[06:16:14] (03PS11) 10Smalyshev: refactoring for better performance [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/194242 [06:20:39] (03CR) 10jenkins-bot: [V: 04-1] refactoring for better performance [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/194242 (owner: 10Smalyshev) [07:46:32] [13ValueView] 15snaterlicious comment on pull request #162 1430687e0: So, you did not test the pull request after your additional changes in DataValuesJavaScript? What do you expect `value.getOption( 'calendarModel' )` to return being aware that `value` is a `TimeValue` object (https://github.com/wmde/DataValuesJavascript/blob/master/src/values/TimeValue.js)? If it is nor the URI, something is wrong as the call to `TimeValue.getCalendarModelKeyByUri [07:48:29] Ola! Why Can't I edit: https://www.wikidata.org/wiki/Q15746737 [07:48:34] I would like to add nlwiki [07:51:48] FakeYourDeath: Is je account al vier dagen oud? [07:52:01] Dunno? Op nlwiki ja? [07:52:14] 3 [07:52:21] Dus weer wachten? :( [07:52:56] Hm, het artikel is semibeveiligd ja dus alleen te bewerken door ingelogde accounts die ouder dan vier dagen zijn. [07:53:12] Ok [07:53:28] Ik had ook wat op je pagina gezet. Ik snap niet wat ik gisteren fout deed? [07:53:50] Ja, ik had al geantwoord. Lijkt mij gewoon een inschattingsfoutje. [07:54:30] ok [08:26:28] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata: Use EntityIdHtmlLinkFormatter for format properties in EntityView - https://phabricator.wikimedia.org/T70020#1103618 (10adrianheine) 5Open>3Resolved a:3adrianheine Done in baf7578c5e675a455a43f3afee3aa59d8fe4c81b [08:26:29] 10MediaWiki-extensions-WikibaseRepository: Use pre-fetched labels throughout EntityView - https://phabricator.wikimedia.org/T70019#1103622 (10adrianheine) [08:29:51] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata, 3§ Wikidata-Sprint-2015-02-25: investigate solutions for items with a large number of aliases - https://phabricator.wikimedia.org/T87579#1103623 (10Snaterlicious) Added a few more alternatives. It would be good to have as few redundant information as pos... [08:30:40] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata: [Monolingual] Can not search for language names in my language - https://phabricator.wikimedia.org/T69394#1103625 (10adrianheine) 5Open>3Resolved a:3adrianheine Done in https://github.com/wmde/ValueView/commit/dceacfc4aff1b794fa90045f738db838f93c88b8. [08:36:14] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata, 7Tracking: Migrate current selenium tests to cucumber - https://phabricator.wikimedia.org/T55849#1103629 (10adrianheine) Are there still unmigrated tests? [08:46:07] jzerebecki: aude: We don't have a move toolbar anymore, do we? [08:48:16] Found it [08:48:36] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata: save button not activated when changing order of statements - https://phabricator.wikimedia.org/T75360#1103638 (10adrianheine) 5Open>3Invalid a:3adrianheine [08:49:44] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata: changes to ordering don't persist under certain condition - https://phabricator.wikimedia.org/T60180#1103641 (10adrianheine) 5Open>3Invalid a:3adrianheine Moving is not supported anymore. [09:00:21] -_- https://www.wikidata.org/w/index.php?title=Q317990&action=history [09:00:43] why does the President of Venezuela only has one page watcher? [09:08:18] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata: editing data values using server-side parser/validation will disable "save" button until server responds - https://phabricator.wikimedia.org/T59665#1103670 (10adrianheine) This has been the case for all datavalue types for a year. I suppose users adapted t... [09:25:33] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata: Use ContentLanguages object for special pages JS - https://phabricator.wikimedia.org/T87370#1103718 (10adrianheine) [09:25:54] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata: Use ContentLanguages object for terms JS - https://phabricator.wikimedia.org/T87369#1103719 (10adrianheine) [09:26:51] sjoerddebruin: probably watchers are afraid to be mistaken for spies [09:27:05] Nemo_bis: :O [09:27:16] just kidding, in case someone takes me seriously [09:27:35] As talking bad of the president of Venezuela is the national sport ;) [09:27:49] former btw [09:36:09] 10Wikidata, 10MediaWiki-extensions-ContentTranslation, 3LE-Sprint-85: Link translated articles with other languages through Wikidata automatically - https://phabricator.wikimedia.org/T87410#1103757 (10Arrbee) [09:40:30] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata, 7I18n, 7Need-volunteer: Don't try to add labels in non-existing languages: restrict to Language::isKnownLanguageTag - https://phabricator.wikimedia.org/T39459#1103764 (10adrianheine) Due to language fallbacks, this looks pretty good now (https://www.wi... [09:43:08] Tobi_WMDE_SW: https://integration.wikimedia.org/ci/view/BrowserTests/view/Wikidata/job/browsertests-Wikidata-PerformanceTests-linux-firefox-sauce/performance/respondingTimeGraph?width=900&height=550&performanceReportPosition=TEST-features-performance_test.xml hmmmm hmmmm [09:45:40] 10Wikidata, 10MediaWiki-extensions-ContentTranslation, 5ContentTranslation-Release4, 7Design, 3LE-Sprint-84: When creating a new page, suggest doing a translation instead - https://phabricator.wikimedia.org/T88600#1103770 (10Arrbee) p:5Low>3Normal [09:47:08] Lydia_WMDE: https://integration.wikimedia.org/ci/view/BrowserTests/view/Wikidata/job/browsertests-Wikidata-PerformanceTests-linux-firefox-sauce/performance/respondingTimeGraph?width=900&height=550&performanceReportPosition=TEST-features-performance_test.xml&ab [09:47:19] :D [09:47:39] not sure what caused it to have longer loading time over several days [09:47:52] just built it again and it seems down again [09:47:57] could be several reasons [09:48:03] but good catch, thx [09:48:20] not that much new in the past days... [09:51:31] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata, 6Wiktionary: Phase 0: Centralize interwiki language links for Wiktionary - https://phabricator.wikimedia.org/T987#1103780 (10Darkdadaah) Is there a reason why only "main namespace" pages are mentioned in the above proposal? Also, does this proposal takes... [09:52:06] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata: Split validators from DataType class - https://phabricator.wikimedia.org/T76866#1103783 (10adrianheine) Possibly done with 28c003dbba7ce80a750a1e8d7f96df993ebdf66d. What do you think, @daniel? [09:55:20] 10Wikidata, 10Citoid, 6Editing, 10Possible-Tech-Projects, and 2 others: Create a system to store and query links to books - https://phabricator.wikimedia.org/T90852#1103801 (10Magnus) You can do that now (and for a while already): https://tools.wmflabs.org/autolist/?language=en&project=wikipedia&category=&... [09:57:32] (03CR) 10Thiemo Mättig (WMDE): [C: 04-1] "The description in the message made me expect a min-width, but there is a max-width in the code. Hm." (031 comment) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/195271 (owner: 10Aude) [09:58:07] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata: Implement new Wikibase Skin - https://phabricator.wikimedia.org/T75653#1103813 (10adrianheine) [09:58:08] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata: Research how we can maximize compatibility of