[00:34:52] i put "foo" in a template but when i call it it has no title? [00:35:05] https://test.wikipedia.org/wiki/Template:Searchengineinstall & its discussion page has an example [00:38:38] sorry, it has $1 as title but does not pick up the actual parameter value [02:04:13] I need help [02:04:13] Hi Guest1343, just ask! There is no need to ask if you can ask [02:04:44] wm-bot: I need to create a wiki [02:37:33] don't we all need to create a wiki :-P [04:18:47] How can you load RDF graphs into tables and infoboxes? [04:38:42] zzo38: https://en.wikipedia.org/wiki/Wikipedia:Wikidata#Infoboxes_.28Phase_2.29 [04:41:09] rom1504: That isn't really what I am looking for; that is specific to Wikipedia and to Wikidata and other things different [04:41:24] Thank you for the information anyways, but tell me if you have something more better [04:45:22] zzo38: http://wikiba.se/ is usable by any mediawiki [04:45:33] so that's not specific to wikipedia/wikidata [04:50:16] maybe that's also interesting https://www.mediawiki.org/wiki/Extension:RDFIO [05:32:59] Well, that's closer, at least [05:37:22] My idea was rather, each article (but not special pages or talk pages) has an associated RDF graph; similar to associated talk pages there can be associated RDF pages which contains a RDF graph in Turtle format, and then templates can query the quads (or take all pages of a certain kind, such as all with a certain category or all user pages or all non-user pages or whatever, and query the triples when all those graphs are merged) in order to constr [05:40:57] Or else if there is some way to do it even if you cannot install an extension, to make some approximation of such too [05:42:56] For example, if I can make up a comparison table automatically by taking RDF data of them to automatically fill it in! Wikipedia has some comparison tables but not quite like that, I think? [05:42:58] I can't see any reason why triples of a page should be in a separate rdf graph [05:43:19] and I can see many reasons why not [05:43:54] You can merge all of them and treat it as a single graph if you prefer too though [05:44:09] (for example : you'd have the triple (chelsea_clinton,daughter,bill_clinton) two times : both in bill_clinton graph and in chelsea_clinton graph [05:44:14] ) [05:44:16] yeah [05:44:49] anyway [05:44:58] I think wikibase is what you want [05:46:24] OK, although one reason I was thinking to have it separate is that you can also have an implicit "default graph" which contains category links and namespace links and so on; therefore, you can query them separately if you want to. It also mean you can cause a query to ignore some data if an optional graph ID is specified [05:49:04] Although RDFIO requires configuration and is an extension, and also does all sorts of additional things I am not looking for; I do not know whether or not these additional things will interfere with my use cases though. [05:50:09] do you already have rdf graph you will use ? do you want the user to be able to edit the data ? [05:50:10] And it says it uses ARC2; I have seen ARC2 before and looks like it contains a bug where if you give the authority component but not the scheme, it assumes http:// instead of using the same scheme as the base URI (my own implementation in C uses the same scheme as the base URI in this case). [05:50:32] rom1504: I want to allow anyone who can edit the wiki can edit the data, whether I have data or not already. [05:50:38] if you want the user to be able to edit the data, I think wikibase is better [05:50:54] (And if the page is locked then it cannot be edited) [05:53:03] But I also want to do it on existing wikis without installing an extension; I have no access to install extensions. Some wikis may already have ParserFunctions and/or Lua installed too though [05:53:29] If I have to run software on my own computer to convert the data it can be OK though, but other people should still be allowed to edit it too [05:55:06] well you can always convert data on your computer yeah sure [05:55:16] but the user won't be editing the graph [05:55:39] I think it's unlikely you can let the user edit the graph without an extension [05:55:55] or even just getting information from a rdf graph without an extension [05:56:31] you can generate wiki source from a rdf graph using any programming language though [05:56:46] Actually you can still make a editable page without an extension, although I suppose it probably won't do anything by itself; if I post the software used then anyone who edits it can use the software too [05:57:38] rom1504: Is SQL a valid programming language to do it? I wrote a RDF Turtle parser in C which can be used as a SQLite extension [05:59:18] SPARQL is the language usually used to query rdf graph [05:59:41] yes you can make a page editable, but that won't change the graph. [06:01:13] I know, if you make a page editable it won't change the graph unless another program reads it. Also I know that SPARQL is the language usually used to query RDF graph, but I don't have any SPARQL implementation in my computer. [06:03:43] you can use a triplestore like virtuoso [06:05:34] Most implementations I find are very large or difficult to compile or something; that's mainly why I wrote my own implementation, but also because I want to use it with SQLite. [06:07:13] Most RDF parsers use a lot of files rather than just one [06:09:23] But the main issues is if I do not have access to install extensions anyways, then I have to do it by the other way [06:12:06] I can't quite see how you can to make a rdf graph fit in sqlite [06:12:23] *want [13:28:52] is there a way to mass-fill info about pics in Upload