[06:28:18] (PS1) Awight: Don't overwrite donor language with fallback value [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/201132 (https://phabricator.wikimedia.org/T94506) [06:33:08] (PS1) Awight: New testing wrapper to circumvent object access [core] (fundraising/REL1_23) - https://gerrit.wikimedia.org/r/201133 [06:33:10] (PS1) Awight: Fix TestingAccessWrapper::__call [core] (fundraising/REL1_23) - https://gerrit.wikimedia.org/r/201134 [06:34:02] (CR) jenkins-bot: [V: -1] New testing wrapper to circumvent object access [core] (fundraising/REL1_23) - https://gerrit.wikimedia.org/r/201133 (owner: Awight) [06:34:25] (CR) jenkins-bot: [V: -1] Fix TestingAccessWrapper::__call [core] (fundraising/REL1_23) - https://gerrit.wikimedia.org/r/201134 (owner: Awight) [07:17:58] (PS2) Awight: Don't overwrite donor language with fallback value [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/201132 (https://phabricator.wikimedia.org/T94506) [07:24:46] (PS1) Awight: WIP Cleanup, notes, explicit fields, rename, attempt AccessWrapper [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/201140 [07:25:12] (CR) jenkins-bot: [V: -1] WIP Cleanup, notes, explicit fields, rename, attempt AccessWrapper [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/201140 (owner: Awight) [07:46:33] Wikimedia-Fundraising, Fundraising Tech Backlog: Backfill donor language information from logs? - https://phabricator.wikimedia.org/T94670#1169480 (awight) NEW [17:00:47] K4-713: ejegg: https://gerrit.wikimedia.org/r/#/c/201132/ with your breakfast? [17:01:00] I have breakfast? [17:01:04] K4-713: :p [17:01:17] K4-713: Do you know of any historical reason we would want to unstage language? [17:01:29] * K4-713 cries [17:01:30] Yes. [17:01:34] noooo [17:01:38] what is it? [17:01:51] You know how GC has mutant codes for Chinese? [17:01:56] >_< [17:01:57] <_< [17:01:58] >_> [17:02:27] yes, but that seems like a similar problem to the one I'm fixing, and the solution is to never unstage language... [17:02:47] Well, it would be if unstage worked like it was supposed to. [17:03:09] You know: Making things available in a normal way, instead of clobbering everything. [17:03:28] You could find out if they switched stuff up on you in a consistent way. [17:04:01] I forget, do we unstage when they come back to decide which language TY page they get? [17:04:19] That shouldn't be a thing. [17:04:23] But it might be. [17:04:30] If it is, it's not right. [17:04:34] I think that's all in the ReturnURL [17:04:44] wait: GC doesn't redirect. [17:04:51] ...for cc. [17:04:57] They do for other things. [17:05:02] But. [17:05:03] thx [17:05:33] I think a lot of people's conflicting opinions have revisions for them, in prod, on the returnURL stuff. It's probably a mess. [17:06:26] Sounds like we should add that logic to the refactor page (if you don't want to tackle it right now). [17:06:41] I see that the return URL is built using the normalized language field. [17:06:59] Is it, by any chance, unbuilt and reconstructed whenever people try to use it? [17:07:06] wat? [17:07:35] I feel like I had to stop somebody doing something overly complicated there. [17:08:29] * awight strolls off to try getting a returnurl on prod [17:08:32] Also... one would hope that every place that generates either a returnURL that ends up being the TY page, uses the same stuff. [17:08:37] Er. [17:08:42] That was almost english. [17:09:23] So, it's the... issue in which cc's returnURL is the cc resultswitcher, and most (if not all) of the other ones just go straight to TY. [17:09:39] Hopefully cc generates the redirect to TY the same way that the others do. [17:09:53] Sensemaking? [17:10:13] Well, I don't imagine that the returnurl parameters are ever