[00:27:25] fr-tech emartin heads-up it looks like the issue with Apple Pay not settling is in fact on our end... just triple-checking, will send an explanation in reply to the e-mail in a bit [00:56:21] oh man, do apple pay txns need to be captured after all AndyRussG ? [01:07:11] ejegg|away: yep it seems so [01:07:57] as it happens, (nearly?) all the test ones in the test console just stayed in Authorized, or went to Expired,s tate [01:07:59] state [01:08:29] I didn't see any doc about capturing anything in the Adyen components doc, only in a different doc about payment statuses [01:12:00] faceplam... [01:12:24] so we should be able to fix that with a small chanhe in one smashpig class [01:12:43] whichever status normalizer we use for apple pay [01:16:26] yep yep [03:29:30] 10Fundraising-Backlog: Apple Pay: Payments require capture - https://phabricator.wikimedia.org/T294721 (10AndyRussG) [03:30:32] 10Fundraising-Backlog, 10fundraising sprint Universal Cereal Bus: Apple Pay: Payments require capture - https://phabricator.wikimedia.org/T294721 (10AndyRussG) [05:30:20] 10Fundraising-Backlog: DonationInterface CI is stuck - https://phabricator.wikimedia.org/T294722 (10AndyRussG) [05:30:52] 10Fundraising-Backlog, 10fundraising sprint Universal Cereal Bus: DonationInterface CI is stuck - https://phabricator.wikimedia.org/T294722 (10AndyRussG) [05:48:54] (03PS1) 10AndyRussG: [DO NOT MERGE] Test CI [core] (fundraising/REL1_35) - 10https://gerrit.wikimedia.org/r/735765 (https://phabricator.wikimedia.org/T294722) [05:52:05] (03CR) 10jerkins-bot: [V: 04-1] [DO NOT MERGE] Test CI [core] (fundraising/REL1_35) - 10https://gerrit.wikimedia.org/r/735765 (https://phabricator.wikimedia.org/T294722) (owner: 10AndyRussG) [06:07:16] (03PS2) 10AndyRussG: [DO NOT MERGE] Test CI: Update SmashPig library and vendor directory [core] (fundraising/REL1_35) - 10https://gerrit.wikimedia.org/r/735765 (https://phabricator.wikimedia.org/T294722) [06:08:33] (03PS3) 10AndyRussG: [DO NOT MERGE] Test CI: Update SmashPig library and vendor directory [core] (fundraising/REL1_35) - 10https://gerrit.wikimedia.org/r/735765 (https://phabricator.wikimedia.org/T294722) [06:10:00] (03CR) 10jerkins-bot: [V: 04-1] [DO NOT MERGE] Test CI: Update SmashPig library and vendor directory [core] (fundraising/REL1_35) - 10https://gerrit.wikimedia.org/r/735765 (https://phabricator.wikimedia.org/T294722) (owner: 10AndyRussG) [06:15:44] 10Fundraising-Backlog, 10fundraising sprint Universal Cereal Bus, 10Patch-For-Review: DonationInterface CI is stuck - https://phabricator.wikimedia.org/T294722 (10AndyRussG) Fiddling with stuff, to no avail... ^ [06:46:00] 10WMSE-Content-partnerships-support-2021-Fundraising, 10User-Jenny_Brandt_WMSE: Prepare document /presentation for Fundraising for WMSE - https://phabricator.wikimedia.org/T292455 (10Jenny_Brandt_WMSE) [06:56:52] 10WMSE-Content-partnerships-support-2021-Fundraising, 10User-Jenny_Brandt_WMSE: Prepare document /presentation for Fundraising for WMSE - https://phabricator.wikimedia.org/T292455 (10Jenny_Brandt_WMSE) Part 1 Will be presented on Personalsamlingen. - Part 1 consist of giving an overview to the staff on differ... [07:01:35] 10WMSE-Content-partnerships-support-2021-Fundraising, 10User-Jenny_Brandt_WMSE: Prepare fundraising presentation for staff. - https://phabricator.wikimedia.org/T292455 (10Jenny_Brandt_WMSE) [07:11:19] 10WMSE-Content-partnerships-support-2021-Fundraising, 10User-Jenny_Brandt_WMSE: Prepare Fundraising PM and Presentation for personalsamling - https://phabricator.wikimedia.org/T294725 (10Jenny_Brandt_WMSE) [08:40:45] (03CR) 10jerkins-bot: [V: 04-1] Localisation updates from https://translatewiki.net. [extensions/DonationInterface] - 10https://gerrit.wikimedia.org/r/735835 (owner: 10L10n-bot) [09:33:53] (03Abandoned) 10Raimond Spekking: Localisation updates from https://translatewiki.net. [extensions/DonationInterface] - 10https://gerrit.wikimedia.org/r/735528 (owner: 10L10n-bot) [09:34:38] (03CR) 10Raimond Spekking: [C: 03+2] "false positive" [extensions/DonationInterface] - 10https://gerrit.wikimedia.org/r/735835 (owner: 10L10n-bot) [09:55:37] (03CR) 10jerkins-bot: [V: 04-1] Localisation updates from https://translatewiki.net. [extensions/DonationInterface] - 10https://gerrit.wikimedia.org/r/735835 (owner: 10L10n-bot) [14:21:21] 10Fundraising-Backlog: SmashPig Recurring Failmail: No translation found - https://phabricator.wikimedia.org/T294763 (10jgleeson) [14:25:13] Hi AndyRussG. As of this morning, Nov 1, I am still seeing Apple Pay transactions unsettled. Is there an ETA for resolution? Thanks [14:25:48] The console is back being slow so settling manually is problematic now. [14:32:16] emartin: we'll have to write up a script to capture them then [14:32:41] Ok. [14:33:21] frtech we are waiting on the Adyen call. Will someone join? [14:33:29] fr-tech ^ [14:33:59] I can't make the new meeting time, it conflicts will my school run sorry! [14:34:07] it used to be half an hour earlier I think? [14:36:49] ahhh the UK clocks went back over the weekend :( so the new time was 3:30pm last week but now it's 2:30pm. Sorry about that! [14:49:53] fr-tech btw here's a task for the CI issues for DonationInterface https://phabricator.wikimedia.org/T294722 [14:50:34] thanks for making that AndyRussG [14:52:18] AndyRussG: thanks! I'll write my theory for how we can fix it [14:54:52] ejegg AndyRussG cstone. I have sent the list of unsettled Apple Pay transactions in email [14:55:00] thanks emartin! [14:55:45] ejegg cstone :) [14:55:51] emartin: thanks!! [14:58:45] 10Fundraising-Backlog, 10fundraising sprint Universal Cereal Bus: Apple Pay: Payments require capture - https://phabricator.wikimedia.org/T294721 (10AndyRussG) We just learned from Adyen that all Apple Pay transactions do need to be captured. We can either do it ourselves, as we do for cards, or set the paymen... [14:59:51] 10Fundraising-Backlog, 10fundraising sprint Universal Cereal Bus, 10Patch-For-Review: DonationInterface CI is stuck - https://phabricator.wikimedia.org/T294722 (10Ejegg) So it looks like other WMF-managed libraries get around this by changing the settings in packagist to point to the github mirror of their g... [15:00:18] 10Fundraising-Backlog, 10fundraising sprint Universal Cereal Bus: Apple Pay: Payments require capture - https://phabricator.wikimedia.org/T294721 (10AndyRussG) [15:03:35] 10Fundraising-Backlog, 10fundraising sprint Universal Cereal Bus, 10Patch-For-Review: DonationInterface CI is stuck - https://phabricator.wikimedia.org/T294722 (10AndyRussG) >>! In T294722#7472003, @Ejegg wrote: > So it looks like other WMF-managed libraries get around this by changing the settings in packag... [15:05:25] fr-tech is it correct that we have no IPN listeners configured on Adyen on