[04:49:25] how cool, Germany is superprotected :) [06:47:33] Uh? Is that related to the SAL entry? [06:55:27] Nemo_bis: yes [06:55:42] https://lists.wikimedia.org/pipermail/wikidata-tech/2014-October/000640.html [06:59:01] https://www.wikidata.org/wiki/Wikidata:UI_redesign_input#Updates [07:00:03] legoktm: thanks; meh [08:26:25] (03CR) 10Adrian Lang: "Badges work for me on master. This patch should not change behavior, though." (031 comment) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/164058 (owner: 10Adrian Lang) [08:35:47] (03CR) 10Aude: [C: 032] Fix / use dieException in SetClaim [extensions/Wikibase] (wmf/1.25wmf1) - 10https://gerrit.wikimedia.org/r/164749 (owner: 10Aude) [08:38:20] (03PS1) 10Aude: Fix reference snaks handling in ClaimHtmlGenerator [extensions/Wikibase] (wmf/1.25wmf1) - 10https://gerrit.wikimedia.org/r/164910 (https://bugzilla.wikimedia.org/71479) [08:38:27] (03PS1) 10Aude: Fix editing of properties, once again [extensions/Wikibase] (wmf/1.25wmf1) - 10https://gerrit.wikimedia.org/r/164911 [08:38:36] (03PS1) 10Aude: First initialize sitelinklistview, then attach events [extensions/Wikibase] (wmf/1.25wmf1) - 10https://gerrit.wikimedia.org/r/164912 [08:38:42] (03PS1) 10Aude: Pass string to wfTemplate in FingerprintView instead of Message [extensions/Wikibase] (wmf/1.25wmf1) - 10https://gerrit.wikimedia.org/r/164913 (https://bugzilla.wikimedia.org/71479) [08:55:18] (03Merged) 10jenkins-bot: Fix / use dieException in SetClaim [extensions/Wikibase] (wmf/1.25wmf1) - 10https://gerrit.wikimedia.org/r/164749 (owner: 10Aude) [09:06:26] (03CR) 10Aude: [C: 032] Fix reference snaks handling in ClaimHtmlGenerator [extensions/Wikibase] (wmf/1.25wmf1) - 10https://gerrit.wikimedia.org/r/164910 (https://bugzilla.wikimedia.org/71479) (owner: 10Aude) [09:06:37] (03CR) 10Aude: [C: 032] Fix editing of properties, once again [extensions/Wikibase] (wmf/1.25wmf1) - 10https://gerrit.wikimedia.org/r/164911 (owner: 10Aude) [09:06:49] (03CR) 10Aude: [C: 032] First initialize sitelinklistview, then attach events [extensions/Wikibase] (wmf/1.25wmf1) - 10https://gerrit.wikimedia.org/r/164912 (owner: 10Aude) [09:06:56] (03CR) 10Aude: [C: 032] Pass string to wfTemplate in FingerprintView instead of Message [extensions/Wikibase] (wmf/1.25wmf1) - 10https://gerrit.wikimedia.org/r/164913 (https://bugzilla.wikimedia.org/71479) (owner: 10Aude) [09:23:05] (03Merged) 10jenkins-bot: Fix reference snaks handling in ClaimHtmlGenerator [extensions/Wikibase] (wmf/1.25wmf1) - 10https://gerrit.wikimedia.org/r/164910 (https://bugzilla.wikimedia.org/71479) (owner: 10Aude) [09:23:24] (03Merged) 10jenkins-bot: Fix editing of properties, once again [extensions/Wikibase] (wmf/1.25wmf1) - 10https://gerrit.wikimedia.org/r/164911 (owner: 10Aude) [09:25:50] (03Merged) 10jenkins-bot: First initialize sitelinklistview, then attach events [extensions/Wikibase] (wmf/1.25wmf1) - 10https://gerrit.wikimedia.org/r/164912 (owner: 10Aude) [09:25:57] (03Merged) 10jenkins-bot: Pass string to wfTemplate in FingerprintView instead of Message [extensions/Wikibase] (wmf/1.25wmf1) - 10https://gerrit.wikimedia.org/r/164913 (https://bugzilla.wikimedia.org/71479) (owner: 10Aude) [09:38:39] hi [09:38:42] hum, used the merge tool, now the obsolte item doesn't have any content left, but doesn't seem to have a "delete me" template or something attached to it, is there a crawler finding these for deletion? [09:39:03] or should I do anything else with it? [09:39:28] empty item being https://www.wikidata.org/wiki/Q15962069 [09:40:36] argh, found it in the Merge how to :S nvm me [09:49:37] Adrian_WMDE: did you have a look at https://github.com/wmde/WikidataBrowserTests/pull/11 yet? [09:49:46] anything that should be changed? [09:52:25] [13WikidataBrowserTests] 15tobijat comment on pull request #11 14bbe23d7: The entity-id and the site is tight together, the ID does not make sense on another site. So I see no way around having the whole url hardcoded.. 02http://git.io/8V3Cgw [09:55:32] (03PS1) 10Aude: New deployment build, partial fix for bug 71479 and editing properties [extensions/Wikidata] (wmf/1.25wmf1) - 10https://gerrit.wikimedia.org/r/164925 [09:58:34] (03PS1) 10Aude: New Wikidata Build - 06/10/2014 [extensions/Wikidata] - 10https://gerrit.wikimedia.org/r/164926 [10:10:11] (03CR) 10WikidataJenkins: "Build Successful" [extensions/Wikidata] (wmf/1.25wmf1) - 10https://gerrit.wikimedia.org/r/164925 (owner: 10Aude) [10:12:16] [13WikidataBrowserTests] 15adrianlang comment on pull request #11 14bbe23d7: But the label is available and should be unique, right? 02http://git.io/K8sFiw [10:20:38] (03PS18) 10Lucie Kaffee: Update ApiModules to conform to api conventions [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/162859 (https://bugzilla.wikimedia.org/69739) [10:21:15] (03CR) 10WikidataJenkins: [V: 032] "Build Successful" [extensions/Wikidata] - 10https://gerrit.wikimedia.org/r/164926 (owner: 10Aude) [10:35:17] (03PS1) 10Adrian Lang: Fix module name [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/164934 [10:36:02] is the "in other languages" section intentionally much larger than it used to be? I thought people wanted to make the page more compact, not less [10:36:44] ^ Thiemo_WMDE: Could you merge that? Quite embarrassing [10:42:15] (03CR) 10Siebrand: [C: 031] Make labelview match the appropriate language direction [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/164743 (owner: 10Ebrahim) [11:10:12] (03CR) 10Thiemo Mättig (WMDE): [C: 032] Fix module name [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/164934 (owner: 10Adrian Lang) [11:12:46] (03CR) 10Thiemo Mättig (WMDE): [C: 032] Update