[09:44:18] <joakino>	 morning
[09:50:30] <phuedx>	 hey joakino 
[09:50:39] <joakino>	 morning
[09:50:43] <phuedx>	 joakino: don't know if you saw this on the twitters: https://aerotwist.com/blog/the-cost-of-frameworks/
[09:52:12] <joakino>	 i did not, looking
[09:57:56] <joakino>	 interesting article phuedx 
[09:59:07] <joakino>	 not sure about the stats data, but definitely interesting
[09:59:28] <joakino>	 i must also say that doing vanilla for mobile devices is kind of a nightmare
[09:59:43] <joakino>	 it's like IE wars again
[10:00:15] <phuedx>	 ^^ that
[10:00:28] <phuedx>	 everything's a tradeoff
[15:30:17] <mdholloway>	 dbrant: hey!  do you have any idea convert-icons.py would be generating smaller-than-expected pngs on my machine(s)?
[15:30:57] <mdholloway>	 dbrant: i made a light version of kaity's icon for the cab menu, but getting the same size irregularity (as expected) when i try to crush it.  
[15:31:43] <mdholloway>	 dbrant: *any idea *why*, that is
[15:32:07] <dbrant>	 mdholloway: is this for all images, or just this one?
[15:35:08] <mdholloway>	 dbrant: lemme do a quick test.  last time i ran the script was around the time DanReyLop was working on his first patch, a month or two ago, and it worked fine
[15:36:29] <mdholloway>	 dbrant: looks like it
[15:36:36] <mdholloway>	 dbrant: looks like it's doing the same for all images
[15:40:10] <mdholloway>	 dbrant: the perplexing thing is that i'm running the same script, from the same repo, on the same files, with the same software dependencies, updated with homebrew, on the same os... and yet it works for both you and bernd and fails for me, in the same way, on two separate laptops.  
[15:40:29] <mdholloway>	 dbrant: i guess i could just try it in an ubuntu vm, but this is really kind of bothering me.
[15:40:45] <dbrant>	 mdholloway: looking at the history of the script, our command line to rsvg_convert hasn't changed in forever, so I doubt it's the script's fault
[15:41:08] <mdholloway>	 dbrant: yeah, probably not
[15:55:01] <mbinder>	 dr0ptp4kt_: RIP my inbox. Did you mean to decline a bunch of apps meetings?
[16:39:29] <mdholloway>	 dbrant: niedzielski: bearND: looks like we've got a little regression
[16:39:41] <niedzielski>	 mdholloway: ?
[16:40:31] <mdholloway>	 when i highlight text in a genymotion 4.1.1 emulator, the cab menu background is transparent
[16:41:05] <niedzielski>	 mdholloway: any idea if that's an emulator gl issue or something that manifests on a physical device?
[16:41:12] <mdholloway>	 niedzielski: not sure yet
[16:41:30] <mdholloway>	 trying on my 4.4 phone now
[16:41:56] <mdholloway>	 at the top of an article it overlays the main toolbar and looks really messed up
[16:42:10] <niedzielski>	 mdholloway: i do not have a 4.1.x physical device
[16:43:08] <mdholloway>	 niedzielski: me neither
[16:43:22] <mdholloway>	 niedzielski: looks ok on my physical 4.4 phoen
[16:43:23] <mdholloway>	 phone
[16:44:28] <dbrant>	 mdholloway: niedzielski: looks fine on my 4.1.2 tablet
[16:44:55] <mdholloway>	 niedzielski: dbrant: hmm, might be a genymotion thing.  let's hope. 
[16:48:18] <niedzielski>	 mdholloway: i'm not seeing anything on the api 15 emulator or api 23 device. i'll keep my eyes peeled tho
[16:48:51] <mdholloway>	 niedzielski: cool, thanks.
[16:49:03] <niedzielski>	 er aosp emulator that is
[16:49:55] <mdholloway>	 i have lately noticed genymotion emulators being flaky around ActionMode stuff, so maybe this is just another instance of that
[16:50:24] <mdholloway>	 i think i might be leaning back toward preferring aosp emulators
[16:58:01] <phuedx>	 going to go feed the kids
[16:58:06] <phuedx>	 will be back for standup
[17:15:32] <dr0ptp4kt_>	 mbinder: i did that to get you to make a funny comment like that. yes, i mean to decline those. sorry for the calendar change email spam
[17:15:51] <dr0ptp4kt_>	 mbinder: i gotta knuckle down on the webz work
[17:27:28] <phuedx>	 backsies
[18:01:08] <phuedx>	 jhobs: jeff "i like my takeovers hostile" hobson
[18:01:19] <jhobs>	 looool
[18:01:26] * phuedx runs to a t-shirt site
[18:01:37] * jhobs follows you
[18:02:02] <jhobs>	 incidentally i was (re-)watching die hard last night
[18:02:12] <jhobs>	 so that may have influenced that comment some ;)
[20:02:27] <jdlrobson>	 jhobs: can you do a Gather release (and I merge?)
[20:02:30] <jdlrobson>	 seems we're a little behind
[20:02:42] <jdlrobson>	 i'm testing the dev branch as we speak to check it's deployable
[20:03:53] <jdlrobson>	 we need one for QuickSurveys too..
[20:04:54] <jdlrobson>	 jhobs: can confirm it's releasable.
[21:13:19] <jdlrobson>	 jhobs: ping me when back :)
[21:14:50] <jhobs>	 jdlrobson: hey sorry about that, irccloud was late with the notifications for some reason
[21:15:28] <jhobs>	 I can do the releases, sure. I assume you want to SWAT them both tonight?