prod? https://ca-live.adyen.com/ca/ca/config/showthirdparty.shtml [15:05:51] they are working somehow [15:06:29] huh that's really weird [15:06:45] I guess I should have asked this before the meeting hehe [15:21:14] fr-tech emartin no objections to e-mailing Adyen to note where their doc needs to say more about capture? It'll be quick, and I might as well do so now while it's fresh in my brain..... [15:22:11] I could also ask about the IPN listeners not showing up on the console [15:22:56] no objections here [15:23:11] emartin: cool thanks much :) [15:28:27] hi fr-tech! [15:29:05] sorry about the timing mix ups on the Adyen call. I wouldn't have been much use anyway as I've been out a week but had hoped to catch up. Did anyone record it by chance? [15:29:31] yep it was recorded jgleeson [15:29:41] thanks much cstone ! [15:31:42] ejegg Jeff_Green anyone have more info on the lack of IPN config in the Adyen live console? maybe it's configured elsewhere somehow? [15:31:52] it's under "devloper/webhooks" [15:32:39] hi AndyRussG [15:32:45] I feel like I've looked at that before [15:32:52] fr-tech analytics meeting? [15:33:12] cstone: no invite for me on that [15:34:27] AndyRussG: looking at what's in puppet, there's an API key and URL for Adyen, but nothing that looks IPN-specific [15:35:27] AndyRussG: then there are some settings in localsettings/SmashPig/local-config/adyen [15:36:11] that's the only config I know about, not sure what's in console, I don't have access to it [15:44:16] AndyRussG: fr-tech looks like we do have webhooks set up in the Adyen console under the WikimediaDonations account. We're currently asking them to send webhooks to https://payments-listener.wikimedia.org/adyen [15:46:18] AndyRussG: in the top left hand corner of the Adyen console you can switch accounts from company to merchant and the webhook settings are under the merchant account [15:58:51] cstone: we made a ticket earlier for that no translation bug you looked into https://phabricator.wikimedia.org/T294763 [15:58:57] it happened again this morning [15:59:37] thanks jgleeson yeah I thought I fixed the smashpig logs last week to give us more info there but I think we need to add more logging still [15:59:52] we were guessing that its a language missing from the new system [16:00:00] ahh [16:01:04] 10Fundraising-Backlog, 10fundraising sprint Universal Cereal Bus, 10Patch-For-Review: DonationInterface CI is stuck - https://phabricator.wikimedia.org/T294722 (10Ejegg) > Ah nice find!! Hmmm that would also potentially work to unstick it in this particular situation, as well as to prevent the issue in gener... [16:14:28] 10Fundraising-Backlog: SmashPig Recurring Failmail: No translation found - https://phabricator.wikimedia.org/T294763 (10Cstone) We were thinking this was from a missing language in the new failmail system but the logging isn't being helpful. I cleaned up the logs some with T294172 but we need more info still [16:17:03] fr-tech I'd like to force-merge the smashpig version bump to be able to move ahead with cstone's patch [16:17:32] as I wrote on https://phabricator.wikimedia.org/T294722 I think maybe the change to the repo URL in packagist will make it not get stuck on the NEXT upgrade [16:18:16] ejegg should we put things in maintenance mode until that DI patch is +2d and deployed? when i tested locally there wasnt an error with ingenico but the form just refreshed weirdly [16:18:41] cstone wait, put production in maintenance mode? [16:18:50] yeah cause itll break ingenico until my patch is deployed? [16:18:54] I don't think so [16:19:11] cstone only if we deploy the SmashPig lib update [16:19:16] I just want to merge it [16:19:19] ahh okay sorry misunderstood [16:19:57] fr-tech. I wanted to mention here that authorized/unsettled Apple Pay transactions are showing up in Civi as settled. I THINK that is ok since the auth sort of guarantees settlement (once we start settling these transactions) but I just wanted to call it out in case anyone sees issues in how Civi is accounting for things. [16:20:11] ejegg: what happened? [16:20:25] DI get out of sync with merged smashpig? [16:20:30] jgleeson there's another slight breaking change [16:20:51] so the current DI won't pass on new SmashPig [16:21:00] and the new DI won't pass on current SmashPig [16:21:09] ahh [16:21:22] on top of that, jenkins still runs composer update, which complains of a missing .git dir [16:21:30] ticket here too jgleeson https://phabricator.wikimedia.org/T294722 [16:21:30] I think I have a solution for the missing .git dir [16:21:47] (see comments on ticket for missing .git dir solution) [16:22:08] gotcha. thanks both [16:23:00] anyway, I can force-merge the SmashPig udpate for now, then maybe the next time around when it's upgrading from zip to zip it won't complain of a missing .git dir [16:23:08] that's the theory anyway [16:24:39] then maybe additionally we will need to start pegging smash-pig to a specific version in DI's composer.json [16:26:00] 10Fundraising-Backlog, 10fundraising sprint Universal Cereal Bus: Apple Pay: Payments require capture - https://phabricator.wikimedia.org/T294721 (10EMartin) I vote we automatically capture within minutes given we do not do fraud checks on Apple Pay transactions. [16:32:43] sounds like a plan ejegg [16:34:14] 10Fundraising-Backlog, 10fundraising sprint Universal Cereal Bus: Apple Pay: Payments require capture - https://phabricator.wikimedia.org/T294721 (10Ejegg) We just need to make sure to save the transaction to Civi under the correct pspReference / gateway_txn_id - I think Adyen changes the pspReference after a... [16:35:49] 10Fundraising-Backlog, 10fundraising sprint Universal Cereal Bus: Apple Pay: Payments require capture - https://phabricator.wikimedia.org/T294721 (10Cstone) +1 for doing it right away as we have seen IPN messages in the past take 5-10 minutes [16:38:14] Jeff_Green jgleeson ahh thanks for that info on the webhooks... yeah I didn't see that account switch jgleeson, good to know... I guess as regards payments though it'll show the same transactions regardless of which account you're on? looks like it I guess? [16:38:26] (03PS1) 10Ejegg: Adyen: Maintenance script to batch aprove payments [wikimedia/fundraising/SmashPig] - 10https://gerrit.wikimedia.org/r/736009 (https://phabricator.wikimedia.org/T294721) [16:38:39] AndyRussG: the account switch is in the top right I think [16:38:45] I think so AndyRussG! [16:38:47] and yeah it shows the same [16:39:08] (03CR) 10jerkins-bot: [V: 04-1] Adyen: Maintenance script to batch aprove payments [wikimedia/fundraising/SmashPig] - 10https://gerrit.wikimedia.org/r/736009 (https://phabricator.wikimedia.org/T294721) (owner: 10Ejegg) [16:39:26] cool thx cstone jgleeson [16:40:47] emartin wrt authorised-but-unsettled Apply things moving