[03:00:10] [[Tech]]; Harpreetsandhux; /* Jobshub.live your job search starts here */ new section; https://meta.wikimedia.org/w/index.php?diff=19893167&oldid=19888137&rcid=15055838 [03:01:40] [[Tech]]; AntiCompositeNumber; Undo revision 19893167 by [[Special:Contributions/Harpreetsandhux|Harpreetsandhux]] ([[User talk:Harpreetsandhux|talk]]) rv spam; https://meta.wikimedia.org/w/index.php?diff=19893170&oldid=19893167&rcid=15055841 [09:01:08] Hi! i'm trying to do a "GROUP BY" in a query, but I just get "Bad aggregate": https://w.wiki/KNH [10:45:27] iridescence: I don't see a GROUP BY clause in that query [10:45:51] you do know you need to use an aggregation function for things that are not being grouped by right? [10:46:09] you might get better responses in #wikidata [10:49:13] JD|cloud: Thanks for your reply. I don't get any feedback at #wikidata. How should I structure the query? [10:50:40] https://www.w3.org/TR/sparql11-query/#groupby [10:53:02] And how can i implement this in "my" query? I wish to use group by to remove duplicate items. [10:53:36] you would not want to use GROUP BY for that purpose [10:53:53] DISTINCT? [10:54:22] on ordered data you can use the MAX() aggregate function but that's pretty hacky and won't work for non-ordered data (plus it's not guaranteed all data will be from a single item) [10:54:48] it should not be possible to have multiple WD items associated with a given page [10:56:35] SELECT DISTINCT also won't do the job since rows with different QID's are not considered distinct [10:57:58] or, items associated with the same page can also have different labels [11:00:17] Hmm.. Should I do a MAX on ?item (Q-id) that is unique value? [11:01:03] Not sure if it’s anything you can help but my train wifi is generating a cert error for commons.wikimedia.org in chrome. Slightly different error in safari. Think it might be blocked though as nationalrail does the same but not all sites. [11:01:09] Any thoughts [11:01:26] RhinosF1: check the issuer of the certificate [11:01:43] if it's not from the WMF you're being MiTM'd and blocked [11:02:21] JD|cloud: I’m mobile, how can you see on chrome for iOS [11:02:52] https://commons.miraheze.org/wiki/Main_Page is fine weirdly but https://commons.wikimedia.org/wiki/Commons:Office_actions/DMCA_notices#The_Weeknd is [11:03:15] iridescence: I'm not entirely sure since I work in SQL more often than SPARQL. It also depends on how you define "duplicate" actually. [11:04:20] https://usercontent.irccloud-cdn.com/file/gl8mYJkc/RPReplay_Final1584183793.mp4 [11:04:39] hehe, true. I got some more ideas now. So I guess I will figure it out. Thanks for the help. [11:04:54] https://usercontent.irccloud-cdn.com/file/wnuIxEA2/IMG_5948.PNG [11:05:14] JD|cloud: them two files show the errors [11:05:36] this is not necessarily a cert error [11:05:42] this could be a simple firewall [11:06:31] JD|cloud: ah, anyway to tell? [11:06:54] it is not necessary to highlight me on every message [11:07:09] what happens if you try on plaintext? [11:07:30] What url would I use for that? [11:08:50] same one but remove the https:// at the beginning [11:09:04] or better, try a browser like chrome or Firefox that isn't so vague with errors [11:09:33] I am using chrome for ios on the mp4 one [11:09:47] And without https:// redirects to add it [11:11:29] Chrome should be able to be more specific [11:12:09] https://www.irccloud.com/pastebin/KDTzfLax [11:12:20] ^ that’s chrome’s error [11:13:00] well that's pretty clear-cut evidence of the wifi hotspot MiTM'ing you b/c it doesn't like the content of your connection [11:13:13] or someone else MiTM'ing you [11:13:21] either way, not much you can do about it [11:13:30] TLS is doing its job [11:15:17] Urgh, suppose i’ll have to wait until I’m at the hotel then. [13:46:34] [[Tech]]; Sarri.greek; /* Edittools at el.wiktionary */ new section; https://meta.wikimedia.org/w/index.php?diff=19893728&oldid=19893170&rcid=15057304 [21:12:22] hi, what is the labs channel? [21:13:03] hi Wilfredor #wikimedia-cloud [21:13:27] thanks Platonides, I remember when it was justo toolserver channel [21:13:37] me too [21:16:14] I jute finish a very nice bot :) [21:17:01] it perform something over the files to improve the overexposition, chromatic aberration, blurring, etc. and reupload [21:18:31] it could too recheck the Quality images and add a note if the bot think that there are a problem with the image [21:19:11] nice [21:20:31] At the moment I think I can only run it locally due to the amount of resources it consumes