ApiModules to conform to api conventions [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/162859 (https://bugzilla.wikimedia.org/69739) (owner: 10Lucie Kaffee) [11:20:35] (03Merged) 10jenkins-bot: Fix module name [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/164934 (owner: 10Adrian Lang) [11:25:59] (03CR) 10Aude: [C: 032] New Wikidata Build - 06/10/2014 [extensions/Wikidata] - 10https://gerrit.wikimedia.org/r/164926 (owner: 10Aude) [11:26:08] (03CR) 10Aude: [C: 032] New deployment build, partial fix for bug 71479 and editing properties [extensions/Wikidata] (wmf/1.25wmf1) - 10https://gerrit.wikimedia.org/r/164925 (owner: 10Aude) [11:31:01] (03PS19) 10Thiemo Mättig (WMDE): Update ApiModules to conform to API conventions [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/162859 (https://bugzilla.wikimedia.org/69739) (owner: 10Lucie Kaffee) [11:32:27] (03CR) 10Thiemo Mättig (WMDE): [C: 032] Update ApiModules to conform to API conventions [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/162859 (https://bugzilla.wikimedia.org/69739) (owner: 10Lucie Kaffee) [11:34:54] (03Merged) 10jenkins-bot: New Wikidata Build - 06/10/2014 [extensions/Wikidata] - 10https://gerrit.wikimedia.org/r/164926 (owner: 10Aude) [11:36:54] (03PS18) 10Thiemo Mättig (WMDE): Add ApiConventionsTest [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/151659 (https://bugzilla.wikimedia.org/67407) (owner: 10Lucie Kaffee) [11:39:19] [13WikidataBrowserTests] 15tobijat comment on pull request #11 14bbe23d7: Only label+description is unique. There might be several Items with a label Italy. Also this is language specific. 02http://git.io/8W9CEw [11:42:19] [13WikidataBrowserTests] 15JeroenDeDauw comment on pull request #11 14bbe23d7: Agree with Tobi; you could have a WIKIDATA_BETA_URL constant or something, but it'd always need to be specific to the site. (Unless of course the test goes create something new, avoiding assumptions on what is already there, though I think that this is not the intention of this test.) 02http://git.io/FpQFpA [11:43:45] (03Merged) 10jenkins-bot: New deployment build, partial fix for bug 71479 and editing properties [extensions/Wikidata] (wmf/1.25wmf1) - 10https://gerrit.wikimedia.org/r/164925 (owner: 10Aude) [11:44:31] (03CR) 10Thiemo Mättig (WMDE): [C: 031] "Almost there. :-)" (034 comments) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/151659 (https://bugzilla.wikimedia.org/67407) (owner: 10Lucie Kaffee) [11:45:44] (03Merged) 10jenkins-bot: Update ApiModules to conform to API conventions [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/162859 (https://bugzilla.wikimedia.org/69739) (owner: 10Lucie Kaffee) [11:46:54] [13WikidataBrowserTests] 15tobijat pushed 1 new commit to 06item_smoke: 02http://git.io/ZgwR8g [11:46:54] 13WikidataBrowserTests/06item_smoke 1499f1ca3 15Tobias Gritschacher: Moved tags [11:56:03] (03PS19) 10Lucie Kaffee: Add ApiConventionsTest [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/151659 (https://bugzilla.wikimedia.org/67407) [12:05:31] Wow, we managed to compress that last dump a lot better than the previous ones! http://dumps.wikimedia.org/other/wikidata/ [12:08:55] [13WikidataBrowserTests] 15tobijat pushed 1 new commit to 06item_smoke: 02http://git.io/qU5SMQ [12:08:55] 13WikidataBrowserTests/06item_smoke 14ebd664c 15Tobias Gritschacher: Make URL to smoke item configurable [12:09:36] aude: do you know what the size of an uncompressed recent json dump is? [12:09:44] estimate [12:10:09] no [12:10:32] dump look suspiciously broken [12:10:34] Adrian_WMDE: ok, changed the url to be configurable now [12:10:49] after the item_smoke thing is merged, I'll chenge the pproperty one [12:11:01] meh. typo-day [12:17:14] (03CR) 10Adrian Lang: "SqlUsageTracker is quite big. Just from the structure I inferred the following possible split-up:" (032 comments) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/161477 (https://bugzilla.wikimedia.org/64586) (owner: 10Daniel Kinzler) [12:19:58] (03CR) 10Adrian Lang: [C: 04-1] "This is not based on top of the current Introduce usage tracking service change." [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/161487 (https://bugzilla.wikimedia.org/64588) (owner: 10Daniel Kinzler) [12:22:04] [13WikidataBrowserTests] 15adrianlang comment on pull request #11 14ebd664c: Don't you have to include URL? 02http://git.io/eYBc-A [12:26:48] (03CR) 10Adrian Lang: Record usage info from ParserOutput to the database. (035 comments) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/162650 (https://bugzilla.wikimedia.org/64589) (owner: 10Daniel Kinzler) [12:28:42] [13WikidataBrowserTests] 15tobijat comment on pull request #11 14ebd664c: No. That happens in env.rb and then strangely in utils.rb... need to clean up that stuff (though not in this commit). 