[00:03:27] Question:can you set a table as a parameter for a template? [00:04:04] I am trying to accomplish that here:https://wiki.gamepaduniverse.com/wiki/Duke_Nukem:_Manhattan_Project_Step_0_Module [00:04:07] no success [00:16:39] any brave souls out there? [00:17:32] sure, just if your table has any = in it then you need to name the parameter [00:18:01] like {{template|1={{{!}} ... {{!}}}}} [00:18:07] also all of the | in the table need to be {{!}} [00:18:27] it's generally a royal pain, you should probably look for another way :P [01:24:52] Skizzerz:you still there? [01:26:55] Yaron:you seem to be a fairly intelligent individual. I am having trouble finding a good solution to the problem of adding a parameter in a template that is a table like the one found here:https://wiki.gamepaduniverse.com/wiki/Duke_Nukem:_Manhattan_Project_Step_0_Module [01:27:23] any ideas? [01:54:58] NotSneeze: What's the problem? [01:55:12] You can use HTML, BTW. [01:55:22]

foo

should work fine. [01:55:50] If you use wiki syntax, you'll need to escape the "|"s using a "!" template/hack (that is, {{!}} ). [01:56:49] so I need to replace every "|" with "{{!}}" , correct? [01:58:54] katie: I updated the page and there are no errors but nothing is displaying [01:59:44] should double pipes "||" be treated differently? [02:03:46] https://wiki.gamepaduniverse.com/wiki/Duke_Nukem:_Manhattan_Project_Step_0_Module [02:04:29] NotSneeze: sorry, I only saw your message now. But I don't think I can help you as much as other people here can. [02:06:36] Yaron: thanks! well if you could look at https://wiki.gamepaduniverse.com/wiki/Duke_Nukem:_Manhattan_Project_Step_0_Module [02:06:42] the table is not displaying [02:30:07] yaron:you still there? [02:48:53] Yaron:? [04:44:33] I need to edit the Vector skin so the class "user-anon" is added to body if the user isn't logged in, and "user-logged" if they are. Does anyone know how to do this? [05:02:25] Nevermind, I got it. :) [05:03:22] CoreyChambers: What are you trying to do? [05:03:35] Editing skin files isn't usually the greatest idea. It can make maintenance annoying. [05:03:59] If you're using a newer version of MediaWiki, we have "MediaWiki:Group-user.css" I believe. [05:04:19] So "Common.css" would apply to everyone, then "Group-user.css" would apply to logged-in users. [05:04:22] Probably. [08:01:30] hello all, i'd like to use mediawiki for managing two scopes - "finished" stuff and "work-in-progress" stuff. former should not be editable by world, latter is supposed to be a kind of collaboration [08:01:42] iirc i should use namespaces, right? [08:02:49] can i configure mediawiki so 'new page' create by group 'users' is always created inside specific namespace? [12:57:06] hi [13:05:58] is there a good way to copy an old wiki? I asked my supervisor if I could have access to the data and he said I should build it by scratch new. Now I'm stuck with importing the data into a new wiki without having access to the old one [13:06:28] I'm talking about this wiki: http://maps.kriegsmaterial.ch/wiki/index.php/Hauptseite Is there a way to export it without admin rights? [13:14:15] Satco: http://maps.kriegsmaterial.ch/wiki/index.php/Spezial:Exportieren [13:18:17] hey thanks, how did you find this link? [13:19:35] By knowing it was there ;) [13:19:37] !export [13:19:37] To export pages from a wiki, navigate to Special:Export on the wiki, type in the names of the pages to export, and hit "export". See for an example of this form. See also: !import [13:21:18] awesome :D I don't know but humans are still so much better than google :P Or i'm bad at googling :D [13:58:33] I imported everything I found into the new wiki, but all the pages still have a lot of stuff like this inside of them: {{#if: | * mit Kriegsmaterial von [[Mit Kriegsmaterial von::{{{firma}}}]] }} [14:00:59] Satco: Have you got the parserfunctions extension installed? [14:02:38] Hmmm I don't think so, let me check [14:41:41] is there an exstension to delete all posts that are in a wiki? [14:42:07] i