[00:04:46] Cheers for the reviews hoo! [00:06:07] :) [00:46:11] hm, should "India" etc be listed as "part of" "Eleventh constituency for French residents overseas"? [06:39:45] New review: LivingShadow; "And thanks for integrating!" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/71713 [08:02:26] liangent: http://www.wikidata.org/wiki/User:Addshore/plan [08:02:54] heh>> http://www.wikidata.org/wiki/User:Addbot/plan [08:08:17] addshore: I wonder what all 'database' mentioned refer to [08:15:37] liangent: how do you mean? [08:15:53] it referes to my database/big list of articles everywhere ;p [09:17:54] liangent: next code branch is on july 17 [09:43:15] hoo: https://bugzilla.wikimedia.org/50718 [09:43:51] doh! :) [09:45:45] :D [09:45:54] aude: Wow... that's weird [09:45:59] thanks for having a look [09:46:02] all solved, though we have a couple bugs or action items [09:46:30] need to handle the exception in a more graceful way when this happens, yet log it so we know about it [09:46:47] or do something [09:52:53] aude: mh [09:52:53] hoo: and we made a branch last week, thinking we'd deploy this week [09:52:55] but we wont :P [09:53:01] Yeah... figured that out yesterday [09:53:03] we'll be using that on monday [09:53:18] and then july 17, make new branch [09:53:33] should be able to get all of liangent's stuff in by then [09:53:59] yeah [09:53:59] about logging somewhere we know about.... if beta labs is not logging warnings, then it's tricky [09:54:05] there are debug logs [09:54:24] i'm not in a habit of looking at those regularly yet [09:54:34] and it's a lot of noise or can be [09:55:01] i think production must be logging warnings, but really don't know [09:55:32] * aude glad we discovered the sites table was empty! [10:11:10] https://www.wikidata.org/wiki/Q34404 so New Orleans is an instance of Government debt? :) [10:11:34] aparently :P [10:13:31] https://www.wikidata.org/w/index.php?title=Q34404&diff=prev&oldid=11827710 [10:13:41] has been since march! [10:14:01] lovely [10:14:28] fixed :> [10:14:36] https://www.wikidata.org/wiki/Q2431020 also [10:14:44] no one fixed that [10:15:34] done :P [10:15:44] thanks [10:16:05] I wonder why no more reviews are coming to my mailbox recently, besides hoo's :p [10:17:33] liangent: i'm slowly making my way through the review queue [10:17:35] heh https://www.wikidata.org/wiki/Special:Contributions/32.151.228.208 [10:18:19] i got a couple patches from daniel, some from hoo and yours to review [10:18:58] :) [10:19:29] heh > https://www.wikidata.org/w/index.php?title=Q6319740&curid=6127100&diff=54345526&oldid=54343907 [10:20:13] interesting... [10:20:42] i saw another page like that some time ago and didn't figure out how/why this occurs [10:20:55] ask Henning_WMDE, maybe he knows? [10:21:29] https://bugzilla.wikimedia.org/46945 [10:23:15] heh, added this case also to that bug [10:23:23] oh, okay :) [10:23:23] Mid-air collision detected! [10:23:25] xD [10:23:32] * aude added it [10:23:42] just so we know it still occurs [10:25:28] heh, i tried and failed to add an empty alias myself :P [10:26:43] me too [10:31:15] Lydia_WMDE: has the community seen the "Data item" link on test2? [10:31:16] e.g. https://test2.wikipedia.org/wiki/New_York_City [10:31:20] any feedback? [10:44:01] Change merged: jenkins-bot; [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/69845 [10:44:50] New patchset: Daniel Kinzler; "Factor string normalization functions out of Utils." [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/72078 [10:52:07] New review: Aude; "I get an exception when I try to create a claim." [mediawiki/extensions/Wikibase] (master) C: -1; - https://gerrit.wikimedia.org/r/71343 [10:55:42] New review: Aude; "array (" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/71343 [10:59:13] New review: Aude; "@jeroen poke! @daniel: what is the status of this?" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/70163 [11:00:36] New review: Aude; "needs a rebase and is blocked by https://gerrit.wikimedia.org/r/#/c/70163/ (which currently has a -2)" [mediawiki/extensions/Wikibase] (master) C: -1; - https://gerrit.wikimedia.org/r/70164 [11:03:54] New review: Aude; "this needs denny, jeroen, danwe or someone to take a look to make sure this approach and any of thes..." [mediawiki/extensions/Wikibase] (master) C: -2; - https://gerrit.wikimedia.org/r/68002 [11:08:14] New patchset: Daniel Kinzler; "Move search key generation to TermSqlIndex." [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/72079 [11:24:37] New review: Hoo man; "You can filter the tests on the JavaScript tests page (there's a module selector on the right)." [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/71713 [11:34:29] New patchset: Daniel Kinzler; "(bug 46867) trim bad utf-8 sequences before normalizing." [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/70139 [11:36:25] New patchset: Daniel Kinzler; "(bug 46867) trim bad utf-8 sequences before normalizing." [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/70139 [11:36:46] New patchset: Daniel Kinzler; "(bug 46867) skip bad search keys and report them." [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/70140 [11:37:33] New patchset: Daniel Kinzler; "(bug 46867) skip bad search keys and report them." [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/70140 [11:50:50] New patchset: Daniel Kinzler; "Factor string normalization functions out of Utils." [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/72078 [11:51:16] New patchset: Daniel Kinzler; "(bug 46867) trim bad utf-8 sequences before normalizing." [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/70139 [11:51:20] New patchset: Daniel Kinzler; "Move search key generation to TermSqlIndex." [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/72079 [11:51:26] New patchset: Daniel Kinzler; "(bug 46867) skip bad search keys and report them." [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/70140 [11:54:04] New patchset: Daniel Kinzler; "Factor string normalization functions out of Utils." [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/72078 [11:54:18] gah! can we require php 5.4 already? clusires are just broken in 5.3! [11:54:24] *closures even [11:54:38] DanielK_WMDE: WMF runs 5.3 still [11:54:48] and so do I on my Cent Os system :P [11:55:04] hoo: i know. that's the problem :) [12:06:51] DanielK_WMDE: how are they broken? [12:07:02] oh about $this? [13:19:57] hah [13:19:57] gah [13:20:04] Guest88939: do /nick aude please :P Confusing [13:20:52] Krinkle wants relase notes for https://gerrit.wikimedia.org/r/71591 that's what I wanted to tell :P [13:21:29] \o/ [13:21:37] ok, yeah i saw [13:22:08] * aude gives caffeineexcess more club mate :) [13:22:27] \o/ You can't even buy that stuff over here :P [13:22:39] At least I don't know where [13:22:39] :( [13:26:41] New patchset: Hoo man; "Allow (optional) title normalization in wbgetentities" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/71543 [13:27:25] New review: Hoo man; "@Danwe: Fixed the commit message now... I guess this should suffice" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/71543 [13:29:51] New review: Daniel Werner; "These modules have been registered in Lib for a reason. They can be useful in Client as well as in R..." [mediawiki/extensions/Wikibase] (master) C: -1; - https://gerrit.wikimedia.org/r/70115 [13:32:53] aude: Can you tell Danwe to get here please? [13:33:12] I disagree with his comment and I don't think that's something we want to discuss in gerrit [13:34:05] addshore: throw something at danwe please :) [13:34:10] poke him [13:34:21] aude: Not in the office either? [13:34:29] he is here [13:34:33] i am here [13:34:42] just not in throwing range [13:35:03] :D Did you change the tables/ places? [13:35:08] yes [13:37:55] New review: Daniel Kinzler; "@aude: please try with usePropertyInfoTable = false, or run update.php to create the table." [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/71343 [13:44:01] New review: Hoo man; "I'm aware that this