[00:05:23] <^demon> You can leave comments on whole files now I believe. [00:06:19] yes, on https://gerrit.wikimedia.org/r/#/c/50004/1/UiFeedback.php [00:06:29] not https://gerrit.wikimedia.org/r/#/c/50004/1/UiFeedback.alias.php [00:06:51] he's not listening [00:09:11] Wait, so he made his honours thesis dependant on us reviewing and deploying his code? [00:09:19] That sounds like a stupid decesion [00:10:21] All the \r are annoying me [00:10:38] 'version' => '0.00001',\r [00:10:43] * @version 0.2\r [00:10:44] gj [00:10:52] Windows linebreaks.. [00:11:30] Indeed [00:11:36] Long live svn:eol-style native [00:13:19] We should mention https://help.github.com/articles/dealing-with-line-endings somewhere [00:14:40] step 1: throw away Windows and never use it [00:14:56] :) [00:16:39] :D [00:17:40] You know, people always criticize linux because they want something that Just Works(tm) - but when it comes to git, it seems linux is the one that Just Works (tm) [00:18:29] git just works on my mac :) [00:18:39] git on windows works [00:18:46] git review however.. [00:19:13] git review works on windows [00:19:21] ... or at least it does for me [00:19:28] people seem to have problems setting up even just normal git on windows [00:19:41] (Not that I normally use it on windows!) [00:20:34] Free as in you're free to try and make it work yourself ;) [00:23:02] i've used git via github's windows package and via msysgit, seems ok generally… but i haven't used git-review on windows [01:01:30] Reedy, this guy is using nyan cat and ascii art text in his code, for his bachelor thesis... Those are the bits which amaze me, I can forgive not being aware of MediaWiki's API system [01:01:48] Really? [01:01:54] It's not like it's hidden or anything [01:01:58] Well. [01:01:59] Sort of. [01:02:12] https://www.google.co.uk/search?q=mediawiki+api&aq=f&oq=mediawiki+api&aqs=chrome.0.57j60l3j0l2.2262&sourceid=chrome&ie=UTF-8 [01:02:15] googling is hard, I know [01:02:26] Okay fine, that's also amazing then :p [01:03:02] Also, to pass a degree, you don't necesserily need anything really to show [01:03:09] just bullshit about what you learnt, etc [01:04:24] particularly at the bachelor level [01:05:10] (or perhaps that's just me feeling like there is a lot of busy work keeping me very very busy but that is not that useful, in my current bachelor program ;) [01:05:26] so he could potentially get it regardless of this... although if it's dependent on deployment by wikimedia then almost certainly not at it's present state [01:06:01] That would be a very strange criteria [01:06:12] Completely outside his control [01:06:16] talk about a big risk [01:07:11] It's usually just about jumping through hoops [01:07:38] Yeah but it seems to be exactly what he's doing, assuming I'm reading his email correctly... [01:10:26] Does feel like wheel re-inventing too [01:12:06] Not to mention if he is planning to use data gathered from this extension in his thesis, even if it was deployed today, 2 weeks is not a long time to get the data and write it up [01:13:42] I'm guessing he's had no/little review from the wikidata guys too [01:31:23] New patchset: Krinkle; "JSHint: Upgrade from 0.9.1 to 1.1.0" [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/53503 [01:31:37] Change merged: Krinkle; [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/53503 [01:40:09] New patchset: Krinkle; "Sync: Commit deletion of jenkins job." [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/53506 [01:40:31] New review: Krinkle; "That's what we have version control for :)" [integration/jenkins] (master); V: 2 C: 2; - https://gerrit.wikimedia.org/r/53506 [01:40:32] Change merged: Krinkle; [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/53506 [02:03:24] New patchset: Krinkle; "Update and clarify some comments." [integration/zuul-config] (master) - https://gerrit.wikimedia.org/r/53509 [02:03:54] New patchset: Krinkle; "Update and clarify some comments." [integration/zuul-config] (master) - https://gerrit.wikimedia.org/r/53509 [02:04:03] Change merged: Krinkle; [integration/zuul-config] (master) - https://gerrit.wikimedia.org/r/53509 [04:26:22] New review: Hashar; "\O/" [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/53503 [04:26:33] Krinkle: congrats on upgrading JSHint :] [04:27:12] Krinkle: if they ever release Release Candidates, we could use them to help test out their final releases [04:50:12] New patchset: Reedy; "Bug 42756 - Add labsconsole extensions to make-wmf-branch" [mediawiki/tools/release] (master) - https://gerrit.wikimedia.org/r/53519 [05:56:21] New patchset: Hashar; "(bug 31518) PHPUnit code coverage for mw/core" [integration/jenkins-job-builder-config] (master) - https://gerrit.wikimedia.org/r/53532 [06:20:29] New patchset: Krinkle; "Update files and add doc.wikimedia.org" [integration/docroot] (master) - https://gerrit.wikimedia.org/r/53533 [06:20:42] Change merged: Krinkle; [integration/docroot] (master) - https://gerrit.wikimedia.org/r/53533 [06:22:47] New patchset: Mwjames; "Add qunit job template for extensions" [integration/jenkins-job-builder-config] (master) - https://gerrit.wikimedia.org/r/53534 [06:22:52] New review: Krinkle; "Merging this as it works and is closest to status quo, we can improve it later." [integration/jenkins] (master); V: 2 C: 2; - https://gerrit.wikimedia.org/r/39212 [06:24:29] New patchset: Mwjames; "Add Qunit tests for Semantic Mediawiki" [integration/zuul-config] (master) - https://gerrit.wikimedia.org/r/53535 [06:25:58] New review: Krinkle; "I don't know what the problem is here. You keep bringing up some kind of problem with zuul's post fe..." [integration/jenkins-job-builder-config] (master) - https://gerrit.wikimedia.org/r/39210 [06:27:27] New review: Krinkle; "As mentioned on I67cf7405, I have no idea how that bug or the fetching of tags are relevant here. We..." [integration/zuul-config] (master) - https://gerrit.wikimedia.org/r/44384 [07:19:03] what was that magic comment that made jenkins re-run 2nd level review? The unit tests timed out for https://gerrit.wikimedia.org/r/#/c/53530 [07:19:25] "recheck" ? [07:24:15] New patchset: Hashar; "(bug 31518) PHPUnit code coverage for mw/core" [integration/jenkins-job-builder-config] (master) - https://gerrit.wikimedia.org/r/53532 [07:26:45] yurik: recheck only run the lint checks [07:27:02] hashar, what about unit tests? [07:27:08] yurik: if you want to retrigger the unit tests you need a new patchset :-/ [07:27:16] bleh! [07:27:41] not a big deal [07:27:48] you can skip them for that kind of change [07:27:54] they will be run before merge anyway [07:27:56] reviewing :-] [07:28:17] hashar, i changed commit message - easy enough, no need to do git-review :) [07:28:28] its recheking now [07:28:31] legoktm, thanks [07:28:46] yurik: oh did you ran the spell checker on comment only ? [07:28:47] np [07:29:10] hashar, i ran them on everything, but changed only comments & a few wfDebug() strings [07:29:14] * YuviPanda waves at hashar [07:29:31] oh, and a few docs [07:29:53] yurik: maybe we can add the spell checker as a jenkins job :-] [07:29:54] ... at least i hope so [07:30:01] hashar, unreal [07:30:06] too many custom words [07:30:12] i already have a dictionary of ~5000 [07:30:21] do [07:30:24] doh [07:30:28] * hashar waves at yurik  [07:30:30] grrr [07:30:31] YuviPanda: [07:30:32] hehe [07:30:50] i do want to commit that monster list ~50KB to some dir in core [07:30:56] haha :D [07:31:04] this yurik confusion will keep happening I guess :) [07:31:25] phpstorm has a good integrated spellchecker [07:31:32] all it needs is an extra dict [07:31:43] YuviPanda, i just can't change it :D [07:31:50] yeah, me neither :D [07:31:57] but it's going to be fun :) [07:32:12] hashar: had time to play with the commons app after that? :) [07:32:13] yurik has been yurik since 2006 i think [07:32:25] hah, YuviPanda has been YuviPanda since before that1 [07:32:26] :D [07:32:29] YuviPanda: not at all. But I got a shinny new camera :-] [07:32:37] oooh, what camera? [07:33:28] some entry level canon DSLR. Canon EOS 650D (or in US Digital Rebel 4) [07:33:48] with a sigma 17-55 opening at 1/2.8 on all the focal range [07:34:11] YuviPanda, just checked - first edits around 20th of march, 2005 [07:34:17] can't claim anything earlier :) [07:34:35] hashar: ah, that's no entry level one :) [07:34:41] that's two levels better than mine [07:34:47] (1100D Is entry level) [07:34:50] but :) [07:34:56] yurik: https://gerrit.wikimedia.org/r/#/c/53530/3/includes/api/ApiMove.php,unified -> replaced Exampel with Exxaammppllee [07:35:16] yurik: hmm, I started using YuviPanda on Yahoo Chatrooms, Programming:3 I think. [07:35:24] but that's okay, there's enough space for all of us :) [07:35:33] hashar, that's an api self documentation [07:35:39] so that's a comment too :D [07:36:48] still that looks wrong to me :] [07:36:51] hehe :) YuviPanda is definitly cooler name choice [07:36:58] because its more unique [07:37:09] there are other yuriks out there [07:37:09] yurik: what about: from=ExampleOldName&to=ExampleNew ? [07:37:41] i could i guess [07:38:21] hashar, sec, will recommit [07:39:36] hashar, take a look at the linker.php [07:39:38] https://gerrit.wikimedia.org/r/#/c/53530/4/includes/Linker.php [07:39:54] line 684 is just wrong [07:40:13] yurik: well fix it in another commit :] [07:40:25] hashar, the thing is that i don't want to fix it [07:40:32] so don't fix it :-] [07:40:41] because i don't understand what the logic was behind it [07:40:42] I am not sure what else I can say hehe [07:41:16] its a clear that someone was trying to do some check there, but would need to figure out why [07:41:37] yurik: yeah in a different commit. One could use git blame to find out why it happened [07:41:48] maybe someone did a massive replace [07:41:59] some git-blame, others blame git! :P [07:42:08] for now just update the ApiMove example and I will be happy to merge the change :-] [07:42:20] * YuviPanda blames ori-l [07:42:37] probably appropriate [07:43:05] hashar, committing [07:44:51] 2b9c22deb3f7031892b3cbef70197c59a3709bc6 [07:45:00] by tpt [07:45:03] na [07:45:04] https://gerrit.wikimedia.org/r/#/c/53530/5 [07:45:05] :-] [07:45:20] +2ed [07:45:29] hashar, thx [07:45:52] https://gerrit.wikimedia.org/r/#/c/20683/ [07:45:55] this is the change [07:47:32] I guess it is an easy fix :-] [07:48:03] probably :) [07:48:12] just don't want to mess with the frontend side of things [07:50:43] New review: Hashar; "That will not work when a commit is tagged :-]" [integration/zuul-config] (master) - https://gerrit.wikimedia.org/r/44384 [08:38:03] saper: help! [08:38:14] or someone who knows about gerrit [09:35:36] New patchset: Hashar; "(bug 31518) PHPUnit code coverage for mw/core" [integration/jenkins-job-builder-config] (master) - https://gerrit.wikimedia.org/r/53532 [09:42:33] aude: do you still need help? hashar is online [09:42:37] in fact, he just helped me :) [09:45:16] aude: i am there :-] [09:57:55] hashar: think i figured it out sort of [09:58:18] it's not possible to amend a gerrit change that's been merged and then reverted [09:58:28] so i removed the change id and submitted it as a new patch [09:59:23] aude: yeah once it is merged, the change is closed :-] [09:59:50] yep [10:00:11] but at least i know how to submit it as a new patch now [10:03:33] annnd [10:03:36] I am off for a nap [13:22:22] hashar- The test "mediawiki-core-jslint" failed on a backport into wmf11 I did earlier (change 53561), for no particular reason considering that all the change did was update an extension. Seems odd. [13:25:19] <^demon> qchris: Did you see the new jgit landing in master? :) [13:27:15] <^demon> garbage collection and recursive merges and new bitmap optimizations, oh my :) [13:39:52] anomie: the jenkins console should give you some information :-] [13:41:28] hashar- Yes. It appears it compared the change in the wmf11 branch to some previous change in master. [13:41:48] Which seems an odd thing for it to expect to work. [13:44:01] anomie: compared? [13:44:46] ahh jshint [13:45:20] anomie: can you fill a bug mentioning the change issue and https://integration.mediawiki.org/ci/job/mediawiki-core-jslint/3455/checkstyleResult/ [13:45:30] anomie: make sure Krinkle is in CC [13:45:34] hashar- ok [13:45:49] that seems to be an issue with the checkstyle plugin [13:46:16] hashar- Component? [13:46:17] or maybe jshint :-] [13:46:25] Wikimedia > Testing Infrastructure [13:46:41] that is the component for anything related to testing from Jenkins to Zuul and Jshint :-] [13:47:32] New patchset: Hashar; "Revert "trigger test pipeline for wikibase (non voting unit tests)"" [integration/zuul-config] (master) - https://gerrit.wikimedia.org/r/52766 [13:47:41] Change merged: Hashar; [integration/zuul-config] (master) - https://gerrit.wikimedia.org/r/52766 [13:52:30] hashar- bug 46066 filed [13:52:36] danke [13:56:12] hashar- I wonder if it will do the same if someone submits a patch to fix some jshint error, failing anything submitted after that until the fix patch is merged and everything else is rebased on top of it. [14:05:53] <^demon> Hmm, wonder if we can sneak a gerrit upgrade in Friday. [14:05:57] <^demon> If not, early next week. [14:06:08] ^demon: I read the message on the mailing list, but have not yet toyed with it (I am just rebasing project renaming) [14:06:30] <^demon> Yeah, I've seen it in action. [14:06:49] <^demon> I saw a presentation Dave gave on it back in like October. [14:06:55] Oh :-) [14:06:57] <^demon> I'm more excited about the bitmap work than anything else. [14:07:13] Is thta presentation available somewhere? [14:08:12] <^demon> I only know of a copy on google docs :p [14:08:15] <^demon> Lemme find the url [14:12:53] <^demon> Ah, here it is: https://docs.google.com/presentation/d/1t2qzSO4z5MbrN21Cz4RrLWz9IZ5ylJNwQassajKyPk8/view#slide=id.g3236f3d6_0_73 [14:13:51] * qchris takes a look [14:14:09] <^demon> So yeah, this plus jgit gc should help alleviate a lot of core's performance problems. [14:18:36] The numbers for the bitmap stuff look great. [14:19:17] So if that is stable too, I'll be all for updating gerrit soon :-) [14:19:39] <^demon> Yeah, that stuff's in jgit stable now, and gerrit master pulled it in. [14:19:51] <^demon> gerrit's wrapper for gc is still in progress. [14:19:52] That's what I've read as well :-) [14:19:57] <^demon> That's https://gerrit-review.googlesource.com/#/c/32531/ [14:20:33] I read that patch this morning :-) [15:10:59] qgil: should we change the channel topic when the event starts later today? [15:11:20] chrismcmahon: ^ [15:31:04] zeljkof_, sure [15:36:13] The "Navigable TOC" idea in http://usability.wikimedia.org/wiki/File:Babaco_ntoc_click.jpg - is that idea (and its bug reports) dead? [15:36:22] qgil: we'll be set up to stream a short presentation on youtube, yes? [15:36:33] chrismcmahon, yes [15:36:45] qgil: awesome, thanks [15:37:07] just checking: we are starting in about 1h30, right? [15:37:26] I'm sorry I won't be able to make it. I'm presenting at a GSoC info session [15:37:47] sumanah, you can still write a scenario later today / this week. :) [15:37:56] :) [15:38:26] sumanah, who organizes that info session, and is someone doing something similar in SF Bay? [15:38:52] qgil: some local Hunter College folks are organizing it, and I don't know whether anyone is doing a similar thing in the SFBay -- we could if we wanted [15:39:16] in fact we could simply