[16:55:12] Fundraising-Backlog, FR-ActiveMQ: Migrate FundraisingEmailUnsubscribe to new queue backend - https://phabricator.wikimedia.org/T145419#2629496 (awight) [16:56:29] (PS1) Awight: Include a null default for $wgStompServer [extensions/FundraisingEmailUnsubscribe] - https://gerrit.wikimedia.org/r/310040 [17:00:40] Fundraising Sprint Muggle Baiting, Fundraising Sprint Nitpicking, Fundraising Sprint Octopus Untangling, Fundraising Sprint Pretending This Isn't Happening, and 9 others: Run the orphan rectifier job from CRM Jenkins - https://phabricator.wikimedia.org/T141487#2629577 (awight) Consolidating the p... [17:01:05] cwd: this one's for you, https://gerrit.wikimedia.org/r/#/c/309523/ [17:01:19] There are also a few Configuration patches if you're interested [17:02:04] cwd: I also updated the SmashPig listener tests to match your newer work--you might want to squash all those patches in. [17:02:12] awight: nice, yes i will do that [17:02:19] Anything I can do? [17:02:46] I was gonna focus on expanding the tests more but I think we need to sync up about implementation before I can do that [17:02:58] nothing from me at the moment, still prying my head out of the weekend [17:03:11] want to get on the phone in a few minutes? [17:03:35] sure! [17:03:47] Just tell me when. I can dial you in [17:07:28] sorry breaker just popped, gimme a few minutes [17:07:32] haha [17:09:39] i hate this place [17:09:41] aaaaarg!!!! [17:10:25] cwd: sorry :( [17:10:36] (CR) Ejegg: "recheck" [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/308800 (https://phabricator.wikimedia.org/T131277) (owner: Ejegg) [17:13:57] cwd: your new house? uh, oh [17:15:07] i think the breaker is fucked [17:17:53] hmm. a 110 breaker or 220? [17:18:11] I think uneven load on the legs of a 220 breaker will do that [17:18:16] (PS2) Ejegg: Update composer libs [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/309516 (owner: Awight) [17:18:24] (CR) Ejegg: [C: 2] Update composer libs [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/309516 (owner: Awight) [17:19:05] it's a 220 breaker [17:19:25] yesterday i was going to insulate the ceiling [17:19:38] then i wondered why this one light wasn't working [17:19:53] spent the rest of the day replacing the switch, fixture, and all the romex [17:20:04] (Merged) jenkins-bot: Update composer libs [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/309516 (owner: Awight) [17:20:27] it finally worked when turned it on today [17:20:54] but then i stuffed all the wires back in and put the plates back on and stuff, and then when i turned it on the breaker popped [17:21:06] and now won't come back on [17:21:38] the water is still not working [17:21:48] things have moved significantly backwards since closing [17:23:00] yargh [17:23:28] i pulled all the wires back out and can't see anywhere it could have possibly shorted [17:24:21] you could check with a multitester [17:24:43] i'll have to rip all the insulation out to the breaker box [17:24:48] it will take a whole day [17:25:06] Is the 220 breaker feeding only 220 stuff, or is it split into one-phase 110 circuits? [17:25:15] yeah split into 110s [17:25:24] well there's also one 220 fixture on the circuit [17:25:29] that's prolly not good practice [17:25:40] I can't find any internets to back that statement up though [17:25:49] i don't know enough to say [17:25:54] but it did strike me as weird [17:26:38] I think people do that, e.g. getting 220 and 110 to an outbuilding. But I don't understand how the breaker will behave [17:27:08] but for the constant tripping, that does sound like a short--the breaker box is buried under insulation? [17:27:51] no but all the wires are [17:28:45] Here's some b.s. which seems reasonable, http://electrical.about.com/od/appliances/qt/Balancing-Electrical-Loads.htm [17:29:11] But the wires should be attached to terminal strips inside the breaker box, which you can access by taking the front panel off the box [17:29:43] some crap like, http://www.allstarselectricheatingandair.com/panel-breaker-box.jpg [17:30:21] ah right, only ground and neutral go to terminal strips, the others go straight into screw terminals on the breaker [17:31:43] (CR) Ejegg: "Oops, I thought I remembered a PS where you'd removed the whole querystring harvest, but I can't see it in this change's history! I think " [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/306852 (https://phabricator.wikimedia.org/T141487) (owner: Awight) [17:36:31] (PS15) Awight: Be less magical about unstaging order status things [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/306852 (https://phabricator.wikimedia.org/T141487) [17:36:43] (CR) Awight: "Good point!" [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/306852 (https://phabricator.wikimedia.org/T141487) (owner: Awight) [17:38:24] (CR) jenkins-bot: [V: -1] Be less magical about unstaging order