[00:01:48] Reedy: around? [00:01:53] No [00:01:54] :p [00:02:05] spagewmf: https://github.com/cheezy/page-object/issues/128 [00:11:47] Krinkle, https://bugzilla.wikimedia.org/show_bug.cgi?id=45526 [00:12:08] superm401: already got 3 e-mails popping up in notification center [00:12:17] :) [00:12:25] Krinkle, well, I know you get CCed on every bug. [00:12:30] That doesn't mean you *read* every bug. :) [00:12:54] actually, I do get an e-mail for every action on bugzilla, but those go away in a filter. [00:13:00] the ones I'm actually CCed on I always read. [00:13:09] anyone here have IE? [00:13:23] I'm getting a report that this page doesn't redirect in IE: https://wiki.openstack.org/wiki/OpenStackUsersGroup [00:13:27] Krinkle, I didn't explicitly CC you. You were auto-CCed. [00:13:37] superm401: there is no distinction [00:13:41] can anyone check this [00:13:42] ? [00:13:54] Ryan_Lane: crossbrowsertesting.com, we have a shared wmf login [00:14:05] ah. cool [00:14:08] good idea [00:14:31] and you can sign up for saucelabs for free if you prefer a personal account [00:14:47] crossbrowsertesting.com goes via vnc, saucelabs does it in-browser with html5 [00:15:37] Ryan_Lane: http://cl.ly/image/1o0X1I0h172H [00:18:07] Krinkle: yep, I'm doing a live browser test [00:18:11] and it's working perfectly fine [00:18:14] thanks [00:19:00] yw [00:19:55] MaxSem: re solr: we had a meeting for figuring out what we want to use solr for. i'll ping you for the technical stuff later. [00:20:06] MaxSem: hm... can you tell me who to talk to for setting up an instance? [00:20:25] labs instance? [00:38:16] New patchset: Spage; "Generalize test to visit URL variants" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/51307 [00:39:56] DanielK_WMDE_, so what do you want with an instance? [00:42:59] Try to take over the world. [00:43:39] They're Reedy, they're Reedy and MaxSem, Sem, Sem, Sem, Sem. (dadadadada, da da) [00:49:31] New review: Cmcmahon; "Create Account test" [qa/browsertests] (master); V: 2 C: 2; - https://gerrit.wikimedia.org/r/51307 [00:49:31] Change merged: Cmcmahon; [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/51307 [00:59:41] Isarra: you were complaining that today was very argumentative -- considering that adam and I just had a 2 hour argument about our payments libraries -- I cannot help but bring new meaning to that statement [01:07:11] I... today... what? [01:07:42] mwalker|RAWR: Sorry, my brain kind of went to sleep and I'm not really sure what anything... any... uh. [01:08:00] I should know who Adam is, shouldn't I? [01:15:09] Isarra: meh; he's just another crazy fundraising guy [01:15:51] * marktraceur encourages mwalker|RAWR to take his anger out with nerf weapons....seriously, all talk and no action, it's been [01:15:51] You're all crazy. [01:16:20] Some day I would like to nerf you. [01:16:26] But yes, this is a good point. [01:16:45] man all this violence [01:16:57] mwalker|RAWR: Oh come on [01:17:00] It's like video games [01:17:04] It doesn't count [01:17:16] Violence is good. Embrace the violence. [01:18:43] I'm curious why marktraceur is trying to get fundraising to instigate the WMF war -- maybe we don't want to be the evil team that strikes pre-emptively [01:18:49] I sense a trap! [01:19:01] * marktraceur would totally accept the attacker position [01:19:20] Just need to get someone else to join the assault....everyone's bloody busy [01:19:32] Help me get a job and I'll join you. [01:19:33] Curse the WMF and their solid work ethic! [01:19:36] * Isarra grins. [01:20:24] why dont you attack during friday drinks? [01:20:28] no one is working then [01:20:31] Then again, currently I'd probably settle for coffee. Hard to prioritise when