[11:36:28] <benestar>	 Lydia_WMDE: I've moved https://phabricator.wikimedia.org/T102888 to consider for next sprint because Wikipedians will complain
[13:13:42] <benestar>	 SMalyshev: do you need unions for WDQ to SPARQL?
[19:14:42] <multichill>	 hi Amir1, I wanted to ask you something :-)
[19:14:58] <Amir1>	 hey :)
[19:15:02] <Amir1>	 happy to see you
[19:15:04] <Amir1>	 Sure
[19:15:33] <multichill>	 You had some code to import from Commons right? Quite a few paintings have been imported from Commons, but often the collection and/or inventory number is missing
[19:16:17] <Amir1>	 I have
[19:16:25] <Amir1>	 I need to search but I can find them
[19:16:40] <multichill>	 Like http://tools.wmflabs.org/autolist/autolist1.html?q=CLAIM%5B31%3A3305213%5D%20AND%20NOCLAIM%5B195%5D%20AND%20CLAIM%5B18%5D
[19:18:09] <multichill>	 Not sure how many hits that will give
[19:19:47] <multichill>	 Amir1: The source metadata on Commons is often quite bad. Only the Google Art ones tend to be quite good
[19:20:08] <Amir1>	 ok
[19:20:17] <Amir1>	 Can we parse google Art ?
[19:20:37] <multichill>	 Pff, haven't tried that yet, but most of it is on Commons in templated format
[19:21:43] <Amir1>	 awesome
[19:21:57] <multichill>	 https://www.wikidata.org/wiki/Special:Contributions/BotMultichill <- I'm just using museum websites, easier and better quality
[19:22:29] <multichill>	 But cleaning up the collections before import is a lot of work, so having the inventory numbers and collections already from Commons would save a lot of time
[19:22:32] <Amir1>	 https://www.wikidata.org/wiki/Q3944483
[19:22:41] <Amir1>	 This didn't return anything good
[19:24:08] <multichill>	 Yeah, that happens to have a category, but extracting that would be a nightmare
[19:24:47] <Amir1>	 hmm
[19:24:49] <Amir1>	 okay
[19:25:15] <Amir1>	 So I need to import data from commons about museums and collections?
[19:28:00] <multichill>	 Amir1: Hmm, clicked about 15 random paintings in those lists and none of these seem to include an institution template so that seems to be a dead end
[19:28:29] <Amir1>	 I think we can work based on categories
[19:29:09] <multichill>	 Amir1: http://tools.wmflabs.org/autolist/autolist1.html?start=350&q=CLAIM%5B31%3A3305213%5D%20AND%20CLAIM%5B195%5D%20AND%20CLAIM%5B18%5D%20AND%20NOCLAIM%5B217%5D seems to be more interesting
[19:31:25] <Amir1>	 hmm
[19:31:27] <Amir1>	 okay
[19:31:34] <Amir1>	 I start with inventory number
[19:33:45] <multichill>	 I would do. Loop over the list. Grab the image. Look for institution and accession number. If institution is the same, add the accession number with the museum as qualifier
[19:34:56] <Amir1>	 okay :)
[19:35:14] <multichill>	 http://tools.wmflabs.org/autolist/autolist1.html?start=550&q=CLAIM%5B31%3A3305213%5D%20AND%20CLAIM%5B195%5D%20AND%20CLAIM%5B18%5D%20AND%20NOCLAIM%5B217%5D%20AND%20NOCLAIM%5B195%3A768717%5D
[19:35:20] <multichill>	 Had to filter out the private collections
[19:36:22] <multichill>	 Amir1: Or we could just wait for the arbitrary access and do some template tricks on Commons to figure out what to import ;-0
[19:36:43] <Amir1>	 I can't wait that much ;)
[19:58:47] * Amir1 is writing the code
[20:11:39] <multichill>	 Amir1: https://commons.wikimedia.org/wiki/File:Edwin_Landseer_-_Scene_from_A_Midsummer_Night%27s_Dream._Titania_and_Bottom_-_Google_Art_Project.jpg <- images like this one are good to import info from
[20:12:00] <multichill>	 https://www.wikidata.org/w/index.php?title=Q18162195&type=revision&diff=223003450&oldid=222801927
[20:12:15] <Amir1>	 I'm working on it
[20:12:22] <Amir1>	 It will finish in half an hour
[20:14:09] <multichill>	 :-)
[20:28:02] <Amir1>	 multichill: https://www.wikidata.org/wiki/Special:Contributions/Dexbot
[20:28:05] <Amir1>	 I just started
[20:28:29] <Amir1>	 I should add sources
[20:28:30] <Amir1>	 brb
[20:28:54] <multichill>	 Amir1: Nice! can you add the collection as a qualifier?
[20:29:08] <multichill>	 yes, imported from Commons sounds like a good thing to do
[20:42:40] <Amir1>	 multichill: I will do
[20:44:55] <Amir1>	 multichill: Can you check this? https://gerrit.wikimedia.org/r/#/c/219244/
[22:02:54] <__mek__>	 If I wanted to access an entity's tatements like https://www.wikidata.org/wiki/Q556 as json, how would I go about this? The desired result would be something like the #infobox of https://en.wikipedia.org/wiki/Hydrogen as json, but it appears much of the info isn't available via the wikidata entry. I'm looking at https://www.mediawiki.org/wiki/Wikibase/API now.
[22:09:17] <nikki>	 __mek__: https://www.wikidata.org/entity/Q556.json will give you the entity's data as json
[22:10:09] <__mek__>	 nikki: Thanks!
[22:11:00] <__mek__>	 I see, and it redirects to https://www.wikidata.org/wiki/Special:EntityData/Q556.json
[22:11:07] * nikki nods