[00:10:30] hmm here's an interesting problem. [00:10:50] we had a request to remove someone's real name that was included w/o permission and managed to hide it everywhere EXCEPT here. [00:10:51] http://doomwiki.org/w/index.php?title=Arjang_Baroumand_%28Dark_Exodus%29 [00:11:01] even logged out, you can still see the old article name. [00:11:14] is there any way to make that not appear there? [00:16:08] Quasar``: there is a maintenance script. https://www.mediawiki.org/wiki/Manual%3aDeleteArchivedRevisions.php that might help [00:33:17] hi [00:33:21] wow there's a lot of people here [00:33:24] but uh [00:33:25] hi [00:33:29] sup [00:33:49] I was wondering if there's any mediawiki 3d model viewers [00:33:51] something like this [00:33:52] opened chrome://net-internals/#events&q=type:SPDY_SESSION%20is:active [00:33:54] uh [00:33:55] oops [00:33:56] wrong link [00:35:01] I have an old mediawiki database dump, whats the best way to restore it to view as a wiki? [00:35:11] eh doesn't matter [00:35:14] not your thing I mean [00:35:16] my thing [00:36:14] anyways [00:36:24] is there any way to view 3d models on mediawiki? [00:36:28] any extensions? [00:36:56] because on TCRF (https://tcrf.net/The_Cutting_Room_Floor) there are a TON of models being rendered as GIFs [00:37:01] and it takes up a ton of bandwith [00:37:19] but mostly 3d model viewers are more fun :) [00:37:19] thanks however it seems log suppression via oversight lets me do it [00:37:41] we just didn't have our oversight group configured properly because the documentation used to reference a 'suppress' group instead of 'oversight' [00:38:08] glad you fixed it Quasar`` :) [00:38:13] :> [00:44:56] http://www.x3dom.org/ [00:45:09] wonder if that could be made into a mediawiki extension [00:45:15] it claims to be easy to use [01:00:51] A database query error has occurred. This may indicate a bug in the software. [01:00:59] I get that error when importing my database [02:11:28] can anyone here answer a quick question about $wgSMTP? I'm trying to configure email I have through my domain registar. According to the registar, information appears correct, but I'm not receiving any of the email. The registar said I should contact my web host, but I'm assuming it's not my hosts fault (I've had email working through gmail SMTP, but [02:11:28] recently reinstalled mediawiki) [02:12:58] for the record using mediawiki 1.27alpha, $wgSMTP with assumingly correct information, and I have php-pear and mail and net_smtp .. trying to think of anything else [02:13:57] oh. the page is showing a different error now. "Mailer returned: Failed to connect to mail.privateemail.com:465 [SMTP: Invalid response code received from server (code: -1, response: )]" hmm [02:22:56] Ignore all that -.- dumb domain support.. was using the wrong port and told it was the correct one.. mails working fine <3 [07:44:34] Is it possible to connect my Template namespace to git? AKA use github to manage a namespace's pages? [07:48:01] voldial: no, what I was doing for some JavaScripts (connected via MediaWiki: namespace or as a user script, but that does not matter): I have a download and upload scripts that synced my local copy with the wiki. They might be plugged in as git hooks, mayme. [07:48:05] maybe [07:48:59] saper: that's the hack I was thinking of... where did you hook MW to get the content for a namespaces pages from flat files? [07:50:02] er, if I understand you have some glue scripts to (maybe use the bot? that's what I do) to edit pages? [07:50:57] I could take my current bot controller's and tell them to replicate some git repo (page per file)... [07:51:24] but that would be cron [07:51:39] not that I have anything agains cron [07:57:47] voldial: https://bitbucket.org/plwiki/bot/src/d9c8a55117a78315fc4c66cd274f972a936ee07f/plwiki/sk/?at=default here's the code [07:57:53] getpage/putpage is just it [07:58:21] it is invoked from the commandline instead of "git fetch" or "git push" [07:58:36] (we actually use hg in this repo but that doesn't matter, either) [07:59:37] neat. it's going to take me a minute to get what this does... [07:59:47] 15 seconds [08:00:06] check script.pt [08:00:07] check script.py [08:00:30] and yes, I am using pywikipediabot to do this [08:01:03] so all login/API stuff/whatever is specific to MediaWiki is done by the pywikibot library [08:01:51] ummmm. my setup uses the bot