a new Wikibase skin with Vector (and Monobook) - https://phabricator.wikimedia.org/T75652#1103810 (10adrianheine) 5Open>3Invalid a:3adrianheine Currently we are not planning to create a new Wikibase skin. [10:00:31] (03CR) 10Thiemo Mättig (WMDE): "A "max-width: 65em;" is already set on all ".wikibase-entityview > *" elements, see lib/resources/jquery.wikibase/themes/default/jquery.wi" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/195271 (owner: 10Aude) [10:01:39] CFisch_WMDE: this requires adjusting browsertests: https://gerrit.wikimedia.org/r/#/c/190799/ [10:01:49] 10MediaWiki-extensions-WikibaseClient, 10MediaWiki-extensions-WikibaseRepository, 10Wikidata: split wikibase.git - https://phabricator.wikimedia.org/T75863#1103824 (10adrianheine) [10:01:50] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata: EntityChange in lib depends on EntityContent in repo - https://phabricator.wikimedia.org/T75643#1103825 (10adrianheine) [10:02:23] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata: Find a home for EntityChange and friends - https://phabricator.wikimedia.org/T75875#1103828 (10adrianheine) See also T75643. [10:02:33] 10Wikidata, 10ValueView: MessageProviders should inherit from util.MessageProvider - https://phabricator.wikimedia.org/T92245#1103830 (10Snaterlicious) 3NEW [10:04:04] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata, 3§ Wikidata-Sprint-2015-02-25: Browsertests for geo value input - https://phabricator.wikimedia.org/T88543#1103849 (10WMDE-Fisch) [10:05:53] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata: move wikibase parsers and validators into a separate repository - https://phabricator.wikimedia.org/T75879#1103855 (10adrianheine) Validators are only used in repo now and could be moved there. [10:10:36] [13WikidataBrowserTests] 15WMDE-Fisch created 06geo_T88543 (+4 new commits): 02http://git.io/p4Vp [10:10:36] 13WikidataBrowserTests/06geo_T88543 14da73170 15WMDE-Fisch: renamed InputExtender steps and elements [10:10:36] 13WikidataBrowserTests/06geo_T88543 14a800f5b 15WMDE-Fisch: initial tests for geo properties [10:10:36] 13WikidataBrowserTests/06geo_T88543 148a1beee 15WMDE-Fisch: missing newline [10:10:50] (03PS3) 10Thiemo Mättig (WMDE): Deprecate getParserClass in favor of getInstance [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/194531 [10:12:21] [13WikidataBrowserTests] 15WMDE-Fisch opened pull request #58: initial browser tests for geo type statements and properties T88543 (06master...06geo_T88543) 02http://git.io/p4wl [10:12:26] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata, 3§ Wikidata-Sprint-2015-02-25: Browsertests for geo value input - https://phabricator.wikimedia.org/T88543#1103871 (10WMDE-Fisch) https://github.com/wmde/WikidataBrowserTests/pull/58 [10:12:33] (03CR) 10Thiemo Mättig (WMDE): [V: 032] "Jenkins is broken. Overriding. Super-trivial change is super-trivial." [extensions/WikibaseJavaScriptApi] - 10https://gerrit.wikimedia.org/r/192362 (owner: 10Hoo man) [10:12:39] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata, 5Patch-For-Review, 3§ Wikidata-Sprint-2015-02-25: Browsertests for geo value input - https://phabricator.wikimedia.org/T88543#1103872 (10WMDE-Fisch) [10:13:48] (03CR) 10jenkins-bot: [V: 04-1] Deprecate getParserClass in favor of getInstance [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/194531 (owner: 10Thiemo Mättig (WMDE)) [10:14:53] 10Wikidata: Adjust browsertests for references - https://phabricator.wikimedia.org/T92249#1103887 (10Tobi_WMDE_SW) 3NEW [10:15:27] (03CR) 10Thiemo Mättig (WMDE): [C: 032] "Cool, thanks. Restoring my +2." [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/189727 (owner: 10Henning Snater) [10:15:29] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata: Browsertests for statements on properties - https://phabricator.wikimedia.org/T74664#1103896 (10WMDE-Fisch) a:3WMDE-Fisch [10:16:05] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata, 3§ Wikidata-Sprint-2015-02-25: Browsertests for statements on properties - https://phabricator.wikimedia.org/T74664#774814 (10WMDE-Fisch) [10:16:07] 10Wikidata: Adjust browsertests for references - https://phabricator.wikimedia.org/T92249#1103900 (10Tobi_WMDE_SW) [10:16:07] 10Wikidata, 7Tracking: Wikidata Browsertests (tracking) - https://phabricator.wikimedia.org/T88541#1103899 (10Tobi_WMDE_SW) [10:16:42] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata, 3§ Wikidata-Sprint-2015-02-25: Smoke browsertests for statements on properties - https://phabricator.wikimedia.org/T74664#1103901 (10WMDE-Fisch) [10:18:24] 10Wikidata, 10WikiGrok, 3Roadmap: Test of WikiGrok --> Wikidata pipeline - https://phabricator.wikimedia.org/T91840#1103905 (10Lydia_Pintscher) [10:19:51] (03PS4) 10Thiemo Mättig (WMDE): Deprecate getParserClass in favor of getInstance [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/194531 [10:20:25] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata: php class namespace errors - https://phabricator.wikimedia.org/T65804#1103908 (10adrianheine) 5Open>3Resolved a:3adrianheine I suppose this is done since we removed all `class_alias` usages. [10:22:39] 10MediaWiki-extensions-WikibaseClient, 10Wikidata, 3Roadmap, 7Tracking: Allow accessing data from an item not connected to the current page - arbitrary access (tracking) - https://phabricator.wikimedia.org/T49930#1103912 (10Lydia_Pintscher) Yeah. April-June is more realistic. March is definitely out at thi... [10:23:35] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata, 7Need-volunteer: Replace usage of SimpleSiteLink with Wikibase\DataModel\SiteLink - https://phabricator.wikimedia.org/T65528#1103918 (10adrianheine) 5Open>3Resolved a:3adrianheine Couldn't find any `SimpleSiteLink` in Wikibase.git. [10:23:42] (03PS3) 10Thiemo Mättig (WMDE): Make options nullable/optional in all ValueFormatters [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/194112 [10:24:59] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata: editing data values using server-side parser/validation will disable "save" button until server responds - https://phabricator.wikimedia.org/T59665#1103921 (10Lydia_Pintscher) I still think we should fix this tbh even if people get used to something broken. [10:25:27] (03PS2) 10Thiemo Mättig (WMDE): Make tests independent of local group permissions [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/194532 (owner: 10Adrian Lang) [10:25:53] (03CR) 10Thiemo Mättig (WMDE): "PS2 is a rebase, primarily done to re-trigger jenkins." [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/194532 (owner: 10Adrian Lang) [10:28:40] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata, 7I18n, 7Need-volunteer: the message that shows the number of sources for a statement seems to use PLURAL incorrectly - https://phabricator.wikimedia.org/T57961#1103926 (10adrianheine) 5Open>3Invalid a:3adrianheine I don't think this is a code iss... [10:30:23] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata: [statements-ui] Qunit tests for the creation and display of statements - https://phabricator.wikimedia.org/T44576#1103930 (10adrianheine) 5Open>3Resolved a:3adrianheine [10:30:39] (03CR) 10Thiemo Mättig (WMDE): [C: 032] Add @group Database where needed in client [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/194336 (https://phabricator.wikimedia.org/T72710) (owner: 