to Civi, that does sound like what's expected, since the Apple Pay code was written on the assumption that there was nothing further to do on those... I also think it's ok [16:41:55] ejegg: sounds like a plan eh :) [16:44:08] AndyRussG. But now something does have to be done on Apple Pay, right? [16:44:20] (03PS2) 10Ejegg: Adyen: Maintenance script to batch aprove payments [wikimedia/fundraising/SmashPig] - 10https://gerrit.wikimedia.org/r/736009 (https://phabricator.wikimedia.org/T294721) [16:44:33] emartin: yes indeed [16:45:38] emartin: yep, we're just discussing on that ticket whether it's best for us to use our credit card flow, calling the capture API ourselves (and maybe skipping the fraud checks), or to set them up for automatic capture [16:45:53] I'm in favor of calling the capture API ourselves [16:46:10] because it means we won't have to wait for an IPN to tell us what the final pspReference is [16:46:52] ejegg. thanks. I follow you. [16:47:02] ejegg: ah good point [16:47:29] also a smaller code change then [16:48:45] AndyRussG: ah yep, let [16:48:57] 's jsut confirm my hunch that the pspReference changes [16:49:22] and that the one which ends up in Adyen's console is the one from the capture [16:50:33] i have definitely had mistaken assumptions before [16:51:11] I can confirm that it changes, looking at logs from my local setup [16:51:18] k, cool [16:51:19] (references below are only testing, not live) [16:51:22] 021-11-01T03:19:27+00:00 deb810dc55bb SmashPig-Adyen: adyen::180:180.1 | Preparing to send POST request to https://checkout-test.adyen.com/v67/payments/862635736536041D/captures | | [16:51:23] 2021-11-01T03:20:52+00:00 deb810dc55bb adyen_gateway: 180:180.1 Returned PSP Reference 852635736768048J [16:53:07] ejegg: however it looks like the main PSP reference is from the payment API call, not the capture one [16:53:11] https://ca-test.adyen.com/ca/ca/accounts/showTx.shtml?pspReference=862635736536041D&txType=Payment [16:53:25] the capture one is still mentioned in the detailed view under "payment events" tho [16:53:34] ohhh [16:53:49] in that case I guess we can use auto-capture just fine! [16:54:09] so we CAN turn autocapture on just for Apple Pay? [16:54:40] Ah right, I think I remember having to change the DonationInterface code to reflect that now [16:54:44] btw appears to be the same for normal cc transactions, too: https://ca-test.adyen.com/ca/ca/accounts/showTx.shtml?pspReference=862635728638153A&txType=Payment [16:55:28] ejegg: we can, so that would be less interaction needed with Adyen servers, so on less place things could go wrong. Also something we haven't tried before, but likely easy [16:55:31] https://gerrit.wikimedia.org/r/c/mediawiki/extensions/DonationInterface/+/715994 [16:56:03] yep, we initially coded it with the capture ID for cards, but then changed it to use the auth ID ^^^ [16:56:28] ok, so now the minimal change is zero code change, just config on adyen's site [16:56:53] makes me feel like the smashpig library should have configuration to match [16:57:02] ejegg: no, that config to change on Adyen's side is for everything, not per payment method, I think [16:57:13] AndyRussG: argh, that's a bummer [16:57:17] haven't checked for sure [16:57:33] ejegg: the change I think we would want is an added property on the payments call [16:57:42] oh? [16:57:53] https://docs.adyen.com/api-explorer/#/CheckoutService/v67/post/payments__reqParam_captureDelayHours [16:58:19] I guess I'm also not positive if that works without a change in the account config [16:58:50] AndyRussG: hmm, yeah, i would interpret "The delay between the authorisation and scheduled auto-capture, specified in hours" to have no effect if there's no scheduled auto-capture [16:59:48] ejegg: yea but Manavi's e-mail seems to imply the opposite [17:00:37] oops, i need to read that [17:01:48] hmm, given that there's a slight chance the auth could succeed and the capture could fail, i'd still opt for calling the capture API ourselves and immediately recording the result (i.e. to the queue) [17:02:09] rather than specifying an auto-capture and waiting for the IPN [17:02:22] or specifying an auto-capture and assuming it will be successful [17:02:53] ejegg: oki [17:03:47] sounds like it's faster to implement just because it's a known flow, that I've already tested locally/with the testing account, and we'd have less questions or things to confirm wrt account setup [17:04:01] yep, that too [17:04:22] +1 for not changing account setup i [17:04:24] -i [17:04:27] though we might want one additional change to skip the cvv/avs for apple pay [17:04:59] sure [17:05:31] the adyen adapter already overrides getScoreAvs and getScoreCvv [17:05:46] so those could check for payment method and return 0 [17:07:05] hmm, but would we want to skip the minFraud call too? [17:07:23] I guess if we're pretty sure apple pay fraud is infinitesimal [17:08:00] and we can set up ad-hoc filters to run on the initial phase to catch script kiddies [17:08:21] then we might just skip the validation phase filters altogether for apple pay [17:09:07] ack, standup [17:10:57] ejegg: yeah apple seems to have that quite locked down [17:12:23] k, we can just put a payment_method check around the $this->runAntifraudFilters() bit in handleCreatedPayment [17:18:10] AndyRussG: mind if I just re-title your test patches to update SmashPig in our mediawiki and vendor branches? [17:18:49] i.e. remove do not merge [17:19:09] oh wait, that one pulls in dev deps too [17:19:16] never mind, will make up a fresh one! [17:21:25] (03PS2) 10Ejegg: Update SmashPig [core] (fundraising/REL1_35) - 10https://gerrit.wikimedia.org/r/735470 [17:26:23] (03CR) 10jerkins-bot: [V: 04-1] Update SmashPig [core] (fundraising/REL1_35) - 10https://gerrit.wikimedia.org/r/735470 (owner: 10Ejegg) [17:27:11] (03CR) 10Ejegg: [V: 03+2 C: 03+2] Update SmashPig [core] (fundraising/REL1_35) - 10https://gerrit.wikimedia.org/r/735470 (owner: 10Ejegg) [17:27:29] ejegg: yeah I mean do what's needed, but I think some of those patches are suspect heheh [17:27:30] (03CR) 10Ejegg: "recheck" [extensions/DonationInterface] - 10https://gerrit.wikimedia.org/r/724233 (https://phabricator.wikimedia.org/T291223) (owner: 10Cstone) [17:33:38] ejegg: regarding the 905 validation error patches, I have an idea of how to do that in a way that's a bit cleaner... though it's probably faster to just code it up than to explain it... mind if I do that, i guess on separate Gerrit changes, at least for purposes of discussing? [17:33:50] sure AndyRussG, go for it [17:33:56] cool thx! [17:41:27] fr-tech. I wanted to point out that Apple Pay transactions are showing up on the stopped fraud list today. As the transactions are saying that "issuer will accept liability" due to all of the authentication, I am