repurpose our Wikipedia meetup in April and then email carol smith and she'd put it on the relevant calendar, send us stickers if we wanted, etc [15:39:44] sumanah, that is a good idea [15:40:15] a very good idea :) [15:40:16] I just forwarded you some info. [15:40:19] thanks! [15:42:46] also qgil yay at the attendance for http://www.meetup.com/Wikipedia-Engineering-Meetup/events/106078042/ [15:44:14] qgil: you might also enjoy reading http://www.icerocket.com/search?tab=twitter&lng=&q=wikipedia+lua&x=45&y=22 which is half in languages you speak & I do not [15:45:10] sumanah, Lua meetup attendance I'm looking forward: almost everybody signing up until now are NOT WMF employees. [15:45:45] YES [15:45:54] sumanah, I like this one: "Vivimos en la Generacion WTF: Wikipedia Twitter y Facebook!" :DDD [15:46:52] ha [15:46:54] :) [15:47:01] https://en.wikipedia.org/wiki/Wikipedia:Village_pump_%28technical%29#Meetup_.26_videostreamtomorrow_-_focus_on_Lua [15:48:56] sumanah, thank you! [15:49:11] I've also posted at en.wikisource and am posting at en.wiktionary [15:55:18] qgil: it might actually be worth sending an announcement of this meetup to wikitech-l and/or mediawiki-l [15:55:29] I can forward it to wikitext-l which will interest certain people [15:55:52] sumanah, will do. After my email to Carol where you are CCed [15:56:18] okay! [15:56:27] ok, off I go to tell kids about GSoC [16:04:52] The "Navigable TOC" idea in http://usability.wikimedia.org/wiki/File:Babaco_ntoc_click.jpg - is that idea (and its bug reports) dead? [16:37:45] Wikimedia Language Engineering team office hours in #wikimedia-office in about 20 minutes. [16:49:56] * qgil starts preparing the hangout / video stream. [16:50:09] My first time organizing video sream, let's see how it goes. [16:50:24] chrismcmahon, zeljkof be ready to receive invitations [16:50:40] qgil: ready :) [16:50:47] qgil: link for hangout I guess [16:51:07] I'll only go about 10 minutes or so, maybe a little longer [16:56:15] Wikimedia Language Engineering team office hours in #wikimedia-office in a few minutes. [17:00:05] qgil- Is there going to be an official IRC channel along with the YouTube stream for the Lua meetup tomorrow? [17:03:38] http://youtu.be/Kad3EUM4GdM [17:04:59] Hello browser testers! [17:05:04] https://www.mediawiki.org/wiki/QA/Browser_testing/Search_features [17:05:21] demo streaming right now at http://youtu.be/Kad3EUM4GdM [17:05:30] hi rachel99 ! [17:05:35] ok, thanks [17:06:28] hi everybody [17:06:33] we are live now! :) [17:06:59] check the channel topic for youtube stream [17:07:40] any questions so far? [17:08:38] hi sucheta ! http://youtu.be/Kad3EUM4GdM [17:10:42] qgil, Thanks. Watching :) [17:11:44] zeljkof: Yes, is it possible to participate sans YouTube? [17:12:00] Hi rihnapstor we are streaming right now: Writing scenarios for test automation http://youtu.be/Kad3EUM4GdM - more https://www.mediawiki.org/wiki/QA/Browser_testing/Search_features [17:12:11] marktraceur: yes, of course, we can talk here [17:12:15] Cool beans. [17:12:23] I will drift in and out :) [17:12:24] marktraceur, YouTibe is for the video streaming now. You can ask questions and get involved. The rest will be done via IRC only. [17:12:37] marktraceur: we will set up a etherpad pad and collaborate there [17:12:57] the streaming will take no more tha 30 mins total [17:13:05] Cool, I'll be around then. [17:13:19] Hi ialex we are streaming right now: Writing scenarios for test automation http://youtu.be/Kad3EUM4GdM - more https://www.mediawiki.org/wiki/QA/Browser_testing/Search_features [17:14:05] the backlog mentioned by chris can be found at https://www.mediawiki.org/wiki/QA/Browser_testing/Test_backlog [17:15:20] ok, streaming complete. Great pitch chrismcmahon ! [17:15:28] And now let's start with some work [17:16:12] Who wants to sart writing an scenario right now? [17:16:13] https://www.mediawiki.org/wiki/QA/Browser_testing/Search_features [17:16:35] qgil: how do we set up etherpad pad? [17:16:55] zeljkof, wait.,.. [17:17:10] Guys [17:17:12] :) [17:17:12] I'd like to do a regression test. When I put that page together, the "umlauts and accents" one is intriguing [17:17:28] because it's a bug too [17:17:41] rihnapstor: do you have questions? [17:17:41] zeljkof, http://etherpad.wmflabs.org/pad/p/Browser-automation [17:17:46] qgil: thanks [17:18:12] I'm curious about https://www.mediawiki.org/wiki/Talk:QA/Browser_testing/Search_features#Scenario_for_transcluded_pages_search_at_Wikisources_25030 [17:18:24] it's not a Search case, but a case nevertheless, and proposed by someone else [17:18:45] I could try with that one or one based in a bug report you give me [17:19:24] should we draft in the etherpad and then move descriptions to the backlog once you decide they are good enough? [17:20:00] that sounds good [17:20:57] ok, can you give me a feature? [17:21:14] * qgil is ready for "My First Time" [17:21:18] I need some time to look at it more properly. Going through the wiki and the video once again [17:21:47] sucheta, so do I. If you want we can work together in one scenario. [17:22:16] qgil, Sounds great :) [17:22:26] I want! I want! [17:22:35] waiting for chrismcmahon zeljkof to give us one task. Unless you have a proposal sucheta [17:22:41] will be using agile scrum concepts here ? [17:23:20] rihnapstor: what do you mean by "agile scrum concepts"? :) [17:23:24] * valeriej will also be participating [17:23:33] valeriej, Yay! Hi :) [17:23:58] This will help me identify bugs that could add to the backlog as well. [17:24:00] qgil: do you need help with writing scenario for the case you have linked to a few minutes ago? [17:24:01] sucheta: Hi! [17:24:05] valeriej, rihnapstor: do you want to team for writing another scenario? [17:24:22] rihnapstor: sort of agile. if you choose a bug to write a test for, that test will be failing until the bug is fixed. if you choose existing functions to test, then that will be a regression test (we can use both kinds) [17:24:33] zeljkof, sucheta and me need one bug or a feature to write the corresponding scenario. Give us one please. [17:25:41] qgil: what about "search ignores user settings" [17:25:49] rachel99: hello, do you have a test that you would like to automate? or would you like guidance? [17:26:23] hi rachel99! (rachel99 has already contributed a Given/When/Then test to the backlog) [17:26:57] chrismcmahon, ok! sucheta let's give us some time to read https://bugzilla.wikimedia.org/show_bug.cgi?id=37871 [17:27:25] chrismcmahon: Hi! [17:28:25] sucheta, valeriej, qgil and everybody else: the channel is a bit chaotic now, anybody needs help? [17:29:09] rachel99, Ryan_Lane: welcome! :) do you need help with writing your first scenario? [17:29:37] zeljkof, qgil and I are working on the bug you pointed us to. Also I am reading up the wikis related, first. [17:29:50] sucheta: great, let me know if you need any help [17:30:01] zeljkof, Sure :) [17:30:17] zeljkof: I guess I could work with someone else on a scenario, if you have an easier one. [17:30:27] zeljkof: I'm checking out bug 44811. rihnapstor can join if available. I think I'll watch the video again. [17:30:30] rachel99: they are all easy :) [17:30:38] fwiw, I think https://bugzilla.wikimedia.org/show_bug.cgi?id=44238 is invalid [17:30:45] or fixed [17:30:57] rachel99: is there a feature of mediawiki or an extension that you use every day? [17:31:07] rachel99: it is a good place to start [17:32:22] DarTar: welcome :) do you need help writing your first scenario? (see this if you do not know what I am talking about: http://www.mediawiki.org/wiki/QA/Browser_testing/Search_features) [17:32:38] ok, so I have added the bug reports about user setting & search at the etherpad and