10Aude) [10:32:48] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata: Make EntityViewFactory - https://phabricator.wikimedia.org/T68486#1103941 (10adrianheine) 5Open>3Resolved a:3adrianheine Done in f027aea0c5d75df239fa3c6dc7a5b899845043f6. [10:33:10] (03CR) 10Thiemo Mättig (WMDE): [C: 032] Add @group Database to repo tests that access the db [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/193133 (https://phabricator.wikimedia.org/T72710) (owner: 10Aude) [10:34:47] (03Merged) 10jenkins-bot: Add @group Database where needed in client [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/194336 (https://phabricator.wikimedia.org/T72710) (owner: 10Aude) [10:35:36] (03CR) 10Thiemo Mättig (WMDE): [C: 032] Add @group Database tags to lib tests that use database [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/193120 (https://phabricator.wikimedia.org/T72710) (owner: 10Aude) [10:35:38] 10Wikidata, 10DataValues: PHP notice: Undefined Offset: 1 - https://phabricator.wikimedia.org/T67432#1103951 (10adrianheine) 5Open>3Resolved a:3adrianheine All our modules now support being but directly in `vendor/`. [10:36:23] (03Merged) 10jenkins-bot: Add @group Database to repo tests that access the db [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/193133 (https://phabricator.wikimedia.org/T72710) (owner: 10Aude) [10:36:47] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata: WikibaseExceptionLocalizer has repo-specific code - https://phabricator.wikimedia.org/T66995#1103965 (10adrianheine) Is this still the case? [10:36:59] (03CR) 10Thiemo Mättig (WMDE): [C: 032] Add more @group Database tags to repo tests [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/193275 (https://phabricator.wikimedia.org/T72710) (owner: 10Aude) [10:37:58] [13WikibaseDataModel] 15thiemowmde comment on pull request #403 14bfd836e: Oh. Yea, that was stupid. Will fix. 02http://git.io/p4Xr [10:38:15] (03Merged) 10jenkins-bot: Add @group Database tags to lib tests that use database [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/193120 (https://phabricator.wikimedia.org/T72710) (owner: 10Aude) [10:39:31] (03Merged) 10jenkins-bot: Add more @group Database tags to repo tests [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/193275 (https://phabricator.wikimedia.org/T72710) (owner: 10Aude) [10:40:54] wikimedia/mediawiki-extensions-Wikibase/master/b240df7 : jenkins-bot The build is still failing. http://travis-ci.org/wikimedia/mediawiki-extensions-Wikibase/builds/53780655 [10:41:05] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata, 7I18n, 7Need-volunteer: the message that shows the number of sources for a statement seems to use PLURAL incorrectly - https://phabricator.wikimedia.org/T57961#1103970 (10Amire80) 5Invalid>3Open Sorry, no, it is very much an issue. Wikibase-stateme... [10:42:30] [13WikibaseDataModel] 15thiemowmde 04force-pushed 06dataProviderDocs from 14bfd836e to 144cc366a: 02http://git.io/p41B [10:42:30] 13WikibaseDataModel/06dataProviderDocs 144cc366a 15Thiemo Mättig: Clean-up dataProvider docs and type hints [10:43:28] [13ValueView] 15thiemowmde closed pull request #163: Removed jQuery.NativeEventHandler (06master...06rmnativeeventhandler) 02http://git.io/xyqA [10:45:38] wikimedia/mediawiki-extensions-Wikibase/master/905061c : jenkins-bot The build is still failing. http://travis-ci.org/wikimedia/mediawiki-extensions-Wikibase/builds/53780904 [10:50:04] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata: False message: List of values is complete. - https://phabricator.wikimedia.org/T71671#1103979 (10adrianheine) 5Open>3Resolved a:3adrianheine Please reopen if this happens again. [10:56:51] wikimedia/mediawiki-extensions-Wikibase/master/8582831 : jenkins-bot The build is still failing. http://travis-ci.org/wikimedia/mediawiki-extensions-Wikibase/builds/53781227 [11:05:57] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata, 7I18n, 7Need-volunteer: the message that shows the number of sources for a statement seems to use PLURAL incorrectly - https://phabricator.wikimedia.org/T57961#1103997 (10adrianheine) You're right, it's like this: `wikibase-ui-pendingquantitycounter-n... [11:08:21] 10Wikidata, 3§ Wikidata-Sprint-2015-02-25: Adjust browsertests for references - https://phabricator.wikimedia.org/T92249#1103999 (10Tobi_WMDE_SW) [11:09:55] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata: wbformatvalue called twice when creating statement - https://phabricator.wikimedia.org/T73100#1104007 (10adrianheine) @Snaterlicious Still the case? [11:12:21] [13ValueView] 15thiemowmde comment on pull request #162 1430687e0: Me dumb. Sorry. 02http://git.io/p4Qb [11:14:19] 10Wikidata, 10DataValues: remove class_alias usage - https://phabricator.wikimedia.org/T87748#1104014 (10adrianheine) [11:14:21] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata: Remove class aliases from data-values/geo - https://phabricator.wikimedia.org/T74043#1104015 (10adrianheine) [11:14:44] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata, 10DataValues: Remove class aliases from data-values/geo - https://phabricator.wikimedia.org/T74043#763064 (10adrianheine) [11:16:54] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata: Split EntityIdFormatter from EntityIdValueFormatter - https://phabricator.wikimedia.org/T65308#1104022 (10adrianheine) 5Open>3Resolved a:3adrianheine Done in a79f1366a8bef7431854e30d89b7e4a312e0fa55. [11:20:03] [13ValueView] 15thiemowmde comment on pull request #162 1430687e0: This is the line that confused me, not the one above. There is an option `calendarModel` now. It contains an URI. I think this line needs to be `value.getOption( 'calendarModel' )`. The rotator changed from using string keys to URIs, right? 02http://git.io/p458 [11:20:07] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata: Cleanup PID code in Utils - https://phabricator.wikimedia.org/T44433#1104032 (10adrianheine) 5Open>3Resolved a:3adrianheine Done in acf54e9dd03a043494bab3c6f482a911092c8eee. [11:21:04] [13ValueView] 15thiemowmde comment on pull request #162 1430687e0: Oh, and no, all I can do at the moment is looking at the code. I can not test anything here because of an "Error: Unknown dependency: mw.ext.dataValues" I do not understand. 02http://git.io/p4dk [11:24:49] Lydia_WMDE: https://twitter.com/filbertkm/status/575255879907471360 :) [11:32:22] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata: create AcceptLanguageSet that is usable in DataModel - https://phabricator.wikimedia.org/T76252#1104039 (10adrianheine) I don't get this. What would it be used for? [11:33:10] [13WikibaseDataModelJavaScript] 15thiemowmde created 06mwExt (+1 new commit): 02http://git.io/p4Nl [11:33:10] 13WikibaseDataModelJavaScript/06mwExt 14ac4ea26 15Thiemo Mättig: Remove mw.ext.dataValues [11:35:59] [13WikidataBrowserTests] 15WMDE-Fisch created 06prop_statements_T74664 (+1 new commit): 02http://git.io/p4NA [11:35:59] 13WikidataBrowserTests/06prop_statements_T74664 1470fcc52 15WMDE-Fisch: added smoke tests for statements on properties [11:38:39] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata, 7Need-volunteer: Special pages should check permissions and read-only mode before showing a form - https://phabricator.wikimedia.org/T45024#1104045 (10adrianheine) Is that still the case? When doing https://gerrit.wikimedia.org/r/#/c/194532/, I got the