[13:06:12] Wikimedia-Fundraising: Fundraising in Latin American Spanish - https://phabricator.wikimedia.org/T199680 (Pcoombe) [13:06:26] Wikimedia-Fundraising: Fundraising in Latin American Spanish - https://phabricator.wikimedia.org/T199680 (Pcoombe) [13:08:32] Wikimedia-Fundraising: Localise most important DonationInterface messages to Latin American Spanish - https://phabricator.wikimedia.org/T199682 (Pcoombe) [13:10:51] Wikimedia-Fundraising: Fundraising banners and donatewiki should pass language es-419 if user is in Latin America - https://phabricator.wikimedia.org/T199683 (Pcoombe) [13:13:28] Wikimedia-Fundraising: Translate Thank You email into Latin American Spanish - https://phabricator.wikimedia.org/T199684 (Pcoombe) [13:17:23] Wikimedia-Fundraising: Translate donation support pages into Latin American Spanish - https://phabricator.wikimedia.org/T199685 (Pcoombe) [14:50:51] Fundraising Sprint Karma chameleons hide amongst us, Fundraising Sprint Lactose is unusually tolerant, Fundraising Sprint Matt Damon to head up Space Force, Fundraising Sprint Naming Sprints Is Not Important, and 2 others: Help switch over foundation pages ... - https://phabricator.wikimedia.org/T193663 [15:02:09] (CR) Reedy: [C: 1] "Will leave it to FR to know if it's ok to merge this atm" [extensions/CentralNotice] - https://gerrit.wikimedia.org/r/445779 (https://phabricator.wikimedia.org/T177809) (owner: Phantom42) [15:06:40] (PS3) Mepps: WIP Move to problem array, split out functions [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/445535 (https://phabricator.wikimedia.org/T194517) [15:06:49] getting closer ejegg ^^ [15:07:01] nice [15:08:31] (CR) jerkins-bot: [V: -1] WIP Move to problem array, split out functions [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/445535 (https://phabricator.wikimedia.org/T194517) (owner: Mepps) [15:13:15] (PS4) Mepps: WIP Move to problem array, split out functions [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/445535 (https://phabricator.wikimedia.org/T194517) [15:16:04] (CR) jerkins-bot: [V: -1] WIP Move to problem array, split out functions [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/445535 (https://phabricator.wikimedia.org/T194517) (owner: Mepps) [15:17:05] (CR) Mepps: "recheck" [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/445535 (https://phabricator.wikimedia.org/T194517) (owner: Mepps) [15:19:48] (CR) jerkins-bot: [V: -1] WIP Move to problem array, split out functions [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/445535 (https://phabricator.wikimedia.org/T194517) (owner: Mepps) [15:20:09] ejegg not sure what's up with this failure ^^ [15:20:33] mepps ugh, looks like a bad translatewiki merge [15:20:40] >> Message "apihelp-donate-param-variant" lacks documentation. [15:20:50] ahh [15:21:17] so something from translator-bot merged a change that undid the qqq.json change to document that param [15:22:29] yeah, weird, it was a change that ONLY deleted that string: https://gerrit.wikimedia.org/r/445811 [15:22:39] (PS1) Ejegg: Revert "Localisation updates from https://translatewiki.net." [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/446046 [15:22:46] (CR) Ejegg: [C: 2] Revert "Localisation updates from https://translatewiki.net." [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/446046 (owner: Ejegg) [15:23:43] strange! [15:24:39] (Merged) jenkins-bot: Revert "Localisation updates from https://translatewiki.net." [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/446046 (owner: Ejegg) [15:28:37] (CR) Mepps: "recheck" [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/445535 (https://phabricator.wikimedia.org/T194517) (owner: Mepps) [15:30:36] might need a rebase too [15:35:25] seems to have worked! [15:37:49] (PS5) Mepps: WIP Move to problem array, split out functions [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/445535 (https://phabricator.wikimedia.org/T194517) [15:39:25] cwd: Jeff_Green fr-tech thinking about how the new Kafka ingress scripts should output detailed warnings... Option 1) All output just goes to stdout and stderr. stdout gets a summary of number of files and events processed, and