[05:52:30] Is there any way to set a reference for a qualifier? [05:52:30] Is there any way to set a reference for a qualifier? [05:53:36] (independently of setting the reference for the statement) [05:53:36] (independently of setting the reference for the statement) [12:12:08] Was this game ever active? https://github.com/dissemin/oa-wd-game [12:12:08] Was this game ever active? https://github.com/dissemin/oa-wd-game [12:29:38] Nemo_bis: no, we haven't finished it yet [12:29:38] Nemo_bis: no, we haven't finished it yet [12:30:14] it's not very far from being ready though! Tpt[m] coded it at the last Wikicite. [12:30:14] it's not very far from being ready though! Tpt[m] coded it at the last Wikicite. [12:30:46] I should finish it (but I'm lazy...) [12:30:46] I should finish it (but I'm lazy...) [12:31:27] well, anybody can finish it :) [12:31:27] well, anybody can finish it :) [12:31:44] (which is probably why nobody does, as usual ^^) [12:31:44] (which is probably why nobody does, as usual ^^) [12:39:43] hehe, it's ok [12:39:43] hehe, it's ok [12:39:59] just looking for URLs to advertise in my next round of Dissemin invitations [12:39:59] just looking for URLs to advertise in my next round of Dissemin invitations [12:40:22] about 100 clicks on oabot turned into 0 edits so I'm not sure that was useful :) [12:40:22] about 100 clicks on oabot turned into 0 edits so I'm not sure that was useful :) [12:42:20] Nemo_bis: yeah, we need to revamp that one too… Ideally I'd love to have time to work on free-form citation parsing for that one [12:42:20] Nemo_bis: yeah, we need to revamp that one too… Ideally I'd love to have time to work on free-form citation parsing for that one [13:06:18] true, that's perhaps where the biggest gains are possible [13:06:18] true, that's perhaps where the biggest gains are possible [13:07:03] pintoch: I feel I'm wasting my time when I add URLs which could be reached just by pointing the existing DOI to oaDOI [13:07:03] pintoch: I feel I'm wasting my time when I add URLs which could be reached just by pointing the existing DOI to oaDOI [13:07:43] Which is why I filed https://phabricator.wikimedia.org/T166959 [13:07:43] Which is why I filed https://phabricator.wikimedia.org/T166959 [13:10:03] Nemo_bis: I was frustrated like you not to be able to run the bot directly… [13:10:03] Nemo_bis: I was frustrated like you not to be able to run the bot directly… [13:13:07] pintoch: consensus can change though :) [13:13:07] pintoch: consensus can change though :) [13:14:55] I guess after one year of discussions at CS1/2, BRFA and off-wiki, I kind of lost hope… [13:14:55] I guess after one year of discussions at CS1/2, BRFA and off-wiki, I kind of lost hope… [13:16:34] no wonder [13:16:34] no wonder [13:16:45] now my procrastination time goes to Wikidata instead, because change is welcomed here :) [13:16:45] now my procrastination time goes to Wikidata instead, because change is welcomed here :) [14:54:53] hawke: references for individual qualifiers, vs. the statement itself - remind me of the use case? I think I've wanted that before too but the reason slips my mind [14:54:53] hawke: references for individual qualifiers, vs. the statement itself - remind me of the use case? I think I've wanted that before too but the reason slips my mind [18:25:35] this has probably been asked a hundred times already, but what is the best way to get Qids from sitelinks? [18:25:36] this has probably been asked a hundred times already, but what is the best way to get Qids from sitelinks? [18:26:34] pintoch: I’m not sure what you mean… like, [[Special:ItemByTitle]]? [18:26:34] pintoch: I’m not sure what you mean… like, [[Special:ItemByTitle]]? [18:26:35] 10[2] 04https://www.wikidata.org/wiki/Special:ItemByTitle [18:26:35] 10[2] 04https://www.wikidata.org/wiki/Special:ItemByTitle [18:26:48] say I have "https://en.wikipedia.org/wiki/Knuth-Bendix" and I want to get "Q2835803" back (ideally, in one request for a batch of them) [18:26:48] say I have "https://en.wikipedia.org/wiki/Knuth-Bendix" and I want to get "Q2835803" back (ideally, in one request for a batch of them) [18:27:06] hmm that looks good indeed [18:27:06] hmm that looks good indeed [18:27:25] I was doing it via SPARQL but it does not handle redirects [18:27:25] I was doing it via SPARQL but it does not handle redirects [18:28:19] yeah I just noticed that [18:28:19] yeah I just noticed that [18:28:30] I was going to suggest http://tinyurl.com/y9hfnrkx for batching but it doesn’t work for redirects [18:28:30] I was going to suggest http://tinyurl.com/y9hfnrkx for batching but it doesn’t work for redirects [18:28:52] yeah that's also the query I was using [18:28:52] yeah that's also the query I was using [18:29:27] so, maybe, do it with SPARQL, and then for each sitelink that did not return anything, use Special:ItemByTitle [18:29:27] so, maybe, do it with SPARQL, and then for each sitelink that did not return anything, use Special:ItemByTitle [18:29:43] I like https://tools.wmflabs.org/wikidata-todo/linked_items.php [18:29:43] I like https://tools.wmflabs.org/wikidata-todo/linked_items.php [18:29:44] there’s probably also an API for it… [18:29:44] there’s probably also an API for it… [18:30:37] Stryn: ah, I did not know this one, thanks :) [18:30:37] Stryn: ah, I did not know this one, thanks :) [18:30:55] looks like action=wbgetentities doesn’t follow redirects either [18:30:55] looks like action=wbgetentities doesn’t follow redirects either [18:31:11] wait, no, it has an option for it [18:31:11] wait, no, it has an option for it [18:31:52] ah, I suppose that’s for redirected entities, not redirected sitelinks [18:31:52] ah, I suppose that’s for redirected entities, not redirected sitelinks [18:32:33] pintoch: you could use this instead of SPARQL (and still fall back to Special:ItemByTitle): https://www.wikidata.org/wiki/Special:ApiSandbox#action=wbgetentities&format=json&sites=enwiki&titles=Knuth-Bendix%7CDouglas+Adams&props=info [18:32:33] pintoch: you could use this instead of SPARQL (and still fall back to Special:ItemByTitle): https://www.wikidata.org/wiki/Special:ApiSandbox#action=wbgetentities&format=json&sites=enwiki&titles=Knuth-Bendix%7CDouglas+Adams&props=info [18:32:47] not sure which one’s better [18:32:47] not sure which one’s better [18:33:00] I suppose SPARQL doesn’t impose a limit of 50/500 titles ;) [18:33:00] I suppose SPARQL doesn’t impose a limit of 50/500 titles ;) [18:33:12] makes a lot of sense! [18:33:13] makes a lot of sense! [18:33:25] thank you so much as always :) [18:33:25] thank you so much as always :) [18:33:31] you’re welcome :) [18:33:31] you’re welcome :) [19:12:44] WikidataFacts, pintoch: the "normalize" parameter for wbgetentities seems to follow redirects, but it seems to only support one page at a time [19:12:44] WikidataFacts, pintoch: the "normalize" parameter for wbgetentities seems to follow redirects, but it seems to only support one page at a time [19:13:41] oh interesting ! but… why ?? oO [19:13:41] oh interesting ! but… why ?? oO [19:14:06] not sure :/ [19:14:06] not sure :/ [19:14:26] huh, okay [19:14:26] huh, okay [19:21:23] > oh interesting ! but… why ?? oO [19:21:23] > oh interesting ! but… why ?? oO [19:21:23] Probably because it increases the number of DB requests [19:21:23] Probably because it increases the number of DB requests [19:22:57] yeah, I guess redirects aren’t even in Wikidata’s database, they have to be checked in the individual wikis [19:22:57] yeah, I guess redirects aren’t even in Wikidata’s database, they have to be checked in the individual wikis [19:23:13] (though I don’t know how expensive that is, it’s all the same DB cluster afaik) [19:23:13] (though I don’t know how expensive that is, it’s all the same DB cluster afaik) [19:25:19] The implementation is probably using API calls because setup shared tables is not easy [19:25:19] The implementation is probably using API calls because setup shared tables is not easy [19:41:05] Tpt[m]: yeah, but my point is : if they already accept a list as input format for some combinations of the parameters, why can't they just loop over the titles server-side ? [19:41:05] Tpt[m]: yeah, but my point is : if they already accept a list as input format for some combinations of the parameters, why can't they just loop over the titles server-side ? [19:41:39] because I am going to end up making individual queries for each redirected title anyway… [19:41:39] because I am going to end up making individual queries for each redirected title anyway… [19:42:17] that's definitely possible but they maybe want to make sure that all API requests respond "fast" [19:42:17] that's definitely possible but they maybe want to make sure that all API requests respond "fast" [19:42:28] yeah, makes sense [19:42:28] yeah, makes sense [19:42:47] so that all API calls have roughly the same cost [19:42:47] so that all API calls have roughly the same cost [19:42:56] okay [19:42:56] okay [19:43:12] > because I am going to end up making individual queries for each redirected title anyway… [19:43:12] > because I am going to end up making individual queries for each redirected title anyway… [19:43:13] What you could do is first resolve the redirect using the wiki API and then call Wikidata API [19:43:13] What you could do is first resolve the redirect using the wiki API and then call Wikidata API [19:43:23] * the redirect -> all the redirects [19:43:24] * the redirect -> all the redirects [19:46:28] hmm, okay, if there is an API that lets me resolve redirects by batch on the target wiki [19:46:28] hmm, okay, if there is an API that lets me resolve redirects by batch on the target wiki [19:46:56] that would be indeed much better [19:46:56] that would be indeed much better [19:52:26] there is such a thing! [19:52:26] there is such a thing! [19:53:10] awesome, I feel like I'm getting to the optimal solution thanks to all of you :) [19:53:10] awesome, I feel like I'm getting to the optimal solution thanks to all of you :) [19:54:13] pintoch: https://en.wikipedia.org/w/api.php?action=query&titles=Main%20page|Lowendal&redirects [19:54:13] pintoch: https://en.wikipedia.org/w/api.php?action=query&titles=Main%20page|Lowendal&redirects [19:54:26] yep [19:54:26] yep