change is big and nasty but I think that it's still the best option (despite jus..." [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/70115 [13:45:02] New patchset: Daniel Kinzler; "(bug 49742) Hook PropertyInfoStore in service reg." [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/71343 [13:49:56] http://meta.wikimedia.org/wiki/Wikidata/Development/Representing_values [13:49:58] New patchset: Daniel Kinzler; "Remove hacks from jenkins entry point." [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/70164 [13:50:52] JeroenDeDauw1: Y U NO REVIEW https://gerrit.wikimedia.org/r/#/c/70163/ ? [13:52:38] http://www.mediawiki.org/wiki/Extension:GeoData [13:52:45] • dim - approximate size (or size's order of magnitude) of an object. Used by GeoData to restrict search and by Geohack for determining appropriate map zoom. [13:54:07] New review: Daniel Werner; "I beg to differ. We are creating a little mess either way. The actual problem is resource loader req..." [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/70115 [13:58:59] DanielK_WMDE_: Denny_WMDE1: Abraham_WMDE1: anyone disagree with stuff in http://pastebin.com/pHQQLX8S ? [13:59:26] caffeineexcess: Danwe_WMDE is here now [13:59:55] oh... hi Danwe_WMDE :) [14:00:12] New review: Daniel Kinzler; "Fine for now." [mediawiki/extensions/DataValues] (master) C: 2; - https://gerrit.wikimedia.org/r/70386 [14:00:17] JeroenDeDauw1: looks good to me. and we're using a new organization in github, right? [14:00:24] caffeineexcess: hi [14:00:25] Change merged: jenkins-bot; [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/70593 [14:00:26] Change merged: jenkins-bot; [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/70386 [14:00:31] Abraham_WMDE1: yeah, WikidataCI [14:02:06] Danwe_WMDE: I'm still not a fan of doing all this... that's only going to reduce the module count by 6% on enwiki, so I really doubt the impact after all... Krinkle|detached suggested that this is a cleaner approach than having modules we don't use, but I'm not sure about that either [14:02:10] JeroenDeDauw1: why pull from github into gerrit? [14:02:59] caffeineexcess == hoo? [14:02:59] other way round [14:03:04] Danwe_WMDE: Oh yes :P [14:03:06] :) [14:03:36] caffeineexcess: so you are suggesting the whole change set might not even necessary? [14:03:41] DanielK_WMDE_: ??? what? [14:04:00] DanielK_WMDE_: ?? [14:04:34] Danwe_WMDE: Krinkle's gonna hate me for that, but I don't think so... at first I thought this has a bigger impact, but given that enwiki has 500+ modules already our 30 don't really matter to much [14:04:37] JeroenDeDauw1: WHUT? [14:05:38] JeroenDeDauw1: never mind, i misread. [14:05:56] JeroenDeDauw1: looks good. I had something slightly different in my mind, but i think what you suggest should work too and be less fragile actually :) [14:06:29] is one branch = one changeset ? [14:06:40] i thought, a branch can live longer and comprise several changesets? [14:07:10] caffeineexcess: that's been my thought since the very beginning. I'd rather look at a way to further improve resource loader. I am not sure I understand the whole problem, but I guess what Krinkle worries about is that the list of resources gets sent to the client on each request and is not cached at all? [14:07:35] Danwe_WMDE: It's only cached for 5 minutes and it's rather big [14:07:59] why not cache it for longer? [14:08:11] could cache it till config changes, no? [14:08:27] My idea to make it smaller (in RL) would be to wander down the dependency trees and kill duplicates... like this: foo.bar needs mediawiki.Api and mediawiki.util... as mediawiki.Api already needs mediawiki.util we can kill it [14:08:45] Danwe_WMDE: No, client doesn't know when we change our stuff [14:09:22] That might be a little expensive server side, though... [14:09:57] memcache it? [14:10:52] Danwe_WMDE: That sounds feasible... we could also make it configurable... so that it's disabled by default (this optimization maybe doesn't make sense for small wikis w/o cache), but