02http://git.io/JSI7og [12:36:12] (03CR) 10Thiemo Mättig (WMDE): [C: 032] Use $.when for promise handling [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/164058 (owner: 10Adrian Lang) [12:36:18] (03PS10) 10Bene: [WIP] Finish refactoring of EntityView [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/160305 [12:36:34] (03CR) 10jenkins-bot: [V: 04-1] [WIP] Finish refactoring of EntityView [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/160305 (owner: 10Bene) [12:39:15] (03CR) 10Thiemo Mättig (WMDE): [C: 032] Replace widgetBaseClass with widgetFullName [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/164064 (owner: 10Adrian Lang) [12:39:44] (03CR) 10Jeroen De Dauw: Introduce usage tracking service. (032 comments) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/161477 (https://bugzilla.wikimedia.org/64586) (owner: 10Daniel Kinzler) [12:41:36] Adrian_WMDE: thanks for looking into more detail at SqlUsageTracker - that's exactly what I suspected when I has a pass at it [12:42:33] (03CR) 10Jeroen De Dauw: [C: 04-1] "Only remaining point from my side are the cohesion and coupling issues in SqlUsageTracker" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/161477 (https://bugzilla.wikimedia.org/64586) (owner: 10Daniel Kinzler) [12:47:24] (03Merged) 10jenkins-bot: Use $.when for promise handling [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/164058 (owner: 10Adrian Lang) [12:48:47] (03Merged) 10jenkins-bot: Replace widgetBaseClass with widgetFullName [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/164064 (owner: 10Adrian Lang) [12:52:08] [13WikidataBrowserTests] 15adrianlang pushed 1 new commit to 06master: 02http://git.io/XTRuEA [12:52:08] 13WikidataBrowserTests/06master 143ec8484 15Adrian Lang: Merge pull request #11 from wmde/item_smoke... [12:53:34] \o/ [12:57:40] (03PS1) 10Adrian Lang: Change badgeselector test to use the new EntityStore [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/164954 [13:01:12] (03PS1) 10Thiemo Mättig (WMDE): Remove unused code from snakview [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/164955 [13:01:38] (03CR) 10Thiemo Mättig (WMDE): [C: 032] Remove unnecessary calls to variation.draw() [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/164307 (owner: 10Adrian Lang) [13:08:45] [13WikidataBrowserTests] 15JeroenDeDauw 04deleted 06item_smoke at 14ebd664c: 02http://git.io/QSpDIg [13:13:58] [13WikidataBrowserTests] 15tobijat pushed 1 new commit to 06property_smoke: 02http://git.io/Jv171Q [13:13:58] 13WikidataBrowserTests/06property_smoke 1424919fe 15Tobias Gritschacher: Merge branch 'master' into property_smoke... [13:14:11] (03PS1) 10Adrian Lang: Fix sitelinklistview and sitelinkgroupview tests [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/164956 [13:14:57] (03Merged) 10jenkins-bot: Remove unnecessary calls to variation.draw() [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/164307 (owner: 10Adrian Lang) [13:16:27] (03CR) 10Thiemo Mättig (WMDE): [C: 032] Change badgeselector test to use the new EntityStore [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/164954 (owner: 10Adrian Lang) [13:16:36] (03CR) 10Adrian Lang: [C: 032] Remove unused code from snakview [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/164955 (owner: 10Thiemo Mättig (WMDE)) [13:21:40] (03CR) 10Thiemo Mättig (WMDE): [C: 032] Fix sitelinklistview and sitelinkgroupview tests [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/164956 (owner: 10Adrian Lang) [13:27:21] Whats happening? Wikidata - The Game is D.E.A.D.! [13:28:15] [13WikidataBrowserTests] 15adrianlang 04deleted 06property_smoke at 1424919fe: 02http://git.io/VRF_7g [13:30:21] (03Merged) 10jenkins-bot: Change badgeselector test to use the new EntityStore [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/164954 (owner: 10Adrian Lang) [13:31:07] [13Wikidata.org] 15Benestar 04deleted 06important at 14b4aba14: 02http://git.io/lT_btw [13:31:17] (03Merged) 10jenkins-bot: Remove unused code from snakview [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/164955 (owner: 10Thiemo Mättig (WMDE)) [13:31:37] [13Wikidata.org] 15Benestar created 06fixselectors from 06master (+0 new commits): 02http://git.io/jNNCSw [13:32:09] (03CR) 10Thiemo Mättig (WMDE): [C: 032] Parse EntityIds via API [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/164053 (owner: 10Adrian Lang) [13:32:14] (03Merged) 10jenkins-bot: Fix sitelinklistview and sitelinkgroupview tests [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/164956 (owner: 10Adrian Lang) [13:33:34] [13Wikidata.org] 15Benestar pushed 1 new commit to 06fixselectors: 02http://git.io/tuDj7Q [13:33:34] 13Wikidata.org/06fixselectors 14aeb16e8 15Bene*: Fix selectors for badge icons... [13:33:50] [13Wikidata.org] 15Benestar opened pull request #14: Fix selectors for badge icons (06master...06fixselectors) 02http://git.io/oAq93A [13:38:49] (03CR) 10Thiemo Mättig (WMDE): [C: 032] Add ApiConventionsTest [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/151659 (https://bugzilla.wikimedia.org/67407) (owner: 10Lucie Kaffee) [13:40:10] (03PS9) 10Bene: Refactor EntityView [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/158877 [13:43:11] (03PS11) 10Bene: [WIP] Finish refactoring of EntityView [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/160305 [13:43:18] (03Merged) 10jenkins-bot: Parse EntityIds via API [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/164053 (owner: 10Adrian Lang) [13:47:39] (03CR) 10jenkins-bot: [V: 04-1] [WIP] Finish refactoring of EntityView [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/160305 (owner: 10Bene) [13:49:54] (03PS12) 10Bene: [WIP] Finish refactoring of EntityView [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/160305 [13:51:49] (03CR) 10Daniel Kinzler: Introduce usage tracking service. (031 comment) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/161477 (https://bugzilla.wikimedia.org/64586) (owner: 10Daniel Kinzler) [13:54:00] <___Chess___> How do I create a Wikidata entry? [13:55:32] (03CR) 10jenkins-bot: [V: 04-1] Refactor EntityView [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/158877 (owner: 10Bene) [13:55:34] (03Merged) 10jenkins-bot: Add ApiConventionsTest [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/151659 (https://bugzilla.wikimedia.org/67407) (owner: 10Lucie Kaffee) [13:58:34] <___Chess___> nevermind, i figured it out [13:58:35] (03CR) 10jenkins-bot: [V: 04-1] [WIP] Finish refactoring of EntityView [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/160305 (owner: 10Bene) [14:04:06] (03PS13) 10Bene: [WIP] Finish refactoring of EntityView [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/160305 [14:12:42] (03PS1) 10Adrian Lang: Revert "Replace widgetBaseClass with widgetFullName" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/164977 [14:13:03] (03CR) 10jenkins-bot: [V: 04-1] [WIP] Finish refactoring of EntityView [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/160305 (owner: 10Bene) [14:13:49] (03PS1) 10Thiemo Mättig (WMDE): Cleanup some EntityStore JavaScript code [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/164978 [14:16:34] (03CR) 10Thiemo Mättig (WMDE): [C: 032] Revert "Replace widgetBaseClass with widgetFullName" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/164977 (owner: 10Adrian Lang) [14:27:31] (03Merged) 10jenkins-bot: Revert "Replace widgetBaseClass with widgetFullName" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/164977 (owner: 10Adrian Lang) [14:34:28] [13Wikidata.org] 15filbertkm 04deleted 06fixselectors at 14aeb16e8: 02http://git.io/x3iYjg [15:05:18] (03CR) 10Thiemo Mättig (WMDE): [C: 032] "Not 100% happy, but good enough. Does this have a bug number?" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/164725 (owner: 10Hoo man) [15:21:30] (03PS1) 10Thiemo Mättig (WMDE): Cleanup UserLanguageLookupTest [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/164989 [15:28:40] (03CR) 10Thiemo Mättig (WMDE): [C: 032] "Manual test succeeded. :-)" (031 comment) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/164059 (owner: 10Adrian Lang) [15:33:09] (03CR) 10Hoo man: [C: 032] Cleanup UserLanguageLookupTest [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/164989 (owner: 10Thiemo Mättig (WMDE)) [15:38:25] hi, is there some easy way to search for 10 entities that have a german label but no english? [15:40:00] Hm, good question. [15:40:54] This is somewhat like that: No English label, but a German sitelink. http://tools.wmflabs.org/wikidata-terminator/?list&lang=en&mode=t1000&q=link[dewiki] [15:41:49] sjoerddebruin: thanks [15:42:04] Also sorted on the number of incoming links. :) [15:43:01] yupp, quite handy ;) [15:43:08] (03Merged) 10jenkins-bot: Use single API request for fetching multiple entities [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/164059 (owner: 10Adrian Lang) [15:43:37] [13WikidataBrowserTests] 15tobijat pushed 1 new commit to 06master: 02http://git.io/A9MAsg [15:43:37] 13WikidataBrowserTests/06master 14a2c2b37 15Tobias Gritschacher: Load config file only if file exists [15:44:04] (i got a report that my altLabels script ( https://www.wikidata.org/wiki/User:Joern/altLabels.js ) doesn't work anymore and wasn't able to easily find entities for debugging ;) ) [15:44:45] (03Merged) 10jenkins-bot: UserLanguageLookup: Only return languages we support [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/164725 (owner: 10Hoo man) [15:44:46] Lydia_WMDE: Got my e-mail? :) [15:45:21] (03Merged) 10jenkins-bot: Cleanup UserLanguageLookupTest [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/164989 (owner: 10Thiemo Mättig (WMDE)) [15:45:47] sjoerddebruin: be specific :p Titles etc might help her rather than names :) [15:46:08] She always forgets about my emails :( [15:46:44] :( [15:49:00] sjoerddebruin: hey [15:49:02] yes i saw it [15:49:06] will take care of it :) [15:49:16] busy with the multimedia bootcamp this week [15:49:20] Ah, nice. [16:07:35] Thiemo_WMDE: aude: I should totally go to amici amici on my own and then subtly take a table as far from where the group is sitting as possible :) [16:09:20] JeroenDeDauw: sit in the table next to the group and ignore them completely. ;) [16:09:33] most evil thing possible. [16:12:30] * ori wishes he was in berlin [16:27:05] (03PS14) 10Daniel Kinzler: Collect usage tracking info in ParserOutput. [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/161487 (https://bugzilla.wikimedia.org/64588) [16:27:25] (03CR) 10Daniel Kinzler: "@adrian clicking the "rebase" button fixed that." [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/161487 (https://bugzilla.wikimedia.org/64588) (owner: 10Daniel Kinzler) [16:27:55] (03CR) 10Daniel Kinzler: Record usage info from ParserOutput to the database. (035 comments) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/162650 (https://bugzilla.wikimedia.org/64589) (owner: 10Daniel Kinzler) [16:33:15] [13Geo] 15brightbyte pushed 1 new commit to 06FixPrecisionInFormatter: 02http://git.io/3bW-NQ [16:33:15] 13Geo/06FixPrecisionInFormatter 14548398a 15daniel: Pass precision from Globe*Formatter to Geo*Formatter... [16:37:58] [13Geo] 15brightbyte opened pull request #19: Fix precision handling in GlobeCoordinateFormatter and GeoCoordinateFormatter (06master...06FixPrecisionInFormatter) 02http://git.io/lIz3eQ [16:39:52] DanielK_WMDE__: Just because I stumbled upon this again... https://github.com/DataValues/Iri [16:39:56] why don't we use it [16:40:10] I think there was a discussion, but can't remember it [16:52:21] hoo: because there was no need/reason too, and re-using the code for StringValue was a lot easier [16:52:42] I'd actually be fine with using it if it extended StringValue [16:53:39] Thiemo_WMDE: rate my effort please :D [16:53:58] Lydia_WMDE: see, it was all Thiemo_WMDE his idea, rage to him [17:03:25] [13Geo] 15brightbyte pushed 1 new commit to 06FixPrecisionInFormatter: 02http://git.io/E5Qgsw [17:03:25] 13Geo/06FixPrecisionInFormatter 14961eb0d 15daniel: Add round trip tests for Globe and Geo [17:05:05] JeroenDeDauw: so, is a change that fixed broken behavior in the public api a