[05:23:38] good morning! [06:39:13] hey phedenskog [06:39:14] :) [06:39:25] hey mister :) [06:42:27] sort of here but not [06:42:33] while i get the children ready for nursery [06:59:56] ok [08:42:32] pheudx: will order the pi today so I hopefully get it to Monday so I can start to test. One question, what about the antenna, anything special I should think about? [08:58:39] phedenskog: ok [08:58:44] sec [08:58:48] i'll dig out the one i'm using [08:58:54] thanks! [08:59:37] https://www.amazon.co.uk/gp/product/B00KNPO6HA/ref=oh_aui_search_detailpage?ie=UTF8&psc=1 [09:01:26] phedenskog: i think (but don't know) that antennas vary on the number of service sets they can broadcast [09:01:32] the one i've just linked supports 8 [09:03:43] thanks I'll try that one [09:04:10] i was going to buy the official raspberry pi one to test it [09:04:31] there's likely a website /somewhere/ that lists the output of `iw list` for these devices [09:43:10] I got everything now, lets see when it arrives, I'm excited :) [09:46:37] woo! [11:38:50] niedzielski: hmm that is strange, the OG file shows the width as 360dp but I will check it out more closely first thing Monday [13:48:47] phuedx: you tend to know odd performance facts sometimes; are you aware of any performance difference between converting a string to a number vs. a boolean in JS? [13:49:50] no, i'm not aware (though i imagine the latter is a lot simpler) [13:50:01] to the jsperf cave! :D [13:50:59] phuedx: I ask regarding my latest comment on the references toggling patch if you're itching for context [13:51:27] the tradeoff of 9-10 characters vs. converting to a number instead of a boolean! [13:51:48] ah such important distinctions ;) [13:52:14] jhobs: what's the string? [13:52:20] "1" [13:52:51] (or potentially non-existant, although I don't think the codepath is hit in that scenario) [13:57:22] joakino or bmansurov you there? [13:57:52] joakino: disregard, sorry, looked at schedule [13:58:38] joakino: that is to say, it can wait until your back. ttyl [13:58:43] s/your/you're/ [13:59:36] dr0ptp4kt: yes [14:00:05] bmansurov: did i pull you from a break or are you actually already working? i don't want to bother you if you're not working! [14:00:24] dr0ptp4kt: i was about to take a break, which can wait [14:00:26] what's up? [14:00:43] dr0ptp4kt: I uploaded a screencast if you haven't seen it yet [14:00:49] bmansurov: you're quick [14:01:03] quick is an underestimation haha [14:01:07] jhobs: see my follow-on comment [14:01:14] thanks for pointing that out [14:01:19] (iirc it was raised before too) [14:02:40] phuedx: is "follow on" the british version of "follow up"? [14:03:30] phuedx: ah, excellent! Do we have a corresponding PHP method of setting attributes without values? We could then leverage HTML's natural true state attributes (such as "checked") to eliminate a few more bytes per page [14:04:37] jhobs: i've never actually checked if setattribute takes an empty string as a value before [14:05:01] bmansurov: the latter allows for repetition of the original, while the former doesn't [14:05:03] or something [14:05:30] i see [14:05:50] the "or something" is key [14:06:03] bmansurov: ok, i see it working on localhost at the end of that screen recording. able to reproduce on reading-web-stage. it can wait until after break, too [14:06:27] dr0ptp4kt: ok, it should be quick, let me do it now [14:06:28] british english maximises ambiguity and characters [14:06:38] original english you mean [14:06:50] un-evolved haha [14:07:16] un-devolved, maybe [14:07:18] dr0ptp4kt: oh wait, staging may not have been updated [14:07:24] facepalm [14:07:41] hrrrm [14:07:44] bmansurov: AH! i should have checked the js source with debug [14:07:52] bmansurov: no, my bad [14:07:58] bmansurov: i'll go look while you go look [14:08:09] i checked the crontab logs this morning and saw the vagrant git-update command was being invoked [14:08:58] dr0ptp4kt: ok, i've updated staging [14:09:16] dr0ptp4kt: please try again, and i'll also try later if it