it probably does for Wikimedia [14:11:03] I guess we could get much more than 6% this way [14:13:22] sounds like something that'd be done upstream, so perhaps suggest it to Krinkle. I am a huge fan of improving resource loader rather than trying to optimize stuff with taking compromises when it comes to maintaining clean, logical structures. [14:13:40] New patchset: Aude; "(bug 50798) Allow dynamic site link group section headings in item view" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/72090 [14:13:54] Going to do that... sounds saner than our hack [14:13:55] New review: Daniel Kinzler; "CR-1 for silent failure if normalization is turned off in the config." [mediawiki/extensions/Wikibase] (master) C: -1; - https://gerrit.wikimedia.org/r/71543 [14:14:13] New review: Aude; "@note: the big todo to refactor the entire entity view remains.... not in scope here" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/72090 [14:14:35] Denny_WMDE1: different how? [14:14:44] caffeineexcess: Also, in PHP, right now we are further separating our code into smaller modules, e.g. Data model. I am intending to do the same in JavaScript. These optimizations would just make it more difficult to properly do that. [14:15:18] I know... Krinkle doesn't like that approach, but it's IMO the way to go... one purpose = one module [14:15:35] DanielK_WMDE_: I don't like that setting after all (on Special:ItemByTitle it's black magic as well)... should we really do dieUsage if it's false... or rather warn? [14:16:40] JeroenDeDauw1: different in sense of: making pull request against the WMF clone on github. but that is more fragile as it depends on the wmf cloning to work in addition to our cloning to work. [14:17:31] caffeineexcess: agreed :) Thanks for tackling this important issue and keeping an open mind for the best solution there! [14:17:48] caffeineexcess: i'd just die. The client explicitly requested normalization, and we don't support it. that'S a reason to fail the request. [14:18:39] DanielK_WMDE_: mh... but that'll render the client widget broken on systems where the setting is false ... gnah [14:19:41] caffeineexcess: the the client system requires the repo to normalize, then it *is* broken if that setting is off [14:19:53] the question is just whether it is silently broken, or "loudly" [14:21:06] Well, normalizing in the client JS is ugly (IMO clients aren't really supposed to query other clients directly...) [14:22:34] New review: Aude; "needs a follow up" [mediawiki/extensions/Wikibase] (master) C: -1; - https://gerrit.wikimedia.org/r/72090 [14:24:09] DanielK_WMDE_: What about moving the setting (normalizeItemByTitlePageNames) into lib and then disabling the client widget in case it's !== true? [14:24:46] Or we could just ignore the setting in the API module as this isn't a black magic fallback like on the special page, but an explicit request [14:24:51] meh [14:25:31] ... and WMDE's net crashed again [14:25:43] (04:24:09 PM) caffeineexcess: DanielK_WMDE_: What about moving the setting (normalizeItemByTitlePageNames) into lib and then disabling the client widget in case it's !== true? [14:25:43] (04:24:46 PM) caffeineexcess: Or we could just ignore the setting in the API module as this isn't a black magic fallback like on the special page, but an explicit request [14:26:02] DanielK_WMDE_: Which channel are we having the scheduled IRC meeting on? Quim, you, me? [14:26:58] New review: Aude; "nevermind... please review" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/72090 [14:27:22] nileshc: good question. i'm fine with here, but i don't know what quim has in mind. [14:27:34] qgil appears not online yet [14:27:39] we could also just set up a private channel for the discussion [14:27:42] aude: yea... [14:28:20] Cool. Let's wait for him to come online, or email then? [14:28:38] nileshc: let's wait a few more minutes [14:28:55] DanielK_WMDE_,: Sure. [14:29:02] New review: Daniel Kinzler; "This seems sane, but I can't claim I fully understand the