breaking change or a fixing change?... [17:20:33] (03PS15) 10Daniel Kinzler: Introduce usage tracking service. [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/161477 (https://bugzilla.wikimedia.org/64586) [17:20:35] (03PS15) 10Daniel Kinzler: Collect usage tracking info in ParserOutput. [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/161487 (https://bugzilla.wikimedia.org/64588) [17:20:37] (03PS13) 10Daniel Kinzler: Record usage info from ParserOutput to the database. [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/162650 (https://bugzilla.wikimedia.org/64589) [17:24:06] (03Abandoned) 10Daniel Kinzler: Draft of subscription and tracking interfaces [WIP] [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/158078 (owner: 10Daniel Kinzler) [17:24:14] (03Abandoned) 10Daniel Kinzler: Initial rough spec for the usage tracking system [WIP] [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/157682 (owner: 10Daniel Kinzler) [17:33:36] (03CR) 10Daniel Kinzler: "This looks like a good start, but doesn't go far enough for my taste. I would suggest to" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/158641 (owner: 10Bene) [17:34:07] aude: Around? [17:34:25] hoo: aude is having dinner with the multimedia team [17:34:35] I feared that :S [17:35:43] DanielK_WMDE__: Did you talk about the Q183 issues today? [17:36:00] * hoo came home late today... [17:36:06] no. [17:36:46] hoo: actually - Thiemo was doing some benchmarking earlier, together with aude. Might have been that [17:37:01] I see... but the issue is (sadly) complexer [17:37:17] regular work is pretty much zero right now though. we have been in sessions with the multimedia folks all day [17:37:18] some revisions of that item segfault, some fatal and some throw an exception [17:37:26] bah [17:37:46] this happens since the switch to DataModel 1.0, right? [17:37:56] That's a good question, actually [17:38:07] sigh [17:38:08] not sure why shit hit the fan exactly now and not earlier [17:38:14] JeroenDeDauw: care to look into that? [17:38:41] hoo: can you collect your findings somewhere, and mail a link to wikidata-tech? [17:38:56] DanielK_WMDE__: Well, everything is bugzillad [17:39:00] just search for Q183 [17:39:35] might be that it's spread across various products, though (Wikimedia and Wikidata repo) [17:41:34] This problem has so many parts :S [17:41:59] But the one that old revisions sometimes can't be viewed/ unserialized(?) is the most minor one IMO [17:42:06] but it seems like all of them are related to the change in the DataModel and/or the serialization format [17:43:16] hoo: my feelign is that we should (partially) go back to deferred unstubbing: have stub implementations of StatementList, etc, that would only instantiate the full structure when needed. [17:43:22] JeroenDeDauw: what do you think? [17:43:45] Probably, yes [17:44:20] even with hhvm we will reach the oom/ time out state again at some point if we keep creating objects we don't use [17:44:22] hoo: would you investigate that option? [17:44:41] StatementList and SitelInks are probably the main condidates. for this. [17:45:45] DanielK_WMDE__: https://www.wikidata.org/w/index.php?title=Q183&oldid=131926878 [17:45:53] that's the bug about viewing old revisions [17:45:56] but I consider it minor [17:46:05] Not sure that one is filed though [17:46:27] hoo: would be good to track it down at least. can you find this in the logs? [17:46:32] * DanielK_WMDE__ should really learn to dig into the logs [17:46:43] Sure, will do in a moment [17:47:09] I planned to do some stuffs today either on this or on the JS patches in gerrit, but I'm just so tired [17:48:04] then leave it for tomorrow [17:48:11] coding when tired is not good [17:48:15] Yeah [17:48:34] .oO( coding when drunk, however, is fun... ) [20:31:55] I'm not able to load https://www.wikidata.org/wiki/Q17939676 [20:32:28] apper: what's the error you get? [20:32:35] I'm getting either "504 Gateway Time-out" or "Our servers are currently experiencing a technical problem." [20:32:40] ok [20:32:42] the rest of wikidata works [20:32:47] known then and being worked on :) [20:32:52] sorry for the issue [20:33:04] aude: hoo ^ [20:33:06] okay, no problem [20:33:18] bye ;) [20:33:21] cya [20:48:42] oh, I am in here [20:49:04] greg-g: leave, then! :D [20:49:08] (no don't) [20:49:08] :P [20:49:20] [13Geo] 15JeroenDeDauw comment on pull request #19 14961eb0d: Fatal if not instanceof `LatLongValue` 02http://git.io/OfkMtA [20:49:20] But I guess you're physically in the office atm? [20:49:22] hoo: ya'll need anything? [20:49:24] yeah [20:49:35] re the above issue? [20:50:02] above? The Q183 thing? [20:50:22] 16:31 < apper> I'm not able to load https://www.wikidata.org/wiki/Q17939676 [20:50:44] Oh, didn't see that [20:51:06] but I briefly lost connection... my home network acts up recently :S [20:51:42] hoo: it might be another instance of what you dealt with this weekend [20:53:29] greg-g: :( http://cdn03.cdn.socialitelife.com/wp-content/uploads/2013/10/11/crying-gifs-10112013-09.gif [20:53:34] That sucks [20:53:45] will have a look of course [20:54:13] moar superprotect! [20:54:18] ;) [20:55:52] hoo: is it time to start thinking about reverting to the old serialization format? [20:56:04] i realize that it's a bit of a drastic move, but things look a bit crappy right now [20:57:23] ori: Not with heavy prior testing [20:57:30] * without [20:57:52] tbh no [20:57:55] And that would be extremely painful [20:57:57] we can not go back to the old format [20:58:02] we need to fix the new one [20:58:07] also there are issue in DataModel probbly that are unrelated [20:58:10] but add up to the