going to settle these items as the console will allow, but thought I'd mention this to you bc we have been scoring Apple Pay transactions it seems. Was that expected behavior? [17:42:10] emartin: what are the criteria for that list? [17:42:24] is it just the authorized-but-not-captured transactions? [17:46:15] https://www.irccloud.com/pastebin/Kf5urajX/ [17:48:29] emartin: so if they're Apple Pay, they weren't intentionally 'stopped', they were just not captured because of the programming error [17:48:35] which we are working to fix [17:48:44] we are also working on a batch capture script [17:48:58] I have it coded up, I'm just about to test it on some sandbox donations [17:49:18] ejegg curious as to why not all of the transactions didn't show up on this list if that is the case. [17:49:33] emartin: how do you generate the list? [17:54:34] ejegg Disregard, I think I understand now. We didn't score these. We just haven't settled yet. Sorry for the confusion. These will be on the list I gave you in email earlier today. Took me a second to absorb. [17:54:58] ok, cool, I think we can have them settled by EOD today [17:55:08] thanks. [17:58:00] ejegg. perhaps I should send an updated list before you run this to include today's transactions [17:58:26] emartin: ok, maybe wait till we have the script tested? [17:58:34] ok! [18:00:07] (03PS3) 10Ejegg: Adyen: Maintenance script to batch aprove payments [wikimedia/fundraising/SmashPig] - 10https://gerrit.wikimedia.org/r/736009 (https://phabricator.wikimedia.org/T294721) [18:00:33] ok fr-tech, I just tested PS3 with a sandbox txn and it seemed to work ^^^ [18:01:00] oh hmm, the API call came back successful [18:01:14] but the test console still has 852635711412149H in 'Authorised' [18:01:21] maybe takes a minute to update? [18:01:38] ah yeah, now it's in SentForSettle [18:01:46] k, that script is ready for review [18:01:54] ejegg: cool got it thx for that! [18:03:34] ejegg: for the actual change to capture Adyen, I was thinking how about separate normalizers for ApplePay, CC and RTBT? [18:03:55] AndyRussG: sure, we can do that [18:04:28] the CC and Apple Pay ones will be identical for now, but just in case Apple Pay adds other underlying payment models? [18:04:58] yeah hmmm I mean consolidating them is also an option, since we were told by Adyen that it works similarly to cards [18:05:18] maybe "CaptureRequiredStatusNormaliser" ? [18:05:26] sure, sounds good [18:05:30] cool [18:05:50] also wondering why RTBT is called DirectDebit at the SP level... it's fully the same thing, right? [18:06:33] yeah, i think DirectDebit was more in line with what you see in payment processor docs [18:06:53] and rtbt was maybe an acronym invented for DOnationInterface [18:07:53] hmm, though DonationInterface has some history of a 'dd' payment_method [18:08:02] so... i guess i don't completely understand [18:08:43] seems only supported in legacy globalcollect [18:11:08] hmmm oki thx! [18:11:18] if DirectDebit is the more standard term then that's also cool [18:11:59] oh interesting, they may indeed be different things [18:12:11] hmmm [18:12:24] though we do use DirectDebit in SP for what's rtbt in DI, no? [18:12:24] maybe this is UK specific, but https://gocardless.com/direct-debit/timings/ says 'direct debit' is processed on a specific daily schedule [18:12:40] while I guess real-time would happen right when the button is clicked [18:12:48] hmmm [18:12:57] welp I guess it does depend on countries [18:13:40] in Canada there's a country-specific debit/bank transfer network which is immediate [18:14:03] which you can also use for POS purchases with your bank debit card, which has no CC logo [18:14:20] yep, Colombia's PSE is more or less immediate too [18:14:31] straight from your bank account [18:14:41] and there's also a time-delayed way for a company to debit your account, which seems to use the same system as cashing a cheque [18:14:57] like with iDEAL, you pick a bank name on the payment page and end up on your bank's page to authorize the payment [18:15:18] whereas here it's all debit cards with visa or mc logos, also immediate [18:15:19] hmmm [18:15:35] maybe "direct debit" is the more general term that can also include real-time? [18:15:55] yeah, i think that makes sense [18:17:14] hmm cool [18:25:47] howdy fr-tech, I was planning to jump back in on the 600=>800 stuff I cookie licked last week. Were there any new decisions on that work last week while I was away or is this ticket still good to pick back up https://phabricator.wikimedia.org/T291226 [18:26:03] unless there's other more urgent stuff [18:26:40] jgleeson I have a couple patches for the risk scorer subtask [18:27:00] https://gerrit.wikimedia.org/r/c/wikimedia/fundraising/SmashPig/+/735428 https://gerrit.wikimedia.org/r/c/wikimedia/fundraising/SmashPig/+/735447 [18:27:32] and we merged cstone's patch to swap from array to the response class [18:29:01] thanks! is risk scorer stuff related to the 600=>800 stuff? sorry if I've missed that [18:29:05] so would you maybe be picking up this WIP and finishing it off? https://gerrit.wikimedia.org/r/c/wikimedia/fundraising/crm/+/722703 [18:29:14] I can review it regardless [18:29:35] jgleeson: yep! cstone's patch is a start, but the risk scorer bit is also needed [18:29:44] ah ok thanks [18:29:54] so the hope is to make the CRM patch only look at normalized properties [18:30:06] while the DonationInterface code just keeps looking at the rawResponse [18:30:41] ok looks like I've not read into this closely enough so I'll reread the tickets to get the full picture. thanks! [18:30:43] anyway, with cstone's patch merged we can update DonationInterface to do that and not have to update it again after follow-on patches [18:31:06] in SmashPig to set the rest of the normalized properties [18:33:26] ah ok it looks like the first ticket https://phabricator.wikimedia.org/T291223 has more tentacles than I was expecting. I had it in my head that cstone's merged patch was all I needed to move on to https://phabricator.wikimedia.org/T291226 [18:34:13] yeah, i had forgotten about the tokens till I re-read that ticket [18:34:40] but fortunately mapping that field was pretty trivial: https://gerrit.wikimedia.org/r/c/wikimedia/fundraising/SmashPig/+/735464 [18:36:38] (03PS1) 10Ejegg: Apple Pay: indicate approval needed after creation [wikimedia/fundraising/SmashPig] - 10https://gerrit.wikimedia.org/r/736028 (https://phabricator.wikimedia.org/T294721) [18:36:50] does that look right AndyRussG ? ^^^ [18:37:15] heh, I guess we'll have a chance to test another smashpig library update sooner than anticipated [18:44:51] i hope the github method works! [18:46:03] (03PS1) 10Ejegg: Apple Pay: Skip validation phase fraud filters [extensions/DonationInterface] - 10https://gerrit.wikimedia.org/r/736029 (https://phabricator.wikimedia.org/T294721) [18:48:34] (03CR) 10jerkins-bot: [V: 04-1] Apple Pay: Skip validation phase fraud filters [extensions/DonationInterface] - 10https://gerrit.wikimedia.org/r/736029 (https://phabricator.wikimedia.org/T294721) (owner: 10Ejegg) [18:51:32] oh heh, I need to rebase that over cstone's patch [18:51:38] and also review cstone's patch [18:54:15] some of the test fixes are kinda gross but I think will be better once the token part is there at least [18:54:18] ejegg: one sec :) [18:54:19] i put notes [18:54:29] cstone: no, they look exactly right! [18:55:01] I mean, we could add some helper methods to cut down on repeated code [18:55:18] one was looking for a token I think [18:55:19] but so far they don't look any grosser than the existing test setup [18:55:37] yep, those TODOs / FIXMEs seem like good tasks for January [18:55:48] ok cool [18:56:10] but for now, I think you've done exactly what's needed to get the expected rawResponse property [19:07:31] (03PS9) 10Ejegg: DI part of switching getHostedPaymentStatus to return a PaymentDetailResponse [extensions/DonationInterface] - 10https://gerrit.wikimedia.org/r/724233 (https://phabricator.wikimedia.org/T291223) (owner: 10Cstone) [19:08:05] (03CR) 10Ejegg: [C: 03+2] "Looks good!" [extensions/DonationInterface] - 10https://gerrit.wikimedia.org/r/724233 (https://phabricator.wikimedia.org/T291223) (owner: 10Cstone) [19:08:53] (03PS2) 10Ejegg: Apple Pay: Skip validation phase fraud filters [extensions/DonationInterface] - 10https://gerrit.wikimedia.org/r/736029 (https://phabricator.wikimedia.org/T294721) [19:10:05] ok fr-tech, here are three patches for the apple pay capture fix: https://gerrit.wikimedia.org/r/c/mediawiki/extensions/DonationInterface/+/736029 https://gerrit.wikimedia.org/r/c/wikimedia/fundraising/SmashPig/+/736028 https://gerrit.wikimedia.org/r/c/wikimedia/fundraising/SmashPig/+/736009 [19:10:26] (03Abandoned) 10Ejegg: Attempt to make SmashPig install from dist [extensions/DonationInterface] - 10https://gerrit.wikimedia.org/r/735468 (owner: 10Ejegg) [19:10:42] (03Merged) 10jenkins-bot: DI part of switching getHostedPaymentStatus to return a PaymentDetailResponse [extensions/DonationInterface] - 10https://gerrit.wikimedia.org/r/724233 (https://phabricator.wikimedia.org/T291223) (owner: 10Cstone) [19:16:04] 10Fundraising-Backlog, 10fundraising-tech-ops, 10fr-donorservices: terminal / fredge access for Amber Johnson - https://phabricator.wikimedia.org/T293679 (10AMJohnson) Yes, sorry! I got... Your Yubikey Public ID is: ccccccnejkhl Let me know if anything additional is needed. Thanks! [19:32:45] (03CR) 10AndyRussG: Apple Pay: indicate approval needed after creation (031 comment) [wikimedia/fundraising/SmashPig] - 10https://gerrit.wikimedia.org/r/736028 (https://phabricator.wikimedia.org/T294721) (owner: 10Ejegg) [19:40:19] 10Fundraising-Backlog, 10fundraising-tech-ops: Dlocal adding 6 IPs - https://phabricator.wikimedia.org/T294606 (10Dwisehaupt) [19:40:49] 10Fundraising-Backlog, 10fundraising-tech-ops: Refresh staging data - https://phabricator.wikimedia.org/T293315 (10Dwisehaupt) [19:52:33] (03PS2) 10Ejegg: Apple Pay: indicate approval needed after creation [wikimedia/fundraising/SmashPig] - 10https://gerrit.wikimedia.org/r/736028 (https://phabricator.wikimedia.org/T294721) [20:21:58] ejegg: cool thx for the update! ^ just gonna smoke test [20:36:46] (03CR) 10AndyRussG: "Cool! One more comment inline :)" [wikimedia/fundraising/SmashPig] - 10https://gerrit.wikimedia.org/r/736028 (https://phabricator.wikimedia.org/T294721) (owner: 10Ejegg) [20:42:00] ejegg: I guess for Apple Pay we don't ever expect getHostedPaymentDetails to be called since there's no redirect to a bank site or the like for Apple Pay? [20:43:39] so another option for ApplePayPaymentProvider would be to throw an "unimplemented" exception if getPaymentDetailsStatusNormalizer is called? as in, maybe there should be no PaymentDetailsStatusNormalized for Apple Pay? [20:44:12] hmmm also maybe getHostedPaymentDetails() could use a rename? [20:44:34] or I guess it's "hosted" as in "externally hosted"' [20:45:12] sounded like a carry-over from old Adyen integration, but maybe I'm wrong? [20:46:03] 10Fundraising-Backlog, 10fundraising-tech-ops: Dlocal adding 6 IPs - https://phabricator.wikimedia.org/T294606 (10Dwisehaupt) 05Open→03Resolved a:03Dwisehaupt As these ip changes are for their **outbound** pool, there is nothing needed on our side to handle this change. [20:47:15] also if it's unused, we can do something with Apple Pay's getPaymentDetailsStatusNormalizer() in a follow-on patch [20:59:42] right AndyRussG it's used in ingenico too [21:00:07] but for apple pay as far as i can tell it's always a one page flow [21:00:17] K I'm gonna +2 in a sec and submit a small thingy for that [21:01:43] might as well smoke test a CC transaction too [21:09:33] (03CR) 10AndyRussG: [C: 03+2] "Looks great, thanks for this!!!! (I'll add a follow-on with an option for the stuff mentioned inline...)" [wikimedia/fundraising/SmashPig] - 10https://gerrit.wikimedia.org/r/736028 (https://phabricator.wikimedia.org/T294721) (owner: 10Ejegg) [21:10:39] (03Merged) 10jenkins-bot: Apple Pay: indicate approval needed after creation [wikimedia/fundraising/SmashPig] - 10https://gerrit.wikimedia.org/r/736028 (https://phabricator.wikimedia.org/T294721) (owner: 10Ejegg) [21:16:34] fr-tech: not sure if I'll ever use this but I just accidentally found out that you can pull down specific patchsets of a patch using the git-review -d cmd. If you pass `git-review 735447/1` or `git-review 735447,1` it will pull down PS1 specifically on that smashpig patch. [21:16:47] oops [21:16:51] I missed the -d arg [21:16:59] Huh, interesting [21:17:01] oh cool! [21:17:13] so it's `git-review -d 735447/3` for PS3 of that patch [21:17:28] eat that, GitLab..... oh wait I bet GitLab can do that, too [21:17:30] yeah not sure if I'll actually use it but good to know [21:23:57] thanks for the CR AndyRussG [21:24:58] thanks for the patchz ejegg :) [21:25:03] just looking at the maintenance script now [21:25:04] fr-tech should we also push out this patch to skip the fraud filters on Apple Pay? they're currently not being hit because they're only called when approvePayment is needed: https://gerrit.wikimedia.org/r/c/mediawiki/extensions/DonationInterface/+/736029 [21:25:12] ok, great! [21:25:40] Oh, I guess I'll tag a minor-minor version and see if jenkins is happy updating it now [21:27:25] (03CR) 10Ejegg: "recheck" [extensions/DonationInterface] - 10https://gerrit.wikimedia.org/r/736029 (https://phabricator.wikimedia.org/T294721) (owner: 10Ejegg) [21:28:45] (03CR) 10jerkins-bot: [V: 04-1] Apple Pay: Skip validation phase fraud filters [extensions/DonationInterface] - 