:28:45] <bgerstle>	 please dont' hide verify buttons for iOS repo
[16:28:54] <Krinkle>	 strlen() was roughly the same length and same first character :P
[16:29:05] <Krinkle>	 bgerstle: OK. Why?
[16:29:29] <bgerstle>	 Krinkle: for the most part, it's a stopgap to side-step Jenkins, which doesn't really do anything for us atm anyway
[16:29:32] <Krinkle>	 Note that if you want to bypass Jenkins, that means the gate job will be disabled. You'll still have patch set testing, but not on CR+2 events.
[16:29:43] <bgerstle>	 Krinkle: we don't even have patch set testing
[16:29:53] <Krinkle>	 You do actually.
[16:30:07] <Krinkle>	 but maybe they're optimised away based on file patterns.
[16:30:18] <bgerstle>	 Krinkle: they're not testing the actual iOS parts of the app
[16:30:27] <Krinkle>	 sure, but that's not my concern.
[16:30:29] <bgerstle>	 so we haven't been paying them much attention
[16:30:34] <Krinkle>	 (not this time anyway)
[16:30:49] <bgerstle>	 Krinkle: i guess this means you're linting the JS in our repo?
[16:31:02] <Krinkle>	 bgerstle: It's been linted for several months yeah
[16:31:03] <Krinkle>	 https://gerrit.wikimedia.org/r/#/c/212030/
[16:31:06] <Krinkle>	 Jenkins does run
[16:31:10] <Krinkle>	 0 jobs, but they pass.
[16:31:12] <bgerstle>	 i see
[16:31:14] <Krinkle>	 and it votes
[16:31:30] <Krinkle>	 I assume the same happens on CR+2, not just patch sets
[16:31:35] <bgerstle>	 right, but its votes are meaningless for now
[16:31:40] <Krinkle>	 which means you'll be fine without V+2 or (force) Submit
[16:31:50] <bearND>	 Krinkle: fyi, I can confirm that the Submit button is gone for the Android app, and I can set verified only to -1 or 0 but no +1 or +2
[16:32:06] <Krinkle>	 bearND: Let me know if the next commit you +2 also merges properly
[16:32:12] <bgerstle>	 i guess what i'm wondering is, what will happen if these buttons are gone for us?
[16:32:22] <bgerstle>	 seeing as we have no jenkins jobs
[16:32:42] <Krinkle>	 bgerstle: the problem is that if a repo is configured with Jenkins/Zuul (specifically, if it has a CR+2 listener) and the commit bypasses Jenkins, things crash badly every time.
[16:33:10] <bgerstle>	 i have no idea what our Jenkins/Zuul config is
[16:33:12] <Krinkle>	 every time you merge, everyone else in Engineering gets their commits depooled from Jenkins and waittime +30 minutes on everage
[16:33:15] <Krinkle>	 For the past 2 months
[16:33:30] <bgerstle>	 orly
[16:33:37] <Krinkle>	 Don't ask me why nobody tells you or is fixing it #damnit
[16:33:54] <bgerstle>	 we're (slowly) working on it, sort of
[16:34:08] <bgerstle>	 but yeah, it's been a source of frustration
[16:34:29] <bgerstle>	 i just need to know whether we can still merge our code
[16:34:36] <Krinkle>	 So, with no authority, I'm demanding we either disable the Jenkins listener (and you'll have to re-enable when you get jobs, and lose jslint for now) - or I just disable the submit button (which means jslint will be enforced, fine; and when you add new jobs, those will be enforced)
[16:35:11] <bgerstle>	 i'm ok w/ jslint being enforced (though i prefer jshint)
[16:35:13] <bgerstle>	 but
[16:35:24] <bgerstle>	 does that mean we have to wait for it to run after we +2?
[16:35:29] <Krinkle>	 bgerstle: You should be able to merge fine still. Let's test it, shall we? Next time you're about to +2 give me a heads up and I'll disable it right before.
[16:35:30] <bgerstle>	 and it only runs *after* we +2?