tried nuke, but it doesn't delete the articels i imported, only the ones i wrote by hand [15:19:41] !deleteBatch [15:19:56] Satco: you can try https://www.mediawiki.org/wiki/Manual:DeleteBatch.php [15:22:58] It seems like this is an old article: Atleast I'm quite sure this command won't work anymore: [15:22:59] mysql -e "select page_title from page where page_touched between 20150431000000 and 20150519000000 into outfile '/tmp/pagelist.csv'" [15:23:53] ok, sorry my bad looked at the wrong table couldn't find the right one [15:30:43] awesome, saved me a lot of trouble the deletebatch script :) [16:38:31] Howdy all! I've got two very different questions to pose that smart people here can hopefully help me with. [16:39:13] I'm in cipher hell, and cannot for the life of me figure out what the latest SSLCipherSuite string is that will make my browser stop complaining about out of date encryption. It's driving me up a wall. Any chance someone can copy/pasta one at me? [16:39:49] ALSO I'm having a lot more trouble than I used to using s3fs to maintain a file upload repository, I was wondering what the wikimedia folks used to keep uploads sync'd across frontend servers [17:00:30] Ok, nix the CipherSuite question. That's less of a priority now. Still would love to know what wizardry the real wikipedia uses to distribute uploads across frontends [17:14:56] Hello, does anyone know if there is a way to download multiple images from a mediawiki installation? [17:15:52] MDL: What's the catch? [17:16:23] So I can resize and convert them. [17:17:23] Well, you could always right click and save as. Without some way of narrowing down which files you want you'd wind up with nothing or the entire wikimedia commons [17:18:41] MDL: you could scrape Special:ListFiles [17:20:30] Ulfr_: wikimedia uses a swift store for files [17:21:21] thedj: a swift store you say. Any chance you could give me one hint further and direct me to the app you're talking about? I very muchly appreciate it [17:23:26] Ulfr_: swift.openstack.org [17:23:32] https://wikitech.wikimedia.org/wiki/Swift [17:25:30] mutante: Much obliged. when I googled that I got sent to Taylor Swift's store. Not that I'm not a fan, but it didn't seem relevant [17:25:53] ^.^ [17:26:40] Ulfr_: here's more about the SSL cipher question [17:26:44] https://github.com/wikimedia/operations-puppet/blob/production/modules/wmflib/lib/puppet/parser/functions/ssl_ciphersuite.rb [17:27:05] see how we have "compat", "mid" and "strong" [17:27:31] :o [17:27:36] You're a gentleman and a scholar. [17:27:41] settings to decide which ciphers to use that depend on the version of Apache etc [17:28:06] I will most certainly run that to make my client happy [17:28:13] just as soon as i finish cleaning up after s3fs -.- [17:28:35] * Ulfr_ has discovered around 30% of uploaded files are inaccessible because they're stored in doubled repos [17:28:45] e.g. /0/0/0f/image.png >< [17:31:39] Ulfr_: here's an example of the settings on an Apache 2.2 [17:31:43] https://phabricator.wikimedia.org/P1999 [17:32:01] if you have Apache 2.4 you can use "mid" [17:32:20] you just have to be ok with excluding some users like Win XP with IE [17:33:00] Honestly, if they're using XP and IE they can call and complain [17:33:12] that way I can laugh at them [17:33:27] yea, at some point you have to draw a line [17:33:36] Ulfr_: rofl. a taylor swift store .. that has to go into the quotes :) [17:34:16] maybe it will be a release name, "taylor" [17:34:51] mutante: Hugs for you. SHA-1 notice is gone [17:35:02] Or if you're not a hugs kinda guy I could arrange for a carbonated beverage [17:35:14] Ulfr_: just be careful with the "Header always set Strict-Transport-Security "max-age=31536000"" part in that paste [17:35:20] if you enable it you can't go back [17:35:54] clients will then only try to use https and never http agan [17:36:09] you might want to lower the max-age thing at first [17:36:14] That actually sounds like a good idea, tbh [17:36:17] and then raise it later [17:36:19] yes, it