[11:42:44] ShakespeareFan00, was there a problem? [11:54:04] Yes [11:54:33] I couldn't get the script mentioned here https://commons.wikimedia.org/wiki/Help:Chunked_upload [11:54:37] to activate [11:54:50] Despite placing the relevant import line in my User js [11:55:26] https://commons.wikimedia.org/wiki/User:ShakespeareFan00/common.js [11:55:48] I am wondering if something changed in the API, which meant the script was using an older method no longer supported. [11:56:28] (Aside : https://commons.wikimedia.org/wiki/Special:UploadWizard ) [11:56:48] (Aside: .. doesn't appear to have options for uploading from URL's) [11:57:39] If upload wizard had options for URL based uploads more generally.. [11:58:06] (like it has for Flikr) I might have more success? [11:58:10] * ShakespeareFan00 AFK [12:11:08] [[Tech]]; BEGJIGIT; /* Bot */ new section; https://meta.wikimedia.org/w/index.php?diff=20168906&oldid=20167309&rcid=15668534 [12:13:00] ShakespeareFan00, so I tried running it in console while on https://commons.wikimedia.org/wiki/File:Sandbox.png [12:13:16] I now see a "Upload a new version of this file (chunked upload)" just above the "File usage" section [12:14:28] Try it with - https://commons.wikimedia.org/wiki/File:Catalog_of_Copyright_Entries,_1978_Performing_Arts_April-June_Fourth_Series_Vol_1_Part_3.pdf [12:15:02] 404 [12:15:22] The file isn't present yet [12:15:33] hm [12:15:36] I was following the instructions at [12:15:43] I suppose we should still expect the chunked upload link I guess [12:15:49] https://commons.wikimedia.org/wiki/Help:Chunked_upload regarding a "New file" [12:16:34] So it seems Chunked upload isn't getting run on a non-existent file [12:16:38] yeah there's a problem here [12:16:46] Despite the documentation saying it probably should [12:16:48] you see that "catch ( ex ) {}" ? [12:16:56] I wasn't reading the console [12:17:08] well this never gets to the console [12:17:12] it's in the source code [12:17:13] anyway [12:17:14] That seems to be catching "TypeError: validTitle.ext is undefined" [12:17:54] I also when loading the page for the non-exsitent file see [12:17:55] This page is using the deprecated ResourceLoader module "jquery.ui". [12:17:55] Please use OOUI instead. [12:18:00] in my console logs [12:18:14] yeah ignore that for these purposes [12:18:32] anyway [12:18:43] lemme dig into core [12:19:50] Also if you are digging into core, letting Chunked Upload use a URL instead of a local file would be VERY useful [12:20:30] https://gerrit.wikimedia.org/r/c/mediawiki/core/+/509775 looks suspect [12:20:57] https://phabricator.wikimedia.org/T232521#5483884 [12:21:48] "..as we may not have caught every case." :eyesroll [12:22:44] Have I found an unknown use case that the patch hadn't considered? [12:23:05] You've found something in userspace [12:23:11] which is unlikely to ever be considered [12:23:20] Ouch [12:23:23] :( [12:23:29] Is it repairable? [12:23:34] yes [12:23:48] What would the ETA on a patch be? [12:23:56] what? [12:24:22] what patch? [12:24:24] As in how long would it take for the script to be made working again? [12:24:36] it's like a one line fix on a wiki page [12:24:43] Okay what's the repair? [12:25:10] s/ext/getExtension()/ [12:25:12] Knowing that I can't edit userspace scripts that aren't mine [12:25:29] mm [12:25:31] I could but [12:25:53] * ShakespeareFan00 is not an Admin [12:26:08] Take it to Village Pump first? [12:26:11] ah hell [12:26:19] it's been edited by a bunch of people other than the user [12:26:28] it's got its own Help: namespace page [12:26:39] it's just mis-namespaced, I'll treat it like MediaWiki: [12:26:54] plus it's broken due to a mistake in MediaWiki development :) [12:27:35] :lol [12:27:56] IAR updates to scripts to make them working :) [12:28:15] Nice when it works in favour of "getting stuff done " [12:28:26] applied [12:29:04] okay so.... [12:29:33] https://commons.wikimedia.org/wiki/File:Catalog_of_Copyright_Entries,_1978_Performing_Arts_April-June_Fourth_Series_Vol_1_Part_3.pdf [12:29:48] Still no script activation... might be a caching issue of the old version [12:38:20] Hmm... [12:38:35] Still not seeing the script activate on a non-existent file page... [12:38:41] Maybe I need to find a different tool [12:38:43] :9 [12:38:45] :( [12:39:39] I can use the workaround of a dummy file. [12:39:44] As the first upload [12:39:48] Done that before [12:50:59] sorry got distracted [12:52:09] ShakespeareFan00, oh I see it [12:52:22] ShakespeareFan00, It changes the 'Create source' button to have a dropdown of its own [12:53:45] That shows up for me if I go to https://commons.wikimedia.org/wiki/File:Catalog_of_Copyright_Entries,_1978_Performing_Arts_April-June_Fourth_Series_Vol_1_Part_4.pdf and run importScript('User:Rillke/bigChunkedUpload.js'); [12:54:57] Okay [12:55:12] I used my own upload a dummy first approach though [12:55:24] It just takes a few attempts to get it to reconstitute [13:01:11] Okay [13:01:14] The script works [13:01:24] It just takes a while to reconstitute [15:25:51] [[Tech]]; Stryn; /* Bot */ re; https://meta.wikimedia.org/w/index.php?diff=20169516&oldid=20168906&rcid=15669492 [20:03:56] Is there a concise explanation of the difference between the js vars wgWikiID and wgDBname ? https://phabricator.wikimedia.org/rMWd45baf7f0734a2506c77925064f42bc13e1bff5e says the former is better, but I don’t really understand the reason [20:11:10] The WikiID is - [20:11:57] on wikimedia, each site has it's own database, so using the db name as an identifier makes sense [20:12:36] however, some setups don't have access to multiple databases, so they can use a table prefix to seperate sites instead [20:12:51] the WikiID will work in both situations [20:13:25] Thanks! [20:13:32] (minor correction: when the table prefix is blank, the WikiID is just the db name)