10https://gerrit.wikimedia.org/r/736029 (https://phabricator.wikimedia.org/T294721) (owner: 10Ejegg) [21:30:38] awww [21:40:23] huh [21:41:02] so where does it still say that's a git package? [21:41:49] dang, yeah, it's not --preferring-dist: Syncing wikimedia/smash-pig (v0.7.6.1) into cache [21:42:44] 10Fundraising-Backlog, 10FR-CiviCRM-Searchkit, 10FR-Tech-Analytics: Date format of date fields when other math / aggregates / field transformations are applied - https://phabricator.wikimedia.org/T294507 (10Eileenmcnaughton) https://lab.civicrm.org/dev/core/-/issues/2932 [21:43:31] (03CR) 10Jgleeson: "Thanks for this. My first impression is that this class might be better placed within the Core/PaymentProviders NS as it feels specific to" [wikimedia/fundraising/SmashPig] - 10https://gerrit.wikimedia.org/r/735428 (https://phabricator.wikimedia.org/T291230) (owner: 10Ejegg) [21:44:11] sounds good jgleeson, will move it there [21:45:56] 10Fundraising-Backlog, 10FR-CiviCRM-Searchkit, 10FR-Tech-Analytics: Default Value in form builder for dates acting as string - https://phabricator.wikimedia.org/T294505 (10Eileenmcnaughton) https://lab.civicrm.org/dev/core/-/issues/2933 [21:46:03] (03CR) 10Jgleeson: "Thanks for this! It looks good but I have a question on the mysterious E response code" [wikimedia/fundraising/SmashPig] - 10https://gerrit.wikimedia.org/r/735447 (https://phabricator.wikimedia.org/T291230) (owner: 10Ejegg) [21:46:17] thanks ejegg ! [21:46:41] and sorry for introducing a whole Helper layer way back when without realising what I was signing us up for [21:47:02] with hindsight it might have been best to go with a different approach [21:48:50] 10Fundraising-Backlog, 10FR-CiviCRM-Searchkit, 10FR-Tech-Analytics: Aggregating fields from searchkit - https://phabricator.wikimedia.org/T294504 (10Eileenmcnaughton) @EYener I logged https://lab.civicrm.org/dev/core/-/issues/2934 but I realise I'm already too vague on this one to create more upstream ticket... [21:48:57] jgleeson: https://www.nytimes.com/wirecutter/reviews/best-lap-desk/ [21:49:51] I'm using my £6.99 lapdesk right now AndyRussG :) [21:50:13] wow some of those look nice [21:50:36] ah cool! [21:50:41] i've used my kids kiwico built lapdesk a few times when the dog wouldn't get off my lap. [21:51:07] and when you search logs on prod you can use lap logdog [21:51:24] (03CR) 10Ejegg: Use risk scorer for Ingenico hosted checkout status (031 comment) [wikimedia/fundraising/SmashPig] - 10https://gerrit.wikimedia.org/r/735447 (https://phabricator.wikimedia.org/T291230) (owner: 10Ejegg) [21:51:24] aww dwisehaupt [21:51:30] you can just put the computer on top of the dog also [21:51:40] LOL [21:51:42] yeah. he isn't a fan of that. [21:51:53] no the fan's in the laptop [21:51:58] extra wide one with dog cushion attachment [21:52:03] especially not when running a full puppet rebuild [21:52:14] so then you can use lap dog logdog [21:52:37] woof [21:53:01] yeah so for me the real big benefit I've found is laptop heat is no longer a problem on my legs [21:53:31] previously after using it for a while the heat of the underside of the laptop would make it uncomfortable [21:53:46] ah I think it's actually even unhealthy [21:55:17] hmmm I guess sp standalone checkout is where I should be looking at the maintenance script [21:56:22] aha, I think it's this line that makes it keep pulling smash-pig from git: https://phabricator.wikimedia.org/diffusion/MWVD/browse/fundraising%252FREL1_35/composer/installed.json$4077 [21:56:45] AndyRussG: yep, we usually run maintenance scripts in standalone sp [21:57:16] ejegg: did we give the E: a score of 100 before also? feels like an instant fail-like score [21:57:30] E:error maybe? [21:57:45] thx [21:57:46] jgleeson: it's in the defaults that way - not sure what our actual configured value is [21:57:53] ahh, E=Error would make sense [21:57:56] ejegg ^ I was able to reproduce the CI error locally [21:58:05] so I could also test after maintenance stuff [21:58:24] *maintenance script stuff CR thingy [21:58:32] AndyRussG: thanks! [21:59:13] (03CR) 10Jgleeson: Use risk scorer for Ingenico hosted checkout status (031 comment) [wikimedia/fundraising/SmashPig] - 10https://gerrit.wikimedia.org/r/735447 (https://phabricator.wikimedia.org/T291230) (owner: 10Ejegg) [22:03:33] (03PS2) 10Eileen: Function extraction in test class [wikimedia/fundraising/crm] - 10https://gerrit.wikimedia.org/r/734789 (https://phabricator.wikimedia.org/T290253) [22:03:57] Ingenico had a make over. It's so green https://epayments-api.developer-ingenico.com/s2sapi/v1/en_US/java/hostedcheckouts/get.html?paymentPlatform=ALL#hostedcheckouts-get [22:04:30] get bought by a company gotta change the css [22:04:38] it seems I was wrong that non of my PRs are review ready - this one is & it's pretty easy https://gerrit.wikimedia.org/r/c/wikimedia/fundraising/crm/+/734789 (just seeing where I got to with them all) [22:06:17] cstone: I'm no designer but there's a odd mix of colour happening here https://epayments.developer-ingenico.com/ [22:08:25] hmm [22:08:28] I take it all back [22:08:39] "go green" is actually a pretty good message for the world right now [22:10:01] i mainly noticed cause they changed the highlight color to green [22:11:59] (03CR) 10AndyRussG: "Thanks for this!! Tiny detail inline...." [wikimedia/fundraising/SmashPig] - 10https://gerrit.wikimedia.org/r/736009 (https://phabricator.wikimedia.org/T294721) (owner: 10Ejegg) [22:15:21] huh I had not noticed [22:15:22] https://techcrunch.com/2020/02/03/worldline-to-buy-ingenico-for-8-6b-in-major-payments-consolidation-play/ [22:15:40] maybe we should start having silly internal names for our PSP integrations [22:15:52] that way we wouldn't have to refactor anything when they buy and sell themselves [22:16:26] yeah they finally changed the css and branding like a month ago [22:16:35] hmmm [22:16:43] So for Braintree, we could scour the 5 years of sprint names, pick the best one, and name the integration that [22:17:47] oh no it never even dawned on me that we have to rename stuff AndyRussG [22:17:51] :((((((((((((((((((9 [22:17:54] x9 [22:18:52] +1 to picking our own PSP code names and then having a config map file we can update as a "real" name reference [22:19:32] for some reason the names of animals for each gateway came to me first [22:19:47] hmmm [22:20:04] or Braintree = KingKongVsMozilla [22:20:17] Queue is pronounced GJi [22:21:05] ChickenTurtleCreatePaymentResponse [22:21:19] https://en.wikipedia.org/wiki/Chicken_turtle [22:21:43] oh man I didn't think it was called Chicken because of that [22:21:56] poor turtles [22:22:15] yum! [22:22:34] Elevators were never intended to go down [22:22:52] LOL @ KingKongVsMozilla [22:23:10] just distracting my brain reminiscing with