[13:57:21] sup [13:57:31] ping [14:04:49] pong [14:06:27] hi) guys im kinda new to wiki engine and wonder how to make my wiki to create links between articles automaticaly [14:07:06] i mean if i have article named FOO and in another article BAR i have word "foo" [14:07:54] i'd like to wiki make link from word foo in article bar to article foo [14:11:08] refreshLinks.php no help [14:50:51] hello [14:51:08] hay alguien acá que escriba y entienda español? [15:06:28] zm: There is no built-in auto-linking. See https://www.mediawiki.org/wiki/Help:Links#Internal_links for how links work in general (but you know that already, I guess) [15:07:39] The Semantic Glossary extension can do it, but that's probably overkill. https://www.mediawiki.org/wiki/Extension:Semantic_Glossary [15:09:22] Or you could write/extend a backend for Lingo, but that requires PHP. https://www.mediawiki.org/wiki/Extension:Lingo [16:27:14] hi all, I was wondering if I could get some help with my module https://dpaste.de/B94h [16:27:26] I need to get the title of the page that includes the template that calls the module [16:28:20] that module (titled "Module:Episodes") is invoked by a template ("Template:Episodebox") that is included in a page with a name like "1: Episode 1" (done with ":1: Episode 1") [16:35:09] khuxkm: so what's currently happening vs what you expect to happen? [16:36:49] `frame:getParent():getParent()` returns nil, and not the page that includes Episodebox [16:37:06] I expected it to give me the page that included Episodebox [16:41:53] khuxkm: looks like that is intended behavior, getParent() only lets you go up one level [16:42:11] ok, but how do I get the title of the page that includes Episodebox [16:42:29] saying it's "intended behavior" doesn't help when my question was "this isn't working, how do I do this" [16:43:30] try mw.title.getCurrentTitle() [16:43:47] (was looking at the docs to find if there was another way) [16:44:41] it will return a Title object, see https://www.mediawiki.org/wiki/Extension:Scribunto/Lua_reference_manual#Title_objects [17:15:50] OK, so that worked, kind of [17:16:36] https://dpaste.de/xc37 here's my code now [17:17:01] the new problem is that it's rendering a link that shouldn't be there [17:17:25] ...gimme a sec [17:18:26] ...wow, I'm an idiot, nvm [17:18:38] need to tostring() the numbers, otherwise it considers them IDs [17:28:24] Skizzerz: thanks for the help! [19:47:17] hi ! anyone from the visualeditor team ? [20:23:58] is there a team? [20:24:09] heh [20:24:14] #mediawiki-visualeditor is probably better [21:15:55] tried the #mediawiki-visualeditor, but seems dead :-P [21:16:09] anyway, I'll throw the comment here and maybe someone will care [21:17:02] for some years now i've been working on a gadget called proveit (https://commons.wikimedia.org/wiki/Help:Gadget-ProveIt) [21:17:10] it's a reference manager for wikipedia [21:17:27] basically it provides a GUI for editing and adding references to the wikitext [21:17:46] recently i completed a grant to make it compatible with the new wikitext editor, among other nice things [21:18:01] Sophivorus: That's funny, Proveit is on my csp log list, so I was just reading about it :) [21:18:01] and now im thinking i could try updating the GUI so that it uses OOUI [21:18:15] anyway, i was thinking that maybe [21:18:19] with some further improvements [21:18:33] proveit could be turned into the default reference manager for the new wikitext editor [21:18:36] since it currently lacks one [21:19:05] and i wanted to talk with someone who may be able to judge how plausible this is [21:20:35] Sophivorus: Have you also tried the #wikimedia-editing irc channel? [21:20:49] not really, i will go there next ! [21:23:52] Sophivorus: btw, there's still some people loading proveit, using http. Which doesn't matter as it is automatically converted to https - but it does generate a warning, and currently ProveIt has the highest number of such warnings of any gadget [21:24:14] hm, interesting ! [21:24:31] Probably not much to be done, as its probably individual people's common.js, but if you happen to come across any sort of central page still loading with http, please change :) [21:24:42] bawolff, we could just go across everyone's common.js files and fix them to https? [21:25:07] Krenair: Yeah, i need a bot and also to get back my global editinterface rights :) [21:26:07] just hijack jquery and have it switch to https if it's hitting one of a certain set of urls regardless of what it actually specified ;) [21:26:15] think that should be possible with ajaxPrefilter [21:26:26] Sophivorus: err, wait. This list is not ordered by count! [21:26:32] hm i think it may be due to proveit being loaded in protocol relative way? see https://en.wikipedia.org/wiki/MediaWiki:Gadget-ProveIt.js [21:26:33] So you're not actually at the top [21:26:42] you might be alphabetically first though ;) [21:26:52] lol [21:26:56] * bawolff feels stupid now [21:27:04] P comes before A? XD [21:27:08] nobody likes my clever workarounds :< [21:27:33] Sophivorus: protocol relative should be fine, as the protocol should always be https [21:27:34] we do like you though, Skizzerz [21:27:42] thanks! :D [21:28:12] Skizzerz: Honestly, maybe we should. CSP (eventually) and HSTS is going to prevent non https requests anyways [21:29:05] ProveIt is now number 5 if i order by count (and only do stuff on enwikipedia) [21:29:09] I've done worse in some (closed source) codebases [21:30:17] http://en.wikipedia.org/w/index.php?title=User:WikiBhasha.MSR/WikiBhasha.js&action=raw&ctype=text/js seems to be number one [21:30:30] eww, ctype=text/js ?? That probably doesn't even work [21:31:18] there're two proveit versions running on enwikipedia, one very old, one very new .. i bet it's the old one [21:31:35] I think it might mostly be people loading it from frwikisource and wikidata [21:32:02] bawolff: yeah just returns text/x-wiki [21:37:47] bawolff: untested super quick draft, but should work ;) https://dpaste.de/PjCD [21:38:40] I wonder if the proveit http, is entirely people's /global.js at meta: https://meta.wikimedia.org/w/index.php?sort=relevance&search=insource%3A%22http%3A%2F%2Fen.wikipedia.org%2Fw%2Findex.php%3Ftitle%3DUser%3AProveIt_GT%2FProveIt.js%26action%3Draw%26ctype%3Dtext%2Fjavascript%22&title=Special%3ASearch&profile=advanced&fulltext=1&advancedSearch-current=%7B%7D&ns2=1&ns8=1 [21:40:08] Skizzerz: I think that would be a bit of a last resort though :) [21:42:26] Hmm, wonder what WikiBasha even is [21:42:29] Copyright (C) 2010, Microsoft (!!) [21:43:15] o.o [21:44:06] seems legit actually [21:44:09] "WikiBhasha is a multilingual content creation tool developed by Microsoft Research." Well that's kind of cool [21:44:13] Didn't there used to be an extension for it? [21:44:35] looks like [21:45:40] MS seems to still own the domain name, but i can't get the site to load [21:46:48] https://phabricator.wikimedia.org/project/profile/417/ [21:52:42] The big CSP violator is really all pathoschild's scripts hosted at toolforge [21:53:04] Honestly, some of them should prbably be straight up merged into CentralAuth [21:54:42] Hmm, maybe its too dependent on tools for that [22:04:52] There is any way to query the _pageID with cargo but remove the comma of large numbers? 1,000 -> 1000 [22:05:33] cargo? [22:05:43] The extension [22:05:49] pagefunctions usually have a |R}} to provide a plain number [22:07:35] Got it with ROUND(_pageID)