the same way.... reading.. [08:05:07] you are calling script.py from mw's php? [08:07:44] wow... what... is this... there's almost 300 files in the bot repo [08:09:14] hello all [08:09:42] found http://link.springer.com/chapter/10.1007%2F978-3-642-04441-0_30http://link.springer.com/chapter/10.1007%2F978-3-642-04441-0_30 [08:09:56] well /2 [08:28:57] voldial: I am not calling script.py from mw, only from the command line - I do "get" before I want to add changes, then I commit it if there were any changes on-wiki, then I do my stuff and then I "put" [08:31:20] somebody can tell me how can i purge my database from bot entries? [08:32:49] the plugins do not help me at all, the db is still full of useless entries [08:35:06] saper: very cool. thank you. I'll give it a try. If the gir repos you pull from are public I would like to see me. [08:35:12] me/em [08:35:25] uug. gir/git [08:35:45] voldial: didn't understand ... [08:36:15] voldial: you can join bitbucket and contribute to our repo if you like [08:36:17] if I get what you said, you call your custom get() to pull changes that were made via MW's interface? [08:36:48] yes [08:37:04] err.. getpage() [08:37:06] it's poor man's interface [08:37:50] so you iterate through every page you are interested in keeping synced with a git repo, first calling getpage() on each? [08:37:55] yes [08:38:06] those were one-page scripts, independet [08:38:08] independent [08:38:14] so that's what I did [08:38:29] then you make your changes... to some dir structure that getpage() writes to which also has a .git dir? [08:38:35] one could possibly write a git-fetch backed "mw:" to do this magically [08:39:22] see the source - there is no magic, no support for version control. It's all done by hand. [08:39:28] where's the code that writes the results of getpage() to the folder structure? [08:39:48] it uses the current directory [08:39:59] (sorry, really I will read this, it's fascianting, I dont meant o bug you for instant answers) [08:40:01] https://bitbucket.org/plwiki/bot/src/d9c8a55117a78315fc4c66cd274f972a936ee07f/plwiki/sk/script.py?at=default#script.py-10 [08:40:22] it's dead simple [08:41:07] except that has no open() [08:41:24] really I sould read more and not ask for a ELI5 [08:42:25] hm. page.put() looks like a open [08:42:30] line 10 [08:42:35] and line 15 [08:43:38] jeesh. ok, I see them now [08:44:15] what calls script.py? [08:45:02] well, it has no sys.argv[] so... I'll just search for imports of script [08:46:55] I don't remember, maybe I was editing it on the fly [08:49:04] it's a great start, much appreciated [08:50:02] have fun [08:51:09] took me a min to see that sk.py defines SCRIPTS. that's kinda what I was missing. so I'll automate that bit to use a namespace and generate it on the fly. I'll send a patch if I get it working nice. [08:51:35] can u please have a private chat? thank you.... [08:53:03] nonse: no, we can't [08:53:16] it's on-topic [08:53:59] nonse: to answer your question you basically cannot. You can manually remove entries from the revisions table but this will more often than not break things. [08:54:10] you can try revision delete or similar tool [08:56:24] * voldial throws charged confetti [09:01:58] despite the risk, can i manually delete unwanted entries in the database tables? [09:02:15] let's say i want to delete bot entries that create new pages [09:03:28] i just want know if is possible to do [09:03:35] the risks are mine [09:05:32] Generally MediaWiki's code will auto-figure-stuff-out if there is at least one revision in the table. No revision = Broken page. [09:06:21] When I saw generally I usually only deleted one spam revision out of multiple containing tons of illegal content. [09:08:15] I also make sure to update any other revisions that referenced the deleted revision. [09:15:59] fkwaexktcxexternallinks [09:16:08] can i empty that table? [09:16:38] we have no links on external sites other the one to the main site [09:17:22] the contents of that table is links on the wiki to external sites [09:17:29] not external sites linking to your wiki [09:17:35] in fact [09:18:02] my question was if i can delete them since we have only a link to the official site [09:18:24] so, can i delete the other ones? [09:18:50] they were added by bots [09:53:36] Hi to all. [09:54:08] First, I apologies in advance, just in case, for any mistake I would be