[00:00:06] this applies when updating mediawiki, and when installing or updating some extensions [00:00:26] i'm about to go offline. there's no good documentation for this i'm afraid... [00:01:05] but now you know what to ask about, here oder on the mediawiki mailing list. [00:01:12] bbl [00:02:00] oh ok [00:02:02] I get an error [00:02:11] that says i havent defineded my server name [00:02:15] in my local settings [00:02:27] and it refers to a line that is part of the dispatcher [00:02:38] DanielK_WMDE: switch ( $_SERVER["SERVER_NAME"] ) { [00:36:32] Hello! I am possibly interested in starting an independent Wiktionary-like project for a Southeast Asian language. I have a lot of lexical data already collected, but I'm unsure if I'll be able to mass-import it into MediaWiki. Is anyone available to chat with me about this for a few minutes? [00:37:42] FKLinguista: I have a question right off the bat: Why not start a Wiktionary, instead of a Wikitionary-like project? :) [00:38:45] marktraceur: That is definitely possibility :) [00:39:03] Cool! [00:39:14] FKLinguista: So you need to import lexical data, how is it stored right now? [00:39:19] Do you mean to say, add my data to wiktionary.org or still create my own wiki? [00:39:39] Right now I have it in Excel format. It is transcribed in IPA and contains tone bars. [00:40:25] I was able to export it to CSV no problem and import it into MySQL for a test. The problem with MySQL when I tried testing the data with a PHP array script, is that the Unicode rendered as '???' [00:40:45] so instead of seeing ə for example, one would see '?' [00:43:07] Hrm [00:43:17] Here is a sample of the data: https://sizangkam.files.wordpress.com/2015/01/luce_tables_test.pdf [00:43:33] So I would say, if you can get it into CSV, you can use that to import [00:43:41] Just need to write a quick bot [00:43:48] So, the problem is, I already have the table set up in one format, but I'm unsure if MediaWiki will be able to use what I have to create new pages. [00:44:02] http://wiki.supaham.com/Template:Table i tried copying mediawiki but it looks weird [00:44:42] marktraceur: Ah, I guess I'll have to learn how to write bots, [00:45:05] It's not hard usually :) [00:46:20] this is my website: http://sizangkam.wordpress.com/ [00:46:52] Sizang is a beautiful language, but there isn't a lot of current research being done on it. [00:47:16] That's why I'd like to compile a lexicon from old sources and hopefully encourage community members to add to it, as well. [00:47:40] That's why I figured a wiki-based dictionary would be an accessible solution. [00:47:57] Let me look for existing wikis in that language [00:48:20] Huh [00:48:23] There are none. :) [00:49:24] marktraceur: this is the closest you'll find [00:49:24] http://en.wikibooks.org/wiki/Bibliography_of_Kuki-Chin_Studies [00:49:29] Yeah, I thought I might get lucky [00:49:31] Oh cool [00:49:33] It's a start! [00:49:52] A colleague wrote that. [00:50:03] FKLinguista: So I don't know if you've seen https://incubator.wikimedia.org/wiki/Incubator:Main_Page [00:50:08] But you could suggest a new wiki there [00:51:08] Wow, that would be helpful. My goal is to have this lexicon up long after I'm dead, haha. [00:51:27] hi [00:51:31] Hello [00:52:27] FKLinguista: The article on Wikipedia says they speak a "dialect", do you know what it's a dialect of? [00:52:35] i have a problem with localurl as the wiki i work on always adds an index.php?title= to its URLs, does anyone have a solution? [00:52:37] Like, my people speak a dialect of English [00:53:00] Sizang? It's not a "dialect" of anything :S [00:53:05] It's a separate language. [00:53:06] !shorturl | Eoro [00:53:06] Eoro: To create simple URLs (such as the /wiki/PAGENAME style URLs on Wikimedia sites), follow the instructions at or try the new beta tool at . There are instructions for most different webserver setups. If you have problems getting the rewrite rules to work, see !rewriteproblem [00:53:23] FKLinguista: Huh, interesting. [00:53:30] do i need to be wiki admin for that? [00:53:35] I have to go for a bit, but talk to you later [00:53:41] Eoro: You need to be basically running the wiki. [00:53:54] okay, so i have to ask my boss [00:54:19] marktraceur: I mean, it is debatable what makes a language a language vs. a dialect [00:54:28] True [00:54:51] I agree with the theory that says Sizang is part of the Northern-Kuki-Chin family, as is Tedim and Thadou. [00:57:01] Just as one wouldn't say "Spanish and Portuguese are dialects of Italian" [01:00:38] marktraceur: Not to be a bother, but could you show me an example of a bot that would do something similar to what I'm looking for? [01:09:18] FKLinguista: an import? [01:09:51] Usually https://www.mediawiki.org/wiki/Manual:Pywikibot/pagefromfile.py is used [01:10:28] If you need examples, here are a few dozens https://stats.wikimedia.org/wiktionary/EN/BotActivityMatrixCreates.htm [01:12:13] If https://sizangkam.files.wordpress.com/2015/01/luce_tables_test.pdf is all you have, you should just import the stuff on en.wiktionary.org [01:12:37] Up to few thousands entries, they will pèrobably not even notice it ;) [01:12:48] If you want to import 100k entries better ask [01:13:51] Nemo_bis oh, okay. [01:14:09] What I wanted to do was make the lexicon searchable from both English and Sizang [01:14:18] and also use more lexical sources than just that one file. [01:14:45] So, for example, the entry for "Boat" may have 4 different sources, depending on who collected the data and how they transcribed it [01:15:49] Nemo_bis: I've added a lot of data since what's in that PDF, but I haven't published it yet. [01:15:56] There is no problem searching either way [01:15:59] I was thinking I could make 1 page per-word. [01:16:19] Where "word" is the English or the Sizang word? [01:16:48] Either way it's not really relevant, just concat the info you have for each Sizang word and import on wiki [01:17:42] But for details on templates to use etc. you should really use #wiktionary instead (or their "grease pit") [01:18:29] Well, I came here first, because I was figuring I should start with my own wiki, but if Wiktionary would accept it, I'd have no problem importing there. [01:18:43] Okay. Since the data is currently in Excel format, and it looks like pagefromfile.py requires it to be .txt, would I have to re-write my DB as a .txt? [01:19:36] Well, if your data happened to be all like https://sizangkam.files.wordpress.com/2015/01/luce_tables_test.pdf that would only take one regex [01:20:11] If your data is more complex, probably better convert everything to CSV and make a custom script to assemble the pieces [01:20:50] Nemo_bis: If it's alright with you, I can PM you and send you the CSV output of the final file [01:21:15] FKLinguista: it's much better if you post it on the grease pit [01:21:35] How many terms are we talking about? [01:21:42] 198 [01:21:46] **189 [01:21:56] https://en.wiktionary.org/wiki/Wiktionary:Grease_pit [01:21:58] possibly more once I compile more sources together [01:22:19] Ok [01:23:02] Then maybe pagefromfile is easiest, unless you're familiar with python [01:24:02] Anyway, when you have merged the data and you know how much and what kind it is just ask at https://en.wiktionary.org/wiki/Wiktionary:Grease_pit [01:25:39] From all sources, you mean? [02:34:19] diff3.exe for windows [04:17:20] Evening all. [04:25:23] Is it feasible to remove the php_uname() from globalfunctions.php and not have anything break? The server is returning that error and I think I'd rather edit it since it doesn't seem like it's anything that is impacting the use of my MW install. [04:25:56] * ElmerG is having the 'php_uname() is disabled for security reasons' message. [04:26:55] ElmerG: yes. is your host running windows? [04:27:15] if yes, change the line: $isWindows = substr( php_uname(), 0, 7 ) == 'Windows'; [04:27:23] to: $isWindows = true; [04:27:28] if not, change it to $isWindows = false; [04:30:17] * Elmer-G returns after getting kicked, and being told the nick is registered. Weird. It's a Linux box. [04:30:29] Okay ,cool, thanks! [04:31:53] You guys have been great help with all my mediawiki noob questions, I just have to say that. [04:32:32] Elmer-G: that's great to hear! :) Thanks! [04:34:32] * Elmer-G downloads the file to edit. I should just need to upload it and refresh for it to take effect right? [04:34:58] yep [04:35:18] Sweet. [04:35:29] * Elmer-G