unstaged, they are probably handled as request params. But I'm trying to verify that. [17:10:39] It would be weird if they had a chance to, anywhere other than a gateway.body page. [17:10:53] You just burned a lot of that to the ground, though, right? [17:11:10] Or I made it really explicit, like "addRequestData" [17:11:49] I'm pretty sure the thing that was saving us up until Mar 12 was that the language unstaging function didn't actually work. [17:12:08] :S [17:12:49] I still have a question. Why just Catalan? [17:13:03] you mean, didn't even work to unmutate the GC language code? [17:13:22] K4-713: I think it was actually affecting anything not supported by GC. Including zh. [17:13:34] Aaaaah. [17:13:45] ejegg: yeah. I'm trying to make a convincing case in code, one moment please [17:14:03] oho, I get you [17:14:18] Okay, here's the old function... http://git.wikimedia.org/blob/mediawiki%2Fextensions%2FDonationInterface.git/0a8731677b1834d62f978d090dccc0db4a5851bc/globalcollect_gateway%2Fglobalcollect.adapter.php#L1862 [17:14:25] crap, no hold on [17:15:19] There it is: http://git.wikimedia.org/blob/mediawiki%2Fextensions%2FDonationInterface.git/d0357f56f5ea37cea578ef19f15213a41e15742a/globalcollect_gateway%2Fglobalcollect.adapter.php#L1862 [17:16:06] yeah: that never worked. [17:16:27] So we should be fine removing the unstaging thing, cos it's always been unused. [17:16:39] because it only affected staged_data? [17:16:42] yep [17:16:46] k [17:16:54] correctly unstaging would have written to normalized [17:17:01] right [17:17:09] You know what I want? [17:17:16] That thing I keep saying I want to do in DI. [17:17:41] hmm. /me shuffles nervously. All the unstaging things write to $staged. I could be wrong here. [17:17:44] So we keep all the alternative versions for all the data and where they came from... and have general rules for what wins with overrides. [17:17:51] K4-713: that. [17:18:01] And then we can sanity check the hell out of everything. [17:18:09] I think we were on a roll with the "Transformer" chat last year. [17:18:14] Oh My Goodness, Can You Imagine The Logs. [17:18:29] diggable! [17:18:31] Debug our whole lives with that stuff. [17:18:45] I wish, right now, that we had a dump of all the variables for each request. sigh [17:19:03] info? [17:19:08] ...debug. [17:19:13] * K4-713 shrugs [17:19:17] Just don't do it forever. [17:21:30] pizzzacat, atgo: Scrum of scrums seems to be a double-sized issue this week - I can't make it to dash check in. Sorry for the late notice! [17:21:38] orly [17:21:47] oh yeah it's the retro thing.. they mentioned that on the last one [17:21:52] Have fun in there! Thanks for representing. [17:21:55] also, anyone want me to poke other teams about things? [17:22:04] enjoy :) [17:22:14] there was a thing.. that i can't remember.. that was brought up last week? [17:22:21] awight, I see you got a bit of help on vagrant [17:22:38] ejegg atgo do you want to skip the dash meeting this week then? I am personally not touching the dash mostly this sprint unless there are bug fixes/etc [17:22:44] atgo: Guess what I got in the mail last night? [17:22:55] ejegg: yes, that's going really well. [17:23:13] pizzzacat: only thing I'm eager to talk about in dash is the filter component, but I can wait on that [17:23:19] a PONY?! [17:23:33] K4-713 ^ [17:23:44] I wish! [17:23:46] ejegg pizzzacat yeah we can skip this one and reconvene next week [17:23:51] that would be a sad mailed-pony [17:23:58] All flat. [17:24:04] K4-713: brb [17:24:06] unless it came by pony express [17:24:22] ejegg: :D [17:24:24] ejegg yes! I haven't forgotten, sorry about that.. want to video chat on that at some point? [17:24:31] All flat and bent, even though it clearly says "do not