also opened a window with sucheta not to bring too much noise to this room [17:34:27] so here is an interesting case: I think https://bugzilla.wikimedia.org/show_bug.cgi?id=44238 because I think searching with "África", "AFRICA", and "Alto" all seem OK to me. I'd like someone else to check out searching with accents on ptwiki though to back me up. rachel99 maybe or valerij? [17:34:49] Yes, I could do that. [17:34:51] that is, I think !b 44238 is not a bug [17:35:04] !b 44238 [17:35:04] https://bugzilla.wikimedia.org/44238 [17:35:08] there we go [17:35:18] so I'll be making a test along those lines [17:35:42] rachel99: let me know if you need help [17:36:20] qgil: I am not sure if I understood you, do you need help or were you just reporting on progress? [17:36:24] I guess a way to do this for us newcomers is to try to describe the feature in ery human"language and then pay more attention to the correct syntax, right? [17:36:48] zeljkof, just reporting progress (before) and asking a question right now ^^^ >) [17:37:20] Guys i am not famililar with medaiwiki notaions syntaxs. [17:37:23] qgil: yes, the first step is to write a story, like you are to write a bedtime story :) [17:37:46] qgil: when that is done, then we can work on given/when/then format [17:37:55] chrismcmahon: is this about search suggestions or after hitting enter? [17:38:00] rihnapstor: neither am I :) [17:38:15] Nikerabbit: either one [17:38:42] Nikerabbit: I would like to have some basic coverage of search features, and some tests for bugs likely to be fixed sooon [17:38:43] rihnapstor: what are you trying to do? are you writing scenarios? http://www.mediawiki.org/wiki/QA/Browser_testing/Search_features [17:39:33] chrismcmahon: I still get suggestions starting with "A" if I type "Ä" into search box in fiwiki (which is absolutely wrong), so accents stripping certainly isn't universally broken [17:40:38] to find that bug search Å in bugzilla [17:40:50] robla: welcome to increase the backlog event! :) do you need help? http://www.mediawiki.org/wiki/QA/Browser_testing/Search_features [17:40:55] https://bugzilla.wikimedia.org/show_bug.cgi?id=24414 [17:40:58] Nikerabbit, note that right here right now we are not triying to fix bugs, but write test automation scenarios for them - http://www.mediawiki.org/wiki/QA/Browser_testing/Search_features [17:40:59] Nikerabbit: I think that is https://bugzilla.wikimedia.org/show_bug.cgi?id=24414 yes? We could either have a failing test for the fix for that, or we could write a passing test that would fail if we ever fix the bug [17:40:59] zeljkof: i referred to that link you previously tagged me in that message,i went through it ,but the content was not making sense to me.I am more into python,so probably it will take some time to get started with bug fixes. [17:41:39] so cool, my test is going to be for Portuguese Wikivoyage :-) [17:41:50] rihnapstor: the channel is pretty chaotic at the moment, I get lost from time to time [17:42:16] rihnapstor: we are writing scenarios for the backlog now: http://www.mediawiki.org/wiki/QA/Browser_testing/Search_features [17:42:54] rihnapstor: scenarios will be automated later, you do not need to know anything besides basic English to help [17:42:54] zeljkof: i referred to that link you previously tagged me in that message,i went through it ,but the content was not making sense to me.I am more into python,so probably it will take some time to get started with bug fixes. [17:43:34] rihnapstor: we are not fixing bugs now, we are writing scenarios for browser automation, would you like an example? [17:43:42] i am not failiar with browser testing concept [17:44:06] i know object-oriented php [17:44:11] rihnapstor: if it sounds interesting to you, I will be more than happy to explain it to you [17:44:48] rihnapstor: interested? [17:44:59] are scenarios the user stories or use cases ? [17:45:07] tewwy: welcome :) are you here for the event? http://www.mediawiki.org/wiki/QA/Browser_testing/Search_features [17:45:34] rihnapstor: what is the difference between user stories and use cases? [17:45:50] zeljkof: Stuck in a meeting :-( This looks far more interesting. [17:45:56] they are same zeljkof [17:45:58] :) [17:46:11] tewwy: let me know if you need help later :) [17:46:33] rihnapstor: we can talk on gtalk/skype/hangout, it could be faster [17:46:54] rihnapstor: would you like to pair on writing a scenario? http://etherpad.wmflabs.org/pad/p/Browser-automation [17:47:23] tewwy: we will be here for the next hour or two, and the entire week (and next week…) :) [17:47:47] aha, I just figured out why https://bugzilla.wikimedia.org/show_bug.cgi?id=44238 is actually a bug. It sort of depends on what the definition of "ignore" is valeriej rachel99 [17:48:21] zeljkof: what about the ongoing project like integrating visual editor into medaiwiki? [17:48:39] rihnapstor: sure [17:48:39] chrismcmahon: not following you on that [17:49:00] rihnapstor: visual editor is a bit buggy, but definitely interesting [17:49:19] rachel99, one sec [17:49:29] rachel99: will put it in the etherpad... [17:49:35] rihnapstor: meet you at ether pad? http://etherpad.wmflabs.org/pad/p/Browser-automation [17:49:52] btw, zeljkof is this the developer channel for mediawiki also ? [17:50:12] medaiwiki powers wiki,so thats why i asked [17:50:36] rihnapstor: I think this is a bit broader channel, mediawiki developers hang at #mediawiki [17:51:10] dschoon: hi! :) are you here for the event? http://www.mediawiki.org/wiki/QA/Browser_testing/Search_features [17:51:49] I'll be listening :) I have a meeting in 10m, tho. [17:51:57] henrique__: hi there and welcome! :) here for the event? http://www.mediawiki.org/wiki/QA/Browser_testing/Search_features [17:54:01] rmoen, tfinc: welcome! :) here for the event? http://www.mediawiki.org/wiki/QA/Browser_testing/Search_features [17:54:08] rachel99: see the note on etherpad.... [17:54:13] COMMENT:  the problem in this bug is that for example on https://pt.wikivoyage.org: Search for "África" and the result goes directly to the page for "África"Search for "Africa" (no accent) and the result shows "África" as the top result, but thinks that "Africa" is a redlink and encourages the user to create the page for "Africa" [17:54:45] chrismcmahon: Ok, I think I see that. [17:55:35] ah, so it is the reverse problem [17:55:55] anybody needs help in writing the first scenario? [17:56:02] (or the second, or the third...) [17:56:36] chrismcmahon: I think that's only part of the issue [17:57:08] because not all special characters in all languages are treated in the same way, according to comments on bug [17:58:25] zeljkof, at the etherpad sucheta and me have written the "bed-time story" about https://bugzilla.wikimedia.org/show_bug.cgi?id=37871 , and now we are starting to write a proper scenario out of it [17:59:40] qgil: great! :) please add a line at the top of scenario qgil/sucheta so I know which one is yours and let me know if you need help [18:02:13] rihnapstor: I think we have a major misunderstanding :) are you here for "write a scenario" event, or are you looking for google summer of code project, or something else? [18:02:39] rihnapstor: this is the event we are in the middle of right now: http://www.mediawiki.org/wiki/QA/Browser_testing/Search_features [18:03:50] i am dreaming of wikipedia as a Saas platform . :) :) [18:04:52] Nemo_bis: is this short enough? https://meta.wikimedia.org/wiki/Global_message_delivery/Spam [18:04:58] zeljkof: Did that :) [18:05:25] sucheta: what did you do? :) [18:05:46] zeljkof, put her name at the etherpad [18:05:46] sucheta: (sorry, a lot of traffic at the moment, got a bit lost) [18:05:59] sucheta, qgil: I see, great :) [18:06:10] zeljkof, Yes. :) [18:06:55] zeljkof, sucheta and me have doubts on how to reference URLs and how specific to be e.g. only English Wikipedia or any Advanced Search, only User-namespace or any... [18:08:00] zeljkof, Also, I am wondering if we have to keep it as precise as possible. As I amfollowing the QA documentation that has been posted here few minutes back. [18:08:01] rachel99 valeriej I think I have my "search for accent" scenarios properly described now, and ready to turn them into Given/When/Then scenarios [18:08:25] sucheta: the more precise the better usually [18:08:30] chrismcmahon: Ok. I am following you on the etherpad. [18:08:43] chrismcmahon, OK. Thanks. :) [18:09:06] qgil, sucheta: "There are only two hard things in Computer Science: cache invalidation and naming things" -- Phil Karlton :) [18:09:29] guillom: looks rather long but I've no idea how to shorten it, especially as most stuff is just links to English-only resources [18:09:32] so rachel99 one thing we should decide is whether we want to put all of those scenarios on the backlog. I think that we should, even though two of them are contradictory [18:09:34] qgil, sucheta: I will take a look at the scenario and let you know what I think [18:10:12] chrismcmahon: I'm following as well. Thanks for walking us through. [18:10:34] chrismcmahon: I see. So you mean write test cases for all 3 of them? [18:10:36] zeljkof, Haha :) Very true [18:10:44] rachel99: I lean toward putting both on the backlog. when it's time to write code for this Feature we can decide then a) whether the bug is fixed and b) which scenario to implement and c) whether we want a failing test in the build [18:10:56] qgil, sucheta: we do not usually specify the site, since the same test should run the same on all sites [18:10:57] chrismcmahon, the bug report describes that in Commons a "MediaWiki:..." search should list pages under the MediaWiki: namespace if the search is restricted to the User: namespace. Should we go only for Commons - MediaWiki: - User: or should we also write scenarios for other combinations, just in case? [18:11:18] Nemo_bis: how about I remove the second paragraph (with all the links) and only keep the first sentence that explains what Lua is? People will find the links in the blog post anyway. [18:11:29] zeljkof, Erm, did I specify site? [18:11:41] zeljkof, ok you answered the site part. What about other combinations in namespace search / results? [18:11:42] qgil: that's a beautiful thing about the backlog, it allows creative decisions like that :-) [18:12:01] sucheta: are we talking about lines 38-46 in the ether pad? [18:12:13] qgil: what do you think the best test is? [18:12:15] chrismcmahon, rigt. Then we will write oine following exactly the bug report and then a copy with other variables [18:12:29] qgil: awesome :) [18:12:51] zeljkof, 39 - 57 [18:13:24] rachel99: would you like to turn those sentences for the accent test into Given/When/Then statements on the etherpad? [18:13:28] zeljkof, Mine is from 52 - 57 :) Ours - 39 - 57 [18:13:31] qgil, sucheta: I have commented on the pad, I would remotve the link from line 44 [18:14:09] qgil, sucheta: the scenario in 39-47 is too generic [18:14:33] qgil, sucheta: I will comment under it [18:14:52] zeljkof, sucheta right, but comment under 54-57 because tat is the last iteration [18:14:54] spagewmf: welcome! :) are you here for the event? http://www.mediawiki.org/wiki/QA/Browser_testing/Search_features [18:16:05] guillom: should work [18:16:11] ok, thanks [18:16:23] zeljkof: Can you check mine out? I think I got ahead of myself. [18:16:25] chrismcmahon: Ok. Let me just start with one of them. [18:16:38] zeljkof, Could you comment under line 57? As 52 - 57 is also ours :) [18:16:53] It doesn't really matter. But.. [18:17:16] valeriej, sucheta: will comment on your scenarios in a minute, finishing one at the moment :) [18:17:22] sucheta, just nmove your lines above zeljkof s [18:17:29] guillom: but then we should probably soft-redirect [[m:Lua]] to [[mw:Lua]], because the links on the blog are to the latter and the Meta page is useless now (the actual page was moved elsewhere) [18:18:18] aude: sorry, was away whole day [18:18:32] saper: not a problem [18:18:58] I see it got solved [18:19:06] http://etherpad.wmflabs.org/pad/p/Browser-automation unavailable... [18:19:10] is ether pad down? [18:19:21] yup [18:19:33] marktraceur: etherpad again [18:19:36] oh crud [18:19:48] Ugh. [18:19:49] etherpad waits for the pope to be announced:) [18:19:53] It'll be back in a sec. [18:19:58] aaand back [18:20:00] I'm back in thanks marktraceur [18:20:07] good news people! we had so much traffic that we broke etherpad! :) bad news is that the work is probably lost :( [18:20:11] Nemo_bis: probably, but for now I'm going to focus on spreading the word, since redirecting the page isn't a blocker for that. [18:20:25] the pad is not lost! :) [18:20:51] zeljkof: The server restarts sometimes when there are a lot of connections at once. [18:21:01] guillom: sure, just saying :) [18:21:05] :) [18:21:32] twooted :) [18:21:37] marktraceur, I'll tell to sumana that the QA weekly was so successful that we brought down the etherpad ;) [18:21:48] Heh. [18:22:08] qgil: She'll maybe not be impressed, since it seems like every meeting she runs brings it down, at least once :) [18:22:34] marktraceur, I wish there would be a special color for ironic comments in IRC :) [18:22:42] Ah. [18:22:44] marktraceur: are you saying that we really caused it? :) [18:22:50] zeljkof: Almost certainly! [18:23:17] qgil: You could use colours, it just takes some weird key combinations that I've forgotten [18:23:20] sucheta, qgil: take a look at my take at your scenario at 58-62 [18:23:26] qgil, I break Etherpad every day at least once. With the help of marktraceur [18:23:27] ok sucheta, nice to see that what you wrote was shorter then mine and what zeljkof wrote is even shorter than yours [18:23:32] good, learning something here [18:23:35] sucheta: High five! :) [18:23:42] To be fair we're also *fixing* it. [18:24:11] marktraceur, Yes! High five :D [18:24:23] ok, people, who else needed scenario reviews? :) valeriej and sucheta, right? [18:24:52] chrismcmahon: I wrote up one of the scenarions on the etherpad [18:24:58] zeljkof: Yes, please! [18:25:20] rachel99:  I think we should be a little more specific [18:25:20] valeriej: your scenario is 7-15, right? [18:25:23] zeljkof, Did you not just review mine? [18:25:42] zeljkof: Yes. [18:25:47] sucheta: I did the one you wrote with qgil, I thought you have another one [18:25:54] if anyone want to follow along, about line 35 in EP [18:26:09] zeljkof, We wrote separately [18:26:33] valeriej: your scenario is next [18:26:45] zeljkof, Mine is from 50 - 55 [18:27:05] chrismcmahon: ok, I will follow [18:27:08] sucheta, qgil: does my take on your scenario (58-62) make sense to you? [18:27:22] zeljkof: No rush. I have to make lunch. :) [18:27:22] zeljkof, You probably just saw Quim's -_- [18:27:26] zeljkof, makes sense and you're "the teacher" [18:27:26] rachel99: for the When statement, let's get very specific about what to search for [18:27:56] sucheta, isn't the same? [18:28:03] sucheta, qgil: I got a bit lost, we had more traffic than I expected :) [18:28:05] rachel99: remember, if we decide we need more than one search term to test with, we can create a "Scenario Outline" and call it with multiple variables [18:28:17] sucheta, I mean, it looks like zeljkof took both into account [18:28:30] zeljkof, Yes :D But, he just asked me if I had written a separate one. [18:29:03] sucheta, ah ok then yes [18:29:07] I did. So, here I claim justice. [18:29:16] Err, not really :D [18:29:20] :D [18:29:29] chrismcmahon: Ok. You