f... [11:38:46] [13WikidataBrowserTests] 15WMDE-Fisch 04force-pushed 06prop_statements_T74664 from 1470fcc52 to 14cf142e9: 02http://git.io/p4AP [11:38:46] 13WikidataBrowserTests/06prop_statements_T74664 14cf142e9 15WMDE-Fisch: added smoke tests for statements on properties [11:39:06] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata, 3§ Wikidata-Sprint-2015-02-25: Smoke browsertests for statements on properties - https://phabricator.wikimedia.org/T74664#1104046 (10WMDE-Fisch) https://github.com/wmde/WikidataBrowserTests/pull/59 [11:39:12] [13WikidataBrowserTests] 15WMDE-Fisch opened pull request #59: smoke tests for statements on properties T74664 (06master...06prop_statements_T74664) 02http://git.io/p4AQ [11:39:16] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata, 5Patch-For-Review, 3§ Wikidata-Sprint-2015-02-25: Smoke browsertests for statements on properties - https://phabricator.wikimedia.org/T74664#1104047 (10WMDE-Fisch) [11:40:36] 10Wikidata, 10MediaWiki-extensions-Scribunto, 7Easy: Detect the edit state to adapt help messages to user errors - https://phabricator.wikimedia.org/T53660#1104054 (10adrianheine) I think this is not Wikidata-specific. @Lydia_Pintscher? [11:41:22] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata, 7I18n, 7Need-volunteer: the message that shows the number of sources for a statement seems to use PLURAL incorrectly - https://phabricator.wikimedia.org/T57961#1104055 (10Amire80) Yes, T72838 is closely related and offers a solution. [11:41:46] 10MediaWiki-extensions-WikibaseClient, 10Wikidata, 7Technical-Debt: Keyword 'use' inconsistency for php namespaces - https://phabricator.wikimedia.org/T45295#1104057 (10adrianheine) I think we are moving towards `use`-ing them, but I'm not sure. @ThiemoWMDE? [11:41:54] [13WikibaseDataModelJavaScript] 15thiemowmde opened pull request #27: Remove mw.ext.dataValues (06master...06mwExt) 02http://git.io/p4x4 [11:43:40] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata: [statements-ui] Display red links for deleted Properties - https://phabricator.wikimedia.org/T47675#1104063 (10adrianheine) This is the inverse to T67533. [11:43:42] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata, 7Need-volunteer, 7Testme: Special pages should check permissions and read-only mode before showing a form - https://phabricator.wikimedia.org/T45024#1104065 (10Lydia_Pintscher) [11:45:24] [13ValueView] 15thiemowmde pushed 1 new commit to 06datavalues070: 02http://git.io/p4p3 [11:45:24] 13ValueView/06datavalues070 14dbdffee 15Thiemo Mättig: Fix mistake [11:51:28] [13ValueView] 15thiemowmde 04force-pushed 06datavalues070 from 14dbdffee to 14929a808: 02http://git.io/xd8k [11:51:28] 13ValueView/06datavalues070 141332c7a 15snaterlicious: Removed time.js dependencies [11:51:28] 13ValueView/06datavalues070 14929a808 15Thiemo Mättig: Fix the tests [11:53:37] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata: Make a separate extension out of the template engine - https://phabricator.wikimedia.org/T43813#1104091 (10adrianheine) 5Open>3declined a:3adrianheine Core now has its own template engine. We should rather switch to that one: T78559 T91067. [11:54:09] [13ValueView] 15thiemowmde comment on pull request #162 14929a808: I was referring to this line. I fixed it in my commit, see https://github.com/wmde/ValueView/commit/929a80853a03c4a83e1d8b23ed5a1ca4454753c6#diff-49060d53e8ad9e0d74b8e4ea672589a5L53. 02http://git.io/p4je [12:18:02] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata, 7Need-volunteer: Wikidata edits do not update "page watched" icon - https://phabricator.wikimedia.org/T87640#1104112 (10adrianheine) We have code in place that tries to do that, so apparently it's broken. [12:26:20] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata: editing data values using server-side parser/validation will disable "save" button until server responds - https://phabricator.wikimedia.org/T59665#1104142 (10adrianheine) I don't think it's broken per se. In any case, supporting saving of unparsed values... [12:54:47] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata, 10Possible-Tech-Projects, 7Tracking: Adding new datatypes to Wikidata - https://phabricator.wikimedia.org/T91505#1104204 (10NiharikaKohli) @Physikerwelt ping! @Lydia_Pintscher ping! [13:10:22] 10MediaWiki-extensions-WikibaseClient, 10Wikidata, 7Technical-Debt: Keyword 'use' inconsistency for php namespaces - https://phabricator.wikimedia.org/T45295#1104240 (10thiemowmde) Is this about `use` keywords or `@uses` tags in tests? The first doesn't need discussion, in my opinion. The second does. I'm p... [13:10:44] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata, 10Possible-Tech-Projects, 7Tracking: Adding new datatypes to Wikidata - https://phabricator.wikimedia.org/T91505#1104241 (10Physikerwelt) @NiharikaKohli I can give guidance and advice with regard to the math part and general concepts in software engine... [13:19:20] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata, 7Need-volunteer: add a new datatype for formulas - https://phabricator.wikimedia.org/T67397#1104252 (10Physikerwelt) @NiharikaKohli A subsequent task would be to allow using more than just meaningless plain text in formulae. I think it would be nice to h... [13:21:38] [13DataValuesJavascript] 15thiemowmde 04force-pushed 06pad from 14691c9c0 to 142c96e34: 02http://git.io/pB4R [13:21:38] 13DataValuesJavascript/06pad 142c96e34 15Thiemo Mättig: Fix and test the TimeValue.pad() function [13:21:42] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata, 10Possible-Tech-Projects, 7Tracking: Adding new datatypes to Wikidata - https://phabricator.wikimedia.org/T91505#1104253 (10NiharikaKohli) @Physikerwelt, understood. The math formulas task alone was small to be a complete GSoC project, hence it seemed... [13:26:45] [13DataValuesJavascript] 15thiemowmde closed pull request #60: Fix some code style warnings (06master...06warnings) 02http://git.io/bJs2 [13:39:14] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata: create AcceptLanguageSet that is usable in DataModel - https://phabricator.wikimedia.org/T76252#1104267 (10JanZerebecki) If we need T73512 (maybe we don't as what Wikibase.git implements is enough) then we would need an argument to it that encodes the crit... [13:41:06] 10Wikidata, 3§ Wikidata-Sprint-2015-02-25: Remove wrong "Gregorian" display from diff views - https://phabricator.wikimedia.org/T92265#1104271 (10thiemowmde) 3NEW a:3thiemowmde [13:42:04] 10Wikidata, 3§ Wikidata-Sprint-2015-02-25: Remove wrong "Gregorian" display from diff views - https://phabricator.wikimedia.org/T92265#1104280 (10thiemowmde) [13:42:06] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata, 5Patch-For-Review: User interface displays false birth and death dates - https://phabricator.wikimedia.org/T87312#1104279 (10thiemowmde) [13:42:27] 10Wikidata, 3§ Wikidata-Sprint-2015-02-25: Remove wrong "Gregorian" display from diff views - https://phabricator.wikimedia.org/T92265#1104271 (10thiemowmde) [13:42:28] 10Wikidata, 7Tracking: Bugs related to time datatype - https://phabricator.wikimedia.org/T87764#1104281 (10thiemowmde) [13:42:48] 10Wikidata, 5Patch-For-Review, 3§ Wikidata-Sprint-2015-02-25: Remove wrong "Gregorian" display from diff views - https://phabricator.wikimedia.org/T92265#1104271 (10thiemowmde) [13:45:12] 10Wikidata, 6Labs: wdq.wmflabs.org does not update (data week old) - https://phabricator.wikimedia.org/T89583#1104295 (10JanZerebecki) 5Open>3Resolved [13:45:44] 10Wikidata, 6Labs: wdq.wmflabs.org does not update (data week old) - https://phabricator.wikimedia.org/T89583#1039840 (10JanZerebecki) Seems to work fine now, please reopen if I missed something. [13:45:53] 10Wikidata, 10DataValues, 5Patch-For-Review, 3§ Wikidata-Sprint-2015-02-25: Remove wrong "Gregorian" display from diff views - https://phabricator.wikimedia.org/T92265#1104298 (10thiemowmde) [14:00:23] 10Wikidata, 3§ Wikidata-Sprint-2015-02-25: Adjust browsertests for references - https://phabricator.wikimedia.org/T92249#1104316 (10WMDE-Fisch) a:3WMDE-Fisch [14:11:12] (03PS1) 10Thiemo Mättig (WMDE): Get rid of SerializerBaseTest::getClass [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/195558 [14:24:39] 10Wikidata, 10DataValues: Get rid of get[Formatter|Parser]Class methods in tests - https://phabricator.wikimedia.org/T92268#1104380 (10thiemowmde) 3NEW a:3thiemowmde [14:25:35] 10Wikidata, 10DataValues, 5Patch-For-Review, 3§ Wikidata-Sprint-2015-02-25: Get rid of get[Formatter|Parser]Class methods in tests - https://phabricator.wikimedia.org/T92268#1104388 (10thiemowmde) [14:32:32] 10Wikidata, 10Wikimedia-Hackathon-2015: Wikidata activities at the Wikimedia Hackathon 2015 - https://phabricator.wikimedia.org/T92270#1104415 (10Qgil) 3NEW a:3Lydia_Pintscher [14:34:19] [13Interfaces] 15thiemowmde created 06getFormatterClass2 (+1 new commit): 02http://git.io/pBFc [14:34:19] 13Interfaces/06getFormatterClass2 14bd31c53 15Thiemo Mättig: Drop ValueFormatterTestBase::getFormatterClass [14:36:31] [13Interfaces] 15thiemowmde opened pull request #10: Drop ValueFormatterTestBase::getFormatterClass (06master...06getFormatterClass2) 02http://git.io/pBbO [14:37:50] 10Wikidata, 10DataValues, 5Patch-For-Review, 3§ Wikidata-Sprint-2015-02-25: Get rid of get[Formatter|Parser]Class methods in tests - https://phabricator.wikimedia.org/T92268#1104447 (10thiemowmde) [14:41:49] (03PS5) 10Henning Snater: Edit References along with Statement [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/190799 (https://phabricator.wikimedia.org/T87759) [14:43:01] [13Interfaces] 15thiemowmde pushed 1 new commit to 06getFormatterClass2: 02http://git.io/pBxI [14:43:01] 13Interfaces/06getFormatterClass2 14c9af195 15Thiemo Mättig: alpha [14:46:04] [13Common] 15thiemowmde created 06getParserClass2 (+2 new commits): 02http://git.io/pBp9 [14:46:04] 13Common/06getParserClass2 144d2d9e0 15Thiemo Mättig: Deprecate getParserClass in favor of getInstance [14:46:04] 13Common/06getParserClass2 14cc8ac09 15Thiemo Mättig: Drop ValueParserTestBase::getParserClass [14:47:04] [13Common] 15thiemowmde opened pull request #25: Drop ValueParserTestBase::getParserClass (06master...06getParserClass2) 02http://git.io/pBhm [14:47:31] (03PS6) 10Henning Snater: Edit References along with Statement [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/190799 (https://phabricator.wikimedia.org/T87759) [14:48:30] 10Wikidata, 10DataValues, 5Patch-For-Review, 3§ Wikidata-Sprint-2015-02-25: Get rid of get[Formatter|Parser]Class methods in tests - https://phabricator.wikimedia.org/T92268#1104503 (10thiemowmde) [14:51:48] 10Wikidata, 10Wikimedia-Hackathon-2015: Wikidata activities at the Wikimedia Hackathon 2015 - https://phabricator.wikimedia.org/T92270#1104512 (10Qgil) [15:03:46] (03CR) 10Daniel Kinzler: [C: 04-1] "Some thoughts/comments from a first pass over the code. Main worry is the "delete all the stuff below later" comment." (0314 comments) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/194750 (owner: 10Smalyshev) [15:23:29] 10Wikidata, 10DataValues: Move Wikibase components from GitHub to Gerrit - https://phabricator.wikimedia.org/T92279#1104644 (10thiemowmde) 3NEW [15:26:00] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata, 7Need-volunteer: "unknown value" and "no value" in diff should be italic - https://phabricator.wikimedia.org/T63662#1104658 (10aude) italics can be weird and not appropriate in some languages (e.g. Japanese [1]) maybe there is another option than italic... [15:33:51] 10Wikidata, 10DataValues, 5Patch-For-Review: Make [Formatter|Parser]Options nullable/optional everywhere - https://phabricator.wikimedia.org/T92280#1104672 (10thiemowmde) 3NEW a:3thiemowmde [15:34:18] 10Wikidata, 10DataValues, 5Patch-For-Review, 3§ Wikidata-Sprint-2015-03-11: Make [Formatter|Parser]Options nullable/optional everywhere - https://phabricator.wikimedia.org/T92280#1104680 (10thiemowmde) [15:36:40] is it possible with Wikibase to do this type of thing out of the box: https://www.mediawiki.org/wiki/Extension:Cargo/Querying_data [15:38:25] 10Wikidata: Browsertests for link-item widget on Wikidata-Clients - https://phabricator.wikimedia.org/T92281#1104693 (10Tobi_WMDE_SW) 3NEW [15:38:39] 10Wikidata: Browsertests for link-item widget on Wikidata-Clients - https://phabricator.wikimedia.org/T92281#1104701 (10Tobi_WMDE_SW) [15:38:40] 10Wikidata, 7Tracking: Wikidata Browsertests (tracking) - https://phabricator.wikimedia.org/T88541#1104700 (10Tobi_WMDE_SW) [15:42:54] 10Wikidata, 10MediaWiki-extensions-Scribunto, 7Easy: Detect the edit state to adapt help messages to user errors - https://phabricator.wikimedia.org/T53660#1104715 (10hoo) >>! In T53660#1104054, @adrianheine wrote: > I think this is not Wikidata-specific. @Lydia_Pintscher? It's not Wikidata specific (at all... [15:49:24] DanielK_WMDE, it's kinda sad to me that there's all this OSM stuff but not one reply to my concerns *sigh* [15:56:06] [13WikidataBrowserTests] 15WMDE-Fisch created 06new_references_T92249 (+1 new commit): 02http://git.io/pR2C [15:56:06] 13WikidataBrowserTests/06new_references_T92249 145b87e16 15WMDE-Fisch: refactored reference browser test to T87759 [15:59:56] 10Wikidata, 10DataValues: Move Wikibase components from GitHub to Gerrit - https://phabricator.wikimedia.org/T92279#1104768 (10Snaterlicious) T74907 All in for merging `DataValues` repositories. However, there should be generic parsers for more complex data structures like `Time`. I could imagine such residin... [16:00:33] Sylvain_WMFr: https://www.youtube.com/watch?v=hjbPszSt5Pc [16:01:38] 10MediaWiki-extensions-WikibaseRepository, 10Wikidata, 7Need-volunteer: "unknown value" and "no value" in diff should be italic - https://phabricator.wikimedia.org/T63662#1104774 (10thiemowmde) No need for a new design decision, there are already CSS classes in wikibase.css for exactly this case: ``` .wb-sna... [16:05:46] [13WikidataBrowserTests] 15WMDE-Fisch pushed 1 new commit to 06new_references_T92249: 02http://git.io/pRKm [16:05:46] 13WikidataBrowserTests/06new_references_T92249 14baba2a2 15WMDE-Fisch: fixed removed step [16:15:00] [13WikidataBrowserTests] 15WMDE-Fisch opened pull request #60: refactored reference browsertests to T87759 (06master...06new_references_T92249) 02http://git.io/pRMT [16:15:18] 10Wikidata, 3§ Wikidata-Sprint-2015-02-25: Adjust browsertests for references - https://phabricator.wikimedia.org/T92249#1104835 (10WMDE-Fisch) https://github.com/wmde/WikidataBrowserTests/pull/60 [16:15:33] 10Wikidata, 5Patch-For-Review, 3§ Wikidata-Sprint-2015-02-25: Adjust browsertests for references - https://phabricator.wikimedia.org/T92249#1104839 (10WMDE-Fisch) [16:16:12] 10Wikidata, 10Wikidata.org, 6Wikisource: mul: interwiki now available for oldwikisource //wikisource.org -> Wikidata - https://phabricator.wikimedia.org/T73406#1104855 (10Aklapper) [16:17:08] is it possible with Wikibase to do this type of thing out of the box: https://www.mediawiki.org/wiki/Extension:Cargo/Querying_data [16:17:38] 10MediaWiki-extensions-WikibaseClient, 6Wikisource, 7Verified: UpdateRepoOnMoveJob not registered in wikidata for wikisource page moves - https://phabricator.wikimedia.org/T64779#1104931 (10Aklapper) [16:20:41] emacsen: i think OSM needs a policy that sais that only *sourced* info from wikidata is acceptable. [16:21:27] or at least only info from statements, not aliases, if quality issuea qwith aliases are as common as you say [16:21:33] DanielK_WMDE: https://phabricator.wikimedia.org/tag/%C2%A7_wikidata-sprint-current/ [16:21:33] that's for the OSM community to decide. [16:24:14] what does pruneChanges.php do? does it change claims or does it shorten the edit history? [16:27:46] despens: neither it only deletes old rows from the table wb_changes, which is used to dispatch changes to client wikis [16:28:25] i see [16:28:52] i don't have change propagation set up anyway [16:30:56] despens: then you can use pruneChanges to empty out wb_changes [16:31:27] it'll just keep growing otherwise [16:32:02] i will set up a cron job then [16:32:32] beluga_: wikibase doesn't have any built in query functionality yet. that'S obviously silly, and we are working to change that, but it's not easy to get it right [16:33:12] DanielK_WMDE: https://phabricator.wikimedia.org/T89175 [16:33:22] DanielK_WMDE: thanks! that's what I gathered from my research today, but wanted to make sure :) I'm from the LibreOffice quality assurance team and we are thinking about ways to enhance our wiki [16:34:25] DanielK_WMDE: so we could f.ex. "make a list of all QA people who have a mac for testing" [16:34:26] beluga_: Wikiabse is a good choice if you want to collect secondary or tertiary data. For maintaining primary data, I'd recommend SMW. [16:34:48] DanielK_WMDE: yeah I guess we have to wait until SMW gets more multilingual [16:35:15] but luckily that seems to be happening http://semanticmediawiki.org/wiki/SMWCon_Spring_2015/SMW_Grammars_%26_Variables [16:35:24] I hope we'll see more synergie between SMW and Wikibase in the future... [16:35:49] spotted that Cargo thing here http://semanticmediawiki.org/wiki/SMWCon_Spring_2015/Cargo_and_the_future_of_SMW [16:35:53] some of the key SMW people are also involved with Wikibase in some way, so the chances are good [16:55:42] (03CR) 10Smalyshev: Refactor tests to be data-driven (0310 comments) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/194750 (owner: 10Smalyshev) [16:56:32] DanielK_WMDE: I've added some comments to the tests patch. Could you point out which issues are the most important ones that prevent going forward with it? [16:59:10] DanielK_WMDE, that seems reasonable. Is there an explanation on a web page somewhere on what sourced means. I think most OSMers think of Wikidata like Wikipedia, "generallly reliable" [16:59:47] emacsen: the same thing "reliable source" means on wikipeda. [17:00:20] DanielK_WMDE, I know but is it spelled out somewhere? [17:00:48] emacsen: https://www.wikidata.org/wiki/Help:Sources [17:00:58] DanielK_WMDE, thx [17:01:03] though that is largely technical [17:01:11] DanielK_WMDE, yeah... I am seeing that [17:01:18] https://en.wikipedia.org/wiki/Wikipedia:Identifying_reliable_sources [17:01:35] this more or less also applies to wikidata. or should, imho [17:01:48] i', just that software guy, you know ;) [17:01:56] SMalyshev: looking at your comments now. [17:03:07] DanielK_WMDE: basically the part parked for deletion is kept only because there are other patches. but if it's a problem I can delete it now and use git history [17:03:50] DanielK_WMDE, the issue is't reliable sources, the issue is "sourced vs not sourced" [17:04:46] s/parked/marked/ [17:04:51] SMalyshev: i'm trying to understand the relationshp to other patches. [17:05:06] SMalyshev: if these other patches do not depend on this one, then they should be unaffected by this one [17:05:34] emacsen: well "source = Wikipedia" wouldn't be helpful, right? we have that a lot. [17:05:34] DanielK_WMDE: they depend on the main RDF patch, so they change the tests for it. [17:06:27] SMalyshev: so, if the data driven test patch goes in first, they would need to be modified to change the data driven tests instead. [17:06:36] but that needs to happen anyway, right? [17:06:40] at some point or other [17:07:10] DanielK_WMDE: right. but it's easier to modify when I have the old infrastructure so I could use old tests to generate data for new tests [17:07:28] DanielK_WMDE: if I just drop the old ones I will have to construct data for the new ones manually [17:08:40] jzerebecki: http://www.saltycrane.com/blog/2014/11/how-install-grunt-ubuntu-1404/ [17:08:48] SMalyshev: ah, i see. this makes for messy code and patches :( [17:09:10] Tobi_WMDE_SW: on the lts before that it was node instead of nodejs [17:09:23] SMalyshev: maybe having a local branch with the old code for generating new test cases would be the best solution [17:09:23] DanielK_WMDE: temporarily, yes. I could merge everything in one patch, but the it'd be even bigger :) [17:09:34] no, that's not a good idea :) [17:09:36] DanielK_WMDE: I can do that too [17:10:03] SMalyshev: would you? it would make it much more obvious what is going on, and would get rid of a ton of legacy code. [17:12:27] SMalyshev: regarding the file names: P_info.nt should be called Q1_info.nt, P_refs.nt should be called Q7_refs.nt, etc [17:12:38] that would make the relationship of the files obvious [17:12:44] DanielK_WMDE: ah, no. Same entity can be used in different tests [17:12:50] jzerebecki: symlink? https://github.com/joyent/node/issues/3911#issuecomment-8956154 [17:13:03] DanielK_WMDE: since it can be serialized in different ways with different data [17:13:10] SMalyshev: sure, no problem, you can have Q1_foo.nt, Q1_bar.nt, etc [17:13:29] SMalyshev: unless you use multiple items as input to generate a single output file. [17:13:44] but then you can still use Q1_Q2_Q3_blabla.nt [17:13:47] DanielK_WMDE: no, right now it's always a single one [17:13:52] I can add prefixes [17:14:02] so, Q1_info.nt should work fine, no? [17:14:08] thanks [17:14:27] DanielK_WMDE: ahh, also RdfSerializerTest relies on old methods [17:15:02] DanielK_WMDE: I didn't change that one [17:15:20] SMalyshev: how hard would it be to change the serializer test to also use the data driven approach? [17:15:26] should be nice & easy, right? [17:15:56] just put the expected output in separate files. [17:16:08] DanielK_WMDE: I don't know I didn't touch this one. But this means the patch gets bigger... [17:16:13] would be sensitive to order & whitespace... but the current tests are too, right? [17:16:29] hm, let me have a look [17:16:32] DanielK_WMDE: order not really, it's sorted [17:16:51] DanielK_WMDE: whitespace yes, but it uses regexps so it can be fixed [17:18:47] SMalyshev: ok, to keep the patch mall and concerns separate, plese move the stuff that RdfSerializerTest needs to a separate class. [17:18:52] Like RdfTestGraphs or something [17:19:02] DanielK_WMDE: about the statics - there's a slight problem in phpunit that generators are called before setUpForClass. which