bend" all over it. [17:24:42] pizzzacat: sure, whenever works [17:24:45] btw how is there a bee in my house? I don't have any windows open [17:24:56] Anyway. No. But close. [17:24:57] It's a jury duty summons. [17:25:04] it's huge, like…how did you get in here dude [17:25:19] K4-713: I think I understand what I changed about unstaging. You might want to do a sanity postmortem. In GatewayAdapter::addResponseData, we end up copying unstaged things back to the DonationData normalized layer. Previously, we would only unstage into... "staged_data". [17:25:19] is it carrying a lockpick set? [17:25:23] K4-713: I think that's the final stage of Dinosaur Flue [17:25:26] *Flu [17:25:30] Oh crikey, a bee! [17:25:31] Jury summons [17:25:32] Jury duty? [17:25:33] haha [17:25:35] yes haha [17:25:39] That's how you know it's almost over with. [17:25:43] hahahaha [17:25:48] Well, that's good news. [17:25:48] oh no i missed something :o [17:25:48] I wish I would get mine already [17:27:43] atgo we were just talking about how jury duty is the final stage of Dinosaur Flu [17:28:58] haha [17:29:05] well CC just got hers, too [17:29:06] for next week [17:29:11] that's 4 fundraisers [17:29:12] whaaaaat [17:29:23] Protect yourself, or you too will catch The Summons. [17:29:46] also ejegg pizzzacat dash checkin is during metrics next week [17:29:55] could we do friday instead? [17:29:58] I mean I'm going to court but it's to hand the cop who pulled me over on my bike a copy of Nonviolent Communication [17:30:04] haha [17:30:12] i dunno man... was flipping him the bird nonviolent? [17:30:14] :P [17:30:23] Just "emphatic". [17:30:32] nope! :) [17:30:38] That's insane. [17:30:47] pizzzacat: LOL [17:31:19] well he was driving through an intersection I was crossing and laying on his horn so technically the person in power started the violence [17:31:41] A nice piece of trivia--it's technically assault if you use four-letter words on regular people, but you're legally entitled to say all kinds of fucking shit to the cops. [17:31:46] the "little lady" has to stand up to that, not be an example [17:31:48] Cos in theory they're trained to only tase you. [17:32:23] oh this guy was not trained in communication between humans, let alone cop-to-noncop communication [17:32:48] atgo I'm confused [17:32:52] I thought metrics is tomorrow [17:33:03] and checkins are wednesdays anyway? [17:33:42] Wikimedia-Fundraising-CiviCRM, Fundraising Tech Backlog: Lock timeout while adding a contribution from the web UI - https://phabricator.wikimedia.org/T94716#1170922 (awight) NEW [17:34:07] Wikimedia-Fundraising-CiviCRM, Fundraising Tech Backlog: Lock timeout while adding a contribution from the web UI - https://phabricator.wikimedia.org/T94716#1170929 (awight) @RLewis mentioned she was doing some big searches at the time... [17:34:12] but doesn't matter, Friday works for me so whatevs [17:34:18] pizzzacat: 'roids... [17:35:38] what a total bozo [17:36:17] maybe if I see him in court and hand him the book and some candy he will be my friend and stop being mean to bikers [17:36:27] April fools! [17:36:56] I've been waiting to hear this story, btw... sounds like a real popcorn-gnasher! [17:38:20] oh nah I was just repeatedly rude to a cop who impatiently didn't want to wait for me and some other cyclists to cross an intersection (he laid on his horn obnoxiously) [17:38:38] not knowing it was a cop, I flipped him off as I crossed [17:38:44] Wikimedia-Fundraising-CiviCRM, Fundraising Tech Backlog: Lock timeout while adding a contribution from the web UI - https://phabricator.wikimedia.org/T94716#1170932 (atgo) p:Triage>Normal [17:38:57] and this action apparently ruined his life [17:39:07] Wikimedia-Fundraising, Fundraising Sprint