[19:12:09] <DanielK_WMDE>	 SMalyshev: hey! i saw you enabled ElasticSearch on Wikidata! Cool!
[19:12:33] <DanielK_WMDE>	 can you drop a note about this to the wikidata mailing list, and tell people where to report any issues?
[19:34:11] <SMalyshev>	 DanielK_WMDE: I've added it to the news. I wondered if I should write on the list since it was supposed to be birthday thing?
[19:49:12] <no_justification>	 DanielK_WMDE: If you're still about...just spotted T179038 in wmf.5
[19:49:12] <stashbot>	 T179038: DomainException / NullResult holds no constraint - https://phabricator.wikimedia.org/T179038
[19:51:49] <hoo>	 no_justification: Lucas is probably the go to person for this. Assigned him.
[19:52:58] <no_justification>	 ty
[19:58:18] <no_justification>	 Also spotted Warning: Wikibase\Rdf\DispatchingValueSnakRdfBuilder::getValueBuilder: No RDF builder defined for data type globe-coordinate nor for value type bad. [Called from Wikibase\Rdf\DispatchingValueSnakRdfBuilder::getValueBuilder in /srv/mediawiki/php-1.31.0-wmf.5/extensions/Wikidata/extensions/Wikibase/repo/includes/Rdf/DispatchingValueSnakRdfBuilder.php at line 75] in
[19:58:20] <no_justification>	 /srv/mediawiki/php-1.31.0-wmf.5/includes/debug/MWDebug.php on line 309
[19:58:53] <no_justification>	 Unrelated: Fatal error: Class undefined: ProofreadPage\Index\MultilineTextInputWidget in /srv/mediawiki/php-1.31.0-wmf.5/extensions/ProofreadPage/includes/index/EditIndexPage.php on line 107
[19:58:58] <no_justification>	 James_F: I thought we fixed that last one ^
[19:59:30] <James_F>	 no_justification: We… did?
[19:59:44] <no_justification>	 Well there was a similar thing with multiline input widget things
[19:59:44] <James_F>	 no_justification: Ultra-cached PHP somehow?
[20:00:12] <no_justification>	 Definitely more than one apache reporting it
[20:00:16] <Reedy>	 You sure?
[20:00:16] <Reedy>	 https://github.com/wikimedia/mediawiki-extensions-ProofreadPage/commit/1b12648e0e5a52ce31248235d390b4e8d635f42e
[20:00:16] <no_justification>	 So not like a single one outta sync
[20:00:22] <Reedy>	 Came in 6 days ago
[20:00:40] <Reedy>	 Phpstorm can't seem to find it in the extension/core
[20:01:21] <no_justification>	 Has that version of OOJS/UI been deployed?
[20:01:23] <Reedy>	 I can only see a js MultilineTextInputWidget
[20:01:40] <James_F>	 no_justification: Yeah, two weeks ago.
[20:02:25] <no_justification>	 Reedy: Github says that commit's only in master
[20:02:57] <James_F>	 Specifically it's in OOjs UI v0.23.4 which was merged into MW master on 2017-10-11.
[20:03:20] <James_F>	 no_justification: You back-ported it?
[20:03:36] <James_F>	 no_justification: https://gerrit.wikimedia.org/r/#/q/I81e50d9585bac8323adf7b8fe7f2e0f662c556e6
[20:03:50] <Reedy>	 https://github.com/wikimedia/mediawiki-extensions-ProofreadPage/commits/wmf/1.31.0-wmf.5
[20:03:52] <Reedy>	 It's in .5
[20:03:53] <no_justification>	 Ok yeah it's in wmf.5
[20:03:57] <James_F>	 So it's in wmf.4/.5 and master.
[20:04:05] <James_F>	 We're not running wmf.3 somewhere?
[20:04:06] <Reedy>	 And .4
[20:04:06] <Reedy>	 https://github.com/wikimedia/mediawiki-extensions-ProofreadPage/commits/wmf/1.31.0-wmf.4
[20:04:09] <no_justification>	 But...it can't find the code?
[20:04:35] <James_F>	 Did it get reverted in production somehow?
[20:05:03] <no_justification>	 Hah
[20:05:05] <no_justification>	 ProofreadPage\Index\MultilineTextInputWidget in /srv/mediawiki/php-1.31.0-wmf.5/extensions/ProofreadPage/includes/index/EditIndexPage.php on line 107
[20:05:16] <no_justification>	 Namespace code lacked a use
[20:05:25] <James_F>	 Ooooh.
[20:05:26] <James_F>	 Meh.
[20:05:44] <James_F>	 One second.
[20:05:58] <no_justification>	 There's already OOUI imports at the top too!
[20:07:33] <James_F>	 no_justification: https://gerrit.wikimedia.org/r/386478
[20:09:14] <no_justification>	 +2'd and cherried to wmf.5
[20:24:27] <no_justification>	 James_F: Live. Seems to be disappearing from logs
[20:24:51] <no_justification>	 Ouch: Notice: Cannot access property on non-object in /srv/mediawiki/php-1.31.0-wmf.5/extensions/GlobalBlocking/includes/api/ApiGlobalBlock.php on line 53
[20:25:30] <no_justification>	 53, 56, 59
[22:50:51] <Krinkle>	 legoktm: Thinking about a sniff that would require a PHPUnit test*() method to have a doc block with at least 1 @covers.
[22:50:52] <Krinkle>	 Thoughts?
[22:51:15] <Krinkle>	 .. optionally tolerated if the class has one already.
[22:51:35] <Krinkle>	 Just finding quite often that when fixing code coverage for an older area, there are actually tests, but just forgotten to add @covers
[22:52:37] <Krinkle>	 opt-out via @coversNothing, although I don't think we need it. We typically do include coverage from integration tests, but when strict we'd specify only the entry method, not the indirect ones. 
[22:58:07] <legoktm>	 Krinkle: how do you determine whether the class is a phpunit test?
[22:58:15] <legoktm>	 otherwise that seems like a decent idea to me
[23:55:08] <SMalyshev>	 should each separate test have @covers? 
[23:55:49] <SMalyshev>	 I put it on class but specific tests would probably just duplicate the class one often. They frequently use more than one method from the class...