trouble [20:58:12] what can we do to help? [20:58:53] it's a site issue; the mediawiki core team's weekly planning meeting is starting now [20:59:05] one of the things we do is talk about things that we didn't plan for but that deserve some focussed attention [20:59:23] so if you think having some of our time to work on this would help it get resolved, say something :) [20:59:30] I'm at hand to help from the ops side in any way we can. [21:00:18] I'm not sure what to do, honestly [21:00:26] these problems are very complex [21:00:42] I can of course implement more Q183 like workarounds, but that's not nice [21:00:56] and downgrading DataModel will heavily break stuff [21:01:10] downgrading the serialization format is not really an option also [21:01:18] so I'm out of things we can do nowish [21:01:29] supporting aude with debugging might be good [21:01:33] nod [21:01:36] but she needs to say more about that [21:01:40] meeting starting now, will ping you after if you're still around [21:01:56] ori: May or may not be... let's see [21:39:09] any devs online? [21:39:22] just wondering if anyone can explain wtf is this: https://bugzilla.wikimedia.org/show_bug.cgi?id=71519 [21:40:36] TimStarling: https://bugzilla.wikimedia.org/show_bug.cgi?id=71542 [21:41:04] is that a duplicate? [21:41:21] I just marked it as such, yeah. [21:42:40] this bad [21:42:51] I'll have a look in a second [21:43:53] Ok, I'm having a look now [21:45:03] TimStarling: Do you know which urls are causing this? [21:45:19] the current spike [21:45:26] nothing obvious in the logs now [21:45:52] what spike? [21:46:34] I thought there was a new fatal spike [21:46:43] ok, there is [21:49:37] "value":"\u0b9a\u0bc6\u0bb0\u0bc1\u0bae\u0ba9\u0bbf" [21:49:43] very efficient encoding there [21:51:46] mh... that's what? utf-16? [21:55:36] well, using UTF-8 only reduces it from 736163 to 721591 [21:55:46] it's JSON, though the JSON specification doesn't require that you encode code points outside the ASCII range; it simply allows it [21:56:43] http://php.net/manual/en/json.constants.php#constant.json-unescaped-unicode [21:57:01] I think FormatJson::encode() supports a similar option, and it's compatible with PHP 5.3 [21:57:35] yes, that's what I just tried, reduces oldid 158433886 from 736163 to 721591 bytes [21:59:00] I haven't reproduced that crash in eval.php yet [21:59:44] hoo: FormatJson::encode( $value, /* $pretty = */ false, FormatJson::UTF8_OK ); [22:00:08] the important thing is to fix the crash [22:00:14] ori: Can you open a bug [22:00:22] There's much more to this issue [22:00:25] sure, I'm trying to repro the crash, too [22:00:28] like DataModel is also inefficent [22:01:08] https://www.wikidata.org/wiki/Q17939676 doesn't segfault but it does hit the memory limit [22:01:29] ah, you just reverted it [22:01:33] ori: Yeah [22:11:14] <_joe_> what is the memory limit exactly? [22:11:55] 'wmgMemoryLimit' => array( [22:11:55] 'default' => 300 * 1024 * 1024, // 300MB [22:11:55] ), [22:12:15] but we also see exceptions, timeouts (also php fatal) and even segfaults [22:12:56] <_joe_> yeah sorry, I meant which is the data size that makes zend segfault [22:13:01] <_joe_> and why [22:13:05] <_joe_> do we know that? [22:14:38] we don't know anything [22:15:10] is there a server with core dumps enabled? [22:16:14] Don't think so [22:16:17] but one would be nice [22:16:33] already knowing what called debug_backtrace would help solving that issue [22:16:41] thus the segfault would go away in prod [22:17:10] <_joe_> TimStarling: no [22:17:19] <_joe_> but it can be done pretty easily [22:17:36] easily as in you can do it more quickly than me? [22:17:40] <_joe_> I think you just need the dbg packages for php5 [22:17:48] I'll have it done in 5 minutes I would say [22:17:52] <_joe_> TimStarling: ok, no :) [22:18:31] <_joe_> TimStarling: I was just offering to help if you didn't have time [22:30:20] TimStarling: _joe_: https://lists.wikimedia.org/pipermail/wikidata-tech/2014-October/000641.html [22:30:26] This should document the current state [22:37:27] well, I have a backtrace [22:37:32] maybe it was slightly more than 5 minutes [22:38:05] I know someone (Roan?) removed my core dump support from the config but I wasn't sure what it was replaced with [22:39:59] TimStarling: Ok... does it show what actually caused this problem? [22:40:06] it is a crash in ByPropertyListSerializer::getSerialized() [22:40:09] Oh [22:40:52] throw new InvalidArgumentException( 'ByPropertyListSerializer can only serialize Traversable objects' ); [22:40:55] zend_hash_num_elements(ht=0xb6b84cc230fcbfff) i.e. ht is garbage [22:40:56] I guess? [22:41:36] Ah yes, I looked into that myself [22:41:46] with Ori's bt [22:42:17] yes [22:42:21] it is throwing that exception [22:43:23] Ok... but I couldn't even figure how it went from debug_backtrace into zend_hash_num_elements [22:43:34] (at least while looking at PHP's source briefly) [22:43:45] http://paste.tstarling.com/p/JpjgYD.html [22:43:59] #2 0x00007fea99b674b0 in debug_backtrace_get_args (curpos=0x7fea9f7f44e0) [22:44:04] ah, that's the missing step [22:44:08] the previous backtrace was missing the static functions [22:44:14] because it had no debug symbols [22:44:15] ori didn't have symbols AFAIR [22:44:23] yep [22:44:29] i have them now, but you're faster >.< [22:45:34] TimStarling: Worth filing that upstream (what's the upstream now? Debian maintains it?)