doesn't work for you [14:09:16] bmansurov: again, that was quick [14:09:19] bmansurov: i'll go check [14:09:29] phuedx: that's weird, popups was behind master [14:11:25] bmansurov_break: yeah, now dwelledButAbandoned working when hovercards disabled [14:14:43] dbrant: nope. :( [14:35:46] ^ intermittent connectivity as we reach the midpoint of our journey south/south-west [15:01:43] dbrant: o/ hey i wasn't surei f you wanted to join us to talk about JPLS but we're about to start [15:57:46] dr0ptp4kt: do you still need me? [15:58:16] joakino: always. but on that specific matter, no, we're good. the patch just needed to be pushed to reading-web-staging [16:00:48] nice [16:39:27] mdholloway bearND: do the content service tests always finish with "85 passing, 1 pending" for you two? is the "1 pending" expected? [16:39:58] niedzielski: one of the tests has a "skip" in there [16:40:23] bearND: ah ok. is that something that should be changed? [16:41:01] niedzielski: not sure at this point who made the change. I'm in a meeting right now. so i can't check myself [16:41:24] bearND: ok cool. no rush :) [16:47:32] niedzielski: bearND: that's an artifact of the service template upgrade. expected, i think (though i'm not sure why a skipped test is included.). /cc mobrovac [16:48:12] niedzielski: i've just checked prod mcs [16:48:21] niedzielski: it seems your fix has never even been deployed [16:48:45] mobrovac: ah nuts :/ [16:48:52] git log in prod gives me eeb2289 Fix tests and release v0.2.0 as the latest commit [17:01:34] jdlrobson: standup [17:11:34] * mdholloway facepalms [17:12:34] niedzielski: bearND: mobrovac: a scheduled services deploy window just started. any objection to my doing a deploy now? i haven't actually deployed with scap3 yet so this might be a good opportunity [17:13:41] mdholloway: i don't know much about deployment at the moment so no objection or endorsement from me :) [17:18:00] mdholloway: sure, go ahead [17:18:13] niedzielski: mobrovac: bearND: cool, doing now [17:20:32] ugh, coffee shop wifi [17:26:57] mdholloway: ok, cool. so you just want to try out the new deploy mechanism? because the only new thing is a change for the tests but not production facing changes [17:27:46] mdholloway: anyways, I updated the instructions on Wednesday to account more for the changes of deploying with scap3 [17:29:31] bearND: seems the latest deploy commit wasn't actually deployed, so i'm going to deploy that and the latest stuff (and try out the new system) [17:30:33] mdholloway: really? hmm [17:31:42] mdholloway: then something must be missing in the instructions. Please update any missing steps [17:32:07] sigh [17:32:50] in my vm the tests pass with 'npm test' but running in docker it's failing on 'should get robots.txt' [17:33:21] anyway maybe i'll deal with that later and just see about deploying what's already in the deploy reop [17:33:24] *repo [17:36:06] mdholloway: ok, thanks. I'll have to go unfortunatley [18:00:07] bearND|afk: tin had the latest but the scb machines didn't. not sure what went wrong. anyway, scap worked as expected this time [19:38:43] mdholloway: thanks for the check and the deployment. I think we should add a deployment verification step to ensure the correct version is deployed. I created T136355 [19:38:46] T136355: Ensure right version of MCS is deployed - https://phabricator.wikimedia.org/T136355 [19:39:54] mobrovac: i'm wondering if there is a scap command for that ^ [19:40:01] bearND: i usually just check 'git log' in the /srv/deployment/mobileapps/deploy on each scb machine before restarting [19:40:48] mdholloway: scap does the restart of services [20:03:39] HaeB: would you please review https://phabricator.wikimedia.org/T134778#2331751 for hovercards? [20:05:18] dr0ptp4kt: ok will try to take a look tonight [20:05:36] jhobs: wikidata description patch ready for review yet? [20:41:33] HaeB: thx [23:02:12] jhobs: did you finish for the day? I have bandwidth to work on the wikidata description patch and not sure whether i should