means I can't use setup to initialize something I will use in generators [17:19:30] DanielK_WMDE: can I just move it to RdfSerializeTest? it's the only thing now that uses it [17:19:37] SMalyshev: i tend to use getters that apply lazy initialization. [17:19:51] DanielK_WMDE: getters? [17:20:07] yea, sure, you can also move it into RdfSerializerTest if it's not a lot. you can use the old test data from before your patch [17:20:36] DanielK_WMDE: you mean like getCodec()? but doesn't phpunit create a new object for each test? [17:21:25] SMalyshev: it does. so what? the performance hit is negligable [17:21:58] i tend to instantiate everythign fresh for every test case. [17:22:17] what makes tests slow is i/o, not object instantiation (if you arn't doing it a few thousand times) [17:22:25] DanielK_WMDE: looks a bit wasteful for me - creating the same object repeatedly [17:22:31] or db interaction. that *sucks* in tests [17:22:44] SMalyshev: yea, but micro-optimization is the root of all evil :) [17:22:48] especially in test cases [17:22:53] DanielK_WMDE: well, we don't have much DB interaction, but we do have performance problems in RDF [17:23:15] yea, i know. one issue at a time. [17:23:37] i'd like to get your patch in, then look at how we can re-structure the code to allow sequential output generation [17:23:38] granted, test performance is less important, but if we'd have lots of tests... ok, I'll see what the effect of that would be [17:24:00] DanielK_WMDE: I'm not sure fully sequential is possible with Turtle [17:24:23] you have to have some intermediate buffer [17:24:31] anyway, let me fix the tests first [17:24:39] SMalyshev: i started using unit tests in earnest only with the wikidata project. in the beginning, i wrote a lot of code to try and re-use objects, to make tests faster (they were crawling). [17:24:52] it only caused pain and did nothing for performance. i have abandoned this approach. [17:25:11] instead, i take care to make everything injectable, and use mocks for everything, to avoid i/o and DB access [17:25:16] that makes a huge difference [17:26:14] SMalyshev: the onyl thing i foudn that needs buffering is namespace declaratiosn for xml output. it means that you have to declare all prefixes before emitting the first triple [17:26:44] turtle lends itself to sequential output very nicely. nt needs limited context (current subject and verb) [17:27:09] xml would need more buffering if you want to make it pretier, e.g. by putting some values into attributes, but it's not needed. [17:28:15] DanielK_WMDE: what did you mean on your comment about getSiteList? I didn't get that [17:33:09] SMalyshev: There is a class called TestSites that provides a standard dummy list. no need to write your own. [17:33:23] Oh, i git the name wrong. TestSites::getSites() [17:33:34] just a thought, not a problem, though [17:56:51] (03PS3) 10Smalyshev: Refactor tests to be data-driven [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/194750 [17:57:34] DanielK_WMDE: I've updated the tests thing [17:58:15] SMalyshev: cool, thanks! I think now i can use the tests to understand the main patch :) [17:59:04] DanielK_WMDE: TestSites::getSites() returns array but I need SiteList [17:59:17] SMalyshev: new SiteList( TestSites::getSites() ) [17:59:30] but yea, it's not really worth it [17:59:32] never mind [18:00:19] DanielK_WMDE: that list also misses ruwiki [18:00:39] it does? booo ;) [18:03:31] ok, it's gerrit-is-dog-slow-in-europe time again [18:03:34] every. day. [18:03:40] why? [18:07:34] DanielK_WMDE: proxy [18:07:44] ? [18:07:49] use a proxy for gerrit [18:07:57] * aude has resorted to doing that [18:08:34] aude: can you put a brief howto on mediawiki.org, on the gerrit page? [18:09:13] DanielK_WMDE: sure [18:16:18] DanielK_WMDE: https://www.mediawiki.org/wiki/Gerrit/Advanced_usage#ssh_proxy_to_gerrit [18:16:39] i'm sure there are other types of proxies / proxy services one can use [18:17:19] suppose one could even use labs for this purpose [18:17:33] that was my first thought :) [18:17:37] yeah [18:17:39] thanks for the howto! [18:17:43] sure [19:12:25] SMalyshev: i don't understand the purpose of RdfSerializer::PRODUCE_PROPERTIES [19:13:17] DanielK_WMDE: when you output the entity (single entity, since that's what RDF builder works on) you have an option to also add definitions of properties to the dump, or not [19:13:40] DanielK_WMDE: usually it's not needed but may be useful on updates, when new properties may be added [19:13:42] SMalyshev: but why treat properties specially? shouldn't that switch be for all "mentioned" entities? [19:14:36] DanielK_WMDE: mentioned entities usually are not very important, as they are just IRIs. Properties may be since they are typed, etc. [19:14:57] DanielK_WMDE: old code had this, so I kept it as an option [19:15:16] old code had an option for this?... [19:15:21] DanielK_WMDE: I can add option for mentioned entities too (it's in TODOs) [19:15:36] well, you can just call the method for resolving them, or not. [19:15:51] actually ,maybe the properties-or-not switch should be there [19:15:56] DanielK_WMDE: I don't think it was an option, I think it always did that. But on dump, you don't need to do this. On single entity, you may sometimes want to do it [19:16:19] DanielK_WMDE: which method? we're talking about dumps [19:16:30] dumps don't have methods [19:16:33] resolvedMentionedEntities() [19:16:48] it's not called by the builder [19:17:12] it'S called from higher up, if mentioned entities are wanted [19:17:21] yes for linked data, no for dumps [19:17:38] could have a switch for "properties only" or something [19:18:01] DanielK_WMDE: that would produce a lot of entities in single entity export, which usually not what we want [19:18:05] right now, you can exclude properties; from what you are saying it seems like it would make more sense to exclude everything *but* properties [19:18:24] DanielK_WMDE: in fact, we may not even want properties depending on why we ask for it [19:18:40] yea - you may want everything, or only properties, or nothing [19:18:54] right now you can get everything, or everything except properties, or nothing [19:19:19] doesn't make much sense to me. but doesn't do much harm, either [19:20:03] DanielK_WMDE: ahh, I've commented out entityMentioned() in addStatementForWikibaseEntityid in later patches I thinl [19:20:06] think [19:20:23] that's not good, should at least be an option [19:20:38] we want them for their labels [19:20:42] ok, I can add an option, though I'm not sure who would use it or why [19:20:55] linked data tools, for display [19:21:03] labels and type [19:21:14] DanielK_WMDE: ok, I'll add an option for this [19:21:16] note that referenced entities are not always items [19:21:20] they could be properties, too [19:21:21] in further patches [19:21:28] yea [19:21:43] DanielK_WMDE: yeah we don't care about which type of entity it is I think [19:21:52] well, i'd actually prefer this to be a parameter to resolveMentionedEntities [19:21:57] they could be grouped by entity type [19:22:26] well, we don't care while we output the rdf, but consumers may care, so they'd want a stub that mentions the type [19:22:31] oh, btw... [19:22:39] DanielK_WMDE: I'd suggest adding a ticket in phabricator describing how you want it to look like [19:22:50] we'd want a test case for statements on