how many (if any) had problems, and stderr gets specific warnings about any lines that had problems [15:40:13] Option 2) stdout same as option (1), and warnings go do a log file or something in a location defined by the script's config [15:40:23] any comments are appreciated, thx in advance!!! [15:40:42] (CR) jerkins-bot: [V: -1] WIP Move to problem array, split out functions [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/445535 (https://phabricator.wikimedia.org/T194517) (owner: Mepps) [15:41:00] AndyRussG: output to stderr sounds good to me - those should be clearly visible in the process-control logs [15:43:13] ejegg: ok yeah sounds good :) [15:43:40] yeah I think the principle would be that process control should be what determines where all output goes, no? [15:44:49] (PS1) Ejegg: Copy es messages to es-419 (Latin American) [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/446054 (https://phabricator.wikimedia.org/T199682) [15:48:36] seems correct to me [15:49:05] it would be nice if the output is clearly filterable by subject line [15:49:41] like "SUCCESS: blah" vs "FAILURE: blah" [15:50:21] (i mean for any case where it's gonna email) [15:59:19] cwd Jeff_Green ah okok that makes sense...! I think a lot of stuff would just be warnings that some data was bad, but it could come from any source (since anyone can hit the EL url). I expect those shouldn't e-mail, though maybe some Prometheus or e-mail alerts if they go above a certain level would be good [16:02:09] cool, sounds good to me! [16:08:07] (PS4) Ejegg: Delete unused extensions and skins [core] (fundraising/REL1_31) - https://gerrit.wikimedia.org/r/440415 [16:08:09] (PS4) Ejegg: Add WMF logos [core] (fundraising/REL1_31) - https://gerrit.wikimedia.org/r/440416 [16:08:12] (PS4) Ejegg: Add payments-wiki extensions [core] (fundraising/REL1_31) - https://gerrit.wikimedia.org/r/440417 [16:08:29] (PS4) Ejegg: Update vendor for merged fundraising deps [core] (fundraising/REL1_31) - https://gerrit.wikimedia.org/r/440418 [16:12:14] :) [16:12:16] (PS1) Ejegg: Add the es-419 language code to support South American Spanish [core] (fundraising/REL1_31) - https://gerrit.wikimedia.org/r/446058 (https://phabricator.wikimedia.org/T112889) [16:13:26] AndyRussG: process-control should log stderr messages with (ERROR) on each line [16:16:09] ejegg: okok... and warnings, for example, from someone hitting the EL endpoint with odd data or unprocessable data (that somehow got past event validation), I imagine with "WARNING" and details on each line? [16:16:56] AndyRussG: that would be nice, but process-control only knows about stdout vs stderr [16:17:11] ejegg: right, so that could go to stderr? [16:17:16] it tags all stdout with INFO and all stderr with ERROR [16:17:32] ah hmmm [16:17:46] so like nowhere for something in between? [16:17:55] (other than a custom logging output place) [16:18:26] the python subprocess module doesn't tell the parent process anything about the log levels, just which socket i/o is being sent to [16:23:28] ejegg: right... so for such warnings (for example, illegal characters in a language code) options are: (1) stderr with a the word "WARNING" at the beginning (2) a log file somehwere, determined in the script's config settings, or (3) nothing (other than summary stats that will be output to stdout) [16:23:47] I'm not especially keen on (3) [16:24:24] hey fr-tech! [16:24:32] good day! [16:24:50] jgleeson: hewwoh! :) [16:25:33] hi jgleeson ! [16:26:25] AndyRussG: I'd still go with (1), even though process-control will add an additional ERROR tag on each line [16:26:44] ejegg: okok gotcha :) [16:27:20] yeah that keeps control of output in the domain of process control and ensures we don't just forget about those warnings [16:27:51] right, my thought is we have log files in enough places already [16:28:02] and throwing them somewhere no-one ever looks is the same as throwing away [16:28:04] or close [16:28:07] yah [16:28:14] okok [16:29:12] you could use return codes [16:30:10] cwd: mm? [16:30:18] http://www.tldp.org/LDP/abs/html/exitcodes.html [16:31:16] cwd: right... the thing I'm thinking of isn't a fatal error