commiting: the last time I used IRC was ages ago ^_^U. [09:54:17] Well, I don't know if this is the appropiate place to ask about this (if not, can you kindly point me to the appropiate one?), but I have a doubt about the Wikimedia API. [09:54:32] I'm trying to do an exercise from FreeCodeCamp (this: http://freecodecamp.com/challenges/zipline-build-a-wikipedia-viewer) and I'm using (for now) this query (link to the API Sandbox): https://en.wikipedia.org/wiki/Special:ApiSandbox#action=query&list=search&format=json&srsearch=lepidoptera&srnamespace=0&srprop=snippet|titlesnippet&generator=search&gsrsearch=lepidopteraI'm trying to do an exercise from [09:54:34] FreeCodeCamp (this: http://freecodecamp.com/challenges/zipline-build-a-wikipedia-viewer) and I'm using (for now) this query (link to the API Sandbox): https://en.wikipedia.org/wiki/Special:ApiSandbox#action=query&list=search&format=json&srsearch=lepidoptera&srnamespace=0&srprop=snippet|titlesnippet&generator=search&gsrsearch=lepidoptera [09:55:05] (Damn middle mouse click) [09:55:43] Well... My doubt is: what thing refers, exactly, the «index» parameter in «pages» objects? At first instance, I believed it was the index of the same page in the «search» array of objects. That seems logic. But, when using that, sometimes certain indexes simply don't match (they pair «pageid» from one page with «titlesnippet», «snippet», etc from another page). So, I don't know if that [09:55:45] «index» parameter is, in fact, about some other thing or if, instead, there is a bug/problem with the API in this matter. And, of course, I know that array index order starts at 0. [09:55:56] If anybody can enlighten me about this, I will be very grateful :). Thanks in advance for reading me :). [09:58:22] And, of course, I searched the wiki pages, and bla bla bla. I couldn't locate a place where this is solved. [10:57:03] hello, bit of a basic wording question, sorry. If a template is included on a wikipage page, and my bot edits a parameter of that template within the page, whats the short way of saying that? Would it be editing the template? Editing the include? Editing the parameter of the template? [10:57:47] "Editing the template" seems to indicate changing the actual Template, so doesn't fit for me [13:34:07] soulchainer: I don't want to be rude, still I'm afraid your posting above might fall into the "too long didn't read" category for some folks :-/ [13:34:30] You might be more successful by saying what you do in one or two sentences, and another sentence about your doubts [13:34:48] and thanks for making me aware of http://freecodecamp.com/challenges/zipline-build-a-wikipedia-viewer - I hadn't heard about that before! [13:47:14] soulchainer: Can you explain which "index parameter" are you talking about? [13:52:00] (I was rephrasing it for making it shorter, after what @andre__ said). Well; I refer... to the «index» properties of objects in «pages» object, inside «query» property. Is about the JSON that request I mentioned returns. [13:53:19] soulchainer: Got it. I dunno either. There seems to be no documentation about it. :( [13:54:03] Yeah :(. [13:54:31] Because of that I asked. I already search in the docs and found nothing >_<. [13:55:03] Anyways, thanks, @Niharika [13:55:25] Ummm... I think I don't remember hoy mention someone in IRC xD. [13:55:29] soulchainer: Unfortunate. Do you know you can use "formatversion=2" in the query to get it in a better format? [13:55:30] *how [13:55:45] soulchainer: Just the nickname, no need for @. [13:55:54] Oh. OK. Thanks :D. [13:56:45] soulchainer: For example https://en.wikipedia.org/w/api.php?action=query&titles=Barack%20Obama&prop=revisions&rvprop=ids%7Ctimestamp&format=jsonfm vs https://en.wikipedia.org/w/api.php?action=query&titles=Barack%20Obama&prop=revisions&rvprop=ids%7Ctimestamp&format=jsonfm&formatversion=2 [13:57:13] Oh. Yeah, yeah. Don't worry, I was already trying it :þ. [13:57:18] Oh good. [13:57:46] Sure that format it's better :D. More clean and less nested :D. [13:58:08] Thanks again ^^. [13:58:42] And thanks, andre__. Sure you're right: I know I tend to write too much ^_^U. [14:00:04] :) [14:02:56] Well, I also sent this to the mail list and... just get a response... And... It doesn't clarify nothing to me xD. [14:03:01] Here: [14:04:16] «You may notice that submitting the same list=search query will sometimes give different