[16:35:34] <bgerstle>	 ok
[16:35:39] <Krinkle>	 I'll leave it as-is for now so you don't depend on me in case it goes wrong
[16:36:07] <Krinkle>	 bgerstle: No, it'll run on patch sets (as it does now) and on CR+2 (as it does now) but instead of CR+2 immediately merging, itwill try Jenkisn first, and then merge if passing
[16:36:39] <bgerstle>	 Krinkle: ok, don't have any patches ready to merge just yet
[16:36:51] <Krinkle>	 cool, I'll be here to try it when you do. no worries :)
[16:36:53] <bgerstle>	 i'll try to ping you before we try (or if something goes wrong)
[16:37:15] <bgerstle>	 i can put up a simple patch to test it now if this is a good time for you
[16:37:42] <Krinkle>	 It's an easy toggle. I'm here for the next ~ 5-6 hours
[16:45:42] <legoktm>	 legoktm@fluorine:/a/mw-log$ grep -c "Special:UserProfile" api-feature-usage.log 928
[16:45:51] <legoktm>	 someone might want to look into that ^
[17:28:22] <Dagy>	 hellow everybody... i am new here i want to know how i can do to participate in development
[17:29:55] <Dagy>	 anyone who knows what i can do please?
[17:31:42] <kristenlans>	 coreyfloyd are you in hangout but AFK?
[17:34:15] <GEOFBOT>	 Dagy, welcome
[17:34:22] <GEOFBOT>	 Have any coding experience?
[17:34:39] <Dagy>	 i have lots of coding experince
[17:34:48] <GEOFBOT>	 like..?
[17:35:02] <Dagy>	 Javascript,
[17:35:05] <Dagy>	 html 5
[17:35:06] <Dagy>	 C#
[17:35:08] <Dagy>	 c
[17:35:12] <Dagy>	 VB
[17:35:17] <Dagy>	 python
[17:35:20] <Dagy>	 mysql
[17:35:27] <Dagy>	 mssql
[17:35:31] <Dagy>	 php
[17:35:53] <GEOFBOT>	 You are primarily interested in mobile?
[17:35:54] <Dagy>	 Thats the experince i have
[17:36:08] <FlorianSW>	 Dagy: For MediaWiki you can start with https://www.mediawiki.org/wiki/How_to_become_a_MediaWiki_hacker
[17:36:32] <FlorianSW>	 iirc we have a little starting guide for mobilefrontend, too... somewhere :)
[17:37:07] <Dagy>	 i just want to contribute in anything that has to do with codin. you suggest for me
[17:37:46] <Dagy>	 i have read that article, and i ended up landing here
[17:38:15] <FlorianSW>	 Dagy: you can find our open tasks for mobilefrontend here: https://phabricator.wikimedia.org/maniphest/query/eQ1qrKjCKp11/ (but you should really start with setting up MediaWiki and MobileFrontend :))
[17:38:35] <FlorianSW>	 some useful (but maybe outdated) information about MobileFrontend development you can find here: https://github.com/wikimedia/mediawiki-extensions-MobileFrontend#development
[17:38:49] <FlorianSW>	 Dagy: btw: Do you already have a gerrit account to submit code? :)
[17:39:26] <FlorianSW>	 GEOFBOT: btw: Hi :) How are you?
[17:39:28] <Dagy>	 No i dont have
[17:39:41] <GEOFBOT>	 Hi FlorianSW, doing well, you?