status things [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/306852 (https://phabricator.wikimedia.org/T141487) (owner: Awight) [17:39:09] ack [17:39:17] (CR) Ejegg: "Gotta hand it to you, cracking open this function takes some guts! Noticing still more oddities." (2 comments) [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/306852 (https://phabricator.wikimedia.org/T141487) (owner: Awight) [17:42:50] (CR) Ejegg: [C: 2] Flesh out required modules for all projects in this repo [wikimedia/fundraising/tools] - https://gerrit.wikimedia.org/r/309080 (owner: Awight) [17:42:56] (Merged) jenkins-bot: Flesh out required modules for all projects in this repo [wikimedia/fundraising/tools] - https://gerrit.wikimedia.org/r/309080 (owner: Awight) [17:45:12] (PS16) Awight: Be less magical about unstaging order status things [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/306852 (https://phabricator.wikimedia.org/T141487) [17:45:30] "If you're part of this movements, don't just be web citin'!" [17:49:28] (PS17) Awight: Be less magical about unstaging order status things [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/306852 (https://phabricator.wikimedia.org/T141487) [17:50:03] (CR) Awight: Be less magical about unstaging order status things (2 comments) [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/306852 (https://phabricator.wikimedia.org/T141487) (owner: Awight) [17:51:03] (CR) Cdentinger: [C: 2] "nice" [wikimedia/fundraising/SmashPig] - https://gerrit.wikimedia.org/r/309523 (owner: Awight) [17:52:07] (Merged) jenkins-bot: Make constructor parameters optional [wikimedia/fundraising/SmashPig] - https://gerrit.wikimedia.org/r/309523 (owner: Awight) [17:54:58] (PS18) Awight: Be less magical about unstaging order status things [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/306852 (https://phabricator.wikimedia.org/T141487) [17:55:05] (PS12) Awight: Don't use frontend classes from fraud filters [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/306853 (https://phabricator.wikimedia.org/T141487) [17:55:21] awight: do you know what status 25 is? [17:55:39] not seeing it in the state transition diagrams [17:56:02] ejegg: Status 25 is when the donor has gotten an iframe but hasn't submitted it yet [17:56:12] I think it's documented in the webcollect appendix [17:57:02] Is that test invalid now? [17:57:09] The HostedMerchantLink transaction is waiting for the consumer [17:57:10] to enter missing data on the payment pages of GlobalCollect [17:57:20] ah wait which test? [17:57:46] oh testConfirmCreditCardStatus25? [17:57:50] Don't fraud-fail someone for bad CVV if GET_ORDERSTATUS comes back with STATUSID 25 and no CVVRESULT [17:57:53] yah [17:58:17] It's already covered by the base class test, and nothing in there is different for the orphan adapter now [17:58:29] (I should have commented in the commit msg...) [17:59:20] ah, cool, got it [17:59:46] There's probably a lot more cleanup to do in there, but I'm starting to really regret this entire segue... [18:00:16] like. I could have just swapped the queue out in the rectifier and called it a day [18:00:57] (CR) Ejegg: [C: 2] "This looks ready for some live fire exercises!" [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/306852 (https://phabricator.wikimedia.org/T141487) (owner: Awight) [18:01:30] It's a good opportunity to render some rancid oleoproteins, but much more so than I'd hoped [18:02:15] (CR) Ejegg: [C: 2] "Eminently sensible!" [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/306853 (https://phabricator.wikimedia.org/T141487) (owner: Awight) [18:02:36] (Merged) jenkins-bot: Be less magical about unstaging order status things [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/306852 (https://phabricator.wikimedia.org/T141487) (owner: Awight) [18:03:03] the rotten streak ran deep... [18:03:04] hehe, well said. The rest of the testing can be on live subjects [18:03:15] clinical trials [18:03:28] (Merged) jenkins-bot: Don't use frontend classes from fraud filters [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/306853 (https://phabricator.wikimedia.org/T141487) (owner: Awight) [18:04:40] https://images-na.ssl-images-amazon.com/images/I/710rKmbjOGL.jpg [18:05:40] nice... [18:07:53] looks like eileen's got a bunch in the review queue... [18:07:59] MBeat: Courtesy warning that I'm going to deploy some stuff that jiggles Ingenico fraud a bit. It isn't *supposed* to change anything, but in case of a bug, there's a possibility we'll be allowing or disallowing CVV/AVS differently [18:08:03] ejegg: good call. [18:08:26] ejegg: Which one you grabbing? [18:08:59] going to warm up my faculties with something easy, haven't even gauged sizes yet [18:09:19] hehe, okay I'll review T143062 [18:09:36] cool [18:09:38] feel free to re-review of course, it sounds like third opinions will be helpful! [18:14:53] (CR) Awight: [C: -1] "What about just removing placeholder emails from our database?" [wikimedia/fundraising/crm/civicrm] - https://gerrit.wikimedia.org/r/308678 (https://phabricator.wikimedia.org/T143062) (owner: Eileen) [18:15:45] (CR) Awight: "Hehe, I see the other patch now..." [wikimedia/fundraising/crm/civicrm] - https://gerrit.wikimedia.org/r/308678 (https://phabricator.wikimedia.org/T143062) (owner: Eileen) [18:15:46] ty awight [18:16:35] MBeat: I'll holler again when I'm ready to deploy [18:16:49] Thanks for keeping us honest! [18:18:57] awight: should i rebase the listener patch on top of your new ones? [18:19:11] but they are built on an old version...might be weird [18:19:28] i'll try to rebase those [18:20:08] nope [18:20:53] yeah i pushed a fairly large update late last week [18:20:59] cwd: I built those as patches on your work, so you could squash down into your patch [18:21:17] IMO you should keep your tests failing until the implementation is actually working [18:21:27] ... rather than have the test fake out the missing pieces [18:21:37] ah ok [18:21:52] "working" is kind of a moving target [18:22:33] (CR) Awight: "Great idea! Other placeholders we should search for and destroy:" [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/308677 (https://phabricator.wikimedia.org/T143062) (owner: Eileen) [18:22:38] hehe [18:23:05] Well either way, it's just a style thing at the end of the day [18:23:08] anyway i will merge the spirit of the new tests into the underlying changes [18:23:35] Personally I'd like my lights to turn green when things are happening [18:24:35] ok, i will try to be better about that [18:24:51] (CR) Awight: [C: 2] "Great idea!" [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/308677 (https://phabricator.wikimedia.org/T143062) (owner: Eileen) [18:25:31] I see what you all were saying about mosh. Awesome! [18:26:12] (CR) Ejegg: [C: 2] "A search for unnecessary inner joins on civicrm_email comes up clean. I think this is safe!" [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/308677 (https://phabricator.wikimedia.org/T143062) (owner: Eileen) [18:26:39] heck yeah mosh! [18:27:53] (Merged) jenkins-bot: Remove placeholder email addresses. [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/308677 (https://phabricator.wikimedia.org/T143062) (owner: Eileen) [18:28:01] i had to give up on mosh [18:28:10] awight: I'll run that update now unless you object [18:30:29] cwd: ooh, why? [18:30:41] ejegg: deleting nobodies? [18:31:14] Looks safe! [18:33:02] the stateless thing breaks some stuff [18:33:05] rockin [18:33:06] tunnels [18:33:08] bastions [18:33:17] oh huh [18:33:17] key forwarding [18:34:23] guess i've only used it to keep a connection open to the vps for irc. never had a problem with that [18:38:13] (PS1) Ejegg: Merge branch 'master' into deployment [wikimedia/fundraising/crm] (deployment) - https://gerrit.wikimedia.org/r/310057 [18:40:17] git commit --amen : expresses the sanctity of the patch [18:40:46] (CR) Ejegg: [C: 2] Merge branch 'master' into deployment [wikimedia/fundraising/crm] (deployment) - https://gerrit.wikimedia.org/r/310057 (owner: Ejegg) [18:40:56] (Merged) jenkins-bot: Merge branch 'master' into deployment [wikimedia/fundraising/crm] (deployment) - https://gerrit.wikimedia.org/r/310057 (owner: Ejegg) [18:43:19] !log updated civicrm from 93091637a0a68950efb8955a2cfef031d6ba7883 to 9cbee66e067f788b2634000a33c010dd3f4606b6 [18:43:24] Logged the message at https://wikitech.wikimedia.org/wiki/Server_Admin_Log, Master [18:46:56] oh boy, these merge patches. one line change, 2k lines of context to get up to speed on :) [19:09:07] (PS2) Ejegg: Fix for form merges not retaining the right email address. [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/308708 (owner: Eileen) [19:09:34] (CR) Ejegg: [C: 2] "Yep, definitely should be the contact ID here, not the contribution ID" [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/308708 (owner: Eileen) [19:13:33] (Merged) jenkins-bot: Fix for form merges not retaining the right email address. [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/308708 (owner: Eileen) [19:18:07] (CR) XenoRyet: [C: 2] "Does what it says, and I like the logic on what to keep." [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/309198 (https://phabricator.wikimedia.org/T145032) (owner: Eileen) [19:22:06] (Merged) jenkins-bot: Resolve conflicts where the name is the same but for lower case vs upper case variation. [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/309198 (https://phabricator.wikimedia.org/T145032) (owner: Eileen) [19:33:53] (PS13) Ejegg: Quit deleting from pending queue, stop saying limbo [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/301630 (https://phabricator.wikimedia.org/T133433) [19:49:12] fr-tech running way late, there in maybe 10 so pls don't wait up :( sorrrrrry! [19:52:10] (Abandoned) Eileen: Exclude placeholder email from dedupe queries [wikimedia/fundraising/crm/civicrm] - https://gerrit.wikimedia.org/r/308678 (https://phabricator.wikimedia.org/T143062) (owner: Eileen) [19:55:27] (CR) Eileen: "Thanks Dylan" [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/309198 (https://phabricator.wikimedia.org/T145032) (owner: Eileen) [20:06:30] Fundraising-Backlog: Impressions slow to update on jaJP dsk campaign - https://phabricator.wikimedia.org/T145446#2630343 (spatton) [20:09:00] Hey FR-Tech, just pinging this channel per Andy's request re: the phab task immediately before this message ^^^^ [20:13:51] spatton: thanks! [20:15:07] Fundraising-Backlog, MediaWiki-extensions-CentralNotice: CN Campaign Setting oddities - https://phabricator.wikimedia.org/T145447#2630364 (spatton) [20:15:27] AndyRussG: thank you in return! I appreciate the help [20:45:00] (PS2) Eileen: Add index to contribution_tracking. [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/307640 [20:46:04] ejegg: I just fixed the dumb typo I did in https://gerrit.wikimedia.org/r/#/c/307640/2/sites/all/modules/contribution_tracking/contribution_tracking.install [20:46:45] Fundraising Sprint Qwerty Thwacking, Fundraising-Backlog, Spike: Spike: Zendesk migration from Web Portal to Helpdesk - https://phabricator.wikimedia.org/T143142#2558203 (Ejegg) Looks like @awight, @MBeat33 and @Ppena have this pretty well figured out! They suggest we delete the spam from the web por... [20:47:01] thanks eileen, looking [20:48:58] (PS1) Eileen: CiviCRM submodule update [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/310085 [20:49:22] (CR) Eileen: [C: 2] "Submodule commit, self-approving" [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/310085 (owner: Eileen) [20:50:33] (CR) Ejegg: [C: 2] Add index to contribution_tracking. [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/307640 (owner: Eileen) [20:53:15] (Merged) jenkins-bot: CiviCRM submodule update [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/310085 (owner: Eileen) [20:53:48] thanks ejegg [20:54:38] (Merged) jenkins-bot: Add index to contribution_tracking. [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/307640 (owner: Eileen) [20:57:58] (PS1) Eileen: Merge branch 'master' of https://gerrit.wikimedia.org/r/wikimedia/fundraising/crm into deployment [wikimedia/fundraising/crm] (deployment) - https://gerrit.wikimedia.org/r/310088 [20:58:36] (CR) Eileen: [C: 2] "Merge to deploy - self- approving" [wikimedia/fundraising/crm] (deployment) - https://gerrit.wikimedia.org/r/310088 (owner: Eileen) [21:02:19] (Merged) jenkins-bot: Merge branch 'master' of https://gerrit.wikimedia.org/r/wikimedia/fundraising/crm into deployment [wikimedia/fundraising/crm] (deployment) - https://gerrit.wikimedia.org/r/310088 (owner: Eileen) [21:03:49] Fundraising Sprint Qwerty Thwacking, Fundraising-Backlog, Spike: Spike: Zendesk migration from Web Portal to Helpdesk - https://phabricator.wikimedia.org/T143142#2630717 (MBeat33) As far as I know no one in DS uses that query, @Ejegg, so I'm fine with removing it if no one in tech uses it. I just tur... [21:07:34] I'm going to deploy the latest code - the index add query might take a few mins but not so interuptive I think it needs an outage [21:10:28] (PS3) Ejegg: Use SmashPig config shortcut, reset Context [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/306972 [21:11:44] (PS1) Eileen: Revert "CiviCRM submodule update" [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/310136 [21:12:29] (PS2) Eileen: Revert "CiviCRM submodule update" [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/310136 [21:12:55] (CR) Eileen: [C: 2] "self-merging (fixing!) submodule commit" [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/310136 (owner: Eileen) [21:16:34] (Merged) jenkins-bot: Revert "CiviCRM submodule update" [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/310136 (owner: Eileen) [21:17:06] (PS1) Ejegg: Remove wmf_zendesk_reports [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/310137 (https://phabricator.wikimedia.org/T143142) [21:22:57] hmm I made a mistake committing a submodule update to civicrm for a patch that I then abandonned rather than was merged [21:23:30] I fixed that all up but because it got to deploy branch it's confusing the fundraising_update script [21:23:49] I feel like it will 'come right' next submodule update [21:23:53] but.... [21:24:04] Entering 'civicrm' [21:24:05] error: fatal: bad object 51079f32513ec33da20c78709a147dd39b8aa2bb [21:24:05] error: Stopping at 'civicrm'; script returned non-zero status. [21:24:05] error: fatal: reference is not a tree: 51079f32513ec33da20c78709a147dd39b8aa2bb [21:24:05] Submodule path 'drupal': checked out 'b4567f5222103fef9a5a17f7dbde91e635b3a792' [21:24:05] Submodule path 'vendor': checked out '6e7a83061dc375f7c2fa285be6c140965d7b6cd8' [21:24:06] error: Unable to