results, i.e. the search engine ordering of results isn't completely deterministic. When you see such a mismatch between the indexes in the pages produces by generator=search and the list produced by list=search, the same thing happened between the two searches.» [14:05:01] O_O. [14:17:49] Well... I still can achieve what I want, just looking at it in other way :þ. And using an extra for loop (that's what I was trying to avoid). Well, thanks again, Niharika. You really helped me. See ya! :D. [14:52:36] :) [14:53:59] And now if somebody could help me. :P I have two revisions with me (in JS), and I want to generate the diff in JS and preview it, just like when viewing a usual diff but without making an API call. How can I do that? [14:54:30] Without doing an API call? [14:54:36] Unless you invent some JS diff library... [14:55:03] Reedy: Ah, so there's no magical method to do that then. [14:55:11] I was hoping for one. [15:33:35] hi all, does anybody know how you can have images in different sizes as we have in wikipedia ? [15:34:09] images where? [15:35:59] shared hosting pains http://www.imagemagick.org/discourse-server/viewtopic.php?t=15534 [15:36:30] Reedy: in the gallery [15:36:31] shirish: thumbnails? perhaps you're looking for https://www.mediawiki.org/wiki/Manual:Thumb.php [15:36:38] which gallery O_o [15:36:50] let me give an example. [15:38:08] damn, ok it seems wikipedia has moved on to using media viewer [15:38:33] I am using this wiki https://wiki.hamaralinux.org/wiki/index.php/File:Bugzilla.png [15:38:57] as can be seen the other possible combos are not generated, just the one which I uploaded. [15:42:13] Reedy: Nemo_bis ^ [15:46:52] (Sorry this is more servery than MediaWiki but it's related): Hi :) Recently happy I got some cool stuff working on my wiki. Could someone maybe help me with the it logbot? dpkg -I package says it's missing dependencies even tho I supposedly just installed python-irclib and mw-client with no errors [15:48:17] shirish: see https://www.mediawiki.org/wiki/Manual:Thumb.php [15:49:17] Nemo_bis: wouldn't it be this rather ? https://www.mediawiki.org/wiki/Manual:$wgUseImageResize [15:57:24] maybe, butunlikely [16:14:35] I am getting "connect to host gerrit.wikimedia.org port 29418: Network is unreachable" , when trying to ssh to @gerrit.wikimedia.org according to https://www.mediawiki.org/wiki/Gerrit/Tutorial#Add_SSH_key_to_use_with_Git. [16:16:12] Are you behind a proxy? [16:16:24] yes [16:16:35] Well that explains it [16:16:40] You'll have to clone using HTTPS [16:17:30] Thanks @Reddy, but how to do that? [16:21:31] I am trying to create a template where the first parameter is part of an if statement and based on the value supplied for that parameter one of three images will appear. The template is https://wiki.gamepaduniverse.com/wiki/Template:Module [16:21:52] I don't understand how to construct the statement [16:22:13] {{#if:{{{1}}}=Active[[File:Active State Arrow.png|center]]
}} [16:22:20] I don't know if that is anything [16:22:42] sneezelextremepl: you'll want to use #ifeq: or #switch: [16:22:55] sneezelextremepl: with #if:, you can only check if a parameter is empty or not [16:23:00] !parserfunctions [16:23:00] "Parser functions" are a way to extend the wiki syntax. ParserFunctions is an extension that provides the basic set of parser functions (you have to install it separately!). For help using parser functions, please see . For details about the extension, see . [16:23:16] hmm, sorry, wrong links. [16:23:39] sneezelextremepl: https://www.mediawiki.org/wiki/Help:Extension:ParserFunctions this is a pretty good help page, i think [16:24:22] cool, i was on the template help page and it did not supply that information, been so long since i installed parser functions extension that I forget about it [16:24:49] sick [16:25:14] and i can nest statements it looks like with #ifeq as the third and fourth parameter [16:25:48] yeah [16:26:17] I kind of have question about this whole IRC channel [16:26:22] do most people just do this for fun [16:26:29] or do they also get compensated by mediawiki [16:26:34] the frequent helpers on here [16:27:28] Or are most people just lone wolfs, mostly likely older, and retired or at least don't need to work a traditional job anymore? [16:34:35] MatmaRex:if you don't mind me asking, where do you fit into the Mediawiki community? [16:35:42] sneezelextremepl: some do it for fun [16:35:46] some people get paid, some do it for fun [16:35:55] Some people get paid, and do it for fun anyway [16:36:08] it's mixed. and about motivations in free software projects there are many research papers :) [16:36:11] but the people on here right now... [16:36:12] lots of different reasons [16:36:32] sneezelextremepl: i'm a mediawiki maintainer (i was a uni student when i got interested), working for the wikimedia foundation now (and still poking MW in my free time) [16:36:41] Most have some interest in mediawiki in some way or another [16:37:04] ...stating the obvious aren't we...we are on the mediawiki IRC [16:37:35] what is the average age? [16:37:43] 40+? [16:37:46] Probably not [16:37:48] That's probably a hard one to answer [16:37:54] Same if you asked M:F ratio [16:38:13] Using ranges always helps [16:38:36] Well, we have numerous that are 18 or younger [16:38:47] And we've got at least one person I know that is over 50 [16:39:24] We've had people under 16 developing mediawiki, who are still involved, but obviously, are now older :P [16:39:34] Reedy: I am really in love with Mediawiki and would like to get involved in some capacity where I can get paid [16:39:38] How do I go about that [16:39:42] sneezelextremepl: many are very experienced contributors here, many general contributors and many new student contributors who have recieved support from this community in learning and who now contribute back! [16:40:03] sneezelextremepl: Well, the WMF do hire people to "develop mediawiki" [16:40:15] There's a few third party companies that do mediawiki support, extension development etc [16:40:26] There are other aspects of managing it besides development of the software, no? [16:40:32] sneezelextremepl: starter https://www.mediawiki.org/wiki/How_to_contribute :) [16:40:37] For the WMF part, https://wikimediafoundation.org/wiki/Work_with_us [16:40:45] Well, the question is where is the money in free software? [16:40:55] sneezelextremepl: if you're in high school, we participate in google code-in. if you're in university, we participate in google summer of code and gnome outreachy. if you're a developer, there are positions opens ;) [16:41:16] I am not a developer really. I am a business person. [16:41:19] Getting experience and involved as a volunteer is a good stead to being employed [16:41:25] Isn't there adminstrative work that needs to be done? [16:41:33] In what sense? [16:41:40] bookkeeping [16:41:49] Finance? [16:41:49] normal business functions [16:41:56] MediaWiki isn't a company [16:42:03] It's a project supported by a charity [16:42:04] that every organization has [16:42:05] lol [16:42:10] of course it is a business [16:42:18] people are much to strict with that term [16:42:32] it is registered as a business i am sure [16:42:33] Most of that sort of function would lie with Foundation staff members [16:42:40] No it's not registered as a business [16:43:27] that can't possibly true. Where any amount of money is involved an organization must be registered with the govt [16:44:02] so the money that you make doesn't get taxed? [16:44:09] non-profits are businesses [16:44:11] The foundation is registered as a charity [16:44:42] k, a charity is still a form of business tho [16:45:03] Right [16:45:09] But MediaWiki isn't a legal entity [16:45:33] interesting [16:45:41] so where is the accountability? [16:45:47] For what? [16:46:00] what gets produced by this "charity" [16:46:03] who is in charge [16:46:08] who is punishable [16:46:24] Punishable for what like? [16:46:34] Have you checked https://wikimediafoundation.org/ ? [16:47:00] say someone wrote malicious code and you made an update available that crashed every person's server that updated it [16:47:09] MediaWiki is open source software. The Wikimedia Foundation is generally considered as the prime maintainer of MediaWiki, and pays developers to work on it to scratch it's own itches [16:47:11] sneezelextremepl: Tough? [16:47:27] MediaWiki is a software under the GPL license. That license covers accountability to some extend - read it :) [16:47:43] https://github.com/wikimedia/mediawiki/blob/master/COPYING#L84-L89 [16:47:47] but in actual practice... [16:47:48] sneezelextremepl: that's why all software comes with a disclaimer [16:48:08] sneezelextremepl: In actual practice, it would break Wikipedia etc before it