makes a backup just in case I explode the wiki. [04:39:13] That worked, thanks so much ori. [04:40:05] Elmer-G: I'm just about to submit a patch to MediaWiki to inspect the PHP_OS constant rather than call that function [08:11:53] good morning :) [08:22:17] i have not success to authenticate my users under ldap. who can help me to find the trouble? my Localsettings are http://pastebin.com/wmCMztfx [09:19:53] How can I deprecate a hook in core? [09:20:45] I want to change core hook name. [09:24:52] I already saw https://www.mediawiki.org/wiki/Deprecation but I can't find about hooks... [09:45:20] any help? [10:48:03] none help? [11:15:52] hi, i have been using vagrant for around 3 months, but for the past two days, every time i'm running vagrant up, i get an error (default: Warning: Connection timeout. Retrying...) .Here is the complete error report: http://paste.ubuntu.com/9915579/ [12:29:14] i updated both virtualbox and vagrant to the latest versions and still get similar errors on vagrant reload http://paste.ubuntu.com/9916525/ [13:25:52] Hello, can anyone recommend a ACL for mediawiki thats working with 1.25wmf14? Preferably similar to HaloACL [13:45:53] I'd like to use the API to find out whether a new transclusion of a specific template was added in some revision, but it seems that's harder than it sounds. The property templates won't tell me how many transclusions of a specific template there are on a revision, and I don't really understand the output format of the rvdiffto property [13:47:28] what is the best way to try and go forward? is the rvdiffto result what I want? If so, is there a specification of the output format anywhere? If not, is there any other way that is better than getting the full revision text of the documents and attempting to parse it? [14:32:13] hello folks, may i ask how i can format a date/time to something user readable? maybe even like "3days ago"? [14:39:47] Is new MWTimestamp()->getHumanTimestamp() the right way to go? [14:54:53] yo guys I am new to open source and want to start contributing to mediawiki [14:56:40] My user page : https://en.wikipedia.org/wiki/User:Abhilash_Mhaisne . I have been contributing to wikipedia for some time, so please help me get started in development of mediawiki [14:59:46] slim_: Hi, have you tried https://www.mediawiki.org/wiki/How_to_become_a_MediaWiki_hacker [15:00:39] marshen: that's what's used most, though it doesn't work well in all languages and situations [15:01:08] Nemo_bis : Yeah I read that, but it confused me. Where do I get started? [15:02:11] slim_: can you please tell us what part confused you? it would help us improve [15:03:01] https://wikitech.wikimedia.org/wiki/Special:UserLogin/signup and https://www.mediawiki.org/wiki/Gerrit/Tutorial seem rather clear step by step process [15:03:08] No particular part, but I could not understand where should I start. I am new to open source. [15:03:14] Oh I see [15:04:00] thanks slim_ [15:16:11] slim (gone, damn), I reshuffled the page https://www.mediawiki.org/wiki/How_to_become_a_MediaWiki_hacker#Get_started [16:14:56] so someone has a site that they want me to fix. their host recently dropped php 5.2 support and what ever version of media wiki they have will not work on a newer version. is there a fast way to get a php 5.2 install that would work? i do not have hours of time to devote to this. would an older ubuntu possibly ship with php5.2 on the disk? [16:18:40] was the installation modified much? if not you could possibly upgrade without much hassle [16:18:57] Nahiyan: it was not changed at all as far as i know [16:19:07] okay [16:19:17] i tried running the update script to update the database and it just dies as the php version is too new [16:19:56] the update script is actually when for updating a database to the mediawiki's version [16:20:16] I mean.. [16:20:18] yes but this version is so old that i have to do incremental upgrades [16:20:27] it is from 2007 [16:20:30] never upgraded [16:20:40] ouch [16:20:50] yes [16:20:55] i may even need php 5.1 [16:21:00] okay, you can download old versions of PHp [16:21:17] Nahiyan: for what os [16:21:19] but it looks like you don't have control of the host.. hmm [16:21:22] or what os do you recommend [16:21:40] for running MW any will work :) [16:21:52] yes but i meant for doing this proceedure [16:22:12] i guess windows. i got php 5.2 to compile on mint the other day but it didnt have mysql support [16:22:14] :-/ [16:22:17] anything you have on hand [16:22:28] you don't need to compile PHP [16:22:55] what do you have on hand? what's the old PHP and MW version and the new PHP version that the host installed? [16:23:02] how else would i get an older version on linux? [16:23:09] 5.5 is on the host [16:23:21] unsure of the media wiki version right now [16:23:34] well PHP releases their bins [16:23:48] I think [16:24:15] i dont think they do for linux [16:24:23] hmm well [16:24:25] let me check [16:24:32] maybe i will download an old version of wamp? [16:24:44] set up a virtual machine running windows 7 and do it that way [16:25:11] huh! [16:25:18] you're right [16:30:51] i think using wamp may be the money here [16:37:07] I asked earlier today, but I guess it wasn't a busy time: I'd like to use the API to find out whether a new transclusion of a specific template was added in some revision, but it seems that's harder than it sounds. The property templates won't tell me how many transclusions of a specific template there are on a revision, and I don't really understand the output format of the rvdiffto property [16:37:14] what is the best way to try and go forward? is the rvdiffto result what I want? If so, is there a specification of the output format anywhere? If not, is there any other way that is better than getting the full revision text of the documents and attempting to parse it? [16:51:48] Syco54645_work, sorry I had to go out, but what's your progress? [16:52:00] none right now [16:52:02] Syco54645_work, also, what distro do youuse [16:52:04] have to do it this evening [16:52:08] ok [16:52:14] generally mint for my home use [16:52:22] wife uses debian [16:52:23] ok [16:53:53] i use mint mate which has been getting worse and worse [16:53:57] she uses debian with kde [16:54:19] i switched out mate to kde and it was night and day different, still dont have much faith in the mint project any longer [16:55:37] JeroenDeDauw: ping [16:56:03] JeroenDeDauw: I have an issue with installing the map extension via composer [16:56:03] ok [16:59:03] I've made a css for category page, I've added it in resources, and added the relevant configuration to Resources.php, where else do I need to add its link? [18:29:04] I asked earlier today, but I guess it wasn't a busy time: I'd like to use the API to find out whether a new transclusion of a specific template was added in some revision, but it seems that's harder than it sounds. The property templates won't tell me how many transclusions of a specific template there are on a revision, and I don't really understand the output format of the rvdiffto property [18:29:11] what is the best way to try and go forward? is the rvdiffto result what I want? If so, is there a specification of the output format anywhere? If not, is there any other way that is better than getting the full revision text of the documents and attempting to parse it? [18:29:46] MartijnH: rvdiffto really just gives you a unified diff of the wikitext [18:30:21] MartijnH: however, you can ask mediawiki to parse the wikitext for you, which is easier than doing that yourself (you'll get an xml parse tree) [18:30:23] So if you want to determine whether a template was added in that diff, you'll have to parse it somehow, which may not be easy [18:31:00] But if all you care about it "this template wasn't here before and it is now" (as opposed to "this template was used 2 times on this page and now it's used 3 times") you can ask action=parse to parse both the before and after versions, and compare the list of templates in each result [18:31:32] it's a bit of a strange format that's returned by rvdiffto; it seems like the body of a html table serialized to a js string [18:32:10] asking mediawiki to parse it like valhallasw`cloud suggests seems much easier; when I have a parse tree it's far easier to traverse [18:32:28] MartijnH: https://nl.wikipedia.org/w/api.php?action=help&modules=parse [18:33:01] MartijnH: generatexml gives you an xml parse tree, but if you just care about 'template used or not', asking for prop=templates probably makes more sense [18:33:42] I do care about the 2 vs 3 transclusions of the same template [18:39:32] being able to get a parsetree like that is really neat! [18:39:36] thanks! [19:24:51] How do I add a new user group to localsettings.php yes I read ghe documentation and tried to add it but it never got added [19:34:55] Repton: Groups magically start existing as soon as rights are assigned to them [19:35:11] You can assign a right to a group with $wgGroupPermissions['groupname']['rightname'] = true; [19:35:22] It didn't though [19:35:31] Examples for 'rightname' would be things like 'edit' or 'delete' [19:36:15] for the past two days, every time i'm running vagrant up, i get an error (default: Warning: Connection timeout. Retrying...) .Here is the complete error report: http://paste.ubuntu.com/9915579/ . I googled the error and updated both virtualbox and vagrant to the latest versions and still get similar errors on vagrant reload http://paste.ubuntu.com/9916525/ [20:10:02] http://pastebin.com/JQi51ctU this is what I did in Localsettings.php but it didn't make the new user group? [20:10:32] Am I missing something in it? [20:12:19] Repton: And on Special:ListGroupRights the group does not appear? [20:12:33] Yes [20:12:44] It does not appear [20:13:12] ragesoss: tried $wgGroupPermissions['Trustworthy'] = $wgGroupPermissions['user']; [20:13:16] ? [20:14:23] Did you mean to ping me? [20:15:02] sorry, it should go to Repton^^ [20:15:38] :) [20:16:21] Repton: I've copy pasted your code into a LocalSettings.php and it works for me [20:16:35] Are you sure that you've saved and uploaded the file into your MW directory? [20:16:56] Repton: strage, I just pasted your config into my LocalSettings.php and it works right away [20:17:19] I thought it just goes into localsettings.php [20:17:30] Which I did and it did nothing [20:17:40] LocalSettings.php [20:17:59] be careful about capitalization [20:18:20] Yes I did into that folder [20:18:27] LocalSettings.php [20:19:03] Are you sure that the lines aren't commented out or something? [20:19:24] are there other lines in this file? [20:19:49] There's the default lines that came with it when I installed it [20:20:08] good, there should be database password at least [20:24:08] Oh it doesn't show in the mwconfig folder [20:24:13] could that be why? [20:24:54] Your LocalSettings.php should be in the base MW directory, not the mwconfig folder [20:25:00] Is there a magic word that outputs the currently displayed title (for pages that are using DISPLAYTITLE) [20:28:42] I'm sorry for asking this noob question but where is the base MW directory at? [20:31:16] Rosencrantz: no [20:31:46] Ok.. so I'll have to duct-tape something together then. thanks [20:55:25] Hi, I have been trying to set up vagrant. When I did “vagrant up”, there was a bloodbath. I am going to share some screenshots. Can someone give some pointers as to how to fix this? [20:55:27] thanks! [20:55:47] you could use a pastebin rather than screenshots [20:57:53] ankita-ks: One known MW-Vagrant bug at the moment is -- https://phabricator.wikimedia.org/T87478#996185 [20:58:34] The bloodbath won’t be visible then. [21:43:48] yay I got it to work now [22:26:27] Does anyone know the help page where I can see how to set up custom protection levels? [22:26:47] I forgot where it is located at in mediawiki.org [22:42:56] Does anyone know how I can create custom page protection levels at I'm sorry for repeating the question I got disconnected [22:46:09] sort of [22:46:16] i can link you to relevant documentation pages :) [22:46:42] Repton: https://www.mediawiki.org/wiki/Manual:$wgRestrictionLevels [22:46:59] it's brief, but hopefully sufficient [22:53:02] Well it added a staff protection level but failed to make a bureaucrat protection level [22:53:03] :/ [22:53:44] $wgGroupPermissions['bureaucrat']['bureaucrat'] = true; $wgRestrictionLevels[] = 'bureaucrat'; what I put for bureaucrats and it failed [22:55:53] Oh wait nvm I found out why [22:59:32] Good. :) [23:07:02] Does this wiki have history? https://wiki.gnome.org/TranslationProject/GitHowTo