will be doing it, and I will follow what you are doing? [18:29:49] I'll deliver some justice whenever zeljkof is at reach - no worries ;) [18:30:21] qgil: I am lucky to be on the other side of the planet :) [18:30:49] btw thanks to writing the scenario sucheta and me found that there was something unclear in the expected result at https://bugzilla.wikimedia.org/show_bug.cgi?id=37871 so I left a comment there [18:31:14] :D This was fun! [18:31:19] zeljkof, you are cornered in Europe, between sucheta and me... [18:31:19] rachel99: I think we should use the example and search for "África" and test that we land on the page for "África" first [18:31:29] rachel99: OK by you? [18:31:52] zeljkof, Hu ha ha. We are watching you! [18:31:52] chrismcmahon: Yes. Ok by me. [18:32:02] ok chrismcmahon zeljkof I'm ready for another one before lunch [18:32:20] Yes yes. Me too ! [18:32:40] rachel99: what should happen on the Then test result line? [18:32:43] qgil, sucheta: any mediawiki feature the two of you use every day? or should we pick something? [18:33:09] zeljkof, pick something search related for me, please [18:33:17] qgil: just the act of writing a test often turns up useful information, good find [18:33:22] chrismcmahon: You will see Africa (with an accent) [18:33:35] qgil, We are not a team anymore? :( [18:33:51] qgil: looking… would you like to pair with sucheta? :) [18:33:53] rachel99: I think we should actually be ON the page title "África". Agree? [18:34:14] chrismcmahon: yes. [18:34:29] sucheta, of course, if you will. I thought you wanted to get rid from old laggers... [18:34:53] qgil, sucheta: does this sound interesting? "Search results should always be shown in the same order" [18:34:53] rachel99: go ahead and write that Then line for us? [18:35:05] sucheta, I'm in [18:35:15] !b 32026 [18:35:15] https://bugzilla.wikimedia.org/32026 [18:35:26] qgil, Why would you think so? We're team! :D [18:35:29] chrismcmahon: Sure. [18:35:32] zeljkof: qgil sucheta Yes! There is a bug about sometimes search results not being in the correct order that I think Ram just fixed, so that is a great test [18:35:58] OK. We're on it ! [18:36:09] sucheta: great :) [18:36:35] * chrismcmahon did a LOT of research into Search bugs last week :) [18:37:25] rachel99: fire at will :) [18:37:36] chris [18:37:59] chrismcmahon: will do. [18:38:13] valeriej: you did a great job! :) I _do_ have a few comments :) [18:38:36] zeljkof: I'd love to hear them! [18:39:02] valeriej: writing the scenario as I would write it in lines 78-84 (in progress) [18:39:26] zeljkof: Yes, I'm looking at that. [18:39:29] rachel99: what do you mean by "be returned"? [18:39:31] sucheta, https://bugzilla.wikimedia.org/show_bug.cgi?id=32026 is an interesting bug. Writing the expecte behavior looks a lot simpler that fixing th bug. ;) [18:40:44] zeljkof: The "or" was because I wasn't sure which sentence to use. So you're saying use both, right? [18:41:09] valeriej: ok, lines 81-88 are my take [18:41:12] rachel99: for the accented character I usually just copy-paste :) [18:41:24] chrismcmahon: ok! [18:41:30] valeriej: yes, if you want to check both things, then check both :) [18:41:42] rachel99: I think we need to say explicitly that the page we arrive on has title "África" [18:41:52] valeriej: if you think just one check is enough, then pick the one that is more important [18:42:20] qgil, Creating bugs is even more entertaining! :D [18:42:44] rachel99: I think that's good [18:42:47] valeriej: I moved my comments to 19-26 [18:42:51] chrismcmahon: Great! [18:42:51] rachel99: let's do another one [18:43:01] chrismcmahon: Ok. [18:43:09] valeriej: done for today, or would you like to continue? [18:43:29] rachel99: do the next one that says THIS TEST WILL FAIL [18:43:31] zeljkof: In the video, there is a positive (what should happen) and a negative test (what shouldn't happen) test case, do always attempt to make both? [18:43:46] zeljkof: And yes, I'd like to continue. :) [18:43:49] chrismcmahon: Ok. [18:43:59] valeriej: it depends :) [18:44:23] valeriej: we test for everything that makes sense, or that we thing could break, or that absolutely has to work all the time... [18:44:38] zeljkof: Heh, not an unexpected answer. [18:44:49] valeriej: you could create a few more scenarios around that topic [18:44:53] rachel99: I think we can just copy-paste the previous G/W/T and change just one thing [18:45:21] chrismcmahon: I just did that, I believe. Is it looking ok? [18:45:29] valeriej: if the page is deleted, what should and should not happen when searching for it? [18:45:53] * zeljkof is back in 5 minutes [18:45:55] rachel99: looks righteous to me. want to do the last one too? then we can get zeljkof to review. (I always like to get zeljkof to review, he is REALLY good at this) [18:46:14] chrismcmahon: Sure. [18:46:42] zeljkof: There should be a message saying that I can create that page. [18:47:43] zeljkof: I'm not sure about what should not happen. The deleted page should not be returned in the search results. [18:48:57] zeljkof, chrismcmahon sucheta the examples provided a https://bugzilla.wikimedia.org/show_bug.cgi?id=32026 are not "purely" browser based but API call based, is that fine? [18:49:11] chrismcmahon: How does the last one look? [18:49:37] I guess they are doing it to get large list of results that you won't easily get searching for a specific word [18:49:59] rachel99: I am going to make a small change to be as specific as we can... [18:50:08] chrismcmahon: sure. [18:50:20] I don't even understand what is the search about... :( [18:50:29] http://en.wikipedia.org/w/api.php?srprop=sectiontitle&srlimit=25&srsearch=10.1371%2Fjournal.pone.0008776&action=query&format=xml&list=search&sroffset=0&srwhat=text [18:51:34] rachel99: I want to have this test a) check that it is on the right page And b) that the result is correct and c) that we have a redlink also. [18:51:51] chrismcmahon: Gotcha. Looks good. [18:52:03] rachel99: because when we run this and it fails (and it will fail) I want the failure to be as specific as possible [18:52:18] * zeljkof is back [18:52:38] chrismcmahon: It is definitely a bug, then? That will need to be fixed? [18:53:00] rachel99: I think it is really a bug [18:53:25] qgil: we are only concerned about stories now, we will think about implementation later [18:53:39] rachel99: I didn't at first, I had to do some exploring to change my mind [18:53:42] ok zeljkof then moving on with the bed story [18:54:01] qgil: if the implementation involves browsers or api or whateverr, it does not matter at the moment [18:54:08] rachel99: OK, I'm going to delete the first-draft sentences since we're happy with our G/W/T statements [18:54:49] valeriej: ok, so the next scenario (or the next check in the existing scenario) could be if there is "create page" link [18:54:55] valeriej: sounds good? [18:55:08] zeljkof: Yes, I'll work on that. Thanks. [18:55:37] anybody else needs help with scenarios? or reviews? (I got a bit lost) [18:56:11] zeljkof: would review 28-55 for rachel99 and me? [18:56:13] zeljkof, I just drafted one at 88-92 [18:56:15] would you [18:56:40] * qgil feels like it's lunchtime [18:56:45] chrismcmahon, qgil: sure, reviewing one by one right now [19:01:09] zeljkof: if 27-54 looks OK to you, I'll ask rachel99 to edit the backlog page with that one [19:01:35] chrismcmahon: will let you know in a minute [19:02:08] hi [19:03:04] hi rihnapstor [19:06:18] qgil, sucheta: I have done my version of 85-96 at 98-104 [19:08:39] chrismcmahon, rachel99: reviewing your scenarios right now [19:09:26] zeljkof, ok works for me [19:10:14] zeljkof, chrismcmahon what I'm not sure is: when I write these bed-time stories are they intended to be read by