properties, including entity references to properties [19:23:01] yea, i'll add a bunch of tickets ;) [19:23:20] DanielK_WMDE: how properties are different? RdfBuilder accepts entity. [19:23:22] i'm currently stepping through the code, adding //FIXME where i think something should be done [19:23:42] i'll submit my patch with the fixmes (plus some comment and style fixes) as a follow-up [19:23:57] if i don't find anything major, we can merge the big patch plus mine, then work on the fixmes [19:24:18] DanielK_WMDE: ok [19:24:29] SMalyshev: the current code accepts it, and they are not different. but that doesn't mean we don't need a test case that makes sure it stays that way. [19:25:03] the unit tests should not use knowledge about the current implementation [19:26:17] DanielK_WMDE: ok, I can add some simple test but it'd be mostly copying existing tests... [19:26:41] with the data driven appraoch, it should be simpel enough [19:26:52] doesn't need to be much, i'd just like to have that case covered somehow [19:27:31] oh, and Property objects don't have sitelinks. if the current code assumes they always exist, it will fail hard. [19:27:56] ( $entity instanceof Item) <--- good. But needs a test case :) [19:48:36] aude: with your proxy hack, how do you tell git review to fetch things from localhost? [19:48:53] i mean, you can just change the remote, but that would mess thigns up badly, no? [19:49:15] DanielK_WMDE: yeah the code checks if it's an item when it uses item's methods. But tests are fine, just add a ticket for it and I'll do that [19:50:16] I;m not sure why it checks EntityDocument though since Entity implements EntityDocument... but maybe just in case it changes [19:50:42] (03PS1) 10Daniel Kinzler: Fixmes and cleanup for the rdf-export patch. [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/195642 [19:51:24] SMalyshev: EntityDocument is a new, narrower interface. It should be used in type hints instead of Entity. [19:52:03] DanielK_WMDE: but that interface has almost nothing useful for us [19:52:12] DanielK_WMDE: it just has id and type [19:52:30] SMalyshev: that's exactly the point: you cannot make any assumptions about entities. next to nothign is guaranteed to be there. [19:52:52] If you want to access statements, check if the entity implements StatementProvider. etc [19:52:55] that's the idea [19:52:59] [13WikibaseDataModel] 15JeroenDeDauw comment on pull request #403 144cc366a: Heh - I totally forgot we had this class 02http://git.io/pEDJ [19:53:26] [13WikibaseDataModel] 15JeroenDeDauw pushed 1 new commit to 06master: 02http://git.io/pED0 [19:53:26] 13WikibaseDataModel/06master 149be0193 15Jeroen De Dauw: Merge pull request #403 from wmde/dataProviderDocs... [19:53:27] DanielK_WMDE: didn't need to tell git review anything [19:53:41] [13WikibaseDataModel] 15JeroenDeDauw 04deleted 06dataProviderDocs at 144cc366a: 02http://git.io/pEDw [19:53:53] the port is forwarded and ssh is configured to use the port to proxy [19:54:00] for gerrit specifically [19:55:37] aude: but gerrit connexts to gerrit.wikimedia.org directly... how would it know to connect to localhost instead? [19:56:05] ssh config [19:56:05] DanielK_WMDE: well, so how I know if it has statements or not? [19:56:24] SMalyshev: instanceof StatementListProvider [19:56:35] i can comment that out in the config and then it no longer proxies [19:58:51] aude: oh, i get it. git is using ssh to connect, of course. [19:58:57] yes [19:59:05] wouldn't work for https remote [19:59:06] silyl me >_< [19:59:16] yea, that's why i was confused [19:59:17] although that can be configured also, i'm sure [19:59:20] :) [19:59:50] DanielK_WMDE: ah, ok, we can do that [20:00:25] (03CR) 10Smalyshev: Fixmes and cleanup for the rdf-export patch. (034 comments) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/195642 (owner: 10Daniel Kinzler) [20:01:01] DanielK_WMDE: I've added some comments to your patch, mostly narrative [20:01:45] DanielK_WMDE: no objections [20:02:00] (03CR) 10Smalyshev: [C: 031] Fixmes and cleanup for the rdf-export patch. [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/195642 (owner: 10Daniel Kinzler) [20:02:23] (03CR) 10Daniel Kinzler: "Reviewed RdfBuilder.php. No obvious blockers. Together with I737de39eed89ee3, test coverage seems adequate." [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/191393 (owner: 10Smalyshev) [20:02:59] DanielK_WMDE: but as soon as we're ready to move forward with the main one, this one can be merged too [20:03:25] DanielK_WMDE: and making tickets to fix the issues would be great too, otherwise i'll forget [20:05:48] SMalyshev: yes. i'll work on it tomorrow. lots of meetings though. i'm confident that we can get it in this week though. [20:06:11] [13WikibaseDataModel] 15JeroenDeDauw pushed 1 new commit to 06master: 02http://git.io/pE7z [20:06:11] 13WikibaseDataModel/06master 149c861bc 15Jeroen De Dauw: Merge pull request #402 from wmde/idParserTests... [20:06:25] DanielK_WMDE: ok, thanks. [20:11:58] DanielK_WMDE: Hmm, arbitrary access seems to have been pushed back a bit. Still the Commons discussion we had in SF or more? [20:13:22] multichill: no exactly that. needs a schema change, see https://phabricator.wikimedia.org/T92288 [20:14:03] joy [20:14:21] multichill: also, https://phabricator.wikimedia.org/T90755 and https://phabricator.wikimedia.org/T90755 are still open [20:14:34] didn't get around to them, busy week [20:21:36] [13Wikiba.se] 15JeroenDeDauw pushed 1 new commit to 06master: 02http://git.io/pEhy [20:21:36] 13Wikiba.se/06master 145828a84 15jeroendedauw: Add ProjetPP/WikibaseEntityStore [20:26:55] SMalyshev: doen for today, see you tomorrow! [20:27:11] DanielK_WMDE: see you, good night [20:27:54] (03CR) 10Daniel Kinzler: Fixmes and cleanup for the rdf-export patch. (032 comments) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/195642 (owner: 10Daniel Kinzler) [20:28:15] (03PS28) 10Legoktm: Implement RDF export for items and RDF dumps (Turtle only for now) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/191393 (https://phabricator.wikimedia.org/T53795) (owner: 10Smalyshev) [20:28:20] (03PS2) 10Daniel Kinzler: Fixmes and cleanup for the rdf-export patch. [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/195642 [20:28:58] multichill: Gezien? https://www.wikidata.org/wiki/Q14042250 [20:29:02] ehm [20:29:07] https://www.wikidata.org/wiki/Property:P1749 * [20:30:09] ah, no [20:30:29] DanielK_WMDE: Sorry, bit distracted. Just booked my flight to the Lyon hackathon :-) [20:30:53] multichill: Als het goed is was huskyr al aan het kijken hoe we die het makkelijkste kunnen vullen. [20:31:22] Ze staan wel in de infobox op nlwiki, maar die links hebben een verouderde indeling maar redirecten wel juist door. [20:33:07] ah, nice [20:34:15] Maar de parameter van de infobox draagt helaas de naam "website", dus er staan volledige URL's en soms zelfs een andere site, hoewel het label van de URL altijd gelijk is (geeft altijd Parlement.com aan) [20:37:37] De boel is iig nu nog een zooitje, daarom wil ik de data straks ook uit Wikidata fetchen. Hopelijk kan ik de hardnekkige Nederlanders (en Belgen) overhalen met de betere foutcontrole (opmaak, duplicaten) [21:24:53] [13Wikiba.se] 15JeroenDeDauw pushed 1 new commit to 06master: 02http://git.io/puud [21:24:53] 13Wikiba.se/06master 144214422 15Jeroen De Dauw: Add missing dot