implications." [mediawiki/extensions/Wikibase] (master) C: 1; - https://gerrit.wikimedia.org/r/71996 [14:29:08] nileshc: how's it going on labs? [14:29:16] anything you need help with? [14:29:28] aude: i don't think i understand the language conversion stuff enough to actually approve that patch [14:29:40] DanielK_WMDE_: ok [14:29:54] aude: Please look for parser cache issues if you test it [14:30:08] aude: It's going swell :) I think the prototype will be up in 4-5 hours..I'll email you all when it's done. [14:30:16] caffeineexcess: ok [14:30:23] nileshc: cool :) [14:30:23] liangent said that core is going to resolve that but I wouldn't bet on that [14:30:25] After that I'll need to play around a bit with puppet. [14:31:00] nileshc: ok [14:31:21] it's a bit more learning curve but nice to have and be able to rebuild new instances easily [14:31:44] nileshc: it's good if you have a look into puppet, but don't sink too much time into it. [14:32:04] having this puppetized will help with deployment, but having a working prototype (including the fron end stuff) is more important [14:32:23] DanielK_WMDE_: i think the issue is that he already needs a new instance [14:32:29] with the security groups assigned [14:32:36] so has to rebuild it [14:32:42] DanielK_WMDE_: ok...about puppet, should I try to focus on having it puppetized, or give more time to actual development and keep the puppet worry for the fag end? [14:32:50] if puppet takes too much time, though then not great [14:32:50] hi DanielK_WMDE_ nileshc Denny_WMDE1 - morning! [14:32:53] ah. having the setup puppetized would help with that, obviously [14:32:56] hey qgil! [14:33:03] Hey qgil! :) [14:33:05] "morning", I mean :) [14:33:06] also think puppet would go a long way to having stuff deployable [14:33:28] someone has to puppetize it eventually [14:33:35] Morning to you too :P [14:33:55] Yes...about the security groups - currently I've opened the ports in 'default'. [14:34:13] nileshc: well, having it puppetized is good, but if it keeps you from coding essential features, it's not worth it. [14:34:14] (there is this IRC tradition to say "good morning" when you join a channel, regardless of your timezone and whether it is actually morning ofr you or not) [14:34:42] qgil: really, is there? [14:34:43] qgil: Ah. I was wondering what it meant. Hmm! [14:34:54] nileshc: that's fine [14:35:03] I'm sure there is a wiki page explaining it. Anyway, sorry for the digression. [14:35:07] aude: ok.. [14:35:11] qgil, nileshc: https://en.wikipedia.org/wiki/Moin [14:35:39] genau [14:35:48] Fast! [14:36:12] nileshc, I dodn't know you spoke German ;) [14:36:40] ok, so how are things going? [14:36:40] qgil: Hehe, how's 'Fast' German? [14:36:46] qgil: but "moin" actually doesn't refer to "morning"... anyway :) [14:36:48] "Almost!" [14:37:10] Ah. Genes maybe, my dad once learnt German. :P [14:37:16] :D [14:37:32] qgil: nileshcjust told us about his propgress on labs, and we were discussion the priority of poppetizing his setup [14:37:36] Things are going great,albeit I admit a bit slow that it should go, but it'll speed up after 15th. [14:37:47] than* [14:37:53] caffeineexcess: resolve what? [14:37:56] ok, good to hear [14:38:03] Denny_WMDE1: then people part of WMF org would get emails [14:38:18] Plus anyone with access there might merge such a pull request [14:38:21] nileshc, sporry if my email sounded a bit catastrophic. I just wanted to sound provocative. ;) [14:38:26] Which could break replication [14:38:32] liangent: Your patch and the parser cache key [14:39:03] qgil: As I said, I totally understood what you meant. I'd have sounded that way too if I were you. :) [14:39:05] nileshc, were you bale to attend last week's meeting? [14:39:07] nileshc, qgil: want to move to a quieter channel? [14:39:14] I mean the IRC allhands nileshc [14:39:18] caffeineexcess: ok [14:39:20] core really resolves it [14:39:23] DanielK_WMDE_, this might be a good idea, yes [14:39:28] JeroenDeDauw: yep, that would be bad. i like your suggestion