or anything bearing on the entire run of the script [16:31:28] aah gotcha [16:31:38] it's more like, in any normal run, there will be a few of these I think, definitely nonfatal [16:32:03] but we want to keep tabs on them, in case there are suddenly a lot, or of a sort that indicates a problem [16:32:20] so maybe an alert or e-mail if they rise above a certain number [16:32:24] or if there's some other issue [16:32:39] in which case we'd want to dig into the details to easily see what was going on [16:33:02] remember this is all coming in via a public EL endpoint [16:33:12] it would be nice to generalize that approach to prevent floods of fail mail [16:33:25] so EL does some validation but can't do it all [16:33:34] hmmm! [16:33:48] yeah I don't know what we may or may not already have in this vein... [16:38:51] hi cwd! [16:39:58] any updates about the access process? [16:44:20] saurabhbatra: hi, yes we will set you up with a temporary password [16:44:26] i am in meetings for a while yet [16:44:37] do you know if adam put in the yubikey request? [16:45:04] I'm not entirely sure [16:45:16] I'll check up with him, he's in the ai room [16:45:22] cool [16:45:27] ping me whenever you get free :-) [16:47:48] will do [16:51:54] saurabhbatra: cwd: fyi, I found my lost yubikey from a few months ago :D [16:52:24] nice [16:53:40] I... have lost a set of keys each time we left the country, so it was nice to at least retrace my steps as far as 3 months ago [16:56:50] Fundraising Sprint Karma chameleons hide amongst us, Fundraising Sprint Lactose is unusually tolerant, Fundraising Sprint Matt Damon to head up Space Force, Fundraising Sprint Naming Sprints Is Not Important, and 2 others: Help switch over foundation pages ... - https://phabricator.wikimedia.org/T193663 [16:58:33] hi awight! [16:58:48] good time to talk about the api? [17:08:06] saurabhbatra: Hi, sure! [17:08:24] I started to skim the link you sent, but hadn't gotten to the punchline yet [17:08:30] i.e. the API design itself. [17:09:46] brb in a couple of minutes [17:10:32] kk [17:11:55] awight, sorry about that, my irc client does some weird stuff some times :-\ [17:14:15] hehe that sounds familiar, I've never heard anyone say they like their IRC client. [17:14:47] ahem [17:14:48] so i was going through the db schema for wmf civi - [17:14:50] https://github.com/wikimedia/wikimedia-fundraising-crm-civicrm/blob/master/sql/civicrm.mysql#L4569 [17:16:16] what part of it do we send to minfraud? [17:16:37] might be a good start to figure out which features are good/relevant [17:21:27] saurabhbatra: the minfraud query is created in our payments form frontend, which is a Mediawiki extension [17:22:08] hi ejegg [17:22:44] https://phabricator.wikimedia.org/diffusion/EDOI/browse/master/extras/custom_filters/filters/minfraud/minfraud.body.php [17:22:47] hi saurabhbatra ! [17:23:00] So, we're using minfraud's composer package [17:23:35] https://github.com/maxmind/minfraud-api-php [17:25:32] the readme on that page shows an example of all the things they can accept [17:25:59] a bunch of those are irrelevant to us, like shipping address (as distinct from billing address) [17:26:26] so we're just mapping our civi_contact and civi_contrib data fields to their data fields [17:26:55] yeah, plus a couple other things about the browser's request that don't get saved in Civi [17:27:41] like the ip and browser info, i see [17:28:17] thanks, that helps! [17:30:01] looks like they take that info and run it through the geoip api to extract info from session information [17:30:04] http://maxmind.github.io/minfraud-api-php/doc/v1.7.0/namespace-MaxMind.MinFraud.Model.html [17:30:45] yep, same company behind the big IP geolocation database and the minfraud service [17:31:11] i think a decent chunk of the score comes from matching the billing address to the request IP [17:31:35] it does make sense, it constitutes almost all of the data they have on the transaction [17:31:45] but of course that score algorithm is a secret [17:33:36] unfortunately the dummy dataset i've been working with does not provide details about what features they've used [17:33:56] "due to confidentiality