half asleep. [01:20:39] * Isarra collapses on mwalker|RAWR. [01:20:46] mwalker|RAWR: That's an excellent point; though most of the remoties will be gone by then methinks [01:20:58] * mwalker|RAWR collapses on marktraceur due to the unexpected weight [01:21:09] (and lack of lunch because I was silly) [01:21:19] hurm... you are correct [01:21:19] damn [01:21:27] * marktraceur doesn't understand what's happening but luckily has eaten and isn't tired [01:21:46] BEASTMAN! [01:23:33] Cowbell. [01:23:41] bwooong! [01:23:50] So... going to throw a war, everyone's invited, or what? [01:26:00] war, check; invitations, check [01:26:04] but you have to be in SF [01:27:53] This can be arranged. [01:28:28] Isarra: Visit SF just for a nerf war? And the subsequent guaranteed beverage-consumption? [01:28:31] SF doesn't have slaughterfish, does it? [01:29:01] Slaughterfish. [01:29:06] Slaughterfish! [01:29:12] * Isarra flails and runs away. [01:29:32] ... slaughterfish? [01:38:32] New review: Krinkle; "(1 comment)" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/51294 [01:42:21] Isarra: you making some sort of elderscrolls reference? [01:43:19] Not entirely. [01:43:25] I think... I'm cold. [01:43:51] * mwalker|RAWR hands isarra a blanket [01:44:40] I can also recomment cuddling with a laptop that's computing pi; and hot coccoa [02:45:53] Pie. [03:31:39] Punch. [18:49:37] YuviPanda : i was planing to make an iphone app for wiki editing (as far as i have researched there is none at the moment) , any tips ? [18:49:56] hey psych0der [18:49:58] hop on to #wikimedia-mobile and we'll talk about it :) [18:50:12] YuviPanda : sure :) [19:03:19] New patchset: Zfilipin; "Use https to access rubygems.org" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/51374 [19:03:19] New patchset: Zfilipin; "Remove x86-mingw32 platform" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/51375 [19:03:20] New patchset: Zfilipin; "Downgrade to page-object 0.8.5" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/51376 [19:04:13] New patchset: Hashar; "jobs for mw/ext/Agora" [integration/jenkins-job-builder-config] (master) - https://gerrit.wikimedia.org/r/51378 [19:04:27] Change merged: Hashar; [integration/jenkins-job-builder-config] (master) - https://gerrit.wikimedia.org/r/51378 [19:05:51] New review: Cmcmahon; "use https not http for gems" [qa/browsertests] (master); V: 2 C: 2; - https://gerrit.wikimedia.org/r/51374 [19:05:52] Change merged: Cmcmahon; [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/51374 [19:07:00] New review: Cmcmahon; "downgrade page-object and i18n for now to clear mysterious error on Jenkins host" [qa/browsertests] (master); V: 2 C: 2; - https://gerrit.wikimedia.org/r/51376 [19:07:34] New review: Cmcmahon; "remove unused gem" [qa/browsertests] (master); V: 2 C: 2; - https://gerrit.wikimedia.org/r/51375 [19:07:35] Change merged: Cmcmahon; [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/51375 [19:08:05] New review: Cmcmahon; "+2 for realz" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/51376 [19:08:06] Change merged: Cmcmahon; [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/51376 [19:10:19] New patchset: Hashar; "triggers for mw/ext/Agora" [integration/zuul-config] (master) - https://gerrit.wikimedia.org/r/51379 [19:10:48] Change merged: Hashar; [integration/zuul-config] (master) - https://gerrit.wikimedia.org/r/51379 [19:30:18] New patchset: Zfilipin; "Use https to access rubygems.org" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/51385 [19:31:15] New review: Cmcmahon; "https not http" [qa/browsertests] (master); V: 2 C: 2; - https://gerrit.wikimedia.org/r/51385 [19:31:16] Change merged: Cmcmahon; [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/51385 [20:33:14] i heard the tech chat is starting - is there a hangout/youtube link yet? [20:33:45] links in http://etherpad.wikimedia.org/linkpass [20:34:01] youtube not live yet [20:35:44] thanks Eloquence [20:37:46] how's the audio [20:37:51] should be live now [20:38:30] audio is fine here [20:46:24] hey all - I'm listening to Filipin now talking about Cucumber & browser tests [20:46:26] doesn't that mean that the dependency list only gets bigger and bigger? [20:46:46] sumanah, link is at the topic [20:47:07] (good to publish it, anyway) [20:47:27] * sumanah looks at https://www.mediawiki.org/wiki/Meetings/2013-02-28 [20:47:31] Why do we need to test cucumber? [20:48:00] it's good that we are told how to run the tests [20:48:57] QA weekly goals - next one is March 11th https://www.mediawiki.org/wiki/Project:Calendar [20:49:38] Reedy: i asked your question [20:49:51] sumanah: I think you missed my point ;) [20:50:03] Reedy: then you oughta ask the question you mean to ask ;) [20:50:15] Reedy: it's been sitting in the fridge for a while; might've gone bad. [20:50:18] xDD [20:58:21] yay we're killing gitweb [20:58:26] gitweb sux [20:58:28] aa that is the reason gravatar icons were visible for a while [20:58:41] http://www.gossamer-threads.com/lists/wiki/wikitech/331992 [20:58:44] gitblit! [20:58:53] I'm surprised by so many links I didn't notice when using the ui [20:58:56] question for chad - are there any plans for mingle integration with gerrit (like bugzilla)? if not and its something up for grabs, where can we find info for hacking that? [20:59:24] Custom dashboards in Gerrit on a per-project basis [20:59:34] awjr: What sort of integration? Auto linking? [20:59:44] Reedy at a minimum yeah [20:59:57] That's pretty simple to add [20:59:58] Gerrit has a stable, public, documented REST API [21:00:05] \o/ [21:00:23] within https://gerrit.wikimedia.org/r/Documentation/index.html [21:00:24] sweet [21:00:29] https://gerrit.wikimedia.org/r/Documentation/rest-api.html [21:00:46] I make my usual pitch for https://gerrit.wikimedia.org/r/Documentation/user-search.html although the search suggestions help a lot [21:01:40] awjr: Chad says, use the plugin interface [21:01:49] I also point people to tweak their settings [21:01:53] like, 100 items per page, etc. [21:02:09] I don't have the plugin link... [21:02:22] https://gerrit.wikimedia.org/r/Documentation/dev-plugins.html [21:02:26] from the Doc index [21:02:46] it would be cool if we could modify card status in mingle from gerrit commits, but at the very least autolinking would be great [21:03:07] autolinking should just be a few lines of copy, paste and replace [21:03:09] mingle does have an API, but, beware, old docs are the easiest to find [21:03:10] yah [21:03:18] code for BZ plugin is with upstream .... didn't catch directions there. [21:03:21] ^demon: ^ [21:03:41] gitblit - no time of delivery predicted [21:03:48] but it's a high priority [21:04:33] awjr: template/gerrit/gerrit.config.erb [21:04:37] in the puppet repo [21:04:38] [commentlink "bugzilla"] [21:04:38] match = "\\b([bB][uU][gG]\\:?\\s+#?)(\\d+)\\b" [21:04:38] link = https://bugzilla.wikimedia.org/$2 [21:04:40] actually autolinking for mingle will probably be trickier than bugzilla - card links looks lik: https://mingle.corp.wikimedia.org/projects/mobile/cards/430 [21:04:44] note /projects/mobile [21:04:55] There's RT and CodeREview etc [21:04:55] where 'mobile' is the specific project name [21:04:59] are those plugins created upstream or are some of them done by us? [21:05:42] "([a-zA-Z]+) card (\d+)" --> /projects/$1/cards/$2 [21:05:57] Reedy, there's also the plugin chad mentioned for posting the gerrit links to bugzilla [21:06:05] but snagging the project name for mingle is the challenge [21:06:38] oh, is ee what you did there marktraceur [21:06:43] awjr: So long as they're sane names it shouldn't be problematic [21:06:45] yeah [21:06:56] i guess it just depends on setting a sane convention [21:06:59] [commentlink "mingle"] [21:06:59] match = "\\b([mM][iI][nN][gG][lL][e]\\:?