? [22:45:54] PHP 5.3 is so outdated [22:46:18] 1089 /tmp/buildd/php5-5.3.10/Zend/zend_API.c: No such file or directory. [22:46:24] '/tmp/buildd'? [22:46:33] I don't really bother filing bugs with PHP unless I have it to the point of "change this line of code to this and then it won't crash or run arbitrary code" [22:46:46] and even then I only get about 50% of them fixed :) [22:47:10] version is usually irrelevant [22:47:12] wow... that's awful [22:47:30] the hhvm team is much better at that [22:47:33] I just isolate the bug, check if it still exists in git master, then file it as a git master bug [22:47:44] TimStarling: This one wont [22:48:01] I think the logic changed much in master [22:48:38] I looked briefly only, but it shouldn't even call zend_hash_num_elements in _zval_copy_ctor_func on mster [22:48:51] doesn't matter [22:48:58] you see it is wrong in frame 5 already [22:51:23] ori: I put the source in ~/src/php5-5.3.10 [22:51:36] on srv1088 [22:51:51] you can install it with apt-get source if you are working somewhere else [22:51:59] use the "directory" command to set the search path [22:54:22] doh, right... obviously it's not supposed to throw an exception :P [22:54:29] * hoo hasn't slept much recently [22:55:27] TimStarling: so for: zend_default_exception_new_ex (class_type=0x7f9b5d64d8c8, skip_top_traces=0) [22:55:41] (gdb) print class_type->filename [22:55:41] $47 = 0xc602a9f56e34a3ef
[22:55:59] is it because that it's trying to unserialize a class the implementation of which is no longer in memory/on disk? [22:56:10] sorry if it's a stupid question, i'm still learning how to use gdb [22:56:33] I am following a suspicion that it is a debug symbol issue [22:56:46] it does this: [22:56:48] Z_OBJVAL_P(arg) = class_type->create_object(class_type TSRMLS_CC); [22:57:04] and supposedly the class_type points to invalid memory [22:57:31] but class_type in frame 6 is correct [22:57:51] how can it get the correct function pointer for create_object but pass the wrong arguments to it? [22:58:04] must be something wrong with the debug symbols or something [22:58:12] gcc is buggy [23:00:54] yeah, so it passes the argument to zend_default_exception_new in rdi [23:01:39] it's apparently not on the stack at all [23:03:41] it's probably the same at frame 0 [23:07:03] no, at frame 0 it is genuinely deferencing a pointer to 0xb6b84cc230fcbfff [23:07:15] (gdb) x/i $rip [23:07:15] => 0x7fea99b63570 : mov 0x8(%rdi),%eax [23:07:25] (gdb) print/x $rdi [23:07:25] $16 = 0xb6b84cc230fcbfff [23:12:11] anyway, if someone can make that exception go away, then the crash will probably go away [23:14:24] TimStarling: Do you think another type of exception would work or shall we just let it go and see what happens (will probably fatal [23:14:25] ) [23:15:00] That could just be a type hint [23:15:02] preferably fix it so that it works [23:15:14] another type of exception probably wouldn't work [23:15:17] TimStarling: Not sure where the invalid data comes from [23:15:25] Ok, will make it a type hint then [23:16:48] TimStarling: Do you see this as urgent? [23:17:01] Eg. shall we fix it *now* tomorrow or with the next regular deploy? [23:17:06] we -> I [23:17:15] yes, I think it's urgent [23:17:31] Ok, I'll prepare the stuff then [23:17:42] will take some time though as master has diverged a bit from what is deployed now [23:18:03] <___Chess___> I got a question. [23:18:10] oh, actually not [23:18:12] hi ___Chess___ [23:18:24] <___Chess___> Alright, in the "in other languages" section. [23:18:37] <___Chess___> How do I get extra languages other than italian, german, and french? [23:19:13] ___Chess___: Put something like {{#babel:de-N|en-3|it-1|fr-1}} onto your user page [23:19:23] just choose whatever languages you want in thee [23:19:24] * there [23:20:13] <___Chess___> So after I put the template, it will show the "in other languages" labels [23:20:33] ___Chess___: Yep, in whateve r languages you specified there [23:21:27] <___Chess___> What does "n" mean? [23:21:34] (03PS1) 10Hoo man: Use typehint instead of manually checking in ByPropertyListSerializer [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/165131 [23:21:36] native [23:21:39] TimStarling: ^ +2 ? [23:21:44] <___Chess___> Alright [23:22:19] Not sure it will make it through the unit tests, though [23:22:21] but we'll see [23:22:49] (03CR) 10Tim Starling: [C: 032] "For https://bugzilla.wikimedia.org/show_bug.cgi?id=71519" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/165131 (owner: 10Hoo man) [23:23:15] (03PS1) 10Hoo man: Use typehint instead of manually checking in ByPropertyListSerializer [extensions/Wikibase] (wmf/1.25wmf1) - 10https://gerrit.wikimedia.org/r/165132 [23:23:27] (03CR) 10Hoo man: [C: 032] "Per master" [extensions/Wikibase] (wmf/1.25wmf1) - 10https://gerrit.wikimedia.org/r/165132 (owner: 10Hoo man) [23:24:51] <___Chess___> If I set my knowledge in a language to zero, will it still show me the corresponding labels in the "in other languages" section? [23:25:01] ___Chess___: Yes it will [23:25:16] <___Chess___> Great! [23:25:23] <___Chess___> Might want to learn another language. [23:26:38] <___Chess___> Thanks for the help, hoo! [23:27:09] You're welcome [23:27:30] <___Chess___> Alright, in the "date of death" property [23:27:38] <___Chess___> A qualifier is "place of death" [23:27:46] <___Chess___> Now, there is also a property named "place of death" [23:28:13] <___Chess___> Do I put both the separate property AND the qualifier? [23:29:06] (03CR) 10jenkins-bot: [V: 04-1] Use typehint instead of manually checking in ByPropertyListSerializer [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/165131 (owner: 10Hoo man) [23:29:44] grrr [23:30:08] ___Chess___: I wouldn't put it as qualifier [23:30:22] (03CR) 10jenkins-bot: [V: 04-1] Use typehint instead of manually checking in ByPropertyListSerializer [extensions/Wikibase] (wmf/1.25wmf1) - 