[21:16:09] <jhobs>	 I won't be around during the window
[21:27:33] <jdlrobson>	 jhobs: didnt want to swat the releases
[21:27:42] <jdlrobson>	 just wanted to make sure they would ride the next train :)
[21:28:00] <jhobs>	 jdlrobson: I thought we hit code freeze already? Or do you mean in January?
[21:28:02] <jdlrobson>	 The Gather one is slowing down some MobileFrontend changes so is the one im keenest to merge
[21:28:26] <jdlrobson>	 i'd do it but then no one would be around to merge it jhobs  :)
[21:28:52] <jhobs>	 yeah i'm doing it now, almost done with Gather
[21:29:01] <jdlrobson>	 marxarelli: having some urgent issues in MobileFrontend with selenium - https://integration.wikimedia.org/ci/job/mwext-mw-selenium/2447/console
[21:29:07] <jdlrobson>	 complaining about no stable Firefox..
[21:29:44] <marxarelli>	 jdlrobson: ok, i'm heading to a meeting but i can help out in 30
[21:32:12] <jhobs>	 jdlrobson: patch is here, but looks like it went to the wrong branch. How do I explicitly set the branch using git review (I don't think it's just git review origin master like it is with push)? https://gerrit.wikimedia.org/r/#/c/259798/
[21:32:24] <jdlrobson>	 git review master
[21:32:46] <jdlrobson>	 you may need to change the change id
[21:33:14] <jdlrobson>	 looks good though - and good catch on the non-version bump :)
[21:33:17] <jhobs>	 oh so it is the same, ok
[21:33:53] <jdlrobson>	 brb
[21:35:57] <jhobs>	 jdlrobson: correct one: https://gerrit.wikimedia.org/r/#/c/259801/. Gonna start on the QS one now
[21:42:05] <jdlrobson>	 jhobs: on it
[21:42:45] <jdlrobson>	 jhobs: thanks dude
[21:43:28] <jhobs>	 jdlrobson: yep np. Currently just double-checking the sha for the last QS release because the 0.6.0 release wasn't named such in its commit for some reason
[21:49:01] <jhobs>	 jdlrobson: what version number are we using for QS? IMHO it should've been 1.0 before it went into production, so I'd like to play catch-up here and bump it to 1.0, but 0.7 is ok if there aren't enough substantial changes
[21:52:45] <jhobs>	 jdlrobson: https://gerrit.wikimedia.org/r/#/c/259860/
[21:56:51] <jdlrobson>	 jhobs: hmm
[21:57:03] <jdlrobson>	 I'm not sure it needs to be 1 before in production
[21:57:49] <jdlrobson>	 jhobs: we're following http://semver.org/ so i'm not sure such a major bump is warranted
[21:58:41] <jdlrobson>	 maybe leave that till tomorrow - i'm sure Baha and phuedx will have some views
[21:59:35] <jdlrobson>	 jhobs: commented on patch
[22:20:00] <dr0ptp4kt_>	 yo bd808, able to talk on video for a few minutes?
[22:20:55] <bd808>	 dr0ptp4kt_: yeah. can you PM me the link. My phone gets them otherwise
[22:21:18] <dr0ptp4kt_>	 bd808: oh right, sorry, i forgot that
[22:29:18] <niedzielski>	 bearND dbrant mdholloway: there's a number of editing tests that use the token "+\\" (literally +\). it seems like a special value. do you know where i can find out more about that? i'm getting a lot of local test failures with the message, "Caused by: badtoken: Invalid token"
[22:32:18] <bearND>	 niedzielski: i recall something special about it, i don't remember all the details about it anymore
[22:42:49] <Deskana>	 niedzielski: https://www.mediawiki.org/wiki/Manual:Edit_token
[22:43:29] <Deskana>	 niedzielski: More specifically, https://www.mediawiki.org/wiki/Manual:Edit_token#The_edit_token_suffix
[22:43:35] <Deskana>	 Seems we do it to break proxies -_-
[22:43:38] <niedzielski>	 Deskana: thanks! i will fix this by making a symbol, ANONYMOUS_TOKEN
[22:44:40] <niedzielski>	 man, these tests are flakier than a wookie with dandruff
[22:45:32] <dbrant>	 rofl
[22:48:16] <Deskana>	 We should add a few more characters in to make it even better.
[22:48:28] <Deskana>	 +\!33th4x0r
[22:48:39] <Deskana>	 It's starting to look like a "strong password"
[23:50:38] <jdlrobson>	 marxarelli: no luck.. https://gerrit.wikimedia.org/r/#/c/259883/ :/
[23:51:30] <marxarelli>	 jdlrobson: are you sure that's not a true positive? or a generally flake test?
[23:51:59] <marxarelli>	 there's only one failure as opposed to the across-the-board failure we were seeing before
[23:52:43] <jdlrobson>	 marxarelli: mm i guess it could be broken by a VE change
[23:53:13] <jdlrobson>	 they have been playing around with switching experiences. I don't have VE enabled and I guess VE doesn't run our tests on a per commit basis..
[23:53:15] <jdlrobson>	 let me dig
[23:53:25] <jdlrobson>	 https://integration.wikimedia.org/ci/job/browsertests-MobileFrontend-en.m.wikipedia.beta.wmflabs.org-linux-chrome-sauce/ is still passing though..
[23:54:45] <jdlrobson>	 well i doubt i'm gonna get to the bottom of it in 6 mins. Will ask for a forced merge
[23:54:50] <jdlrobson>	 and suffer consequences