[14:24:49] each wikipedia category page has a wikidata item, but the item doesn't include links to the parent (or children) category. is that something that just hasn't been done yet, or are you supposed to derive those relationships in a different way? [15:25:46] LotR: as far as I know, you're supposed to derive them in a different way [15:26:20] I haven't paid much attention to it, but the argument is usually that wikis don't all structure their categories in the same way [15:45:16] LotR: yeah, it’s not really integrated (yet? no idea) [15:45:19] there is some information in Wikidata [15:46:29] you could try to guess category relationships based on “category combines topics” statements [15:46:33] but that would be unreliable [15:46:54] however, you can also query the category tree of each wiki via the Wikidata Query Service https://www.mediawiki.org/wiki/Wikidata_Query_Service/Categories [17:34:11] WikidataFacts: ah, cool. thanks! [19:31:15] Does it exists any Discord rooms for Wikidata? [21:22:31] iridescence: if you want something more modern than old good IRC you can use Riot (matrix). All Freenode is bridged. [21:22:41] https://matrix.to/#/!yRuAwUYSCXNJhsRNRB:matrix.org?via=matrix.org&via=matrix.allmende.io [21:30:26] Some time ago i created a bot that copies items from wikidata on test.wikidata. It works but have some issues: [21:30:27] It looks like pywikibot is unable to create new property. I workaround it by POSTting raw api call. The problem is that wikidata log user's IP becouse call is unauthicated. [21:30:28] Does anyone know how to get API token? Is it possible to get token from current wikibot session (without new login)? [21:38:22] PS. I'm writing this bot to make testing another bots easier. Bot takes list of items to copy as input and outputs json with 'wikidataItem': 'test.wikidataItem' pairs. [23:12:21] What's the best way to wait for the wikidata page to be fully loaded? https://stackoverflow.com/questions/55072278/how-to-wait-for-wikibase-edit-links-to-finish-loading [23:15:26] there's a hook, I can look it up in a minute [23:16:08] takitam[m], i do many modifications via the api, and i use the pywikiapi library - https://pypi.org/project/pywikiapi/ [23:16:29] it is far more straightforward than using pywikibot if you want to call api directly [23:16:57] aha: mw.hook('wikibase.entityPage.entityView.rendered').add(function () { ... }); [23:17:05] thanks nikki !!! [23:17:36] (i'm trying to implement custom edit summaries for wikibase) [23:18:09] takitam[m], so just use site.login(), and afterwards call site(...) with whatever parameters you need. [23:18:42] nikki, want to answer stackoverflow question and get some points? :D [23:23:02] I'm not sure if I even have a stackoverflow account :P [23:24:56] anyway I'm happy enough just knowing I could help without getting any points :) [23:28:35] nikki, it worked beautifully, thank you! where did you find that info? there might be other good stuff there :) [23:35:11] I had been wanting something like that for years and lucas implemented it a while back [23:36:48] https://phabricator.wikimedia.org/T169775 is the ticket that implemented it, there are some others mentioned there [23:38:33] hm. probably not that one which *implemented* it, I know what I mean... [23:50:59] yurik: thanks! [23:50:59] So we have 3 python wrappers around wikibase api.... [23:51:54] well, technically there are many more libraries out there, but unlike other libs, pywikibot was written by the original author of the Mediawiki API itself :D [23:53:35] plus pywikiapi is not really a wrapper, because it does not re-implement api concepts, but rather allows you to use the api directly. Essentially the extra layer of documentation is minimal, you just have to use mw api docs for all the stuff you want to do