other humans (you) or should I aim to write them directly for machine processing? [19:10:18] rachel99: zeljkof is making the spacing conform to our standard arrangement. I'll show you how to preserve that when we update the backlog. [19:10:23] qgil: would you and/or sucheta cut/paste your scenarios to backlog page? [19:10:43] chrismcmahon: ok [19:10:53] qgil: cucumber is meant to be read by people [19:11:00] qgil: write them for humans, zeljkof and I will make the machine do the right thing (and maybe that will be our next session?) [19:11:06] zeljkof, for insance"And I got this in the search results: John, A. J. John, Anaparambil, John, King of England" is something "true" for a human because you get the point, but there is no way a machine will get those exact results [19:11:07] qgil: there is an easy way for it to make sense for machines [19:11:29] you guys thinking of semantic search ? [19:11:34] qgil: of course the machine will know how to do it, you "just" have to tell it how :) [19:11:52] qgil: that is a topic for another event [19:12:08] rihnapstor: the tool we're using Cucumber has some pretty heavy regular expression conventions [19:12:17] zeljkof, I thought that no providing examples I can't reproduce was a better way to descibe a scenario, as I did at 92-96 [19:13:59] ok chrismcmahon [19:14:22] zeljkof: Can you check lines 20-24? [19:14:27] qgil: no, scenarios should be as detailed as possible [19:14:52] qgil: with concrete examples, "when I search for john, I should see john in search results" [19:15:06] rachel99: do you have an account on http://www.mediawiki.org [19:15:06] ? [19:15:16] valeriej: sure, in approx 5 minutes :) [19:15:25] zeljkof: Thanks! [19:15:28] whoops rachel99 left [19:16:03] chrismcmahon: is rachel somebody we know? [19:16:41] zeljkof: rachel99 is a volunteer with a testing background who is looking for experience in browser automation [19:16:59] chrismcmahon: great [19:17:06] zeljkof, yes and this is why I asked before about that search query in the bug report (that I don't understand) and whether this can be reproduced from the regular Search interface - see https://bugzilla.wikimedia.org/show_bug.cgi?id=32026#c8 [19:17:28] anyway, I'm hungry and I'm getting my lunch now. See you later. :) [19:17:30] chrismcmahon: I think 33-60 are good, as far as I am concerned you can move them to the backlog [19:17:45] qgil: thanks and see you later [19:18:07] sucheta: would you move the examples you and qgil made to the backlog page? [19:18:14] thanks zeljkof I was hoping to get rachel99 to edit the backlog page too, I'll wait a few minutes to see if she returns [19:18:39] valeriej: reviewing lines 20-24 :) [19:19:37] valeriej: lines 20-24 look good to me! :) [19:20:05] feel free to move 12-24 to the backlog (the link is at the top of the pad) [19:20:19] zeljkof: Thanks! Will do! [19:20:22] valeriej: let me know if you need help with moving to backlog [19:20:32] Ok. [19:20:47] chrismcmahon: do we call it a day? [19:21:06] zeljkof: looks that way [19:21:18] anybody else needs help of reviews? [19:21:45] chrismcmahon: this lasted longer than I thought :) and it was more productive [19:22:12] zeljkof: we can update and clean up http://www.mediawiki.org/wiki/QA/Browser_testing/Search_features at some point but no hurry [19:23:13] zeljkof, I could, yes :) How do I do that but? [19:23:57] sucheta: do you have account at mediawiki.org? (or any wikipedia I guess) [19:24:27] zeljkof, Everywhere :D [19:24:53] sucheta: just edit the page then :) cut/paste the scenarios anywhere in the backlog [19:25:04] sucheta: we will move it around if needed [19:25:13] sucheta: let me know if you need help [19:25:50] rachel99: both chrismcmahon and I think the scenarios that the two of you have written are ready to be moved to the backlog page [19:26:05] oh hi rachel99 you're back [19:26:09] zeljkof: Ok. I can do that now [19:26:10] rachel99: would you do that (the link to the backlog page is at the top of the pad) [19:26:22] rachel99: let me know if you need help with that [19:26:23] chrismcmahon: yea, I got knocked off the channel [19:26:31] rachel99: do you have an account on http://www.mediawiki.org ? [19:26:47] chrismcmahon: yes [19:27:21] rachel99: great! go ahead and log in, so you can update the page at http://www.mediawiki.org/wiki/QA/Browser_testing/Test_backlog#Search [19:27:50] zeljkof: I will need the etherpad link again, I don't have it anymore [19:28:04] rachel99: http://etherpad.wmflabs.org/pad/p/Browser-automation [19:28:06] rachel99: http://etherpad.wmflabs.org/pad/p/Browser-automation [19:30:09] rachel99: when you edit http://www.mediawiki.org/wiki/QA/Browser_testing/Test_backlog#Search take a look at how valeriej used the
 tag to keep the spacing and make the text look nice
[19:30:29] 	 rachel99: (I'm not sure how well you may know Mediawiki markup)
[19:30:33] 	 ok browser automation people, we did a good job! :) let me know if anybody needs help or reviews
[19:30:38] 	 chrismcmahon: ok.  I will look.
[19:30:59] 	 zeljkof, Under miscellaneous ?
[19:31:14] 	 sucheta: there should be search section
[19:31:27] 	 sucheta: let me check (but anywhere is fine)
[19:31:41] 	 sucheta: here is is: http://www.mediawiki.org/wiki/QA/Browser_testing/Test_backlog#Search
[19:31:46] 	 There is one
[19:31:47] 	 Yes
[19:33:53] 	 rachel99: looks good, thanks
[19:34:29] 	 chrismcmahon: you're welcome.  I'll have to drop off now..
[19:34:40] 	 rachel99: thanks for helping us! :)
[19:35:03] 	 zeljkof: Sure, and thanks to both you and Crhis for holding this workshop.
[19:35:12] 	 rachel99: more to come :)
[19:35:30] 	 zeljkof: Great.  Will see you then.
[19:36:15] 	 chrismcmahon: so we had 4 active participants
[19:36:33] 	 chrismcmahon: valeriej, rachel99, qgil and sucheta
[19:37:07] 	 good night everybody, I have to go now, real "bed time story" time :)
[19:37:17] 	 thanks zeljkof, see you tomorrow
[19:37:28] 	 chrismcmahon: see you tomorrow :)
[19:45:20] 	 New patchset: MarkTraceur; "Add parsoid tests to the YAML configs." [integration/jenkins-job-builder-config] (master) - https://gerrit.wikimedia.org/r/52750
[20:59:07] 	 qgil- Is there going to be an official IRC channel along with the YouTube stream for the Lua meetup tomorrow?
[21:17:15] 	 anomie|away: If qgil hasn't planned that I'd happily flash-mob an IRC channel with you. Maybe this one would work, or #wikimedia-office
[21:18:30] 	 hi anomie|away we can have IRC yes. I tried (improvising) in the last meetup and there was not much activity at the time so I just forgot about it
[21:19:07] 	 but if marktraceur can volunteer then great. I will be busy passing the mic. wikimedia-dev is a good one, yes
[21:19:20] 	 I'll gladly yell at you when there are questions from IRC.
[21:19:40] 	 marktraceur, you never lose a chance to yell at me snif snif
[21:19:51] 	 qgil: Aww, yes I do.
[21:20:23] 	 All of this brilliant testing/QA stuff that happened today, that was great
[21:42:21] 	 marktraceur: \o/ glad you liked it.  wish we'd had a couple more participants, but it was nice.
[22:06:43] 	 sumanah, about the security training session. csteipp sent some links to videos and slides and he is proposing to do an office-hour next monday
[22:07:01] 	 sumana csteipp in the meantime I got from HR the list of engineers hired in the past 6 months
[22:07:36] 	 sumanah, the idea is that people would watch the materials before the office hour and there csteipp would answer questions
[22:07:44] 	 sumanah,  csteipp is this what you had in mind?
[22:08:18] 	 qgil_: That sounds right to me, thanks!
[22:08:24] 	 sure!
[22:08:51] 	 great, sumanah is there any process to book wikimedia-office and check if thee are other sessions booked?
[22:08:57] 	 not that I know of.