https://etherpad.wikimedia.org/p/fundraising_shrimp [22:27:12] ejegg: I have a tiny request for the maintenance script which is not even important enough to put on Gerrit, sooo not a blocker [22:27:28] oh, i saw the comment on the commit message [22:27:35] oh this is another one heheh [22:27:41] even less important [22:27:42] i'm just fixing something in the patch jgleeson reviewed [22:27:50] ejegg: just been scanning the ingenico docs and if I can make my indents out correctly, I think 'tokens' is a property of 'createdPaymentOutput' element and not at the root. I'm looking here https://epayments-api.developer-ingenico.com/s2sapi/v1/en_US/json/hostedcheckouts/get.html?paymentPlatform=ALL#hostedcheckouts-get-response-200 [22:27:50] yeee no worries [22:27:51] k, what's the other request? [22:27:57] sorry to distract! [22:28:20] I'll add the comment to the patch [22:28:26] jgleeson: ooh, good catch [22:28:30] I should add a test for that [22:29:29] it is: how would you feel about adding some blank lines here and there in execute() ? I feel they make it much easier to read code, if they divide it into logical sections, kinda like having many short paragraphs vs one long one [22:29:35] jgleeson: no worries! [22:29:48] sure thing AndyRussG [22:29:56] ejegg: thanks!!!! sorry for the bother [22:31:05] (03PS2) 10Ejegg: Move RiskScorer to PaymentProvider namespace [wikimedia/fundraising/SmashPig] - 10https://gerrit.wikimedia.org/r/735428 (https://phabricator.wikimedia.org/T291230) [22:33:06] (03PS4) 10Ejegg: Adyen: Maintenance script to batch aprove payments [wikimedia/fundraising/SmashPig] - 10https://gerrit.wikimedia.org/r/736009 (https://phabricator.wikimedia.org/T294721) [22:33:59] (03CR) 10Jgleeson: [C: 03+2] "Looks great. Thanks much!" [wikimedia/fundraising/SmashPig] - 10https://gerrit.wikimedia.org/r/735428 (https://phabricator.wikimedia.org/T291230) (owner: 10Ejegg) [22:34:29] (03Merged) 10jenkins-bot: Move RiskScorer to PaymentProvider namespace [wikimedia/fundraising/SmashPig] - 10https://gerrit.wikimedia.org/r/735428 (https://phabricator.wikimedia.org/T291230) (owner: 10Ejegg) [22:35:36] hmm gerrit won't rebase the follow-on patch in the UI [22:35:48] lemme pull it down and see what the conflict is [22:35:53] jgleeson: it's ok, it needs a little edit anyway [22:36:00] just rebasing locally now [22:36:12] (03PS5) 10Ejegg: Use risk scorer for Ingenico hosted checkout status [wikimedia/fundraising/SmashPig] - 10https://gerrit.wikimedia.org/r/735447 (https://phabricator.wikimedia.org/T291230) [22:36:27] I got rid of that undocumented 'E' code too [22:36:50] ah ok [22:37:09] I guess we can see if it comes up in the logs [22:37:11] ejegg: thanks so much for that change, yeah for me it's much easier to understand like that [22:37:17] cool cool AndyRussG [22:37:56] ejegg: do you feel like throwing an error if it's a method that doesn't take an approve step (i.e., rtbt)? also not a blocker of course [22:38:37] AndyRussG: hmm, and just hardcoding the list of 'em ? [22:39:16] I guess... I mean we don't have that elsewhere? it's only rtbt, I think? also not sure if it's useful [22:39:24] I was wondering about splitting the approve() off the base PaymentProvider interface actually [22:39:36] right [22:40:22] so we'd have a way to check if $provider instanceof ApprovePaymentsProvider [22:40:30] but... seems a bit much for this patch [22:40:34] yeah for sure [22:41:07] (03CR) 10Jgleeson: [C: 03+2] "Looks good to me! thanks for the test update also!" [wikimedia/fundraising/SmashPig] - 10https://gerrit.wikimedia.org/r/735447 (https://phabricator.wikimedia.org/T291230) (owner: 10Ejegg) [22:41:15] (03PS2) 10Jgleeson: Map tokens to RecurringPaymentToken field [wikimedia/fundraising/SmashPig] - 10https://gerrit.wikimedia.org/r/735464 (https://phabricator.wikimedia.org/T291230) (owner: 10Ejegg) [22:41:18] thanks jgleeson ! [22:41:29] That tokens one needs the positional fix, right? [22:41:36] (03Merged) 10jenkins-bot: Use risk scorer for Ingenico hosted checkout status [wikimedia/fundraising/SmashPig] - 10https://gerrit.wikimedia.org/r/735447 (https://phabricator.wikimedia.org/T291230) (owner: 10Ejegg) [22:41:51] yep ejegg and then that's good to go [22:42:03] happy to do it if you're working on the other patch with AndyRussG [22:42:20] heh, i'd just switched back to the tokens branch [22:42:28] eh I don't see anything else to gripe about also [22:42:34] just gonna smoke test [22:45:15] (03PS3) 10Ejegg: Map tokens to RecurringPaymentToken field [wikimedia/fundraising/SmashPig] - 10https://gerrit.wikimedia.org/r/735464 (https://phabricator.wikimedia.org/T291230) [22:46:11] ah right we already have IPaymentProvider [22:49:20] (03CR) 10Jgleeson: [C: 03+2] "Looks good to me! The plural thing is slightly worrying but I can't any examples of multiple tokens anywhere. Let's hope they don't surpri" [wikimedia/fundraising/SmashPig] - 10https://gerrit.wikimedia.org/r/735464 (https://phabricator.wikimedia.org/T291230) (owner: 10Ejegg) [22:49:51] (03Merged) 10jenkins-bot: Map tokens to RecurringPaymentToken field [wikimedia/fundraising/SmashPig] - 10https://gerrit.wikimedia.org/r/735464 (https://phabricator.wikimedia.org/T291230) (owner: 10Ejegg) [22:49:56] (03PS1) 10Ejegg: Update vendor for smash-pig source force [core] (fundraising/REL1_35) - 10https://gerrit.wikimedia.org/r/736076 [22:50:58] fr-tech sorry about the mixup in timezones for the adyen meeting earlier today. I noticed in our calendar we're on DCT(Daylight Confusion Time) so kinda explains things [22:51:21] last week it was 3:30pm my time but this week it's 2:30pm [22:51:33] and I pick Oscar up from school at 3:00pm [22:51:42] usually get back by 3:15pm [22:51:51] Ah yeah I also just got switched off DST... probably it'll move back to 3:30 for you in a few weeks, when SF also switches [22:51:59] though we can ask to move it meanwhile [22:52:06] fr-tech i'm sorry i missed the other meeting this morning - i don't even have a timezone-related excuse, just baby-related distraction [22:52:20] oh heheh that's always a good excuse [22:52:42] (03PS3) 10Eileen: Function extraction in test class [wikimedia/fundraising/crm] - 10https://gerrit.wikimedia.org/r/734789 (https://phabricator.wikimedia.org/T290253) [22:52:44] (03PS1) 10Eileen: regenerate phpunit & bootstrap files (using civix) [wikimedia/fundraising/crm] - 10https://gerrit.wikimedia.org/r/736077 [22:52:56] you can keep using excuses like that for the next 20 years or so [22:54:06] AndyRussG: no rush to change it.. again.. as long as it's not a burden to other folks getting up really early. happy to wait! [22:54:28] ah oki thanks... for me it's fine either way [22:54:28] oh nice, changing that one line in installed.json did it! [22:54:30] 18:51:29 [50.0MiB/8.09s] - Upgrading wikimedia/smash-pig (v0.7.6 => v0.7.6.1): Extracting