[17:39:55] <FlorianSW>	 GEOFBOT: great, me to :)
[17:40:36] <GEOFBOT>	 Dagy, https://www.mediawiki.org/wiki/Gerrit/Tutorial
[17:40:53] <FlorianSW>	 Dagy: a good start would be https://www.mediawiki.org/wiki/Gerrit/Getting_started
[17:41:02] <FlorianSW>	 or the link from GEOFBOT :D
[17:41:10] <Dagy>	 okay thanks
[17:41:14] <GEOFBOT>	 I think the second link is more straightforward
[17:41:50] <Dagy>	 guys can you please give me you mails so that when i am stuck i can let you guys know.Please
[17:42:09] <GEOFBOT>	 You can just ask on IRC, someone will answer :)
[17:42:19] <Dagy>	 i want to do this so bad
[17:42:30] <FlorianSW>	 Dagy: GEOFBOT is right, or you can use our mailing list: mobile-l@lists.wikimedia.org :)
[17:42:47] <GEOFBOT>	 Where did all these new people show up from?  don't remember having this much volume
[17:43:01] <Dagy>	 hahahahaha
[17:43:06] <Dagy>	 Passion 
[17:43:11] <FlorianSW>	 :P
[17:43:15] <Dagy>	 :)
[17:43:37] <GEOFBOT>	 well, good luck on your adventure
[17:43:51] <FlorianSW>	 hmm, does this work? [[mediawiki:Gerrit/Tutorial]]
[17:43:53] <FlorianSW>	 @link
[17:43:53] <wm-bot2>	 http://enwp.org/mediawiki:Gerrit/Tutorial
[17:44:43] <GEOFBOT>	 oooh
[17:45:27] <wm-bot2>	 This user is now online in #wikimedia-dev. I'll let you know when they show some activity (talk, etc.)
[17:45:27] <FlorianSW>	 @notify GEOFBOT :P
[17:45:45] * FlorianSW should use wm-bot much more :D
[17:45:49] <GEOFBOT>	 :O
[17:46:54] <jdlrobson2>	 Dagy: hey :)
[17:47:09] <Dagy>	 Hellows
[17:47:30] <GEOFBOT>	 wm-bot doesn't seem to understand gratitude :o
[17:48:11] <jdlrobson2>	 Dagy: so you want to help with wikimedia web experience or the app experience?
[17:49:50] <Dagy>	 I am intrested with its app experince but i wouldnt mind to help with its web experince
[17:50:47] <jdlrobson2>	 FlorianSW: can you think of any web bugs that might be easy enough for Dagy to get started with to start understanding the mobile experience?
[17:51:10] <jdlrobson2>	 obvious thing is for him to get setup with vagrant and a local mediawiki mobile instance first
[17:51:38] <mhurd>	 Dagy: regarding apps, do you have Android or iOS interest?
[17:52:12] <FlorianSW>	 jdlrobson2: maybe [[phabricator:T101313]]
[17:52:13] <FlorianSW>	 @link
[17:52:13] <wm-bot2>	 http://enwp.org/phabricator:T101313
[17:52:16] <FlorianSW>	 ??
[17:52:46] <Dagy>	 I developed in house Android Apps
[17:53:04] <Dagy>	 so thats the link i should follow?
[17:54:10] <jdlrobson2>	 Dagy: https://phabricator.wikimedia.org/T101035 would be a great easy bug to work on
[17:54:22] <jdlrobson2>	 you'll need to get a mediawiki instance with Thanks and MobileFrontend up and running
[17:57:00] <Dagy>	 Thank you so much... Am not gonna sleep until i setup up my environment and get ready to work. 
[17:57:23] <mhurd>	 Dagy: a mirror of the android wikipedia app is here: https://github.com/wikimedia/apps-android-wikipedia
[17:57:30] <GEOFBOT>	 Hmm, just wondering what is your timezone?
[17:58:08] <mhurd>	 Dagy: dbrant|food, bearND, mdholloway|lunch all work on it as well and should be able to help with setup questions
[17:59:38] <bearND>	 Dagy: Hey welcome! The Readme on https://github.com/wikimedia/apps-android-wikipedia has set up instructions on how to get started developing the Android app, including the Gerrit set up, etc.
[18:00:07] <bearND>	 We push code to Gerrit but that gets mirrored to Github in case you were wondering.
[18:00:28] <bearND>	 Dagy: Sorry, I gotta run to a meeting but will be back later if you have any questions
[18:00:38] <Dagy>	 Okay
[18:01:18] <GEOFBOT>	 Dagy, what's your timezone?