broke other sites [16:48:08] sneezelextremepl: if you want software with full liability, you'll have to pay someone a lot of money [16:48:33] but yea, what Reedy said. The code runs on wikipedia first. [16:48:40] Well, I lie [16:48:45] It'll probably break translatewiki first [16:48:46] xD [16:48:50] hehe [16:49:24] To me, that is an internal process thing that people choose to follow, not an actual legal what-if thing that could happen [16:49:27] but i get your point [16:50:29] sneezelextremepl: the internal process is "second pair of eyes", and "trust via reputation". [16:50:49] ...and trust is just a belief [16:51:12] sneezelextremepl: pretty much anything is just a belief, if you look closely [16:51:20] exactly my point [16:51:24] money is just a beleif. [16:51:59] contingency plannign [16:52:19] that is what the legal angle i am discussing [16:52:33] who should plan for which contingency? [16:52:52] that's my point [16:52:55] accountability [16:53:26] what is the deal with http://www.mediawikiwidgets.org/ [16:53:32] i know its not apart of the foundation [16:53:37] but its been down for days [16:53:45] never heard of it [16:53:46] i don't even know what that is [16:54:03] are you familar with the Widgets extension? [16:54:10] vaguely [16:54:18] I think it's relatively popular [16:54:28] as in "it exists" and "it has widgets". that's about it [16:54:32] basically the site hosts all these code examples that make it work [16:54:46] afaik, it's not an extension used or maintained by wikimedia [16:54:51] allow for copy and paste for outside services [16:55:02] nice [16:55:25] it seems pretty widely used outside of Mediawiki [16:55:26] that is all [16:55:32] The whois contains a lot of google [16:55:35] and I think Yaron was in charge of it [16:56:21] according to https://www.mediawiki.org/wiki/Extension:Widgets is is one of the authors of the extension, at least [16:56:39] I actually know he is solely in charge of it now actually [16:56:55] I had him update something there not too long back [16:57:21] Where is that dude? [16:58:19] Yaron is on IRC from time to time [16:58:27] recently? [16:58:36] !seen Yaron [16:58:36] Did you mean @seen Yaron? [16:58:36] I sent him an email about this but have not heard back [16:58:37] Yaron (~Yaron@cpe-74-64-108-63.nyc.res.rr.com) was last seen quitting from #mediawiki 14 hours, 4 minutes ago stating (Quit: Leaving.). [16:58:43] cool [16:58:46] thanks for that [17:00:18] I think what you folks do is great [17:00:23] keep doing what you do [17:01:34] thanks :) [17:04:39] Good ole' Sneezel over here doesn't have much going on in his life [17:04:56] so it's nice to see some decent people out there helping when and where they can [17:05:33] Ok wikifellas [17:05:49] I don't understand how I can include a wikitable as apart of a template parameter [17:05:58] seems fairly difficult [17:06:03] https://wiki.gamepaduniverse.com/wiki/Wikipedia_Step_%E2%88%9E_Module [17:06:16] on this page I wan the arrow to come after the table [17:06:48] the table isn't presently apart of the template:module [17:07:00] but I would like to be able to copy and paste it in as a parameter [17:11:43] something with the pipe icons? [17:12:17] what is a direct substitute for a pipe symbol in wikitext? [17:18:39] any suggestions? [17:26:10] sneezelextremepl: {{!}} [17:26:40] sneezelextremepl: (assuming that you're on a fairly new MW, where this is built-in, or created Template:! on your own wiki) [17:27:13] sneezelextremepl: with tables inside parser functions, it's usually easier to use the HTML syntax.

…

[17:27:35] yes unfortunately I have already made dozens of these tables [17:27:49] and think it would be much more effort at this point going back and putting it into html format [17:28:01] than replacing the indivdual pipe symbols [17:44:21] MatmaRex:Thanks for the help. Doesn't seem to be working for me. Created template:! with value | [17:44:43] changed all the pipe symbols [17:44:49] now the table is not showing at all [17:45:03] https://wiki.gamepaduniverse.com/wiki/Template:! [17:45:07] https://wiki.gamepaduniverse.com/wiki/Wikipedia_Step_%E2%88%9E_Module [17:45:58] are you using MediaWiki tables instead of HTML? If you are, are you using | where you should be using !? (Templates require both iirc, in different places) [17:47:37] PuppyKun: I had the wikitable display correctly beforehand, but then I wanted to add it as a parameter to a template [17:47:59] so i changed all the pipe symbols to {{!