10https://gerrit.wikimedia.org/r/165132 (owner: 10Hoo man) [23:30:34] <___Chess___> Just as a separate property? [23:30:51] Yeah, just add it as a statement, I'd say [23:31:01] TimStarling: 23:29:17 Fatal error: Declaration of Wikibase\Lib\Serializers\ByPropertyListSerializer::getSerialized() must be compatible with that of Wikibase\Lib\Serializers\Serializer::getSerialized() in /mnt/jenkins-workspace/workspace/mwext-Wikibase-client-tests/extensions/Wikibase/lib/includes/serializers/ByPropertyListSerializer.php on line 104 [23:32:28] so you don't know where that data is coming from, right? [23:32:47] and you don't have a PHP backtrace? [23:32:49] TimStarling: From the serialization component [23:32:52] Yep [23:32:57] php backtrace would help here [23:33:09] probably the serialization component [23:34:33] I reloaded the core file because things seemed to be going wrong in gdb [23:34:44] and it has changed, it is a new crash -- that's what you get for having no core pattern I guess [23:34:54] how's it going? [23:35:13] aude: hi :) [23:35:15] I mean, it is not a crash I injected, it is some other crash [23:35:21] sounds suspicious liek the thing i fixed in the weekend [23:35:34] maybe a SnakList being cast to an array? [23:35:49] but i don't know where this is happening exactly now [23:35:55] :S [23:36:15] * aude finds link to my change [23:36:20] ok [23:36:28] Didn't look at stuff much [23:36:34] as I was on that German Wiki conference [23:36:37] if we can find where else this might happen.... [23:37:07] https://gerrit.wikimedia.org/r/#/c/164910/ [23:37:28] qualifiers are also SnakList [23:37:50] Oh I see [23:37:52] * aude thinks these list objects are evil and would liek to move away from them (but not somethign quick) [23:37:53] kind of [23:38:02] yeah [23:38:06] it was a SnakList when passed into the function [23:38:09] then became an array [23:38:21] I will get a PHP backtrace [23:38:23] and then the HashArray thing triggered an exception [23:38:24] and the referenced changed it int the object? [23:38:28] TimStarling: Awesome [23:38:44] ByPropertyIdArray casts it to an array [23:39:05] which by reference seems to also change the $snaks within the $reference itself [23:39:20] but not always, it seems as but only some items encounted this [23:39:45] Q72 was Zurich, so could be that it was made super awesome at the hackathon with fancy references and such [23:39:57] and so it had the problem [23:40:06] aude: Seeing it [23:40:15] the cause [23:40:19] i twasn't the only issue, btw [23:40:51] https://gerrit.wikimedia.org/r/#/c/164913/ was also problematic and helped trigger the exception [23:41:13] i don't quite understand how exactly but fixing that made things work also [23:41:52] backtrace: http://paste.tstarling.com/p/mcPmWY.html [23:42:07] https://bugzilla.wikimedia.org/show_bug.cgi?id=71479 has some stacktraces [23:42:19] https://bugzilla.wikimedia.org/show_bug.cgi?id=71479#c1 [23:42:30] might not be exactly related but maybe [23:42:54] TimStarling: thanks [23:43:47] aude: Will upload fix shortly [23:44:51] $serialization['snaks'] = $listSerializer->getSerialized( $reference->getSnaks() ); [23:44:55] bad [23:45:21] (03PS2) 10Hoo man: Don't alter the object passed to ByPropertyListSerializer::getSerialized [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/165131 [23:45:49] (03PS2) 10Hoo man: Don't alter the object passed to ByPropertyListSerializer::getSerialized [extensions/Wikibase] (wmf/1.25wmf1) - 10https://gerrit.wikimedia.org/r/165132 [23:46:15] aude: Ok? [23:46:28] i think [23:46:35] According to Tim's trace it hits Wikibase\Lib\Serializers\ByPropertyListSerializer::getSerialized twice [23:46:38] so that makes sense [23:47:16] maybe 3 times, if it can call itself [23:47:25] I thought I was doing something wrong so I discarded the first entry [23:47:41] aude: Can you take care to get this deployed tonight? [23:47:55] I'm half dead after having to deal with stuff yesterday night [23:47:57] (03PS1) 10Aude: Call $reference->getSnaks() separately in ReferenceSerializer [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/165133 [23:48:04] (and need to get up for university) [23:48:07] i think ^ also might help [23:48:09] hoo: ok [23:48:33] (03CR) 10Hoo man: [C: 032] "Wont harm" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/165133 (owner: 10Aude) [23:48:43] thanks [23:48:43] (03PS1) 10Hoo man: Call $reference->getSnaks() separately in ReferenceSerializer [extensions/Wikibase] (wmf/1.25wmf1) - 10https://gerrit.wikimedia.org/r/165134 [23:48:50] this is how it looks from the core dump: http://paste.tstarling.com/p/Nzgqmi.html [23:49:14] with Wikibase\Lib\Serializers\ByPropertyListSerializer::getSerialized as the top two calls [23:50:32] Ok, leaving for today [23:50:44] aude: Thank you for taking it tonight [23:50:52] I'll do it tomorrow again, if needed :S [23:50:53] hoo: thanks for your help [23:51:06] If you need to reset an item, you should have got a manual by mail [23:51:11] yep [23:51:41] * aude waits for jenkins [23:52:15] do you need anything else from this core dump? arguments etc.? [23:52:28] TimStarling: No, I think we got the patches up now [23:52:52] how do we verify the fix? [23:53:00] try Q183 again? [23:53:12] or the other item [23:53:17] my test case is http://www.wikidata.org/w/index.php?title=Q183&oldid=143201634 [23:53:22] aude: Q183 history [23:53:24] ok [23:53:34] please don't change its state (as in edit it) [23:53:48] * aude nods [23:54:04] :) [23:56:47] TimStarling: also if someone could look at https://gerrit.wikimedia.org/r/164976, i think that would be helpful in future situations [23:58:10] thanks :)