[18:03:02] <Dagy>	 hahahaha
[18:03:19] <Dagy>	 UTC + 2:00
[18:03:22] <FlorianSW>	 jdlrobson2: that looks like a lot of things Gather loads on every page view: https://github.com/wikimedia/mediawiki-extensions-Gather/blob/8381f357775115a3ff02d89b803ae640c381a04b/extension.json#L368-L370 :/ Is it really needed? :)
[18:03:50] <Dagy>	 Central Africa Time ZOne
[18:04:05] <FlorianSW>	 Dagy: me too ;)
[18:04:14] <jdlrobson2>	 FlorianSW: ?
[18:05:16] <FlorianSW>	 jdlrobson2: ext.gather.watchstar, which depends on ext.gather.api, ext.gather.routes, ext.gather.collection.base, which again depends on 				"ext.gather.logging", 				"mobile.contentOverlays", 				"mobile.toast", 				"ext.gather.toastpanel", 				"ext.gather.api", 				"mediawiki.util"
[18:05:16] <GEOFBOT>	 cool, I'm UTC-4 I think? because of daylight savings
[18:05:19] <FlorianSW>	 and so on :/
[18:05:31] <jdlrobson2>	 it loads at the bottom of the page
[18:05:38] <jdlrobson2>	 and there's a patch to shrink it
[18:05:54] <jdlrobson2>	 FlorianSW: https://gerrit.wikimedia.org/r/212585
[18:06:49] <FlorianSW>	 jdlrobson2: ahh! Again you're faster :P
[18:18:13] <jdlrobson2>	 FlorianSW: did you see my comment about making a wrapper for MobileContext::singleton()->getMFConfig() ?
[18:18:24] <jdlrobson2>	 e.g. $config = MobileContext::config()
[18:18:39] <jdlrobson2>	 it would be a lot cleaner than MobileContext::singleton()->getMFConfig()
[18:18:53] <FlorianSW>	 jdlrobson2: yeah, read it, and answered :)
[18:19:43] <jdlrobson2>	 well we could call it whatever we want :)
[18:19:51] <jdlrobson2>	 FlorianSW: did you see krinkle comment
[18:20:23] <FlorianSW>	 jdlrobson2: reading it and trying to understand the context, maybe i should turn of the movie in the background to get what he want's to say :D
[18:20:49] <jdlrobson2>	 haha
[18:20:51] <jdlrobson2>	 what movie?
[18:21:43] <jdlrobson2>	 fyi FlorianSW when bmansurov gets https://phabricator.wikimedia.org/T98518 fixed i'll look to merge talk
[18:21:49] <FlorianSW>	 jdlrobson2: because of missing alternatives (at the moment) i'm watching the beginning of Eureka, a great series from 200X :D
[18:22:09] <FlorianSW>	 jdlrobson2: ah, great! :D
[18:22:35] <FlorianSW>	 jdlrobson2: yesterday i watched the great midseason-final of Person of Interest
[18:23:18] <FlorianSW>	 jdlrobson2: ok, i give up, i don't understand what Krinkle means :( Where i want to set a config option? And how is that related to a second getter function? :(
[18:24:12] * Krinkle is here to possibly enlighten you  (sorry, was in this channel earlier for CI, hadn't left yet)
[18:24:38] <Krinkle>	 Aye, watched it earlier as well. Loooving PoI this season. Can't wait!
[18:24:46] * FlorianSW would welcome Krinkle's light :D
[18:24:52] <Krinkle>	 FlorianSW: link?
[18:24:58] <FlorianSW>	 https://gerrit.wikimedia.org/r/#/c/210649/
[18:25:39] <Krinkle>	 FlorianSW: I'm referring to the two cases where you left comments for todo
[18:26:29] <FlorianSW>	 Krinkle: ok, the change is old, let me look into it :) (Do you still watch Season 4, or do you have S05 already in the US?)