H, Fundraising Tech Backlog: Backfill donor language information from logs? - https://phabricator.wikimedia.org/T94670#1170935 (atgo) [17:40:29] he asked me 'what if it had been a gang-banger you flipped off' and I said 'I would prefer to flip you off because gang "bangers" are nice to me' [17:40:44] and other things I prob shouldn't have said [17:41:08] WOOT [17:41:11] which require that I buy him a copy of NVC because I should reread mine [17:41:15] :p [17:41:16] That is the greatest thing you could have said [17:41:26] haha [17:42:39] my favorite was when he was leaving and didn't acknowledge things were over and I said "Have a lovely day, sir, out there fightin' all that CRIME!" (in a southern accent because I was triggered) and then blew him a kiss [17:43:13] my official crime was "not stopping at a stop sign" [17:43:19] oh damn. You are a huge punk. [17:43:27] You'll need the Nolo press books... [17:44:02] haha no I'm just mean to people who are mean to me, I don't discriminate [17:44:24] I had a crazy success once, fighting a stop sign thing. I sent them a check for the bail amount, and a letter explaining how I wasn't in the mood to go to court but I would fight the ticket to the bitter end. [17:44:38] They sent me the check back uncashed... [17:45:05] Turns out, the trick was that my letter had a header block formatted exactly like their own, so they probably assumed I had a lawyer? [17:45:06] my grandma was a deputy sheriff and I knew bunches of cops and the sheriff of our town. she took us to school in a cop car and we hung out in the jail (in an interrogation room) after school while she finished her shift. [17:45:26] I have nothing against cops and was shocked at how rude he was out of the box. [17:45:43] huh weird! [17:46:32] anyway that's the story about that… I left out some other sick burns but that's the gist [17:46:54] I think of them as, ordinary people who are corrupted by the power. The banality of evil. [17:47:10] want more sick burnz! [17:47:12] yeah clearly that was the main ingredient in this interaction [17:48:50] good thing this wasn't in Ferguson and this guy's self-respect for the day had already been worn down by a dozen other punkass, mustache-twirling cyclists :p [17:52:48] oof :( [17:53:45] reverse police brutality! [17:57:43] Wikimedia-Fundraising, Fundraising Sprint H, Fundraising Tech Backlog: Backfill donor language information from logs? - https://phabricator.wikimedia.org/T94670#1170973 (atgo) p:Triage>Normal [17:59:25] pizzzacat: lost you [17:59:30] metrics is on a weird date this month [17:59:42] i spaced out from IRC for a while there [18:01:52] awight: Wait 1. Campers in my room. [18:02:01] k thx [18:02:58] atgo I have it as tomorrow and next wednesday, weird [18:03:38] Wikimedia-Fundraising, Astropay Integration, Fundraising Sprint H, Fundraising-Backlog, and 2 others: Write adapter class API definitions for New Invoice and Payment Status - https://phabricator.wikimedia.org/T90503#1171001 (atgo) [18:13:03] weird indeeeed [18:16:38] atgo: Metrics is next week, right? [18:17:15] yeah on wednesday [18:33:52] hey k4-713 - want to make a lunch run at some point before group lunch? [18:34:42] atgo: ... you have to see this. [18:36:57] ejegg pizzzacat - how about we just cancel until you're back from vacation? [18:37:05] seems silly to make a plan and then have everybody leave [18:37:13] and forget the plan while they're galavanting :P [18:37:18] sounds good atgo [18:37:44] ok, cool [18:37:57] K4-713: awight: this is relevant to a discussion we had yesterday: http://www.nsa-cloud.com/ [18:38:23] LOL--did you just set that up? [18:38:47] " » all your social interactions," [18:40:28] I wish [18:40:40] haha [18:40:41] K4-713: apropos checkin, I