checkout '51079f32513ec33da20c78709a147dd39b8aa2bb' in submodule path 'civicrm' [21:28:44] eileen: I think fundraising_code_update -p civicrm=rollback shouldwork to recover from that [21:31:14] thanks! [21:32:23] fr-tech review requests (small): [21:32:23] update CRM .gitignore - https://gerrit.wikimedia.org/r/309207 [21:32:23] refactor common duplicate contribution checking - https://gerrit.wikimedia.org/r/307427 [21:32:26] SmashPig defaults for more queues - https://gerrit.wikimedia.org/r/309181 [21:32:29] Use SmashPig config shortcut, reset Context - https://gerrit.wikimedia.org/r/306972 [21:33:45] (CR) Awight: [C: 2] Use SmashPig config shortcut, reset Context (1 comment) [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/306972 (owner: Ejegg) [21:33:55] Reject isRequeue messages after expiry - https://gerrit.wikimedia.org/r/306450 [21:34:00] thanks awight ! [21:35:06] (CR) Ejegg: Use SmashPig config shortcut, reset Context (1 comment) [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/306972 (owner: Ejegg) [21:35:10] (Merged) jenkins-bot: Use SmashPig config shortcut, reset Context [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/306972 (owner: Ejegg) [21:35:18] ejegg: that seems to have worked but now I see 9cbee66e067f788b2634000a33c010dd3f4606b6 (locked) [21:35:29] (CR) Awight: [C: -1] "Sorry, needs a tricky rebase now" (1 comment) [wikimedia/fundraising/SmashPig] - https://gerrit.wikimedia.org/r/309181 (owner: Ejegg) [21:35:35] (PS4) Awight: Defaults for more queues [wikimedia/fundraising/SmashPig] - https://gerrit.wikimedia.org/r/309181 (owner: Ejegg) [21:35:49] eileen: yep, that's to make sure rolled-back changes get noticed [21:36:03] once the deployment branch is in a good state, use -p civicrm=HEAD [21:36:15] cool I was just concluding that myself [21:36:16] thanks! [21:38:02] (PS5) Ejegg: Defaults for more queues [wikimedia/fundraising/SmashPig] - https://gerrit.wikimedia.org/r/309181 [21:38:18] ejegg: heads up, in-browser edited that one ^ [21:38:38] (CR) XenoRyet: [C: 2] Refactor common duplicate-checking code [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/307427 (owner: Ejegg) [21:38:56] (CR) Cdentinger: [C: 2] Update .gitignore [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/309207 (owner: Ejegg) [21:38:57] of *course* gerrit authors thought it was prudent to override how -F works [21:39:12] thanks all! [21:39:25] (PS6) Ejegg: Defaults for more queues [wikimedia/fundraising/SmashPig] - https://gerrit.wikimedia.org/r/309181 [21:39:26] awight: firefox? [21:39:43] it's just supposed to be F, but in FF somehow ctrl gets ignored [21:39:55] cwd: oh hah this is part of a browser war? I hadn't thought of that [21:40:05] click on the url and then ctrl+f :S [21:40:47] (CR) Ejegg: "Thanks for the fixup awight!" [wikimedia/fundraising/SmashPig] - https://gerrit.wikimedia.org/r/309181 (owner: Ejegg) [21:40:59] hmm this is still not working - it's being incremental about the submodule updates so having a wrong one then reverted is still messing with fundraising_update_code [21:41:27] Jeff_Green: Can you rm that directory for us? ^ [21:41:42] org.wikimedia.civicrm/civicrm I Think [21:42:10] gonna bounce for a bit, will check back in on things in a couple few hrs. [21:42:18] (Merged) jenkins-bot: Refactor common duplicate-checking code [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/307427 (owner: Ejegg) [21:42:23] awight: will that work? The problem is it has 2 commits to pull - one is a civicrm submodule update but to this abandoned patch [21:42:24] https://gerrit.wikimedia.org/r/#/c/308678/ [21:42:34] maybe I should revive & merge that? [21:42:44] then I can clean up & finally revert? [21:42:45] (Merged) jenkins-bot: Update .gitignore [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/309207 (owner: Ejegg) [21:43:21] eileen: oh are you saying that the issue is that it's doing a submodule update of old stuff before checking out the target revision? [21:43:30] so I did a submodule commit [21:43:34] on that patch [21:43:39] thinking it was merged [21:43:43] but thinking badly [21:43:58] & then realised & did a revert [21:44:24] but because it was never merged the fundraising_update can't get past the first commit to the merge [21:44:42] but, I think if I merge it & then revert that would be quite clean [21:44:57] * awight looks at repo state [21:45:49] Fundraising-Backlog, FR-Paypal: PayPal audit issue: reversals that don't reach Civi - https://phabricator.wikimedia.org/T144555#2630964 (MBeat33) Another type not reaching Civi as reversals are **canceled e-checks**. ID #9K340606GS2017717 is an example of one we identified as fraudulent in Civi, but when... [21:48:01] (PS1) Awight: Revert "CiviCRM submodule update" [wikimedia/fundraising/crm] (deployment) - https://gerrit.wikimedia.org/r/310143 [21:48:06] eileen: How about just ^ ? [21:49:19] awight: it's the same as