issues" :-( [17:34:15] From minfraud? [17:34:27] nope, using a sample dataset from kaggle [17:34:41] https://www.kaggle.com/mlg-ulb/creditcardfraud/home [17:34:45] oh cool [17:35:41] which is good for the people involved, nobody wants a bunch of data scientists working with your credit card information :-) [17:35:46] but makes life tough for me [17:35:53] yah, yah [17:37:38] hey ejegg, thanks for the note. just looking over your patch now [17:37:59] Thaks jgleeson ! [17:38:30] saurabhbatra: hmm, i'd guess those were probably labeled as fraud because they led to a chargeback [17:39:24] ejegg: are you talking about the kaggle dataset? [17:42:57] saurabhbatra: yeah, I'd guess they classified them that way [17:43:21] but right, having none of the demographic data would make it a lot less useful [17:43:23] yes i think so [17:44:41] the model i've built on that data performs wonderfully [17:44:55] nice [17:45:31] the only problem being it's impossible to find out what pieces of information are contained in the data [17:46:02] Fundraising-Backlog, Wikimedia-Fundraising, Wikimedia-Fundraising-CiviCRM: Add es-419 as a supported 'preferred language' in Civi - https://phabricator.wikimedia.org/T199733 (Ejegg) p:Triage>Normal [17:47:24] but that can be figured out, no problem [17:47:51] huh, can't get to github right now [17:48:13] the idea is, any transaction that happens, goes to our new fraud detection api to be validated before being processed [17:48:27] yeah, that'll be great! [17:48:46] would it be possible to do a 3rd party geoip request in-between the processing of the transaction? [17:49:28] saurabhbatra: we've actually got a local copy of the geoip database that we use for lookups [17:49:31] how feasible would that be in terms of latency, the external APIs rate-limiting policy etc. [17:49:34] oh [17:49:37] (so we don't have to leak the info to others) [17:49:46] that solves a lot of problems :-) [17:49:56] and the data is consumed through an internal api? [17:50:08] or db lookups? [17:50:50] It's done via a Varnish module on our cache servers [17:51:04] they add a cookie which we can use in PHP or javascript [17:52:56] So, the simplest way for us to integrate your service would be to add another filter class like the one we use to call minfraud [17:53:14] we could send whatever features your service can consume [17:53:23] including results of geoIP lookups [17:54:28] and this would work in the dev environment as well? [17:54:39] silly question, but better to be sure :-) [17:56:23] hmm, so in the dev environment we DO use a 3rd-party geoip call to set the cookie [17:57:12] but I think the part that takes the cookie data and sends that to your service would work the same way [17:58:03] so it shouldn't matter in my use case [17:58:07] right [17:58:53] so api integration would not be a problem [18:00:05] but w.r.t. transactions which've already taken place, do we store their geoip details as well? [18:00:53] no [18:00:56] i'm assuming no [18:01:25] any other info from which i can re-generate these details? [18:10:50] fundraising-tech-ops, Operations, netops: NAT and DNS for fundraising monitor host - https://phabricator.wikimedia.org/T198516 (Jgreen) DNS is done! ;; ANSWER SECTION: frmon.wikimedia.org. 3600 IN CNAME frmon-eqiad.wikimedia.org. frmon-eqiad.wikimedia.org. 3600 IN A 208.80.155.9 [18:16:00] hey cwd, sorry kinda skimped on your message there [18:16:11] (if it was meant for me) [18:17:03] saurabhbatra: which message? [18:18:30] the "ahem" [18:18:59] oh hehe no i was just reminding awight that i like my irc client [18:19:16] damn, i thought that was a "ping" :-) [18:19:31] what client are you on though [18:19:55] i use weechat [18:23:02] i'll definitely consider making a switch, limechat's been giving me problems as of late [18:23:50] cwd Jeff_Green for versions of Python dependencies for the new ingress scripts, I guess I should try to stick with what's available as a Debian package for Jessie? [18:24:50] cwd: :) you probably noticed how carefully I hedged my unfair statement [18:25:06] I use irssi, it's... okay. Better than the other ones I've used, at least. [18:25:29] AndyRussG: yeah, but if there's a particular newer version you need we can look at whether it's available in backports [18:30:52] awight: have you tried weechat? pretty much the same as irssi just a little less clunky imo [18:33:43] cwd: I haven't had the opportunity to try it, but you make it sound better than I remember! [18:33:59] I'm terrified of learning all the workarounds for a new program, though. [18:34:50] yeah i have never used an irc client where the default config was not terrible [18:35:43] !log updated payments-wiki config to 4e0c6c3cfd [18:35:46] Logged the message at https://wikitech.wikimedia.org/wiki/Server_Admin_Log [18:36:47] Fundraising Sprint Karma chameleons hide amongst us, Fundraising Sprint Lactose is unusually tolerant, Fundraising Sprint Matt Damon to head up Space Force, Fundraising Sprint Naming Sprints Is Not Important, and 2 others: Help switch over foundation pages ... - https://phabricator.wikimedia.org/T193663 [18:37:02] Fundraising Sprint Karma chameleons hide amongst us, Fundraising Sprint Lactose is unusually tolerant, Fundraising Sprint Matt Damon to head up Space Force, Fundraising Sprint Naming Sprints Is Not Important, and 2 others: Help switch over foundation pages ... - https://phabricator.wikimedia.org/T193663 [19:16:49] Jeff_Green: okok thx! [19:18:45] awight: weechat + relay plugin + tmux on a vps, then ssh tunnel and glowing-bear locally, ftw! [19:18:52] O_O [19:18:56] AndyRussG: o/ [19:19:01] I wish I were that organized [19:19:09] I barely have a VPS to my name, these days. [19:19:13] awight: ;) it was cwd's suggested setup [19:19:47] Google Cloud Platform gives you a $300 credit just for signing up, and an eternal free tier [19:20:05] not meaning to promote any ahem [19:20:15] multinationals [19:20:18] so to speak [19:21:13] Indeed I guess if my life ran as well as my IRC setup, I'd be doing well [19:21:54] I like DigitalOcean's ethos more so give them a few bucks a month, but don't have one where I run random stuff to make my life better [19:22:23] What I really wish I could figure out is an IMAP server. [19:23:00] Used to have that back in the day, but various new fraud things having complicated network architecture and difficult server config kick my butt every time. [19:23:05] i recently put a computer at my rental house in CO where there is GB internet [19:23:12] that is what i'm signed on through now :) [19:23:14] :100%: [19:23:31] I thought you seemed snappier than usual, but wasn't gonna mention it :p [19:23:38] hehe [19:24:09] you still on that busted $hromebook? [19:25:49] ¢hrome, I suppose [19:25:51] nah i have the company issue X1 [19:26:07] it's pragmatic [19:26:20] aside from the loose screws [19:26:45] I found an old stinkpad "p" for a hundred bucks [19:28:03] there is no one within 300miles of here who would do the recall [19:28:13] i did it myself, i hope [19:28:29] I feel like I'm missing out not being based in the SF office https://upload.wikimedia.org/wikipedia/commons/9/92/Office_planking.jpg [19:29:02] lol [19:29:14] doing that on hangouts would be weird [19:29:15] :) [19:29:20] pillow company? [19:31:35] cwd: I was looking for this great cover of Crazy Train as banjo roots, and learned that this is its own genre of variotonen. [19:32:36] fundraising-tech-ops, Operations, netops: NAT and DNS for fundraising monitor host - https://phabricator.wikimedia.org/T198516 (ayounsi) NAT created: ```lang=diff [edit security nat static rule-set static-nat] rule frbast1001 { ... } + rule frmon1001 { + match { + de... [19:36:49] * awight squeezes eyes and attempts to unsee. [19:37:09] Fundraising-Backlog: Org details not exporting - https://phabricator.wikimedia.org/T199589 (DStrine) Open>declined [19:39:44] Fundraising-Backlog, Wikimedia-Fundraising: Move fundraising thank you pages to donate.wikimedia.org - https://phabricator.wikimedia.org/T189664 (Pcoombe) Open>Resolved a:Pcoombe Payments config is updated and this is done! I also redirected the old foundationwiki pages to donatewiki. @MBeat... [19:40:11] Fundraising-Backlog, fundraising-tech-ops: Figure out where json_encode will screw us in php 7.1+ - https://phabricator.wikimedia.org/T199471 (DStrine) [19:41:05] Fundraising Sprint Karma chameleons hide amongst us, Fundraising Sprint Lactose is unusually tolerant, Fundraising Sprint Matt Damon to head up Space Force, Fundraising Sprint Naming Sprints Is Not Important, and 2 others: Help switch over foundation pages ... - https://phabricator.wikimedia.org/T193663 [19:42:34] jgleeson: To go with the theme, we'd have to act out a world without Wikipedia. Maybe a brick factory wall with a drainpipe pouring a large droplet of clippy onto a row of identical books, and an isolated office worker staring into the dreary tomes [19:43:36] Wikimedia-Fundraising: Create translated versions of new Thank You page - https://phabricator.wikimedia.org/T185333 (Pcoombe) Open>Resolved a:Pcoombe Much delayed, but I finished putting all of these live as part of {T189664}! Full list can be found at https://donate.wikimedia.org/wiki/Category:T... [19:47:57] (PS6) Mepps: WIP Move to problem array, split out functions [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/445535 (https://phabricator.wikimedia.org/T194517) [20:09:01] (PS1) Hashar: Jenkins job validation (DO NOT SUBMIT) [extensions/ContributionTracking] - https://gerrit.wikimedia.org/r/446114 [20:09:19] (CR) Hashar: "check experimental" [extensions/ContributionTracking] - https://gerrit.wikimedia.org/r/446114 (owner: Hashar) [20:10:45] Fundraising Sprint Karma chameleons hide amongst us, Fundraising Sprint Lactose is unusually tolerant, Fundraising Sprint Matt Damon to head up Space Force, Fundraising Sprint Naming Sprints Is Not Important, and 2 others: Help switch over foundation pages ... - https://phabricator.wikimedia.org/T193663 [20:18:22] ejegg: I'm so puzzled by FundraisingMessageGroup. Why did I need to wrap all the functions? [20:19:00] Whatever's happening, now I don't think traits will help, since they only work to export methods as a trait rather than import from another class. [20:19:26] Was it really just about private methods?? [20:19:40] oh man, I was just trying to shut up the errors and figured it was something horrifying that I didn't have time to redo [20:19:47] So.... I dunno [20:19:52] awight, I like the idea lol [20:20:40] ejegg: It seems to be a workaround for now being able to instantiate the base class as a true parent, but needing to use composition. [20:20:43] *not [20:21:08] because methods we needed to mess with were private and not protected? [20:21:42] okay, it's because a translation group already exists, and I'm wrapping it. [20:21:47] insane. [20:22:13] Extension:Translate is challenging. [20:26:47] Ideally, we would just modify the behavior of the one method getMessageGroupStates. Even better would be if Translate supported that natively. [20:33:30] (PS1) Raimond Spekking: Revert "Localisation updates from https://translatewiki.net." [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/446183 [20:46:47] Fundraising Sprint Naming Sprints Is Not Important, Fundraising-Backlog, MediaWiki-extensions-ContributionTracking, MW-1.32-release-notes (WMF-deploy-2018-07-17 (1.32.0-wmf.13)), Patch-For-Review: [ContributionTracking] default DB settings differs when ... - https://phabricator.wikimedia.org/T195814 [20:46:50] (Abandoned) Hashar: Jenkins job validation (DO NOT SUBMIT) [extensions/ContributionTracking] - https://gerrit.wikimedia.org/r/446114 (owner: Hashar) [20:51:49] awight: btw dunno if u saw, live bug involving translated messages was fixed, it was a core thing, deploy coming up via the train... https://phabricator.wikimedia.org/T198869 mmm yeah the whole translate stuff is challenging fer surez... [20:56:59] I hadnt followed, but glad to hear it. The extension exists just to do a very small thing to Translate permissions, really should be supported upstream. [20:57:17] !log re-enabled payments listener failmail stream [20:57:20] Logged the message at https://wikitech.wikimedia.org/wiki/Server_Admin_Log [21:22:05] (PS1) Mepps: Trying to simplify logic [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/446187 [21:22:49] fr-tech this isn't really ready for review (i might wrap it into the previous patch), but as my end of day update, just continuing