is [17:36:34] Right now we're getting more tickets about people not understanding why they're getting errors about using http [17:36:36] except if you tell them once and then for whatever reason change your mind there is nothing to do [17:36:48] because you already told all the browsers to remembers this for a full year [17:37:13] so when we first enabled that we only did a week or so [17:37:19] then later moved it up to a year [17:37:20] TBH I really hate that we have https, but now that I've got it running I will be DAMNED before I take it away [17:38:06] so I just set it to 999999999999999999999999999 [17:41:10] Ulfr_: you never know what happens in a few years with protocols (http/2, spdy) [17:43:07] Pfft. Once this contract goes through I somehow doubt I'll ever look at this server again [17:45:05] * Ulfr_ knows it's shortsighted, but just spent a perfectly good monday holiday twaddling with stuff [21:08:22] Hi [21:08:41] Is there a way to import from Wikispaces to Mediawiki (with syntax conversion?) [21:09:09] Wikispace is changing it's hosting feees, and there's a wiki I'd like to rescue ( it's Creative Commons Attribution) [21:19:16] ShakespeareFan00: Don't think so. As a hack you could copy the read HTML into VisualEditor and save, but that will be a lot of manual effort and won't work for special kinds of content which display differently from their core (like images/etc.) [21:24:01] hey guys, having a really weird time installing latest mediawiki (not from github) [21:24:04] on lemp [21:24:40] just gets a blank page when i go to /mw-config/index.php [21:24:47] nginx logs say 200 OK [21:24:47] zach_, that's a php error [21:24:59] what does php error log say? [21:25:03] it doesn't [21:25:47] display_errors = On [21:25:50] and no errors appear [21:25:54] i'm using php 5.6.x [21:25:56] on debian [21:26:12] clean debian box too [21:27:29] however, i'm doing the install over https [21:28:02] only because chrome won't let me go to the http:// version for some stupid reason [21:28:48] no ssl errors, signed wildcard [21:29:17] debian jessie? [21:29:25] yes [21:29:33] and mediawiki from tarball [21:29:38] tar.gz ya [21:30:23] strange [21:30:47] shall i try github version? [21:30:47] using apache is more common than nginx, but there shouldn't be issues [21:31:05] yeah it has to be nginx [21:31:17] by that i mean, i can only use nginx [21:31:33] it's perfectly fine [21:31:48] if i were using apache it would be done by now lol [21:31:57] but i'm not so it isn't [21:32:25] not that you are attempting to use some odd web server [21:32:50] try setting on php.ini [21:32:53] log_errors = On [21:32:56] it's on [21:33:01] error_log = somepath [21:33:08] and it's set to log to /var/log/nginx/phperrors.log [21:33:13] what should the perm/owner be [21:33:20] error_reporting = E_ALL | E_STRICT [21:33:55] the user php-fcgi is running us should be able to write it [21:34:02] php5-fpm [21:34:03] you can set it to 666 while testing [21:34:13] *running as [21:35:04] k got those set, error log on, error reporting e_all | e_strict [21:35:09] restarted fpm [21:35:14] refreshed a few times, no logs [21:35:34] wait [21:36:14] thought fastcgi_intercept_errors on; might help [21:36:15] but did not [21:37:12] is there a recommended nginx vhost for the latest version [21:37:21] i'm using the one from http://wiki.nginx.org/MediaWiki [21:40:01] there's an example on https://www.mediawiki.org/wiki/Manual:Short_URL/Page_title_-_nginx,_Root_Access,_PHP_as_a_CGI_module [21:42:11] hey got it [21:42:33] needed fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name; [21:43:01] thought that was already in include fastcgi_params; [21:43:06] guess not [21:44:11] * Platonides sees double steipps [22:42:25] Hi when I try to upload a picture I get this mistake: https://dpaste.de/WrUk [22:42:54] Google couldn't really help me. And looking at the code in LocalFile.php I see this line: // Bug 54736: use simple lock to handle when the file does not exist. [22:43:07] What does that realy mean, I'm not sure