archive [22:54:36] ahhh yay! [22:54:36] https://integration.wikimedia.org/ci/job/quibble-donationinterface-REL1_35-php73-docker/1041/console [22:54:51] nice! [22:55:02] woohoo [22:56:21] (03CR) 10AndyRussG: Adyen: Maintenance script to batch aprove payments (032 comments) [wikimedia/fundraising/SmashPig] - 10https://gerrit.wikimedia.org/r/736009 (https://phabricator.wikimedia.org/T294721) (owner: 10Ejegg) [22:56:27] heheh found something else [22:57:22] yep, major units. will update [22:58:33] oki thx :9 [22:58:35] :) [22:58:36] (03PS5) 10Ejegg: Adyen: Maintenance script to batch aprove payments [wikimedia/fundraising/SmashPig] - 10https://gerrit.wikimedia.org/r/736009 (https://phabricator.wikimedia.org/T294721) [22:59:04] (03CR) 10Ejegg: "recheck" [extensions/DonationInterface] - 10https://gerrit.wikimedia.org/r/736029 (https://phabricator.wikimedia.org/T294721) (owner: 10Ejegg) [22:59:45] srsly, if we don't have dash to work on, December SHOULD be unit tests month [23:00:24] delayed wooo! [23:01:25] that'd be cool :) [23:04:47] have a good one fr-tech. bye for now! [23:04:51] cya! [23:07:14] (03CR) 10AndyRussG: [C: 03+2] "Yay works great!!! Thx!!!!!!!!!!" [wikimedia/fundraising/SmashPig] - 10https://gerrit.wikimedia.org/r/736009 (https://phabricator.wikimedia.org/T294721) (owner: 10Ejegg) [23:07:58] (03Merged) 10jenkins-bot: Adyen: Maintenance script to batch aprove payments [wikimedia/fundraising/SmashPig] - 10https://gerrit.wikimedia.org/r/736009 (https://phabricator.wikimedia.org/T294721) (owner: 10Ejegg) [23:08:15] ^ maintenance script merged [23:09:20] thanks AndyRussG ! [23:09:41] thank u :) [23:09:48] looks like the patch to skip fraud checks is passing tests now: https://gerrit.wikimedia.org/r/c/mediawiki/extensions/DonationInterface/+/736029/ [23:09:48] K looking at the risk scorer now [23:09:54] yep yep [23:14:05] (03CR) 10Ejegg: [C: 03+2] Update vendor for smash-pig source force [core] (fundraising/REL1_35) - 10https://gerrit.wikimedia.org/r/736076 (owner: 10Ejegg) [23:22:27] (03Merged) 10jenkins-bot: Update vendor for smash-pig source force [core] (fundraising/REL1_35) - 10https://gerrit.wikimedia.org/r/736076 (owner: 10Ejegg) [23:29:16] so there /is/ one unrelated CI issue updating the vendor dir: wikimedia/composer-merge-plugin v1.4.1 requires composer-plugin-api ^1.0 -> found composer-plugin-api[2.1.0] but it does not match the constraint. [23:29:34] I think that composer-plugin-api got bumped up by mistake a few commits back [23:29:53] (03CR) 10AndyRussG: [C: 03+2] "Yaaaayy thanks!!! (P.S. feeling such positive equilibrium vibes now that I see we have both fraud and antifraud filters...)" [extensions/DonationInterface] - 10https://gerrit.wikimedia.org/r/736029 (https://phabricator.wikimedia.org/T294721) (owner: 10Ejegg) [23:30:36] hmmm is that blocking anything ejegg? [23:31:00] AndyRussG: I still have to force-merge the patch in the vendor repo [23:31:17] aww hmmm [23:31:18] lolol fraud and antifraud filters [23:31:33] btw I did also get that error [23:31:35] yeah, i briefly noticed the mismatch [23:31:43] in method names [23:31:54] eh it's okok [23:34:22] hmm, should we update the merge plugin? [23:34:53] oh hey, it is supposed to be updated in core's composer.json [23:34:58] will just do that then [23:36:13] huh? we do have version 2.0.1 installed [23:36:31] so... what the heck is CI talking about 1.4.1 for? [23:37:00] hmmm maybe we need to update something else too? one of the submodules? Vector? [23:37:35] oh wait, what the heck does vendor have its own composer.json and composer.lock for???? [23:37:50] does the main branch vendor have those? [23:38:04] yeah just noticed the 1.4.1 is in vendor/composer.json [23:39:00] huh, so it does have it [23:39:04] v odd [23:39:12] hmmm [23:39:33] mebbe readme explains? [23:39:44] (03Merged) 10jenkins-bot: Apple Pay: Skip validation phase fraud filters [extensions/DonationInterface] - 10https://gerrit.wikimedia.org/r/736029 (https://phabricator.wikimedia.org/T294721) (owner: 10Ejegg) [23:40:53] okay, it looks like they more or less keep it in sync with the parent dir's composer.json [23:40:58] seems duplicative... [23:41:01] "vendor-dir": "." [23:41:52] well, I guess I'll update our branch to match the base REL1_35 one [23:44:10] the original REL1_35 branch also has vendor as a submodule? [23:44:19] yep yep [23:44:22] hmm [23:44:37] and the original vendor/REL1_35 has the composer.json and .lock [23:44:45] so I'm just updating our .json and .lock to match those [23:44:52] it'll pass tests that way at least! [23:45:01] cool thanks much [23:45:16] (basically just linting the vendor dir, which seems like a worthwhile test) [23:45:42] hmmm [23:45:53] not sure how it's deployed currently on the main cluster also [23:46:26] vendor has the non-dev deps checked in [23:46:35] right [23:46:38] so the main cluster can just deploy from git too [23:46:46] right [23:46:59] and for dev setups we run composer to get those dev deps [23:47:24] yep yep [23:48:17] hmm [23:48:32] well that's nice to finally get resolved! We were kicking those cans down the road for a while [23:48:45] yeah absolutely [23:49:11] K gonna take a short break to walk the dog and do a few non-work things here, then back in 60-90 minutes or so... thx again for all this! [23:49:17] rockin! [23:49:36] I'mma just do a lil smoke test locally and prep a deploy of cstone's update and the apple pay capture fix [23:50:05] then we can figure out the list of auths that need capture with that script [23:51:27] (03PS1) 10Ejegg: Update SmashPig + vendor [core] (fundraising/REL1_35) - 10https://gerrit.wikimedia.org/r/736086 [23:51:36] (03CR) 10Ejegg: [C: 03+2] Update SmashPig + vendor [core] (fundraising/REL1_35) - 10https://gerrit.wikimedia.org/r/736086 (owner: 10Ejegg) [23:53:36] yeah it'll be great to have the ApplePay capture stuff fully fixed [23:54:00] AndyRussG: did you have a new port forward script patch? [23:54:38] ejegg: only in my brain [23:54:43] ah cool [23:55:14] but try this, it works a charm (substituting the right remote port numbers): autossh -M 0 -N -o "ServerAliveInterval=30" -o "ServerAliveCountMax=3" -o "ExitOnForwardFailure=yes" -R 8001:localhost:9009 -R 8101:localhost:9006 payments.fr-tech-dev [23:55:46] (and also don't forget to manually turn off ssh on the smashpig apache config, we gotta fix that too) [23:57:49] ohhh right [23:57:57] i still just have that hacked locally [23:58:04] i can update that in dev-images [23:58:17] ah cool thx, hehe I think I still can't [23:59:13] hrm? that doesn't look like it actually captured... [23:59:34] did I actually pull all the updates into my docker src dirs? [23:59:59] (03Merged) 10jenkins-bot: Update SmashPig + vendor [core] (fundraising/REL1_35) - 10https://gerrit.wikimedia.org/r/736086 (owner: 10Ejegg)