to try to simplify the logic here [21:26:41] thanks mepps! [21:27:35] ditto! [21:28:33] awight: cool, thx!! :) [21:40:54] (PS9) Jgleeson: Add opt_in field for selected countries [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/445327 (https://phabricator.wikimedia.org/T199278) (owner: Ejegg) [21:42:29] (CR) jerkins-bot: [V: -1] Add opt_in field for selected countries [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/445327 (https://phabricator.wikimedia.org/T199278) (owner: Ejegg) [21:44:16] hey ejegg, I fixed one test on the OptIn patch but the other, which as far as I can see behaves the same way, is giving an ambiguous "general" error during the gateway validation process (I think it's something to do with the zero passed to opt_in) I've emailed you what I see anyway. I'll try and patch that in the morning if you don't get to it today. [21:44:42] thanks jgleeson, I'll take a look! [21:45:15] I wonder if we're using !empty for fields marked required, and failing on zero [21:45:23] maybe! [21:48:43] have a good evening fr-tech, see you tomorrow! [22:01:44] Fundraising-Backlog, Wikimedia-Fundraising-CiviCRM: slow anonymous save - https://phabricator.wikimedia.org/T199753 (Eileenmcnaughton) [22:02:30] awight: so obviously I’ve been offline & you have been working with Saurabh - is there anything I should be doing to help? [22:03:50] on that note, awight did you request his yubikey? [22:04:47] cwd: ah rats, I never did. I can do that now—Do you agree that it's most expedient to have a local vendor ship to him? [22:05:26] awight: we will make him a temporary password so i don't think it matters [22:09:45] cwd awight let’s do the temporary [22:09:57] cwd are you able to organise access with him directly? [22:11:52] (PS1) Ejegg: Revert "Hack out some failing tests on fundraising/REL1_27" [core] (fundraising/REL1_27) - https://gerrit.wikimedia.org/r/446192 [22:12:02] eileen: yeah i've been talking to him [22:12:09] cwd great [22:12:43] cwd: Is this a loophole around 2FA? [22:12:50] temporary as in one-time, maybe? [22:13:15] it's still 2fa, ssh key and password [22:13:26] O_o [22:13:31] but let's get him the yubikey in a timely manner [22:13:46] cool, I just wrote OIT about it [22:13:51] ty [22:13:58] it's definitely worse 2fa [22:17:05] Yeah I suppose it's something you know & have, but it feels cheaty :-) Thanks for finding a temporary workaround, though! [22:18:46] if we find out we can get him one more quickly that would of course be preferable [22:18:52] but we don't want to hold the whole thing up [22:20:41] exactly. [22:20:49] (CR) Ejegg: [C: 2] Revert "Hack out some failing tests on fundraising/REL1_27" [core] (fundraising/REL1_27) - https://gerrit.wikimedia.org/r/446192 (owner: Ejegg) [22:23:26] (Merged) jenkins-bot: Revert "Hack out some failing tests on fundraising/REL1_27" [core] (fundraising/REL1_27) - https://gerrit.wikimedia.org/r/446192 (owner: Ejegg) [22:34:08] dstrine - I’ve just added Coleman to MG call next week [22:34:24] he was interested in attending - at least for the part where we present his work [22:41:34] eileen: cool and thanks. It seems everyone was OOO this week [22:42:01] Fundraising Sprint Naming Sprints Is Not Important, Fundraising-Backlog, MediaWiki-extensions-ContributionTracking, MW-1.32-release-notes (WMF-deploy-2018-07-17 (1.32.0-wmf.13)), Patch-For-Review: [ContributionTracking] default DB settings differs when ... - https://phabricator.wikimedia.org/T195814 [22:42:21] (PS1) Ejegg: Get rid of extra check / set on DB globals [extensions/ContributionTracking] - https://gerrit.wikimedia.org/r/446195 (https://phabricator.wikimedia.org/T195814) [22:42:38] XenoRyet: do you have a sec to try that one out ^^^ ? [22:43:01] I was getting all my DonationInterface tests failing with the parent patch but not that one [22:43:24] Yea, let me see what I can see. [22:44:01] so, to replicate the failures, just try updating ContributionTracking extension to master [22:44:13] and then running the DonationInterface unit tests [22:45:23] the failures would be a lot of test errors that mention somthing about unittest_contribution_tracking [22:54:57] ejegg: I'm not