[18:26:38] <Krinkle>	 As general tech-debt, that code should not be setting globals. So instead of documenting it as todo for Config::set (which isn't going to happen in that way), it should probably use something else entirely. Nothing complicated.
[18:26:43] <coreyfloyd>	 kristenlans: dbrant|food added ether pad to Phab ticket
[18:26:47] <Krinkle>	 You'd cache it in a static $var in the function, or in a class member.
[18:27:02] <kristenlans>	 coreyfloyd: dbrant|food phabuluous
[18:27:05] <dbrant|food>	 coreyfloyd: thx!
[18:27:32] <FlorianSW>	 Krinkle: ah, ok, now i see, what you mean :)
[18:28:20] <Krinkle>	 config should be considered immutable. There is local setttings, and then there is setup/onsetup/extensionfunctions hooks to pre-process, but beyond that, it should not change. Gives a wrong reflection of the config, causes interoperability issues too. 
[18:28:40] <Krinkle>	 I'm sure it works fine here, but simply by being there, it's a risk and a confusion.
[18:28:53] <Krinkle>	 Anyway no -1, not for this patch :)
[18:29:22] <Krinkle>	 all green ahead on that hygiene patch as far as I'm concerned.
[18:29:25] <FlorianSW>	 Krinkle: i think i will change the inline-code-comments to reflect your comment and do that in a later change :)
[18:40:27] <joakino>	 i'm about to leave folks
[18:41:14] <joakino>	 jdlrobson2: i've put the related pages patch into code review, it is a bit big, i'm sorry, but there's several moving parts
[18:41:46] <joakino>	 jdlrobson2: i want to wrap that one up to show it to kaity and JK before the next train
[18:42:46] <joakino>	 jdlrobson2: i'm going to get back on to the reviewin' queue tomorrow. ping me for anything you need.
[18:42:51] <jdlrobson2>	 joakino: np
[18:42:55] <joakino>	 have a nice day everybody
[18:42:57] <jdlrobson2>	 all is good
[18:47:28] <Krinkle>	 jdlrobson2: No need to duplicate all the karma code. Adding the relevant Jenkins job will install MediaWiki with the relevant extensions and run karma for you already.
[19:09:58] <jdlrobson2>	 Krinkle: oh really? how do i invoke that in Gather from my own Gruntfile (still a bit new to Grunt)
[19:10:02] <jdlrobson2>	 (to run locally)
[19:10:13] <Krinkle>	 jdlrobson2: cd mediawiki; grunt qunit;
[19:10:37] <jdlrobson2>	 Krinkle: ah.. but we have to run our unit tests against the mobile site not the desktop..
[19:10:40] <jdlrobson2>	 does it have support for that?
[19:11:35] <Krinkle>	 define 'the mobile site'
[19:11:36] <jdlrobson2>	 useformat=mobile (targets=mobile)
[19:11:36] <Krinkle>	 Hm.
[19:11:37] <Krinkle>	 useformat shouldn't matter since tests need to mock anything they need on the page
[19:11:37] <Krinkle>	 target.. will need to see.
[19:11:55] <jdlrobson2>	 yeh useformat isn't the issue - it's the fact that some RL modules are mobile only
[19:12:01] <jdlrobson2>	 so won't get run in desktop qunit runner
[19:12:05] <Krinkle>	 It's only about loading the code, so we can just override it in test mode to alow loading any module
[19:51:59] <kristenlans>	 bgerstle, do you want in on the kanban chat on Monday or are you planning to go the larger tech talk?
[19:52:28] <jdlrobson2>	 dr0ptp4kt: ping
[19:55:11] <dbrant>	 kristenlans: count me in on that, too
[20:02:09] <dbrant>	 kristenlans: having some hangout issues...
[20:04:19] <kristenlans>	 bgerstle: coreyfloyd mhurd u retro?
[20:05:52] <dbrant>	 kristenlans: still can't seem to join... feel free to start w/o me
[20:07:41] <kristenlans>	 dbrant gonna try to invite you
[20:11:22] <dbrant>	 kristenlans: I'm seeing the invite call, but nothing happens when I try to join..