also wanted to mention that I've been really happy lately about the stuff I'm working on, and direction it going in (-> open source!). Hopefully that isn't too inversely proportional to helpfulness or bug fixing. [18:40:48] » For your own convenience, we already signed you up! [18:41:14] "Unfortunately, the following services have not been released for all of our customers yet:" [18:41:18] sign out. [18:42:53] :p [19:05:11] (PS2) Ejegg: Use transaction-specific URLs for Astropay [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/201048 (https://phabricator.wikimedia.org/T90503) [19:13:51] atgo: I just added stuph to the roadmap on Mediawiki, namely the UI revamp we talked about yesterday, plus some error logging, some of which has been mentioned (briefly as a TODO) in CR [19:13:53] https://www.mediawiki.org/wiki/Extension:CentralNotice/Notes/Campaign-associated_mixins_and_banner_hisotry [19:14:48] awight: ^ See the Client-side error logging section, I included a reminder about the choiceData bork TODO u spoke uv [19:18:33] I'm siiiiiiiiiiiiiiinging in the rain... [19:18:33] Ah no, wait, it's sunny [19:24:27] I'm self-taaaaaaaaaaaaaaaaaaaaaaaalking-on-IRC in a room with a window looking at the sun... [19:29:45] AndyRussG: Hi, thanks for the snow update :) [19:30:08] ;p [19:30:09] Hope you can peek out the hatch some time soon. [19:30:32] My eyes are peeking but my brain's still in the loony bin straight jacket [19:30:48] thanks tho :) [19:54:34] (PS4) Ejegg: Add skeleton of Astropay gateway [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/200766 (https://phabricator.wikimedia.org/T90501) [19:55:54] (CR) jenkins-bot: [V: -1] Add skeleton of Astropay gateway [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/200766 (https://phabricator.wikimedia.org/T90501) (owner: Ejegg) [20:04:39] awight: ejegg: any have any objections to a feature branch in CN for campaign-associated mixins? It's just that this first patch has some placeholders that will get reverted out when the real mixins are implemented, and they include i18n strings that I don't want to see people translating 8p [20:05:14] no objections here! [20:06:16] (PS3) Ejegg: Use transaction-specific URLs for Astropay [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/201048 (https://phabricator.wikimedia.org/T90503) [20:06:18] (PS5) Ejegg: Add skeleton of Astropay gateway [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/200766 (https://phabricator.wikimedia.org/T90501) [20:07:43] ejegg: fantastic, thanks! [20:13:11] blurg, still can't upload to git review because other stuff that goes in master isn't merg't yet [20:13:15] AndyRussG: awesome. Thanks for modernizing us! [20:13:33] AndyRussG: git push gerrit HEAD:refs/for/BRANCHNAME [20:13:47] Huh? [20:13:55] u can cheat out the .gitreview default by using that commandline directly [20:14:12] Mm no it's the history I'm concerned 'bout [20:14:30] e.g., git checkout -b campy_mixins; git push gerrit HEAD:refs/for/campy_mixins [20:15:02] I think that will check in the master WIP patches, so u can rebase later. [20:15:37] I want to start on top of the poof Special:BannerRandom and Special:CentralNotice: code cleanup, but I want those to get merged 2 master first [20:15:38] nvm, it sounds silly when I say it out loud [20:15:42] cool [20:16:24] ur commands look and sound legit! [20:16:45] I think I've cleaned up the client-side patch enuf for another round, whenever you get bored. [20:19:32] awight: thanks! On that patch, this is the diff I don't understand, though I admit it could be due to laziness... ;) https://gerrit.wikimedia.org/r/#/c/190239/9/tests/BannerChooserTest.php [20:22:12] That was mergefail... [20:22:24] Oh, and PS10 is pushfail [20:24:27] awight: ah