[09:19:57] heya. when logging in, a token has to be retrieved and then submitted (wpLoginToken) - what's the purpose of it ? [09:20:25] Richlv: it protects against https://en.wikipedia.org/wiki/Cross-site_request_forgery attacks [09:25:57] legoktm, thanks :) [09:27:39] legoktm, it specifically protects login from cross site request forgery ? [09:27:51] as in, why is it already required for login, not after user has been logged in [09:29:54] Richlv: https://static-bugzilla.wikimedia.org/show_bug.cgi?id=23076#c0 [09:32:28] legoktm, uh, great, thank you :) [10:35:30] hello everyone [13:00:38] heh [13:00:43] I think InstantCommons is broken on my dev wiki [13:06:56] I thought this was fixed on master with php5-curl? [13:21:06] does anyone know if mwoffliner can handle authentication (shib/saml in this case) ? [14:03:43] Hello! ...I am Vritika. Please let me inform how to get start my Wikimedia project. [14:04:18] What should I do for starting the project? [14:23:26] Vritika1, Wikimedia or Mediawiki? project as in filling/setting up a wiki or software development? [14:23:44] have you tried supplied instuctions? [14:49:28] Project on Mediawiki for Wikimedia. [14:50:25] Vritika1, hi! What would you like to work on exactly? [14:50:32] What are your interests? [14:51:07] What is "your Wikimedia project" exactly? Do you have an idea already? Or do you ask how to contribute in general? [14:51:41] For general info, https://www.mediawiki.org/wiki/How_to_contribute and https://www.mediawiki.org/wiki/How_to_become_a_MediaWiki_hacker [14:55:10] No...I have no any idea about this project. [14:56:14] Vritika1, what are you interested in? [14:56:20] Please see https://www.mediawiki.org/wiki/How_to_contribute and https://www.mediawiki.org/wiki/How_to_become_a_MediaWiki_hacker [14:56:33] I have an interest in Open source.. [14:56:58] Okay..... [14:58:09] See the links :) [14:58:23] Okay. [14:58:30] Get inspired by some ideas, and if you feel like writing code, check out the second link [14:59:07] Vritika1, if you have specific questions after reading that, please don't hesitate to ask! Happy to help! [15:03:13] Okay...This is really nice of you...:) I would never hesitate. [15:14:21] !g [15:14:21] https://gerrit.wikimedia.org/r/#q,,n,z [15:14:31] !gerrit [15:14:31] https://gerrit.wikimedia.org/r/#q,,n,z [16:07:11] does anyone know if mwoffliner can handle authentication (shib/saml in this case) ? [16:51:13] hello, i recently upgraded from 1.22.5 to 1.25.2 and i'm having issues with html cache updates not running on edit - they're being added to the job queue instead. any idea what would cause this? [17:18:34] Hey, I was reading about setting up multiple wikis using the same codebase, but the Wiki Family manual page seems outdated. [17:18:43] Am I simply mistaken? [17:18:56] I wouldn't be surprised if it was outdated. [17:19:07] I see. [17:20:37] I´m trying to do it by myself, but I´m not sure how to include the shared codebase while still making sure that the installation is using the local LocalSettings.php [17:22:24] i have a multi wiki setup, i use a single LocalSettings which changes the database/name/etc based on the host [17:29:21] Hi guys :) I'm trying to install MediaWiki master on some DO droplets. I'm trying to set up MySQL server on one VPS and MediaWiki on another, and after several tries, the web installer still says it can't connect to db server [17:35:47] mwtl: Hm, I might have to resort to that. [17:36:29] mwtl: I was mostly trying to make it so that I didn´t have to edit an extra file on top of the webserver conf. [17:36:57] Ideally I liked the idea of having everything packed in its own directory [17:37:19] I guess I could setup an automatic directory selection based on the host [17:37:48] Or maybe a enviroment variable I set up in the httpd.conf and read in the file. [17:46:14] NDK|Cloud: firewall issue? [17:46:27] and/or mysql not listening on right interface(s) [17:48:31] Reedy: quite possibly. It should be listening on the default port. Maybe private networking isn't on on my mw1 server [17:48:53] I tried using my db servers public and private ip, neither worked :// will look into it when I'm home I guess [17:54:50] NDK|Cloud: I don't know if anyone around here is particularly familiar with DO etc [17:55:30] But if you can get the webserver to talk to the remote mysql (cli from one to the other), but can't get MW to work still, we can help further [19:00:00] Reedy: thanks. Will look into it and see if it's mw or server. Assuming the latter [19:39:13] why does rebuildFileCache.php loading my auth plugin? [19:40:17] what do you mean by loading? any script will read LocalSettings.php and thus loading any extension you have installed [19:41:57] Vulpix: i see. i guess i don't understand why a maintenance script would load an auth plugin, as it would fail to auth since it cannot accept credentials. the script caches the 'Permission error' page and then exits. not very useful [19:43:02] well, think about it [19:43:14] a maintenace script will load a lot of mediawiki internal code internally [19:43:27] the extension using it should probably check if it's in maintenance mode before trying to authenticate anything... [19:43:36] Loading articles will do a permission checks [19:44:21] Reedy: but server-side? what's the point? and Vulpix: probably. unfortunately it looks like this particular extension was cowboy coded [19:44:42] Why not? [19:44:53] Maintenance scripts are the minor usage [19:46:12] you could just wrap the loading of the auth extension if running CLI [19:46:45] Reedy: can you elaborate? i am running CLI [19:47:26] if ( php_sapi_name() !