[20:12:47] <bgerstle>	 dbrant: try restarting chrome and/or your machine?
[20:20:20] <mhurd>	 vibha: r31 too far away for ya? ;)
[20:30:03] <vibha>	 mhurd: please don't freak out but I might have some mild conjunctivitis. Trying to stay away from people and closed rooms. It's not contagious.
[20:32:17] <Krinkle>	 jdlrobson: fyi, I'll have a patch making target= transparent in core tomorrow.
[20:32:46] <jdlrobson>	 Krinkle: what will that to?
[20:33:22] <Krinkle>	 jdlrobson: it'll keep the assertion that targets are validated in phpunit to be well balanced. But within the test suite you can load any module, not filtered down by target.
[20:33:44] <Krinkle>	 So the unit tests registered in the onResourceLoaderTestModule hook will just run on Special:JavaScriptTest/qunit/plain, and via grunt.
[20:33:53] <Krinkle>	 no useformat= or target= parameters.
[20:34:44] <Krinkle>	 Most if it is there already, just 2 lines of code block it from working.
[20:35:19] <jdlrobson>	 oohh nice.
[20:35:42] <jdlrobson>	 Krinkle: <clap clap> :)
[20:35:53] <jdlrobson>	 i guess that will make the tests run a little slower though?
[20:36:05] <jdlrobson>	 (not complaining just saying)
[20:47:55] <kaity>	 kristenlans: hey I have a request :)
[20:48:19] <kaity>	 kristenlans: can you help us remember to discuss if features should go into stable or beta in planning meetings?
[21:00:34] <jdlrobso_>	 JonKatz: apps have my room
[21:06:05] <Krinkle>	 jdlrobso_: Nah, not slower. It will just include all modules you want to test.
[21:06:24] <Krinkle>	 I assume right now you run the test twice?
[21:06:48] <Krinkle>	 That runs some stuff twice, and double init. Which makes it a bit slower.
[21:25:33] <vibha>	 mhurd: shall we meet at my desk?
[21:25:50] <mhurd>	 vibha: ya!
[21:26:56] <vibha>	 ok ready when you are
[21:30:54] <JonKatz>	 jdlrobson http://www.wikiwand.com/en/Tom_Selleck?in_iframe=1#/Commercials
[21:46:42] <jdlrobson>	 bmansurov: kaldari let's get https://gerrit.wikimedia.org/r/214113 swat deployed today
[21:47:08] <jdlrobson>	 http://en.m.wikipedia.org/wiki/Carrier-based_aircraft < made it on to enwiki
[21:47:31] <bmansurov>	 ok i'll add it to the wiki page
[21:54:51] <jdlrobson>	 thanks bmansurov
[22:06:34] <bmansurov>	 kaldari: i see you already added it
[22:06:37] <bmansurov>	 thanks
[22:17:46] <kaldari>	 jdlrobson: https://en.m.wikipedia.org/w/api.php?action=query&prop=pageimages%7Cinfo&piprop=thumbnail&pithumbsize=80&format=json&generator=watchlistraw&gwrnamespace=0&gwrlimit=51&continue=-%7C%7C&gwrcontinue=0%7CAlan_LeQuire
[22:57:57] <mhurd>	 JonKatz: hey you around?
[23:11:01] <kristenlans>	 kaity I saw your request earlier re: stable and beta. And the answer is "yes" :-)
[23:11:43] <kaity>	 kristenlans: thank you!
[23:13:28] <kristenlans>	 kaity jdlrobson has also been historically very good about making sure that is clear. So between me reminding us to indicate what goes where, jdlrobson on the team, and the meeting we're having next Friday to talk through some of this, I am hopeful that much confusion will be alleviated.
[23:14:39] <kaity>	 kristenlans: thanks!
[23:14:57] <kaity>	 kristenlans: yes its good that we're taking a look at this