\\s+) ([a-zA-z] (\\d+)\\b" [21:06:59] link = https://mingle.corp.wikimedia.org/projects/$2/cards/$3 [21:07:03] mingle mobile 430 [21:07:19] probably want to drop off the : etc [21:07:33] "mobile card 430" would probably be more prose-y, Reedy [21:07:35] <^demon> Platonides: Most were started by others (except bugzilla and github-create). [21:07:39] match = "\\b([mM][iI][nN][gG][lL][e])\s+([a-zA-z]\s+(\\d+)\\b" [21:07:43] some of the mingle project names are obnoxious [21:07:50] <^demon> But we've had our hands in most of them. Deleteproject was started by someone else, but I've done a lot of work on it. [21:07:55] anyway, doable. [21:08:02] where are those stored? [21:08:12] Platonides: template/gerrit/gerrit.config.erb [21:08:14] oh these templates are gonna scare me [21:08:15] puppet [21:08:50] <^demon> Platonides: https://gerrit-review.googlesource.com/ [21:09:01] <^demon> Projects beginning in plugins/* are what you're looking for. [21:09:06] ok [21:11:00] <^demon> There's no real distribution of plugins yet, so they have to be built by hand :\ [21:12:58] comparison with {{str len}} isn't too fair, since there's Extension:StringFunctions (not enabled) [21:13:37] it's fair enough since those templates were live :D [21:13:53] nobody enabled Extension:StringFunctions [21:14:20] lol [21:17:29] sure, then we would be showing screenshots of the Extension:StringFunctions version of the convert template [21:17:56] or the citation template, which would have been pretty much the same as without StringFunctions [21:18:14] they did really crazy things [21:19:32] I was impressed some time ago when I saw that they did with timezones [21:19:40] they moved them into template-dbs [21:20:02] then extracted then by the zone name into a table which was like the original data [21:20:18] but then it didn't work for them because they had too many transclusions [21:20:33] yet, what they wanted would have been solved by just pasting the original data! [21:22:30] how did the name 'scribunto' come about? [21:22:45] "We write", IIRC [21:22:52] https://en.wiktionary.org/wiki/scribunto [21:22:54] i see. [21:23:10] "they shall write" [21:23:14] Curse latin conjugation. [21:23:21] they shall [21:23:31] so it's a goal statement and a cool name in one [21:23:35] build it and they will come [21:23:36] it sounds like it should be accompanied by maniacal laughter [21:23:47] "they shall write.... yes, yes! they shall!" [21:23:59] with templates they wrote too much [21:24:03] :) [21:24:21] the name suggested by victor was "scripts" or something generic and boring like that [21:24:46] so I did a bit of surfing in wiktionary looking at the cognates of the word "script" for inspiration [21:24:46] right, scripts is totally boring [21:24:53] it's hard to get good names [21:24:55] you got a cool one [21:26:05] <^demon> You can always name things after furniture designers. [21:26:32] the repository should be accessible from the outside [21:26:57] having hundreds of wikis forking enwiki is not the Right Thing to Do [21:28:21] Platonides: cross-site code execution is a giant bag of giant worms with giant teeth [21:33:29] anomie: http://en.wikipedia.org/wiki/Module:Yesno fwiw, but see the note in the talk page [21:37:00] lol, sumanah [21:37:12] hmm, she disconnected [21:37:25] She does that. [21:37:39] * marktraceur doesn't