how to fix this [22:50:12] hmmm: MWException from line 1856 of /var/www/html/mediawiki-1.25.2/includes/filerepo/file/LocalFile.php: Could not acquire lock for filename.jpg' [22:50:28] it's failing on LocalFile->lock() [22:50:38] I guess you are using a simple db? [22:50:46] no fancy things like memcached [22:51:44] yes [22:51:56] no chaching enabeled at all [22:52:06] could be permissions on /tmp [22:52:13] it should perform the lock through the db [22:53:02] oh wait, shouldn't all the mediawiki files owner be the same as the apache server ? [22:53:11] Right now all folder belogn to me, this seems wrong [22:53:12] not really [22:53:20] oh ok [22:53:27] it only needs to read the php files [22:53:39] however, the images folder and subfolders must be writable [22:53:58] that is, if you expect the apache user to create files there [22:54:18] (also, depending on your sapi, mediawiki may be running as you or as apache) [22:54:37] man fucking linux file permissions have to look them up every time i forget which one is which :P [22:54:42] what's a sapi? [22:56:06] I have these permissions on the images folder drwxr-xr-x [22:56:14] As far as I understand that's fine right? [22:56:33] there are no subfolders in images [22:57:40] Satco: we can't tell with just that info. also need to know which user/group owns it and what user Apache runs as [22:58:15] Satco, mediawiki will create them when you start uploading files [22:58:15] what user owns the image directory? [22:58:54] My user owns all files and folders in mediawiki, like the one i log into the server [22:59:53] Apache is running once as root, 9 times as www-data and 1 time as my user [23:00:36] But how do I find out which one is the one hosting mediawiki. I will post what I get when I type ps aux | egrep '(apache|httpd)' [23:01:33] https://dpaste.de/VSDU [23:04:46] Satco: it runs as www-data [23:05:22] Yeah, just read online, that it only runs as root to spawn other children to handle the requests. [23:05:25] the one showing your user name is just the "grep" finding itself [23:05:42] yes, what you said [23:05:44] Oh ok. I guess it running as www-data is fine and should be that way? [23:05:53] it's normal, yea [23:06:04] any other ideas then ? [23:06:14] the image directory is not writable by www-data [23:06:48] you said above drwxr-xr-x [23:07:01] so that just lets the owner write to it [23:07:07] and you said that's you [23:07:09] Oh right, while I'm at it, should I change the owner of any other folders? [23:07:18] Yeah, your totaly right thanks. [23:07:36] i _think_ it's just the upload dir [23:08:13] I don't have that one. Is it a subfolder in images that will be created? [23:09:29] Oh man you're the best :) thanks for the fast fix. It works now [23:09:52] oh, it works ? nice [23:09:57] so i was about to say that depends [23:10:08] https://www.mediawiki.org/wiki/Manual:Configuring_file_uploads#Upload_directory [23:10:11] ^^ [23:10:29] ok, cool then :) [23:11:27] I have the stupid task of migrating an old wiki into a new one :( A lot of copy paste. [23:12:08] I know about the import and export utilities but we only need some small part of the old wiki, but it's still enough to be annoyingly much :D [23:12:30] So i gotta find out what we need and ctrl + V that :D [23:17:31] Satco: but copy /paste does not save history. why not just paste the list of pages into Special:Export [23:19:46] Because it has a lot of unecessary information in the old wiki, I would need to delete. [23:21:22] i would question why it has to be deleted, and if you really have to do it in the new place [23:22:13] That's what my supervisor told me. I was also for export import but he was like nah we don't need all that old stuff [23:22:31] until you need [23:24:10] yea, you always do .but only if you delete it:) [23:24:41] ^^ [23:28:25] "don't need" probably means for the supervisor "don't spend more time importing old stuff",but it takes you longer to remove it than just importing everything and outdated stuff (c|sh)ould be replaced via normal wiki edits