Wizard? [13:29:00] when uploading multiple of them [14:10:45] nvm about UploadWizard, founded toggles under first image [15:40:41] how do I delete history or a page permanently? [15:48:44] I have a large CSS file for adjusting Monobook (65 KB). Should I add it to the Resource Loader or include it as file using addStyle()? [17:03:49] Hi there! In an extension, is it possible to use skin-defined background colors in the extension css? I can see they're in a variables.less files in the active skin, but can't figure out how to import the active skin into an extension CSS without hardcoding... [17:09:56] Ruad: The body tags throws a skin class name, e.g. body.skin-monobook [18:04:28] Ruad: have you looked at https://www.mediawiki.org/wiki/Manual:$wgResourceModuleSkinStyles ? [18:04:38] Ruad: the extension defines a stylesheet and then the skin overrides it [18:59:29] Hey! [18:59:41] There's something really-really odd that's happening to me. [19:00:22] I have a MediaWiki website, and I am now trying to make it more mobile-friendly [19:00:43] So I installed the Mobile Frontend extension, and I'm just fine with that [19:01:25] Now: if I were to view the wiki from an iPad with Google Chrome, it loads the mobile view and everything is okay; but if I were to load it from Safari, the Desktop view shows up. [19:01:38] The thing is, it's using the Wikipedia logo. [19:01:39] WTF? [19:03:01] It uses a different logo [19:03:08] $wgMobileFrontendLogo [19:37:37] Reedy, I am talking of the Desktop view. [19:38:09] Now, as for the Mobile view, I provided $wgMobileFrontendLogo a path to the mobile version of my logo [19:38:16] But it ain't showing it up anywhere [20:46:53] now that we have tags in edit cation of the API, is there a similar way to indicate tags from the edit form (some shouldn't someone delete https://www.mediawiki.org/wiki/MediaWiki:I_hope_this_doesn%27t_work [21:35:50] We can't, solarquest [22:02:44] eranroz: The API has revision tagging now? [22:03:35] Fiona: In the API it is possible to assign edit to one of the [[Seacial:Tags]] [22:03:56] It is very useful to assign gadget edits with tags [22:04:21] Hmmm. [22:04:44] I wonder if you can tag edits as mobile. [22:04:51] Or as one of the AbuseFilter tags. [22:05:42] mobile is already tagged I think using the extension [22:09:31] Right. But it sounds like you're saying arbitrary tagging is allowed? [22:09:36] That sounds easy to abuse. [22:18:48] except for tags defined by extensions [22:19:09] but yes it can be abused... [22:19:14] Krenair: why couldn't you delete that? i could [22:19:34] I don't know [22:19:39] it didn't give me the toolbar option [22:19:54] oh, i url-edited in action=delete [22:20:00] i wonder where the discussion itself went though... [22:20:57] probably just poor mw integration, lacking a deletion option in the ui.. [22:30:09] The worse MediaWiki integration seems to be allowing pages to be created in that namespace. [22:30:12] Hopefully that got fixed... [22:32:51] haha [22:33:01] that was me [22:34:48] https://phabricator.wikimedia.org/T62218 [22:40:02] "Sir Arthur Conan Doyle yields no record but Arthur Conan Doyle does. You've got to be kidding me. Something smells in Denmark and I don't think it's the cheese. You guys and gals need to get busy and fix this. I love Wikipedia but exact input by the exact name, place or thing is completely outmoded. Get your shit together." [22:40:13] People have been extra snooty in their email to us lately. [22:40:59] what the hell [22:41:09] what an asshole [22:41:13] RD: is that from info-en ? [22:44:10] Is he saying a search for "Sir Arthur Conan Doyle" doesn't have the article in the search results? [22:44:22] That would be a pretty serious bug. [22:48:53] yes [22:48:56] Krenair [22:49:35] ori, that's really not the worst I've seen from info-en@. seriously. [22:50:13] Yes, somebody else was very upset at Fiona because his script broke. [22:50:15] :P [22:50:30] It's your script too. [22:50:33] SHARED OWNERSHIP. [22:50:45] According to the license it's not mine. [22:50:57] The public domain belongs to everyone. [22:50:57] According to the "license," it belongs to all of us. [22:51:01] It's in the public domain! [22:51:36] https://github.com/mzmcbride/daily-article/graphs/contributors [22:51:40] NEES MORE PEAKS [22:51:43] NEEDS* [22:51:56] Who's mzmcbride? [22:53:50] On that note, [22:54:29] If anybody here likes to mess around with python - there's a script that operates the daily-article-l mailing list, which 35,000+ people are subscribed to..but it needs updating. [22:54:32] See https://github.com/mzmcbride/daily-article/blob/master/dal.py for the source [22:54:39] Poke Fiona if interested. :p [22:55:00] step 1) create task [22:55:05] step 2) assign to kunal [22:55:09] step 3) ??? [22:55:16] Heh. [22:57:06] who? [23:01:11] communism = freedom [23:01:25] so? [23:02:02] just sayin' [23:02:41] RD: what's wrong with it? [23:03:34] It's a hack and is extremely dependent on everything on each wiki to be exactly the same every day. [23:03:40] Fiona can elaborate, perhaps. [23:04:04] ori: https://meta.wikimedia.org/wiki/User_talk:Rjd0060#daily-article-l_delivery_failed_.28March_29.2C_2015.29 been having issues lately [23:05:48] ori: If you read the script, it's pretty easy to see the fragility. [23:16:45] Fiona: why doesn't it log exception details? [23:17:44] ori: ikr [23:18:05] I said if it failed again, I'd change the wiki talk page message to be more useful. [23:18:09] And include a traceback. [23:20:42] Or we could do it now [23:21:42] RD: Sure, go ahead. [23:22:15] That's all you. Or whoever knows anything about python. [23:28:02] Fiona: https://github.com/mzmcbride/daily-article/pull/1