[22:09:02] 	 Oliver might know.
[22:09:08] 	 if you don't know then I don't need to know  ;)
[22:10:15] 	 :)
[22:20:29] 	 qgil_: I think this is what you're looking for: https://meta.wikimedia.org/wiki/IRC_office_hours
[22:20:47] 	 thanks valeriej|away!
[22:20:57] 	 thank you valeriej|away ! Even being away you are so helpful.
[22:21:17] 	 sumanah, qgil_: No problem!
[22:23:52] 	 New patchset: MarkTraceur; "Add some more Parsoid tests for other components" [integration/jenkins-job-builder-config] (master) - https://gerrit.wikimedia.org/r/53687
[22:30:49] 	 hi chrismcmahon & greg-g - you have never been to a Wikimedia hackathon before
[22:31:01] 	 here is what to do:
[22:31:11] 	 1) make sure you own a passport that actually works
[22:31:11] * chrismcmahon  follows along... 
[22:31:27] 	 check, passport good into 2015
[22:31:38] 	 2) every once in a while, look at the list of the people who have said that they are going  https://www.mediawiki.org/wiki/Amsterdam_Hackathon_2013
[22:32:04] 	 3) reach out to specific people to say "hey, at the event, would you like to collaborate on _____?"
[22:32:22] 	 4) respond and have useful conversations, and prep appropriately for what you want to do when you're together
[22:32:48] 	 ok :)
[22:32:55] 	 (in some cases this may spread into "let's have it be a big group activity" in which case you should post on the talk page https://www.mediawiki.org/wiki/Talk:Amsterdam_Hackathon_2013 to let others know)
[22:33:15] 	 5) show up and be very proactive and take down people's contact info and keep it in a safe place so you can actually contact them again later
[22:33:35] 	 6) after the event, actually follow through and contact them and continue doing things together
[22:33:50] 	 TimStarling: liangent & design stuff. What's the bug # or gerrit changeset?
[22:34:08] 	 chrismcmahon: greg-g - also good ideas:
[22:34:33] 	 Look at the schedule as it coalesces and think about what you would like to attend or lead, and plan to do so
[22:35:31] 	 https://bugzilla.wikimedia.org/show_bug.cgi?id=44667
[22:35:32] 	 Encourage people you know, who you think would get to learn stuff or teach stuff, to register and ask for a scholarship
[22:37:44] 	 chrismcmahon: do you still have questions?
[22:40:07] 	 sumanah: not for now, maybe more later
[22:40:16] 	 oK1
[22:40:19] 	 er, OK!
[22:40:27] 	 ok2!
[22:40:31] 	 ha!
[22:40:39] 	 There are people here who have been to several Wikimedia tech events -- more than I have
[22:41:02] 	 like Tim & Tomasz and Mazeland
[22:46:52] 	 anomie|away: TimStarling: robla - is it possible to get *any* benchmarks on how much faster a Lua module is than the same template was in wikitext? maybe we saved the rendertime stats for one that we've already converted?
[22:47:10] 	 sumanah: the users have done benchmarks
[22:47:33] * robla  digs up what Tim wrote a while back
[22:48:06] 	 check WP:VPT, Wikipedia_talk:Lua, the talk pages of the big modules, etc.
[22:48:07] 	 http://www.mediawiki.org/wiki/Lua_scripting/Benchmarking
[22:48:17] 	 and yes, there was the benchmark I did prior to deployment
[22:49:30] 	 TimStarling: hi, did you notice my question about frame:getArgument( arg ) pls?
[22:49:51] 	 sumanah: http://test2.wikipedia.org/wiki/Module_talk:Citation
[22:49:52] 	 Danny_B: no
[22:50:03] 	 aha. thanks
[22:50:05] 	 TimStarling: In thinking about the Lua meetup tomorrow, I realized I wanted to ask you - did you consider using another language or framework for extensions? (just wondering) (and sorry we all sort of poinced on you)
[22:50:12] 	 pounced*
[22:50:30] 	 I considered using JavaScript, at length
[22:50:46] 	 I reviewed all the major JavaScript engines before I settled on Lua
[22:50:51] 	 https://www.mediawiki.org/wiki/User:Sumanah/Lua_vs_Javascript
[22:50:57] 	 marktraceur: ^^
[22:50:59] 	 well, except Trident ;)
[22:51:11] 	 all the major *open source* JS engines
[22:51:43] 	 TimStarling: Was Guile ever on your radar?
[22:51:52] 	 no
[22:52:36] 	 Thank you robla & TimStarling - I'm just gonna go with those for now, for journalist use
[22:52:42] 	 Lua is very similar to JavaScript in terms of its features and structure
[22:53:22] 	 TimStarling: I guess I figured you may have reinvented parts of Guile for use in Scribunto - I could be wrong, as my attempts with Guile have been mostly on the surface so far
[23:25:38] 	 New patchset: MarkTraceur; "Add some more Parsoid tests for other components" [integration/jenkins-job-builder-config] (master) - https://gerrit.wikimedia.org/r/53687
[23:30:09] 	 New patchset: MarkTraceur; "Add more parsoid tests" [integration/zuul-config] (master) - https://gerrit.wikimedia.org/r/53700
[23:38:38] 	 arch question:
[23:38:40] 	 need a second opinion on a new patch - commiter insists on creating protected member variables. I don't like having protected vars because the derived classes might be in extensions and will be impossible to change internal logic...
[23:38:46] 	 https://gerrit.wikimedia.org/r/#/c/34677/13/includes/Throttle.php
[23:40:00] 	 i am insisting that those vars should be wrapped in accessor methods
[23:42:35] 	 oh, you're arguing to be more conservative
[23:45:28] 	 exactly - need someone else comment that it should be with accessor methods
[23:46:01] 	 I don't have an opinion on protected vs private.  also, I'm probably the wrong person to ask about what "our" stand is on getters/setters vs properties
[23:47:06] 	 New review: Krinkle; "So what?" [integration/zuul-config] (master) - https://gerrit.wikimedia.org/r/44384
[23:48:13] <^demon>	 I prefer accessors. Then you can control what people do with your data :)
[23:48:34] 	 ^demon, 100% agree - could you comment in the patch?
[23:48:45] 	 i already -1 it several times
[23:48:46] <^demon>	 private vs. protected -- Think..."do I want a child to be able to modify this?"
[23:49:00] <^demon>	 If not, opt for private. You can always make it protected later if you change your mind.
[23:49:20] 	 i would much rather the derrived class just use a protected function to change that data
[23:49:28] 	 than go to the var directly
[23:49:54] 	 negligent cost (sometimes 0 if the JIT is smart enough)) - but much better structure
[23:49:55] <^demon>	 Well, that doesn't make all that much sense.
[23:50:12] <^demon>	 If a child needs it, why not let the child use it?
[23:50:39] 	 if the dervd needs it - why not use a getter/setter functions?
[23:50:40] 	 negligible*
[23:50:52] 	 marktraceur, hehe
[23:51:12] 	 My English syntax alarm went off :)
[23:51:19] 	 thx :)
[23:51:20] <^demon>	 Pointless overhead of function calls?
[23:52:02] <^demon>	 So yeah, I should've put a big *** after "use accessors."
[23:52:16] <^demon>	 In children, I much prefer protected and skip the accessors.
[23:52:57] 	 ^demon, but if the tiny, very infrequent (a few times per execution) function call allows us to later change the structure of the base class???
[23:53:17] 	 and the children are frequently part of extensions
[23:53:24] <^demon>	 So?
[23:53:33] <^demon>	 Update users if you change them.
[23:53:37] <^demon>	 Life goes on.
[23:53:39] 	 having an accessor allows us to change internal strructure
[23:53:51] 	 you can't - often extensions are not part of our code
[23:54:00] <^demon>	 Anyway, the bikeshed should be orange. I'm going to eat dinner.
[23:54:41] 	 New patchset: Krinkle; "Build script for mediawiki documentation" [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/39212
[23:54:48] 	 Change merged: Krinkle; [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/39212