[07:58:26] Good morning. :) [09:04:59] multichill: https://www.wikidata.org/wiki/Wikidata:Database_reports/Constraint_violations/P1774 :( [09:05:31] Catalans making a mess again? :P [09:06:39] Can you contact him? He is also using imported from for sourcing..... [09:06:50] Yeah, no URL or something... [09:07:13] Point him to https://www.wikidata.org/wiki/Help:Sources [09:07:27] All my reports are clutered with Catalan stuff [09:08:14] Part is even from MNAC! I imported all the paintings from that collection and added url's everywhere! [09:08:47] I'll remove the mandatory status for now [13:37:39] 21:22:19 The lack of plural words there annoys me more. :P [13:37:40] ikr [13:38:02] "90 minute" [15:55:21] hoo, nikki: the weighting hack is still online? :D some things survive much longer than expected [16:04:41] dennyvrandecic_: it's not the only hack that is still life... [16:04:44] *sigh* [16:05:05] basically, it'S blocked on integration with Elastic. Which is blocked on implementing language fallback in Elastic. [16:05:09] which is not easy [16:05:38] just fully materialize the language feedback on write, or too expensive? [16:05:51] *fallback, not feedback [16:06:52] (it's crazy how much legacy one can create in such a short time :D ) [16:09:32] Nemo_bis: my daughter is Nemo right now [17:05:52] dennyvrandecic_: the captain? ;-) I've never watched the cartoon [17:58:06] I think everyone is going to miss this, not sure how many people have the rfc page on their watchlist. https://www.wikidata.org/wiki/Wikidata:Requests_for_comment/Editing_descriptions_from_Wikipedia_Android_app [18:08:21] Nemo_bis: the Pixar clown fish character :D [18:10:27] sjoerddebruin: I think so too. RFC isn't a very well established process at all on Wikidata. [18:17:08] dennyvrandecic_ sorry, i forgot, what were you talking about ? :p [18:17:58] Alphos: when? (not sure what you are refering too :D ) [18:18:11] oh look, corral ! [18:18:18] :D [18:21:36] dennyvrandecic_: I think he's more Captain ;-) [18:27:58] dennyvrandecic_: quite hard to dress as a clown fish :o [18:28:14] Perhaps easier than dragging around a submarine full of precious jewels, though [18:28:36] she certainly looks super cute :) but I admit I might be biased [18:29:20] hehe [18:29:31] the KPI will be in the amount of sugar collected? [18:30:21] over the years, yes :) today it is her very first time (and mine too, actually), so merely experiencing it at all is sufficient for today [18:30:39] but from next year, yes, that sounds like a decend objective function to optimize on [18:47:33] hoo: <3 [18:48:29] hi sjoerddebruin :) [19:25:44] SMalyshev: hey. i'm looking at writeQuantityValue() at the moment. I have some ideas on how to make it a little saner [19:26:12] shall i just amend your patch, or make a follow-up? [19:26:19] we may get conflicts either way... [19:26:24] DanielK_WMDE: wait for a min, I have a follow up [19:26:39] that patch wasn't complete, I just was too tired yesterday to finish it [19:26:39] that patch wasn't complete, I just was too tired yesterday to finish it [19:27:04] so I checked in where I stopped, but it wasn't all [19:27:08] yea, i saw [19:27:25] I should've put WIP on it, but I was too tired :) [19:27:57] no worries. [19:28:12] i'm just saying, if you don't want to mess with QuantityRdfGenerator, let me do it :) [19:28:33] if you have something to push, i'll wait a few minutes. i'll take care of the test, anyway [19:29:30] DanielK_WMDE: ok, I'll finish converting tests to nt soon and check it in and then you can have a go [19:31:13] oh, ntriples was just a thought, not a blocker. but i'm happy about it anyway :) [19:35:48] DanielK_WMDE: ok, I've checked it in [19:38:15] i'll have a look [19:38:22] SMalyshev: want me to amend, or follow up? [19:38:38] DanielK_WMDE: feel free to amend [19:38:45] ok, will do [19:48:18] SMalyshev: QuantityRdfBuilderTest fails after your patch, it seems. I'll check it out. [19:48:41] DanielK_WMDE: hmm looks like I missed something, I just checked the maint test [19:48:42] ok [19:52:54] SMalyshev: yea, it has to do with the write being stateful. before your change, the unconverted value was written first, now it is written last - but without setting the subject again. [19:53:01] i'll see what i can do... [19:53:36] hmm... yeah that may be a problem [19:53:58] I think the if then needs to be split and the call moved to before the sub-write [19:54:06] I was afaird there's something like that there [19:54:13] there's various ways this could be done [19:54:19] i'm thinking about it [21:15:18] SMalyshev: well, took a bit longer. I re-structured QuantityRdfBuilder a bit, because the way that addValue was calling itself confused me... [21:20:12] ok I'll take a look now [21:21:30] hm... when runningthis manually, i'm not sure i'm getting the right result... [21:27:31] DanielK_WMDE: I'm worried about using getValueNodeWriter inside writeQuantityValue... I'm not sure that's what is supposed to happen in addUC script [21:29:01] also, don't you want to reuse linkNormalizedValue in the script too? [21:34:24] getValueNodeWriter is indeed what needs to happen, to make sure the correct writer is used [21:34:39] if you use another writer, things fail, because writeQuantityValue needs the writer to be in a specific state [21:35:28] oh, i see what you mean. in the main script, that state is controleld outside the function call [21:35:32] yea, not so pretty [21:36:27] i could re-use linkNormalizedValue, and we could have similar utility functions for writing the type. These should really be in the COmplexValueHelper. [21:36:42] If you want that, drop a note on the patch, i'll look into it tomorrow. [21:37:04] i'm still at the office, and it's way too late... i should go home now [21:37:14] I don't have strong opinion, just a suggestion [21:37:38] it definitly could be polished further. i kind of stopped myself there. [21:37:46] but... i found some odd stuff while testing manually [21:37:51] i wrote about it on the patch [21:38:08] i don't quite understand what's happening there. maybe you can try the test I ran, and see for yourself [21:38:30] ok, which test? [21:38:43] see my last comment [21:40:42] DanielK_WMDE: ok, I'll take a look after lunch [22:19:17] PROBLEM - Response time of WDQS on einsteinium is CRITICAL: CRITICAL: 11.11% of data above the critical threshold [300000.0] [22:30:16] PROBLEM - Response time of WDQS on einsteinium is CRITICAL: CRITICAL: 11.11% of data above the critical threshold [300000.0] [22:42:17] RECOVERY - Response time of WDQS on einsteinium is OK: OK: Less than 5.00% above the threshold [120000.0] [23:09:06] PROBLEM - Response time of WDQS on einsteinium is CRITICAL: CRITICAL: 11.11% of data above the critical threshold [300000.0] [23:39:06] RECOVERY - Response time of WDQS on einsteinium is OK: OK: Less than 5.00% above the threshold [120000.0] [23:42:16] PROBLEM - Response time of WDQS on einsteinium is CRITICAL: CRITICAL: 44.44% of data above the critical threshold [300000.0] [23:46:46] PROBLEM - Response time of WDQS on einsteinium is CRITICAL: CRITICAL: 30.00% of data above the critical threshold [300000.0] [23:52:56] RECOVERY - Response time of WDQS on einsteinium is OK: OK: Less than 5.00% above the threshold [120000.0] [23:54:26] PROBLEM - High lag on wdqs1001 is CRITICAL: CRITICAL: 93.33% of data above the critical threshold [1800.0]