https://gerrit.wikimedia.org/r/#/c/310136/ [21:49:26] but that didn't quite work :-( [21:49:44] oh weird, I looked for a commit like that and didn't see one! [21:49:47] hold on... [21:49:50] presumably because the commits still get processed in order [21:50:02] ah--did you merge that to the deployment branch? [21:50:23] was just checking that myself!!!! [21:50:26] Maybe now [21:50:27] not [21:50:31] There should be nothing like an order to commits, from the f_c_u tool's perspective [21:50:35] ok cool that's simple then [21:52:17] (PS1) Eileen: Revert "CiviCRM submodule update" [wikimedia/fundraising/crm] (deployment) - https://gerrit.wikimedia.org/r/310145 [21:52:58] (CR) Eileen: [C: 2] Revert "CiviCRM submodule update" [wikimedia/fundraising/crm] (deployment) - https://gerrit.wikimedia.org/r/310145 (owner: Eileen) [21:53:04] (Merged) jenkins-bot: Revert "CiviCRM submodule update" [wikimedia/fundraising/crm] (deployment) - https://gerrit.wikimedia.org/r/310145 (owner: Eileen) [21:53:05] yeah retrying [21:53:33] I cherry-picked it to deployment due to merge conflict on something else recently merged (just on .gitignore) [21:53:45] (and wanting to keep things simple) [21:54:38] !log from 3f01d93237c2b8af4fbda5e629a4a59c77dab3c1 to f381bd1c15e30e4d47fe372e128143baee6a7c7a [21:54:43] Logged the message at https://wikitech.wikimedia.org/wiki/Server_Admin_Log, Master [21:54:49] sounds good [21:54:49] Thash better [21:54:53] whew! [21:55:09] well good to make mistakes & have to learn about more about it [21:55:32] FrDeploy.pm is a good read, if you even need... [21:55:37] *ever [21:55:40] nighttime story [21:55:45] haha [21:56:14] if you like monster stories :p naw, it's actually nice Perl [21:56:40] I suspect JGreen of writing Perl since it was invented ;) [21:57:01] :-) [21:59:22] (PS1) Awight: Lighten contract for jobs, they only have to be Runnable now. [wikimedia/fundraising/SmashPig] - https://gerrit.wikimedia.org/r/310147 [21:59:44] cwd: we should be able to merge that before the followup work... [22:00:00] (CR) jenkins-bot: [V: -1] Lighten contract for jobs, they only have to be Runnable now. [wikimedia/fundraising/SmashPig] - https://gerrit.wikimedia.org/r/310147 (owner: Awight) [22:00:01] (which I'm start now anyway) [22:00:03] hehe [22:00:08] (PS17) Cdentinger: WIP smashpig paypal listener [wikimedia/fundraising/SmashPig] - https://gerrit.wikimedia.org/r/305761 (https://phabricator.wikimedia.org/T141654) [22:00:17] (CR) jenkins-bot: [V: -1] WIP smashpig paypal listener [wikimedia/fundraising/SmashPig] - https://gerrit.wikimedia.org/r/305761 (https://phabricator.wikimedia.org/T141654) (owner: Cdentinger) [22:00:21] awight: nice [22:00:50] (PS2) Awight: Lighten contract for jobs, they only have to be Runnable now. [wikimedia/fundraising/SmashPig] - https://gerrit.wikimedia.org/r/310147 [22:01:29] awight: i don't actually see the curl code in https://gerrit.wikimedia.org/r/#/c/306601/ -- am i looking in the wrong place? [22:01:37] or did you want me to just move the curl stuff into there? [22:03:41] cwd: the latter [22:03:49] if you agree it belong in a helper... [22:04:04] ah cool cool [22:04:08] yeah i think that makes sense [22:07:04] (CR) Cdentinger: Lighten contract for jobs, they only have to be Runnable now. (2 comments) [wikimedia/fundraising/SmashPig] - https://gerrit.wikimedia.org/r/310147 (owner: Awight) [22:08:26] Fundraising-Backlog, Wikimedia-Fundraising-CiviCRM: Fix soft credit to not require an email - https://phabricator.wikimedia.org/T145470#2631002 (Eileenmcnaughton) [22:09:42] (CR) Awight: Lighten contract for jobs, they only have to be Runnable now. (2 comments) [wikimedia/fundraising/SmashPig] - https://gerrit.wikimedia.org/r/310147 (owner: Awight) [22:10:16] (PS18) Cdentinger: WIP smashpig paypal listener [wikimedia/fundraising/SmashPig] - https://gerrit.wikimedia.org/r/305761 (https://phabricator.wikimedia.org/T141654) [22:11:02] (CR) jenkins-bot: [V: -1] WIP smashpig paypal listener [wikimedia/fundraising/SmashPig] - https://gerrit.wikimedia.org/r/305761 (https://phabricator.wikimedia.org/T141654) (owner: Cdentinger) [22:14:39] (CR) Cdentinger: [C: 2] "Looks great, will be happy to use this producer interface." (2 comments) [wikimedia/fundraising/SmashPig] - https://gerrit.wikimedia.org/r/310147 (owner: Awight) [22:14:59] Fundraising-Analysis, Wikimedia-Fundraising-CiviCRM: Add links to contact record from dedupe exceptions screen civicrm/dedupe/exception?reset=1 - https://phabricator.wikimedia.org/T145471#2631021 (Eileenmcnaughton) [22:21:55] (Merged) jenkins-bot: Lighten contract for jobs, they only have to be Runnable now. [wikimedia/fundraising/SmashPig] - https://gerrit.wikimedia.org/r/310147 (owner: Awight) [22:23:28] (PS19) Cdentinger: WIP smashpig paypal listener [wikimedia/fundraising/SmashPig] - https://gerrit.wikimedia.org/r/305761 (https://phabricator.wikimedia.org/T141654) [22:23:36] (CR) jenkins-bot: [V: -1] WIP smashpig paypal listener [wikimedia/fundraising/SmashPig] - https://gerrit.wikimedia.org/r/305761 (https://phabricator.wikimedia.org/T141654) (owner: Cdentinger) [22:43:28] (PS20) Cdentinger: WIP smashpig paypal listener [wikimedia/fundraising/SmashPig] - https://gerrit.wikimedia.org/r/305761 (https://phabricator.wikimedia.org/T141654) [23:01:23] (PS1) Awight: [WIP] Higher level serialization codec [wikimedia/fundraising/SmashPig] - https://gerrit.wikimedia.org/r/310163 [23:01:29] cwd: if you wanna take a peek? ^ [23:02:45] yessir [23:04:24] awight: my only question would be whether the ideal serialization needs any logic at all [23:04:31] hmm? [23:04:41] quite the gauntlet to throw :p [23:04:59] if it is just sticking an array in a store, what is being "serialized" ? [23:05:18] we can either assume the class of everything coming out of an array, or we include a type param [23:05:25] then yeah the parameters are an array [23:05:35] that's what the "Construction" style we were discussing would be. [23:05:50] I suspect that if we get rid of "type" we'll end up adding it again [23:06:24] ... but in some sketchy ad-hoc way like as a member of the payload array [23:07:09] If we don't have type, then all jobs need to be sent through their own queue, is what you're thinking? [23:08:22] i guess i was thinking the stored thing is multidimensional [23:08:37] 'classname' => 'class', 'args' => [] [23:09:30] I don't follow [23:09:30] ah. [23:09:30] ok that's the same deal then [23:10:09] what is the advantage of wrapping that format in a class? [23:11:13] I'm not! ok lemme kick out some of the function bodies and resend [23:11:28] hehe I do not like what I just added though-- a "serialization-format" key. [23:11:53] hmm, i must not be understanding [23:12:16] Fundraising-Backlog, FR-Paypal: PayPal audit issue: donation not in Civi - https://phabricator.wikimedia.org/T143903#2631264 (MBeat33) Transaction 71L83076GE8946146 Aug 31, 2016 06:44:55 PDT looks like a spurious charge to a recurring donor, though there's a small chance it was donor error. It reached Ci... [23:12:17] Now I need some way of telling the consumer what type of unpacking to do [23:12:39] awight: in order to handle the old format and new format? [23:12:39] I'd prefer to not do that, feel free to tell me why this is a bad idea [23:12:42] yeah [23:12:45] and future formats. [23:13:01] i guess ideally there'd be only one future format? [23:13:33] sure, but in case we need to rollback code across one of these migrations... [23:14:10] the only thing that seems fucky about the current one is the "meta" keys (e.g. php-message-class) are mingled right in with the data [23:14:33] yeah [23:14:56] I think that's actually caused by the queue overhaul, by accident. [23:15:19] isn't it the default smashpig serialization behavior for those storable objects? [23:15:47] I was just looking and think that's actually sent over the "headers" for Stomp queues [23:16:20] oh right... but we flatten them for redis? [23:16:37] yeah [23:17:15] is it too late to change that? [23:18:59] good question--yeah I think it's too late. [23:19:08] But I'm wrapping the fix up in a reusable place [23:23:55] Am I adding an extra layer of cruft? [23:25:19] awight: i doubt it, i'd bet i'm misunderstanding [23:25:27] huh, JsonSerializableObject already uses the syntax we're talking about [23:26:03] on third reading, no it does not [23:26:12] it actually circumvents a constructor and sets fields directly. [23:27:00] I was thinking, that *should* be provided, but as a class you can extend. Its contructor will populate own fields from a parameter. [23:27:17] nasty affair. [23:28:18] let's say for the sake of simplicity we keep that format for now [23:28:28] where the fields are flattened [23:28:38] I'm currently stabbing it ;) [23:28:51] Listener.php can write that format without extending anything [23:28:55] (PS1) Eileen: Clean up legacy junk data from preferred_language field. [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/310168 (https://phabricator.wikimedia.org/T96410) [23:29:13] and the hypothetical consumer can read it and execute the method it describes [23:29:14] * awight gives the piñata an extra thwack after own turn has clearly ended [23:30:01] would it be nicer if execute was static? [23:30:32] circumvent any instantiation and constructors [23:30:51] Fundraising Sprint Pretending This Isn't Happening, Fundraising Sprint Qwerty Thwacking, Fundraising-Backlog, Unplanned-Sprint-Work: Civi email merge won't retain new email address - https://phabricator.wikimedia.org/T143236#2561596 (Eileenmcnaughton) @CCogdill_WMF Can you confirm if this works n... [23:31:34] How do you pass it parameters then? Directly? [23:31:51] Do php interfaces do statics? I think so... [23:32:03] Inheriting from a thing with statics is balls tho [23:32:20] what