cool that explaines! [20:24:35] s/e// [20:25:27] xplains [20:26:06] * awight demandes an explonation [20:26:53] are you guys thinking we'll get any more of those "review" column items out in the current sprint? [20:27:02] writing the summary [20:28:27] atgo: hmmm I was kinda thinkig of growing that column a bit instead :) [20:29:24] atgo: It's likely that I'll be able to deploy that bugfix today, T94506. [20:30:39] cool, since everythign else is in review, i assume nothing would go with it? [20:32:17] Nope, these are the only things in the queue, shouldn't have much impact: [20:32:20] + 0c198ec166d36662799257b7098a67a53a918edf Remove some donation premium language [20:32:24] + e2d726996e4c6d13e475f535548a1b66b63c8f36 Localisation updates from https://translatewiki.net. [20:36:13] donation premium language...? what? [20:37:31] (PS1) Awight: WIP Don't overwrite normalized data with unstaged [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/201318 [20:38:03] atgo: exactly. Archaic verbiage from before our time... [20:38:10] ha ok [20:39:56] K4-713: What do u think about https://gerrit.wikimedia.org/r/201318 (in principle!)? If we really mean to replace normalized data with unstaged incoming response data, we can do this explicitly in the doTransaction implementation. [20:41:16] Oh hey, we probably want to deploy these soon-ish too: https://gerrit.wikimedia.org/r/201067 [20:42:22] awight: Looking now... [20:42:25] ejegg: sure, i can do that! You want a review of that? Looks like a trivial ffwd. [20:43:02] almost trivial - I maybe shouldn't have deleted the gruntfile + package.json in the merge [20:43:04] awight: The description is totally scary. [20:43:09] But I'll keep going. [20:43:15] ejegg: are you sure about the deleted files? I noticed those too, when playing with a merge to REL1_24. I had to leave them in place. [20:43:29] It's more like... fear of ghosts than fear of car crashes. [20:43:36] got to get your lulz somehow [20:44:00] Okay. I just read the whole thing. [20:44:27] It's totally scary, and I can't immediately figure out how to convince myself that the ghosts aren't real. [20:44:41] I mean, something is going to break. Definitely. [20:44:52] Probably a lot of somethings that you can't search for in one grep. [20:44:53] K4-713: Oh that's for sure. But is making this change an improvement/ [20:45:03] * K4-713 makes 20 faces [20:45:12] I... don't know. [20:46:04] Actually, I think this will take it in to some kind of half-out-of-phase state. [20:46:38] In which the DonationData object and the local copy that the adapter uses for everything, disagree with eachother. [20:46:57] I hate that they're different, btw. [20:47:47] The only reason the gateway has its own copies of the data, is... [20:47:47] I'd like to get to a place where each time data is transformed, there's some explanation that looks like transform("unstaged", "normalized") [20:47:55] ...because... [20:48:27] ...it's because [faint beeping noise] [20:48:44] Honestly, DD should have it all, all the time, and that's it. [20:48:57] sold. [20:49:08] And then: That thing you just said. [20:49:27] DD::get or something [20:49:37] Wait. Get is a terrible word. [20:50:02] But you g... et the idea. [20:51:33] Yeah, adding data to that guy should be just that easy, too. [20:51:43] Array of data, where it came from. [20:51:52] No normalization / whatever until you ask for something. [20:52:11] Can I get demoted for several weeks and write that? [20:52:15] +1 the "where it came from" [20:52:19] yes. [20:52:25] Do I have to steal a starship to do that? [20:52:32] please don't. [20:52:40] * K4-713 looks disappoint [20:53:31] although however you do eventually go out, ask to have that engraved on the headstone. [20:53:45] I'm