== 'cli' ) { require_once( "$IP/extensions/Whatever/Whatever.php" ); } [19:48:03] and where would that go? [19:48:17] i don't know where extensions get loaded, i'm not terribly familiar with MW codebase [19:48:27] It'll be listed in oyur LocalSettings.php [19:48:33] interesting, alright i'll check [19:48:41] Wrap the auth loading in that if [19:49:09] hot dang you're right, let me try this.. [19:53:00] still permission error.. [19:53:34] what did you add? [19:53:50] /what is it showing? [19:53:56] Reedy: i had to wrap not only the call to require_once but a lot of the ensuing configuration [19:54:13] Reedy: it's printing the permission error page to stdout and then exiting [19:54:47] https://github.com/wikimedia/mediawiki/blob/master/maintenance/rebuildFileCache.php [19:54:52] It's not using any explicit user [19:54:54] Reedy: if it helps, this is configured as a 'private' wiki, so no articles are accessible without some sort of auth. i'm not sure how this is being accomplished as i've just recently been thrown in here [19:55:34] hmm [19:55:36] let me try something [19:56:52] Out of interest, do you really need wgUseFileCache? [19:56:58] If it's a private wiki... Is the load that high? [19:57:55] Reedy: no, but what i'm actually trying to accomplish is a static/offline version of the thing that can be thrown on thumb drives and, since apparently dumpHTML is broken beyond repair, and mwoffliner doesn't support auth...this is the only avenue i have left to try [19:58:21] could you not use a portable XAMPP etc? [19:58:44] Reedy: and what if it's a linux box? [19:59:04] wine? [19:59:05] :D [19:59:08] must be a way [19:59:10] Actually [19:59:14] Reedy: this is getting ridiculous :| [19:59:16] OpenZim [19:59:25] rawsted: I was trolling with the wine selection [19:59:38] http://www.openzim.org/wiki/OpenZIM [19:59:51] Reedy: i looked at this but i didn't understand how to use it on anything other than wikipedia. [20:00:16] There's a few "3rd party" offline wikipedia projects [20:00:20] I don't know how open code there is [20:01:08] Reedy: tbh i don't think this helps, it's just a file format spec and lib [20:01:20] There is software to go with it [20:01:35] ah, kiwix [20:01:47] Reedy: yeah, which mwoffliner is part of ;) [20:01:47] http://www.kiwix.org/wiki/Main_Page [20:01:55] kiwix uses mwoffliner. [20:01:57] "Kiwix enables you to have the whole Wikipedia at hand wherever you go! On a boat, in the middle of nowhere or in Jail, Kiwix gives you access to the whole human knowledge. You don't need Internet, everything is stored on your computer, USB flash drive or DVD!" [20:02:08] yeah, it's a no go [20:02:30] i was already given the bad news in #kiwix regarding mwoffliner [20:02:39] ah :/ [20:02:45] by kelson himself actually lol [20:03:10] so ok, it looks like at best, this will be really difficult and at worst is impossible given our use case [20:03:31] rawsted: why not a crawler like HTTrack? [20:03:51] Vulpix: does it support shibboleth authentication ? =P [20:03:58] or even local mw auth [20:04:34] I'm not sure the filecache is going to do what you're going to think [20:04:42] I'm not sure the filecache is going to do what you think it is [20:04:45] Reedy: probably not, but it was worth a try [20:04:51] It's going to still need webserver/php infront of it [20:05:02] Reedy: well than that answers it then [20:05:42] HTTrack provides a proxy so you login first and then let HTTrack connect through your cookies. Not sure if that would suffice for shibboleth authentication [20:06:09] This is silly [20:06:41] Vulpix: this is sometihng that would have to be, for example, crontabbed eventually, so no, that wouldn't work i'm afraid [20:07:38] having a strong authentication method and also want to make an offline copy of the content to share with probably unauthorized users doesn't seem right [20:07:51] Vulpix: tell me about it ;) [20:08:15] one of the many stupid policies around this place [20:08:27] i'm just going to tell them it can't be done [20:08:35] sounds sane :D [20:10:14] thanks anyway Reedy, Vulpix [20:21:50] evening all. Is there a way to change the word "Summary" for the Edit Summary, to "Edit Summary" or something else? [20:25:01] Person: yes, interface texts are editable [20:25:20] !interface [20:25:20] You can edit the text which appears in most parts of the interface by editing the right pages in the MediaWiki: namespace. See [20:25:23] Thanks Vulpix . Where would I do it? I'd guess the Language file. [20:25:54] just editing the corresponding MediaWiki: page [20:26:13] to locate the name of the message, you can use the qqx trick [20:26:15] !qqx [20:26:15] If you want to find a particular system message being used on a page, you can append the uselang=qqx parameter to the URL. That will replace all system messages used on the page by their message names. See https://www.mediawiki.org/wiki/Qqx [20:26:34] $:andre\> grep -r "Summary" languages/i18n/en.json [20:26:34] "summary": "Summary:", [20:28:11] Awesome, thank you [20:28:14] all sorted :) [20:30:00] I want to be evil and create a language that uses qqq as its abbreviation. [20:30:50] You wouldn't get to choose the language code ;) [20:30:54] I think it's reserved too :P [20:33:47] Hmm, I can not actually find any standard that lists it as registered as reserved. [20:34:02] I always thought it was a MediaWiki centric ideal. [21:11:15] Trela: https://www.mediawiki.org/wiki/Localisation#Message_documentation says it's reserved for private usage by ISO 639 [21:12:43] I guess it is just not listed on any of the publicly available lists. [21:28:15] anyone have any idea about my html cache update issue? i managed to work around it by forcing all the update jobs to run when they get added, but would like to know what caused it (maybe it's a bug?)