see her in the immediate vicinity, sorry [21:38:38] don't worry marktraceur [21:39:42] ori-l, there may be an official process for publishing modules to the central repository [21:40:07] so only the best modules get there [21:40:12] and not the crappy ones [21:40:31] which would also help to get modules with good interfaces [21:56:50] Platonides: oh you there [21:56:57] Platonides: so I reverted your "no email" patch [21:57:05] Platonides: I want it to be generic for all tests [21:57:15] will just copy paste your code in the phpunit.php wrapper [21:58:41] than do both in one change :P [21:58:54] *then [21:59:05] that gerrit discussion is silly btw [21:59:35] silly ? [22:00:04] Etherpad labs server going down for upgrade now [22:02:47] hashar, unproductive [22:03:01] and why did you merge it?? [22:03:27] cause I voted -1 that change [22:03:40] and that only fix that particular issue when I want it to be generic [22:04:00] I voted -2 your revert, and you overrode it anyway [22:04:09] twic [22:04:13] twice [22:04:19] twice what? [22:04:44] you removed the -2 twice [22:06:17] I wouldn't mind if you had moved the code into phpunit.php [22:06:27] heh, I probably would have approved it myself [22:06:43] but I don't like a revert with little reason [22:06:51] it pollutes the history [22:07:07] moreover, now those tests should be failing [22:07:48] I don't think you should have removed the -2 [22:08:26] if you didn't like that I didn't like your revert of my change, you should have asked another developer to weight in [22:08:46] sorry [22:08:47] so [22:08:51] here is my POV [22:09:04] some change has been submitted despite my -1 on it requesting to make it a generic fix [22:09:14] I revert the change that should never have landed [22:09:21] then submit a new change that does what has been requested [22:09:30] aka in phpunit.php do something like: [22:09:31] Hooks::clear( 'AlternateUserMailer' ); [22:09:44] Hooks::register( 'AlternateUserMailer', function () { return false; } [22:10:00] someone else approved the code despite your -1 [22:10:04] also note it was not a -2 [22:10:18] it fixed the issue [22:10:26] it was only a partial fix? granted [22:10:34] but then you should make the wider change [22:10:42] not revert it before adding the new one [22:11:03] there was nothing wrong in that code [22:11:50] if a piece of information was added in the article body while it should go in the infobox [22:11:56] that is what I said [22:11:59] I am working on the fix [22:12:08] do we revert the edit, then make another one to add it to the infobox? [22:12:09] no [22:12:32] the "fix" (move the code to phpunit) should have be one changeset [22:12:43] not one revert + one feature addition [22:13:01] gerrit problem [22:13:07] see https://gerrit.wikimedia.org/r/#/c/51394/ [22:13:08] again this annoying [22:13:12] according to your logic [22:13:17] should I have reverted https://gerrit.wikimedia.org/r/#/c/51302/ [22:13:27] then readded it with one extra wfProfileOut ? [22:13:38] that's silly [22:14:10] it would have been preferable to fix it in the original commit [22:14:15] but now that it's into core [22:14:19] just fix it [22:14:27] Etherpad lite back up and on latest stable version [22:14:48] it's not like this was a very complex code [22:15:19] and that you autoapproved it doesn't help [22:17:59] maybe we should create a wikitech thread of 'How to do reverts' [22:19:22] easy [22:19:26] press revert button [22:19:26] submit [22:19:27] period [22:19:42] Platonides: https://gerrit.wikimedia.org/r/51576 [22:19:45] that is the generic version [22:20:27] Platonides: I have made you the author of that change :-] [22:20:32] since I merely copy pasted your code [22:20:38] and yeah [22:20:43] hashar, the revert gets stored in the history [22:20:49] it's polluting for nothing [22:20:53] I should probably have sent one change having both the revert and the generic alternative [22:21:07] but then if the generic alternative ends up being reverted, the revert will no more be there [22:21:27] which would be good [22:21:38] since the parser tests currently are wrong [22:21:45] then that means we will still have the original patch which I do NOT want [22:21:47] as they shouldn't send an email [22:23:10] and I'm having this discussion with you, which complained that I split a change into c37510 and c37511 [22:23:32] meeting time [22:23:41] as penance, you approve them [22:24:02] (I don't know what's your big issue with the original patch, btw) [22:26:36] TimStarling: you had mentioned to pgehres that 'City' probably wasn't the best field label for our donation forms in AU. We have a patch that changes it to 'Suburb / Town' but that doesn't feel right to me either -- do you have any thoughts? [22:27:42] New patchset: Platonides; "PEAR mail appeared twice." [mediawiki/tools/code-utils] (master) - https://gerrit.wikimedia.org/r/48245 [22:27:42] New patchset: Platonides; "Add support for anonymous functions (closures)." [mediawiki/tools/code-utils] (master) - https://gerrit.wikimedia.org/r/37510 [22:27:42] New patchset: Platonides; "Add one indentation level to the body of the do while added in 051f1853." [mediawiki/tools/code-utils] (master) - https://gerrit.wikimedia.org/r/37511 [22:27:43] New patchset: Platonides; "Improve handling of extension_loaded()" [mediawiki/tools/code-utils] (master) - https://gerrit.wikimedia.org/r/48242 [22:27:43] New patchset: Platonides; "Improve checks of wfProfileIn-wfProfileOut." [mediawiki/tools/code-utils] (master) - https://gerrit.wikimedia.org/r/42960 [22:27:43] New patchset: Platonides; "Separate the warnings for not running wfProfileOut() before throwing an exception." [mediawiki/tools/code-utils] (master) - https://gerrit.wikimedia.org/r/42961 [22:27:44] New patchset: Platonides; "Readd $runningQueuedFunctions" [mediawiki/tools/code-utils] (master) - https://gerrit.wikimedia.org/r/48254 [22:27:44] New patchset: Platonides; "There are too many false positives on deprecated-might due to homonym methods." [mediawiki/tools/code-utils] (master) - https://gerrit.wikimedia.org/r/42964 [22:27:44] New patchset: Platonides; "Do not complain of the use of User::decodeOptions in User.php:1089" [mediawiki/tools/code-utils] (master) - https://gerrit.wikimedia.org/r/47800 [22:27:45] New patchset: Platonides; "dirname(dirname(__DIR__)) and JobQueueAggregatorRedis.php" [mediawiki/tools/code-utils] (master) - https://gerrit.wikimedia.org/r/51033 [22:27:59] New patchset: Platonides; "Whitelist getcwd()" [mediawiki/tools/code-utils] (master) - https://gerrit.wikimedia.org/r/50590 [22:27:59] New patchset: Platonides; "Take into account files which include another file that then dies." [mediawiki/tools/code-utils] (master) - https://gerrit.wikimedia.org/r/50589 [22:28:00] New patchset: Platonides; "Whitelist phpversion()" [mediawiki/tools/code-utils] (master) - https://gerrit.wikimedia.org/r/51027 [22:28:00] New patchset: Platonides; "Check the whole history of included files for Maintenance.php" [mediawiki/tools/code-utils] (master) - https://gerrit.wikimedia.org/r/50918 [22:28:00] New patchset: Platonides; "Add support for dirname( __DIR__ ) to find-entries." [mediawiki/tools/code-utils] (master) - https://gerrit.wikimedia.org/r/50117 [22:28:01] New patchset: Platonides; "Instead of having multiple $cliSapiAutomaton, make it an array." [mediawiki/tools/code-utils] (master) - https://gerrit.wikimedia.org/r/50707 [22:28:01] New patchset: Platonides; "Use stream_resolve_include_path to detect files loaded from pear" [mediawiki/tools/code-utils] (master) - https://gerrit.wikimedia.org/r/50055 [22:28:01] New patchset: Platonides; "Add a couple of automatons" [mediawiki/tools/code-utils] (master) - https://gerrit.wikimedia.org/r/50708 [22:28:02] New patchset: Platonides; "Allow whitelisting additional functions on the command line." [mediawiki/tools/code-utils] (master) - https://gerrit.wikimedia.org/r/50813 [22:28:02] New patchset: Platonides; "Allow a php shebang" [mediawiki/tools/code-utils] (master) - https://gerrit.wikimedia.org/r/50812 [22:28:02] New patchset: Platonides; "Accept arbitrarily long lists of dirname() calls" [mediawiki/tools/code-utils] (master) - https://gerrit.wikimedia.org/r/50811 [22:28:03] New patchset: Platonides; "Skip the $evaluatedFiles cache for Benchmarker.php" [mediawiki/tools/code-utils] (master) - https://gerrit.wikimedia.org/r/50122 [22:31:49] marktraceur: have you finished the ether pad lite upgrade ? :-) [22:32:19] hashar: I have! [22:33:51] marktraceur: Did you also turn it off an on again? [22:34:30] Reedy: Yes, did that many times [22:36:27] marktraceur: nice -:] Thanks for taking care of that installation!!! [22:37:05] ^demon: poke poke [22:37:08] meeting time [22:38:16] brion: good to see you at a mediawiki meet up :-] [22:38:18] \O/ [22:38:26] 🍪 🍪 🍪 [22:39:25] :D [22:40:01] whatg has Platonides done O_o [22:42:36] Pushed all of the commits [22:43:02] heh [22:47:54] hi, can someone of you experts have a look to this problem please https://bugzilla.wikimedia.org/show_bug.cgi?id=45586 [22:48:31] hashar: Oh, there is one thing I also wanted to do on Jenkins if you have time - use the test results of a child job as the results of the parent. [22:48:59] marktraceur: I think there is an option for it [22:49:05] something like "aggregate downstream results" [22:49:07] or something [22:49:08] cant remember [22:49:18] must be some random checkbox that need to be ticked in the parent job [22:50:10] hashar: Yes, I see that now [22:50:30] that is supposed to make the parent ask its children for their tests results [22:50:36] hashar: But both of the jobs I run have the same name, and I only want to aggregate one of them....should I just split them? [22:50:37] and that *might* properly aggregate them [22:50:41] you will have to test it out :-] [22:50:44] Hrm. [22:50:51] ahh [22:50:51] yeah hmm [22:51:21] hashar: "Activate Chuck Norris"? [22:52:39] marktraceur: oh no, I should remove that plugin :-] [22:52:46] marktraceur: that is an easter egg plugin [22:52:59] I just don't know how to remove plugins haha [22:53:33] so yeah aggregating both jobs is going to be problematic :( [22:55:03] hashar: I can live with it. The idea is to show long-term trends anyway, so having both is fine. [22:56:07] marktraceur: if you want a dashboard, you could have a job that it is only triggered on merges [22:56:20] that would also let you spot regression that lands in `master` [22:56:44] * hashar wishes that MediaWiki parser had support for `foo` ==> foo [22:57:18] I think we have, or had, that [22:57:30] Not the parser thing [22:57:32] The merged thing [22:58:07] that will be way easier to setup :-] [22:58:12] Yuuup [22:58:12] since the changes are most probably linear [22:58:17] hashar: https://integration.mediawiki.org/ci/view/Parsoid/job/Parsoid-parserTests-merged-regressions/ [22:58:25] ahhh [22:58:38] Using old-style Gerrit interface [22:58:40] of course you don't want to use the Gerrit Trigger plugin anymore :-] [22:58:57] hashar: "want" is a strong term. I can't do anything with Zuul unless you configure it for me! [22:59:07] yeah that is true [22:59:23] though you could be brave and look at integration/zuul.git : layout.yaml :-] [22:59:23] * marktraceur passive-aggresses with the best of them [22:59:25] and add it up there [22:59:29] Righto [22:59:40] I just want you to know you've brought it upon yourself [22:59:52] eyah [22:59:55] Hey; where's the best place to ask a technical wiki coding question? [23:00:03] I will have to look up how to trigger a post merge job [23:00:11] I think there are some setup for mediawiki/core already [23:00:20] xen: "Wiki coding" meaning what? Extension development? [23:00:47] Well, usage, but I think I may have a bug on my hands [23:00:49] xen: either here (most of Wikimedia developers are there) or #mediawiki which his the primary channel for any wikiq uestions [23:01:08] xen: for user support you can try out in #wikipedia which regroups ton of experimented users [23:01:27] experienced, too :) [23:01:53] I've asked in #mediawiki but no one's answered [23:01:53] Ahh thanks [23:02:22] Heheheh, I'll try there [23:02:29] Thanks fellas [23:04:33] hashar: https://gerrit.wikimedia.org/r/gitweb?p=integration/zuul.git isn't the right repository, is it? [23:05:12] marktraceur: oh no that is zuul itsef [23:05:18] marktraceur: integration/zuul-config.git is [23:05:26] Ah. [23:05:40] ah [23:05:46] for merge you want to use the post-merge pipeline [23:05:58] K [23:06:07] so under the mediawiki/extensions/Parsoid part [23:06:13] postmerge? [23:06:16] you already have 'check' 'test' and 'gate-and-submit' [23:06:22] you just need to add a postmerge entry [23:06:28] that will run mwext-Parsoid-jslint [23:06:41] I am in a meeting right now though [23:07:05] K [23:09:38] marktraceur: yeah postmerge will do [23:09:43] with just the job you want to trigger [23:09:51] the mediawiki/core has it [23:10:05] that is a nice way to have regression dashboard [23:10:22] maybe one day I will make it to send some email or write to a wall of shame [23:12:52] Is there any easy way with gerrit to get a list of recent things someone has done there? [23:13:05] Also how did I wind up in this channel? [23:14:31] Isarra: You were here yesterday too. [23:14:46] Oh. [23:15:35] Isarra: you can do reviewer: [23:15:54] or owner: [23:16:09] (this is in the search box) [23:17:33] New patchset: MarkTraceur; "Add Parsoid postmerge regression testing" [integration/zuul-config] (master) - https://gerrit.wikimedia.org/r/51587 [23:17:48] hashar: That may be correct [23:18:38] marktraceur: I usually have those jobs named something-regression [23:18:46] Hrm. [23:18:53] but that is not much of a problem hehe [23:19:02] since you have a Parsoid view already https://integration.mediawiki.org/ci/view/Parsoid/ [23:19:18] regressions jobs: https://integration.mediawiki.org/ci/view/Regressions/ [23:19:29] hashar: I'm changing it, actually [23:20:00] New patchset: MarkTraceur; "Add Parsoid postmerge regression testing" [integration/zuul-config] (master) - https://gerrit.wikimedia.org/r/51587 [23:20:37] hashar: Can I add jobs to the Parsoid dashboard post-creation? [23:20:59] marktraceur: sorry I don't understand your question [23:21:08] Ooh, found it [23:21:10] Never mind [23:21:46] Change merged: Hashar; [integration/zuul-config] (master) - https://gerrit.wikimedia.org/r/51587 [23:21:49] deploying [23:21:58] Cool, thanks [23:22:31] zuul reloading [23:22:51] marktraceur: if you ever need to have a change deployed on the zuul server, the doc is at https://www.mediawiki.org/wiki/Continuous_integration/Zuul#Changing_the_configuration [23:23:07] basically just need someone to git pull on the server and reload the service [23:23:07] Noted [23:23:19] Chad would be able to do so [23:23:22] as well as Timo [23:23:29] or any ops)