totally good with that. [20:53:47] "done in by: stealing a starship" [20:54:54] (PS10) Awight: Remove server-side banner random [extensions/CentralNotice] - https://gerrit.wikimedia.org/r/190239 [20:55:58] K4-713: do you think it makes sense since it's just us 2 to use the room upstairs or maybe we should just all call into the meeting separately? [20:56:03] the room doesn't have a TV [20:56:10] What room are we in? [20:56:13] 67 [20:56:19] Guh. [20:56:24] yep [20:56:29] I would very much prefer not to call in separately. [20:56:44] Because my call-in machine is not great at that. [20:56:54] right. bummer [20:57:10] Seems to degrade with more participants, too. [20:57:18] 1:1 usually works just fine. [20:57:19] GOOD BY WIKIBUGS! [20:57:23] BYE* [20:57:39] (CR) jenkins-bot: [V: -1] Remove server-side banner random [extensions/CentralNotice] - https://gerrit.wikimedia.org/r/190239 (owner: Awight) [20:57:42] pahaha [20:58:46] (PS1) AndyRussG: Check choiceData for staleness [extensions/CentralNotice] - https://gerrit.wikimedia.org/r/201349 (https://phabricator.wikimedia.org/T88276) [20:59:44] awight: ejegg: ^ Quick! +2 FTW, before Jenkins finds it! [20:59:59] hahaha [21:00:02] heh [21:01:23] The burndown chart will be forever indebtted [21:01:28] (CR) jenkins-bot: [V: -1] Check choiceData for staleness [extensions/CentralNotice] - https://gerrit.wikimedia.org/r/201349 (https://phabricator.wikimedia.org/T88276) (owner: AndyRussG) [21:01:33] Aaaaaaaawwwwwwwwww [21:07:28] ejegg: Did they get you? [21:09:05] RIP ejegg. [21:09:22] * K4-713 takes invisible hat off [21:36:43] Wikimedia-Fundraising-CiviCRM, Fundraising Sprint I, Fundraising Tech Backlog, Fundraising-Backlog, and 2 others: Upgrade Civi to 4.4 & integrate new reporting - https://phabricator.wikimedia.org/T77910#1172116 (atgo) [21:37:16] § Fundraising Sprint Abba, § Fundraising Sprint Beastie Boys, § Fundraising Sprint Devo, § Fundraising Sprint the Cure, and 4 others: CentralNotice bucket improvements, step 3 bis part 2: moar better cross-browser tests - https://phabricator.wikimedia.org/T86092#1172121 (AndyRussG) [21:42:03] Wikimedia-Fundraising-CiviCRM, Continuous-Integration, Fundraising Sprint H, Fundraising Tech Backlog, and 2 others: Write Jenkins job builder definition for CiviCRM CI job - https://phabricator.wikimedia.org/T91895#1172167 (awight) [21:45:04] http://etherpad.wikimedia.org/p/fundraising_shrimp [21:54:51] https://www.youtube.com/watch?v=Crjnz3YeiEo [21:55:05] It's like the white smoke. [22:10:25] (PS2) Ejegg: Merge branch 'REL1_23' into fundraising/REL1_23 [core] (fundraising/REL1_23) - https://gerrit.wikimedia.org/r/201067 [22:10:51] ok, no longer deleting those 2 files. [22:11:12] awight or K4, wanna take a quick peek? [22:11:21] What am I peeking at? [22:11:30] merge? [22:11:37] (CR) jenkins-bot: [V: -1] Merge branch 'REL1_23' into fundraising/REL1_23 [core] (fundraising/REL1_23) - https://gerrit.wikimedia.org/r/201067 (owner: Ejegg) [22:11:38] I think keeping those files is right. donno why we both saw that conflict. [22:11:55] and the -1 is cos there are new CI jobs that aren't compatible with our 1.23 branch :( [22:11:55] yep, merge, but looks like Jenkins hates it [22:11:58] lemme see [22:12:55] grr. Warning: Task "karma:main" not found. Use --force to continue. [22:13:04] it's from the future. [22:13:24] i need to make a similar tweak for the CRM CI... cos the deployment branch doesn't have tests. [22:13:25] why was it hanging out in REL1_23 then I wonder... [22:15:41] awight: tweaking the jenkins job yaml, not the gruntfile, right? [22:16:20] wait, where do you see that this job is targeting REL1_23? integration/config? [22:16:54] no, I thought the added gulpfile was the issue [22:17:08] gruntfile [22:18:26] ejegg: here's a link to a potentially similar thing while it's fresh, http://git.wikimedia.org/blob/integration%2Fconfig.git/caa59fedd93a5ed2391ea705530cf399dc1e2da4/zuul%2Flayout.yaml#L530 [22:18:50] i'm not sure if the gruntfile is the issue or not. diff rel1_23, maybe? [22:19:22] was just going by jenkins' silence on PS1 where i deleted it [22:21:50] ejegg: git log -p origin/REL1_23 -- Gruntfile.js [22:22:32] oh hey, we're missing a bit [22:23:15] guess that might explain the conflict [22:23:37] ah [22:25:59] (PS3) Ejegg: Merge branch 'REL1_23' into fundraising/REL1_23 [core] (fundraising/REL1_23) - https://gerrit.wikimedia.org/r/201067 [22:26:18] ok, just clobbered 'em both with the REL1_23 version [22:28:06] V=0; the picture of health! [22:28:11] seems to appease zuul [22:29:05] Do you think you'll have time to review the Catalan thing today? [22:29:26] oh yeah, did you decide it was good as is? [22:29:56] I think i decided, i'm okay with finding out. [22:29:59] oh, i see ps2 [22:30:03] cool, looking [22:30:18] The change I make here is much less scary than the one that caused this bug originally :) [22:30:54] test looks good! [22:31:11] ejegg: I may have mentioned this to K4 and not you, but this is what i realized: https://gerrit.wikimedia.org/r/#/c/201318/ [22:32:05] but i need to dig around to see if anything is relying on the side effect of unstaged data overwriting the normalized stuff. [22:32:07] ooh, that one will need some scrutiny [22:33:32] (CR) Ejegg: [C: 2] Don't overwrite donor language with fallback value [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/201132 (https://phabricator.wikimedia.org/T94506) (owner: Awight) [22:33:39] The place I'd like to go is, data includes metadata about its own type/staginess. [22:33:51] yep, definitely [22:34:16] And, when you do something to unstage data after an API call, you'll have to explicitly merge a few unstaged-data response values back into the normalized-data. [22:34:20] and whether it came from GET/POST/XML, what the key was there [22:34:26] (Merged) jenkins-bot: Don't overwrite donor language with fallback value [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/201132 (https://phabricator.wikimedia.org/T94506) (owner: Awight) [22:34:40] full pedigree [22:34:59] OK with me self-CRing the patch merge? [22:35:04] sure! [22:35:11] here goes... [22:35:38] I can do this deployment, it's dinner time there 'n' stuff... [22:35:43] (CR) Ejegg: [C: 2] Merge branch 'REL1_23' into fundraising/REL1_23 [core] (fundraising/REL1_23) - https://gerrit.wikimedia.org/r/201067 (owner: Ejegg) [22:36:00] sure, be my guest! [22:41:19] (PS1) Awight: Merge master into deployment [extensions/DonationInterface] (deployment) - https://gerrit.wikimedia.org/r/201366 [22:41:30] (CR) Awight: [C: 2] Merge master into deployment [extensions/DonationInterface] (deployment) - https://gerrit.wikimedia.org/r/201366 (owner: Awight) [22:41:45] (CR) Awight: [V: 2] Merge master into deployment [extensions/DonationInterface] (deployment) - https://gerrit.wikimedia.org/r/201366 (owner: Awight) [22:46:53] (PS1) Awight: update DonationInterface submodule [core] (fundraising/REL1_23) - https://gerrit.wikimedia.org/r/201368 [22:47:06] (CR) Awight: [C: 2] update DonationInterface submodule [core] (fundraising/REL1_23) - https://gerrit.wikimedia.org/r/201368 (owner: Awight) [22:55:59] ccogdill: fyi, I just saw this go by in the payments logs: "utm_source":"sp%%MAILING_ID%%..cc" [22:56:37] that was me... [22:56:38] sorry [22:56:40] awight [22:57:19] ccogdill: oh, good! [22:57:29] testing some things out over here :D [22:57:36] ccogdill: Also, I'm about to deploy an attempted fix for https://phabricator.wikimedia.org/T94506 [22:57:48]