}} [17:48:04] Oh so you have a table inside of a template? [17:48:10] correcy [17:48:12] correct [17:48:28] Jw why the template is {{!}} instead of {{|}} or will {{|}} not work? Been a while xD [17:48:48] the name of the template shouldn't matter, right? [17:48:56] the value contained in it does, right? [17:50:36] Yes true enough :p [17:51:12] Tbh idk atm. Not gonna open safari on phone bc battery rn xD I'll look later. Dunno why it no worky [17:51:47] i think it awesome that you are using an iphone to communicate with me right now [17:52:11] not to be pushy, but what time is later? [17:52:29] * PuppyKun isn't gonna ask about the iPhone thing [17:52:55] Later will be in a little less than an how if you're on. If not and your registered I can use memoserv. [17:53:00] I'm out right now [17:53:28] ill be on then [18:54:03] * sneezelextremepl slaps PuppyKun around a bit with a large fishbot [19:13:53] hi all, i'm having a localization-related issue after upgrading one of my wikis to 1.25, where strings like "" that should be replaced with localized text like "Version" are not being replaced [19:14:25] Do you have anything in your config for localisationcache? [19:14:28] i upgraded two english wikis just fine and then used rsync to copy one of them from dev to stage, making sure not to change LocalSettings.php [19:15:14] no, and my LocalSettings.php files are identical between dev and stage except for the things that should differ, e.g. IPs, URLs, etc. [19:16:34] once i got the dev english wiki working, i used rsync to copy the whole directory to the stage server, ensuring not to overwrite LocalSettings.php, so everything should be identical to dev [19:17:04] i also deleted the cache/*.cdb files and restarted apache and varnish [19:17:04] I'm trying to remember what MW uses by default as l10n cache [19:17:45] also enabling the mediawiki debug log hasn't provided any insights either [19:18:56] https://www.mediawiki.org/wiki/Localisation#Caching [19:18:57] Reedy: sql queries? [19:19:06] Yeah, that's what I was thinking [19:19:10] In which case, rsync is useless [19:20:28] it's worked in the past from stage to live, so i'm just trying to streamline my entire procedure to also be able to rsync dev to stage [19:20:40] i'm looking at that page you sent [19:21:12] You can run the rebuildLocalisationCache maintenance script to make sure all the caches are built [19:22:59] I did that too after first just deleting all of the .cdb files in the cache directory and letting them (well, l10_cache-en.cdb) get recreated [19:24:05] i think the cache table is okay, e.g. select * from l10n_cache where lc_lang = 'en' and lc_key like '%version-license%' returns ('en', 'messages:version-license', 's:7:"License";') [19:24:25] It should only be using one method [19:24:49] fwiw, $wgCacheDirectory is set to $IP/cache, same as on dev and live, not even been changed [19:25:10] i was just verifying based on what you sent me :) [19:26:05] so could something cause wfMessage() to return the base key from the l10n file rather than a translated string? I did verify that $wgLanguageCode is still set properly [19:26:47] or could the localization code simply be being bypassed for some reason? [19:27:59] i also re-ran php ./maintenance/update.php just to be sure there weren't any issues there [19:29:10] Like I say, it should be using the database, or the files [19:29:30] I'm guessing you have the cache directory set? [19:29:42] yes, it's never been changed from $IP/cache [19:32:51] l10n is a mess when it fails... I've seen several reports of it failing on some installations, and I think I've never found a reason, or I don't remember... [19:34:39] everything seems to be identical in my stage environment to the working dev environment, as far as i can tell, so i'm trying to figure out where to look for discrepancies that i might not have thought of already [19:35:45] again, this was an upgrade from 1.24.2 to 1.25.2, so there were LocalSettings.php changes required (mostly to how some extensions were loaded) [19:36:29] I've not changed my dev wiki and how it loads extensions [19:36:35] I guess permissions of $IP/cache are okay, since cdb files are created... [19:37:03] yeah, it does indeed recreate the en cache file, perms are 755 owned by www-data:www-data [19:45:24]