better [14:39:31] a quieter channel that'd be great... [14:40:06] qgil: I'm following up on your last question on the new channel if we're moving now. [14:41:18] DanielK_WMDE_, er... #wikidata-gsoc ? [14:41:41] qgil: sounds good :) [14:42:12] Lydia_WMDE, aude: --^ [14:49:05] New review: Daniel Kinzler; "looks good in general, -1 for missing wfProfileOut calls." [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/71778 [14:51:25] DanielK_WMDE_: Saw my questions about the normalization? [14:54:16] caffeineexcess: no - where? [14:54:36] well, i saw several questions... which one didn't i answer? [14:54:57] "DanielK_WMDE_: What about moving the setting (normalizeItemByTitlePageNames) into lib and then disabling the client widget in case it's !== true?" [14:55:04] "Or we could just ignore the setting in the API module as this isn't a black magic fallback like on the special page, but an explicit request" [14:58:58] caffeineexcess: ignore the setting, or fail if it's off. [14:59:13] don't disable the feature, that would be a very surprising implication of that setting. [15:01:15] DanielK_WMDE_: Ok, so you'd say: Either ignore the setting in the API module or let the client widget fail in that case... I'd go for the first one [15:04:00] ok [15:04:08] i was surprised that you used it at all :) [15:04:32] I was unsure about that at first :P [15:04:49] Removing it is also going to make the tests less werid [15:13:46] New review: Aude; "(1 comment)" [mediawiki/extensions/Wikibase] (master) C: -1; - https://gerrit.wikimedia.org/r/71343 [15:20:52] Change merged: jenkins-bot; [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/71336 [15:23:02] New review: Denny Vrandecic; "(1 comment)" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/71182 [15:28:22] New patchset: Addshore; "Adding and fixing API examples" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/71610 [15:44:52] New patchset: Liangent; "Make MultiLangSerializationOptions aware of fallback chains" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/71182 [15:45:20] New patchset: Liangent; "Improve class LanguageFallbackChain and its factory" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/71778 [15:46:28] New patchset: Liangent; "Make MultiLangSerializationOptions aware of fallback chains" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/71182 [15:50:39] New review: Daniel Kinzler; "Minor code hygiene issues. Could be merged as-is, but I'd prefer these to be solved. Accessing a glo..." [mediawiki/extensions/Wikibase] (master) C: -1; - https://gerrit.wikimedia.org/r/71845 [15:51:56] New review: Daniel Kinzler; "(1 comment)" [mediawiki/extensions/Wikibase] (master) C: -1; - https://gerrit.wikimedia.org/r/71865 [15:52:31] Change merged: Denny Vrandecic; [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/71778 [15:58:14] New review: Daniel Kinzler; "The test still needs to be registered in RepoHooks::registerUnitTests" [mediawiki/extensions/Wikibase] (master) C: -1; - https://gerrit.wikimedia.org/r/71281 [16:04:07] New patchset: Daniel Werner; "Implemented MessageProvider for valueView experts" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/70611 [16:04:17] Change merged: Jeroen De Dauw; [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/70163 [16:04:57] New patchset: Jeroen De Dauw; "Remove hacks from jenkins entry point." [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/70164 [16:08:26] Change merged: jenkins-bot; [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/70164 [16:09:54] hoo: https://integration.wikimedia.org/ci/job/mwext-Wikibase-client-tests/1309/console [16:10:09] New patchset: Hoo man; "Allow (optional) title normalization in wbgetentities" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/71543 [16:10:19] hoo: Wikibase\Summary doesn't exist on the client [16:10:39] DanielK_WMDE_: didn't I set @group WikibaseRepo? [16:10:42] More particular, it is used in lib, while defined in repo [16:10:43] * hoo checks [16:10:59] We now have a cyclic dependency [16:11:06] hoo: there is no @group WikibaseRepo [16:11:09] at least not logically [16:11:09] * @group WikibaseRepoTest [16:11:16] ah crap [16:11:23] hoo: that means nothing for testing, no filtering is applied [16:11:32] mh [16:11:56] hoo: just mark the respective job as skipped if the class isn't there (or WB_VERSION isn't defiend) [16:12:16] probably that's the way to go, yes [16:12:54] New review: Hoo man; "Addressed Daniel's comment + one more test case" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/71543 [16:13:02] hoo: please do this soon - this failure essentially means that master is broken. Jenkins just doesn't notice because it'S currently not running client tests :) [16:13:16] on it [16:13:48] DanielK_WMDE_: markTestSkipped or silently skip? [16:13:55] Shouldn't really occur in the logs IMO [16:13:55] hoo: no, mark it [16:14:02] sure it should [16:14:03] ok :) [16:15:17] hoo: i have to run and fetch a train. once this is up, please poke people until someone merges it :) [16:15:26] New patchset: Liangent; "Remove Utils::lookup[User]MultilangText() and related functions" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/71845 [16:15:35] hoo: once it'S merged, please hit "rebase" on https://gerrit.wikimedia.org/r/#/c/70164/ [16:15:46] * DanielK_WMDE_ is off for now [16:16:42] New patchset: Hoo man; "Fix lib UpdateRepoOnMoveJobTest to run on clients" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/72106 [16:16:53] DanielK_WMDE_: https://gerrit.wikimedia.org/r/72106 there you go [16:16:54] ok [16:20:11] New patchset: Liangent; "Show entity label with language fallback in action=history" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/71865 [16:21:54] New patchset: Hoo man; "Fix lib UpdateRepoOnMoveJobTest to run on clients" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/72106 [16:22:40] New patchset: Liangent; "New special page Special:MyLanguageFallbackChain" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/71281 [16:24:12] New patchset: Liangent; "Remove Utils::lookup[User]MultilangText() and related functions" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/71845 [16:24:59] New patchset: Liangent; "Show entity label with language fallback in action=history" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/71865 [16:27:33] New patchset: Liangent; "New special page Special:MyLanguageFallbackChain" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/71281 [16:27:44] New patchset: Liangent; "Remove Utils::lookup[User]MultilangText() and related functions" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/71845 [16:27:53] New patchset: Liangent; "Show entity label with language fallback in action=history" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/71865 [16:28:37] PHP Fatal error: Class 'Wikibase\Summary' not found in /srv/ssd/jenkins-slave/workspace/mwext-Wikibase-client-tests/extensions/Wikibase/lib/includes/UpdateRepoOnMoveJob.php on line 137 ? [16:33:45] New patchset: Hoo man; "Fix client and lib tests" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/72106 [16:33:50] liangent: working on that [16:34:20] do you have merge access? [16:34:42] If so you might want to look at https://gerrit.wikimedia.org/r/72106 [16:37:36] [17:33:03] (mod) Fatal error on login attempt "WikibaseClient.php line 112: Class 'ValueParsers\ParserOptions' not found" - https://bugzilla.wikimedia.org/50807 summary (Chris McMahon) [16:37:36] [17:36:33] I have a feeling a Wikidata merge might have just killed beta labs [16:39:17] gnah [16:41:21] Probably DataValues isn't up to data there... weird [16:41:43] Could anyone merge https://gerrit.wikimedia.org/r/72106 please? Unbreaks the master unit tests... [16:45:00] aude: ^ [16:53:07] hello [16:53:14] is anyone working on badges? [16:53:52] Lydia_WMDE: ^ [16:54:44] New patchset: Addshore; "Adding and fixing API examples" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/71610 [16:55:24] addshore: Can you merge the unit tests fix? https://gerrit.wikimedia.org/r/72106 [16:59:32] hoo: aude gave me the permission to merge but I'm not sure if I understand everything there.. [17:00:32] yeah... that phpunit mock stuff is weird (and IMO rather bad documented) [17:03:50] New patchset: Bene; "fixed error message in Special:SetSiteLink" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/68043 [17:32:40] danwe? [17:32:55] yes? [17:33:03] in the office? [17:33:10] yep [17:33:15] can you let me in? :-) [17:33:27] sure [17:33:32] :-) [18:00:56] lazowik_: no :( someone volunteered and then vanished again [18:01:42] pragunbhutani: hey :) please add your stuff to http://www.wikidata.org/wiki/Wikidata:Status_updates/Next next time - otherwise it will not be sent out to the newsletter subscribers for example [18:12:02] Lydia_WMDE: mayba I could [18:12:24] lazowik_: you'd be willing to give it a try? that'd be awesome [18:12:25] *maybe [18:12:44] thre is a bug for it with some more information that i can dig up for you if you want [18:13:27] actually I came from this bug here [18:13:50] oh [18:13:52] ok :D [18:13:59] cool [18:14:13] if you have any questions about it here is a good place to ask [18:14:18] yep [18:14:23] though it is weekend now so people might not be around [18:14:25] there are two approaches [18:14:42] first proposed by Denny, which I understand [18:14:48] ok [18:14:57] and second by jeblad [18:15:04] go for whatever denny wrote [18:15:06] which I don't understand :) [18:15:08] ignore the rest [18:15:41] oh [18:15:41] ok [18:15:44] will do [18:16:18] :) [18:25:17] Lydia_WMDE: https://bugzilla.wikimedia.org/show_bug.cgi?id=50812 what do you think about this? [18:25:51] benestar: i have looked at it a minute ago in fact ;-) [18:26:01] :) [18:26:09] it would be useful but from my understanding this is not trivial [18:26:17] and what phase 3 is about basically [18:26:29] so this will have a lot of performance implications [18:26:46] but i suggest you have a chat with denny or danielk about it [18:26:48] maybe i am wrong [18:26:53] ah, ok [18:26:59] Lydia_WMDE: no, I didn'T [18:27:09] I've talked with hoo yesterday a bit [18:27:13] ah [18:27:26] however he didn't know much about it, too [18:27:35] *nod* [18:34:25] Lydia_WMDE: know that I want a t-shirt like yours (I mean the "talk to me about Wikidata") [18:34:36] SannitAway: ;-) [18:41:42] New review: Legoktm; "Thanks for working on this Addshore :)" [mediawiki/extensions/Wikibase] (master) C: -1; - https://gerrit.wikimedia.org/r/71610 [18:46:13] why are bugzilla components called "WikidataRepo/Client" instead of WikibaseRepo/Client? [18:50:02] * legoktm huggles addshore [18:50:26] liangent: iirc originally it was called "WikidataRepo/Client" and they changed the name of the extension, but it never got changed in bugzilla [18:51:38] legoktm: file a bug to change it then? [18:52:07] sure, just make sure all the links dont break :) [18:52:42] legoktm: I can't think of any existing links [18:53:01] it there some kind of link to ask bugzilla to preselect a component? [18:53:06] Maybe [18:53:12] I was thinking of the link that's always in the status update [18:53:22] Oh, seems it uses the email, so nothing should break [18:57:42] New review: LivingShadow; "Yes, but when I visit the page, the tests start right away and there is no way to tell it to just ru..." [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/71713 [19:16:33] New patchset: Siebrand; "Update punctuation" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/72129 [19:28:56] New review: Hoo man; "You have to be fast... or manully construct the URI. In general it runs smoother in WebKit/Blink bas..." [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/71713 [19:37:05] Change merged: jenkins-bot; [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/72106 [19:38:54] Change merged: jenkins-bot; [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/72129 [19:41:03] New patchset: Liangent; "New special page Special:MyLanguageFallbackChain" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/71281 [19:41:15]