would inherit from it? [23:33:07] donno :) crappy job subclasses... [23:33:15] yeah let's pretend I didn't say that [23:33:55] hehe [23:34:48] Fundraising Sprint Pretending This Isn't Happening, Fundraising Sprint Qwerty Thwacking, Fundraising-Backlog, Unplanned-Sprint-Work: Civi email merge won't retain new email address - https://phabricator.wikimedia.org/T143236#2631305 (CCogdill_WMF) Just merged two of my own records, and it retaine... [23:35:45] I'm just noticing a lot of $messageArray = json_decode( $message->toJson(), true ); in the code :( [23:36:04] (PS2) Eileen: {WIP} Clean up legacy junk data from preferred_language field. [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/310168 (https://phabricator.wikimedia.org/T96410) [23:36:06] yeah [23:36:26] the only db interface seems to be things that extend that strange json thing [23:37:19] Fundraising Sprint Pretending This Isn't Happening, Fundraising Sprint Qwerty Thwacking, Fundraising-Backlog, Unplanned-Sprint-Work: Civi email merge won't retain new email address - https://phabricator.wikimedia.org/T143236#2631324 (Eileenmcnaughton) Open>Resolved Yay [23:37:22] predis does arrays natively right? not sure if we even need to eff with json [23:37:29] that's right [23:37:40] but they're equivalent. [23:37:46] yeah [23:37:49] just less mess [23:37:53] json is handled by php-queue already so we shouldn't have to worry about that [23:38:04] it's JsonSerializableObject which is the wart [23:38:11] yeah [23:38:24] you sign up for a lot when you extend that baby [23:39:36] simplest possible case, merely for thought and not meant to be the solution [23:39:46] (PS3) Eileen: {WIP} Clean up legacy junk data from preferred_language field. [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/310168 (https://phabricator.wikimedia.org/T96410) [23:40:11] would be "producers" that just write arrays to redis, containing a static method name and arguments [23:40:27] and one big dumpster class of static methods, the "consumers" [23:41:04] does that sound vaguely on target? [23:45:18] That sounds cool--I notice it prevents us from doing anything with the job after it's executed, not that it matters yet. [23:45:54] what kind of things? [23:45:54] I also have a vague affection for being able to sanity-check the object that we pull from the \queue [23:46:52] Can't think of any valid uses :) [23:47:02] yeah we definitely want that. something needs to call pop() so maybe there? [23:47:30] I suppose it could sanity-check that the static function is callable [23:47:38] (PS4) Eileen: {WIP} Clean up legacy junk data from preferred_language field. [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/310168 (https://phabricator.wikimedia.org/T96410) [23:48:41] i'm wondering if a class to house each consumer function has a benefit [23:48:46] (PS2) Awight: [WIP] Higher level serialization codec [wikimedia/fundraising/SmashPig] - https://gerrit.wikimedia.org/r/310163 [23:49:08] cwd: We'll be writing that class anyway, is one way to look at it [23:49:21] how come? [23:49:28] (CR) jenkins-bot: [V: -1] [WIP] Higher level serialization codec [wikimedia/fundraising/SmashPig] - https://gerrit.wikimedia.org/r/310163 (owner: Awight) [23:49:49] cos executing a callback like that is probably its own responsibility, which should be encapsulated [23:50:20] Definitely are testability advantages to a first-class object, too... [23:50:24] ah yeah, i mean the callbacks themselves [23:50:32] do they each need a class? [23:50:32] but that's a poor argument in production design [23:50:44] yeah I think so [23:51:02] yeah I mean, each callback is its own job, and probably deserves a class [23:51:10] (PS5) Eileen: Clean up legacy junk data from contribution_tracking.language field. [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/310168 (https://phabricator.wikimedia.org/T96410) [23:53:00] (PS3) Awight: [WIP] Higher level serialization codec [wikimedia/fundraising/SmashPig] - https://gerrit.wikimedia.org/r/310163 [23:53:25] cwd: *that*'s the part I actually care about: https://gerrit.wikimedia.org/r/#/c/310163/3/Core/QueueConsumers/JobQueueConsumer.php [23:53:58] (CR) jenkins-bot: [V: -1] [WIP] Higher level serialization codec [wikimedia/fundraising/SmashPig] - https://gerrit.wikimedia.org/r/310163 (owner: Awight) [23:55:10] (PS4) Awight: [WIP] Higher level serialization codec [wikimedia/fundraising/SmashPig] - https://gerrit.wikimedia.org/r/310163 [23:55:45] (CR) jenkins-bot: [V: -1] [WIP] Higher level serialization codec [wikimedia/fundraising/SmashPig] - https://gerrit.wikimedia.org/r/310163 (owner: Awight) [23:56:47] (PS5) Awight: [WIP] Higher level serialization codec [wikimedia/fundraising/SmashPig] - https://gerrit.wikimedia.org/r/310163 [23:58:33] cwd: Hey on the bright side, note that none of the new queues or PendingDatabase use that KeyedOpaque madness [23:59:30] yeah that is excellent