getting failures with the parent patch. [22:55:48] XenoRyet: weird, so with ContributionTracking at master/HEAD (without the one in CR), you can run all the DonationInterface tests? [22:56:31] what's the last patch you see in git log in ContributionTracking? [22:56:55] Yea, only getting one failure and it's not related. [22:57:54] git log on C_T has a patch by you on Jul 12 DB settings default to null [22:58:04] ahhh, ok, one more thing... [22:58:42] can you try temporarily commenting out the $wgContributionTrackingDBxxxxx settings in your LocalSettings.php [22:58:55] or the vagrant equivalent? [22:59:03] Yea, let me try. [22:59:32] It's in one of the settings.d files, right? [22:59:48] I think so, let me find it. [23:00:17] seems buried in settings.d/wikis/paymentswiki/settings.d/puppet-managed [23:00:25] 10-ContributionTracking.php [23:01:17] you might want to run php maintenance/update.php after commenting those out [23:01:25] to get the table created in the wiki db [23:02:08] Yea, getting a truckload of errors now. [23:04:06] ok, and if you pull down that ContributionTracking patch do they go away? [23:04:13] https://gerrit.wikimedia.org/r/446195 [23:04:21] trying now [23:06:38] Yea, looking good [23:06:53] Code looks fine too, I'll +2 [23:07:06] thanks! [23:07:10] (CR) XenoRyet: [C: 2] Get rid of extra check / set on DB globals [extensions/ContributionTracking] - https://gerrit.wikimedia.org/r/446195 (https://phabricator.wikimedia.org/T195814) (owner: Ejegg) [23:08:03] No worries [23:17:01] (CR) jerkins-bot: [V: -1] Get rid of extra check / set on DB globals [extensions/ContributionTracking] - https://gerrit.wikimedia.org/r/446195 (https://phabricator.wikimedia.org/T195814) (owner: Ejegg) [23:21:15] (PS10) Ejegg: Add opt_in field for selected countries [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/445327 (https://phabricator.wikimedia.org/T199278) [23:21:17] (PS1) Ejegg: Use !isset() instead of empty() in DataValidator [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/446205 (https://phabricator.wikimedia.org/T199278) [23:21:42] ooh, V -1 on that CT patch? [23:22:45] XenoRyet: ah crud, something was looking for that extensionFunction [23:22:53] one sec, will zap that too [23:23:12] (CR) jerkins-bot: [V: -1] Use !isset() instead of empty() in DataValidator [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/446205 (https://phabricator.wikimedia.org/T199278) (owner: Ejegg) [23:23:53] (CR) jerkins-bot: [V: -1] Add opt_in field for selected countries [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/445327 (https://phabricator.wikimedia.org/T199278) (owner: Ejegg) [23:23:55] (PS2) Ejegg: Get rid of extra check / set on DB globals [extensions/ContributionTracking] - https://gerrit.wikimedia.org/r/446195 (https://phabricator.wikimedia.org/T195814) [23:25:46] jeez, that one little validator change was pretty disruptive too [23:30:34] (CR) XenoRyet: [C: 2] Get rid of extra check / set on DB globals [extensions/ContributionTracking] - https://gerrit.wikimedia.org/r/446195 (https://phabricator.wikimedia.org/T195814) (owner: Ejegg) [23:36:52] (Merged) jenkins-bot: Get rid of extra check / set on DB globals [extensions/ContributionTracking] - https://gerrit.wikimedia.org/r/446195 (https://phabricator.wikimedia.org/T195814) (owner: Ejegg) [23:46:24] Hmm so my take on the Silverpop workings for deleting contacts (https://phabricator.wikimedia.org/T199747#4429171) is that we can add a ‘query’ or list of contacts but not an individual one without making some sort of list [23:52:24] (PS2) Ejegg: Don't mutate $data in DataValidator loop [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/446205 (https://phabricator.wikimedia.org/T199278) [23:55:52] XenoRyet: does ^^^ look reasonable to you? [23:56:13] The issue is that to show the opt-in checkbox conditionally, we need to mark it as required [23:56:51] but the old logic in DataValidator was mutating '0' to 'null' [23:57:02] which then failed the validate_not_empty check [23:57:42] I've gotta go pick up the kid, but I'll take a look at that first thing in the morning. Presuming Maggie or Jack don't beat me to it. [23:58:21] ok, thanks