[02:02:25] on the wiki i administer, a user has forgotten their login. is there any way to see what email address they registered with? [02:04:27] look in the database :) [02:05:37] ok [02:44:21] Hi, I'm looking at the pages which are part of the NPOV dispute category - https://en.wikipedia.org/wiki/Category:NPOV_disputes because they have NPOV tags. I'm interested in getting pages where such tags (e.g. {{POV}}) were added in the past but not removed. Is there a way to get those using the API? [04:20:55] When parsing templates does Mediawiki cache things? [04:21:35] Like suppose I had a template called {{lookup|value=XYZ}} that was either a simple switch - or even something expensive using if pages exist ect, if it's used several times on the same page, will it be expanded once, or at each use? [04:23:50] Yeah it seems it does not [14:30:54] Hi, I'm trying to setup a mediawiki site on the Tor network. A problem I've run into is that with my current editor/moderation model as I have it planned out, users would be required to sign up with an email address to verify their identity in an attempt to not be spammed to hell by bout accounts etc or bot edits. But in order to send email over Tor, I apparently have to purchase a domain for the email [14:31:00] to actually send. This is undesirable as it would leave a paper trail, breaking the anonymity granted by hosting the site on the Tor network [14:31:54] so I'm curious; are there alternative registration models or user/moderation models I could use instead to get around the email over Tor problem? [14:37:36] for instance, I've heard ssh or GPG keys could be used as a form of potential registration? [14:38:16] Or perhaps I could create accounts with passwords and give them out to certain select users? [14:38:38] I'd really appreciate some input on this :) [14:40:26] Hello guys and girls! [14:41:21] recently today I found out my mediawiki with a 500 error [14:41:41] AH00124: Request exceeded the limit of 10 internal redirects due to probable configuration error. Use 'LimitInternalRecursion' to increase the limit if necessary. Use 'LogLevel debug' to get a backtrace. [14:43:55] any idea where i could find some ideas to debug this issue? [14:44:14] I'm pretty sure googling AH00124 will give some answers [14:44:27] Usually it's something is wrong with your redirects/short url config [14:45:17] it's weird I've have my config for several years [14:45:23] *had [14:45:34] and just today I found this [14:45:40] ok, thanks [14:46:44] Hey Reedy, Could I have your opinion on what I wrote above? I'd really appreciate any input [14:46:52] You've either hit some edge case your config never took account of, or some package upgrade [14:47:06] !e ConfirmAccount | burning [14:47:06] burning: https://www.mediawiki.org/wiki/Extension:ConfirmAccount_ [14:47:12] stupid bot [14:47:16] !e ConfirmAccount [14:47:16] https://www.mediawiki.org/wiki/Extension:ConfirmAccount [14:47:52] IIRC you don't have to use the email functionality [14:47:57] Does the Confirm Account extension send out SMTP at all in any way in order to verify the account? [14:48:54] Oh [14:48:55] if ( !$wgEnableEmail ) { [14:48:55] echo "ConfirmAccount extension requires \$wgEnableEmail set to true.\n"; [14:48:56] never mind [14:49:02] It usually emails out passwords [14:49:17] Basically, the other option is allowing registration, but removing most of the user rights [14:49:24] Then add them manually to a group that grants them those rights [14:52:08] Reedy, seems like I have a "visitor" triggering multiple redirections [14:52:57] a very systematic behavior visitor xD [14:57:06] thanks Reedy, I'll try something like that out [17:35:40] I can't find any references to the performance testing of hooks in extensions. Does anyone have a link to one? [19:03:34] Hi, I am writing a dokument. That document references a wiki page. I need to make that page available to the reader. I can not give him access to the wiki. What do you recommend to do? [19:03:59] the reader = a person reading that document and verifying references [19:05:04] I am aware of mediawiki export, but I can not use it, because the export format XML is not really human readable. [19:05:35] I am also aware of the publish function and "save as..." with the webbrowser [19:07:53] nabblet: copy&paste the page into the document (inside an annex for example) [19:10:42] Vulpix: I am allowed to store the pages as separate files in HTML format. That's better to read than forcing the page into A4 paper format. [19:11:13] So I'd rather use "Save as..." . I would like to export the media referenced in that page too. [19:20:53] maybe via API ? [20:32:27] I'm trying to do the simplest thing ever, but getting stymied at every turn. [20:32:53] I just need to add an external Web Analytics script to every page, and send page view data [20:33:23] I've heard that writing an extension is the correct way to do this for performance reasons [20:33:42] So I wrote one, based on Extension:BoilerPlate [20:34:05] But I get: [20:34:05] MWException from line 164 of /opt/htdocs/wiki/mediawiki/includes/Hooks.php: Invalid callback AdobeLaunchHooks in hooks for BeforePageDisplay [20:34:05] when I enable the extension [20:34:38] Code at https://github.com/freephile/AdobeLaunch [20:36:37] That means the callback is not callable [20:37:36] but in extension.json I have: [20:37:37] "HookHandlers": { [20:37:37] "AdobeLaunchHooks": { [20:37:37] "class": "MediaWiki\\Extension\\AdobeLaunch\\Hooks" [20:37:38] } [20:37:38] }, [20:37:39] "Hooks": { [20:37:41] "BeforePageDisplay": "AdobeLaunchHooks" [20:37:44] }, [20:38:10] see https://github.com/freephile/AdobeLaunch/blob/main/extension.json#L26-L33 [20:38:33] I thought this is valid/correct syntax [20:57:27] The includes/Hooks.php file is https://github.com/freephile/AdobeLaunch/blob/main/includes/Hooks.php [20:58:08] which sets the namespace: [20:58:08] namespace MediaWiki\Extension\AdobeLaunch; [20:59:12] and extension.json specifies which Namespaces to Autoload [20:59:12] https://github.com/freephile/AdobeLaunch/blob/main/extension.json#L13-L15 [21:01:54] Technically, I don't even need a hook handler. But because the code was already there from the BoilerPlate extension, and puts output on every page, I figured it would at least be a good way to see that the extension is loaded (aside from visiting Special:Version) [21:03:46] freephile: perhaps you'll wanna model your code after AbuseFilter or somesuch? see eg. https://gerrit.wikimedia.org/g/mediawiki/extensions/AbuseFilter/+/master/extension.json#282 (full disclaimer: I'm not familiar with the new hooks system, hence why I'm not able to help out much with it :/) [21:04:07] DELETING the HookHanders and Hooks specification from my extension.json allows the extension to load [21:04:41] freephile: i am curious of which version of MW this is [21:05:15] My /real/ question aside from understanding what went wrong with the hooks is how to properly write the page view data [21:05:31] grknight: this is MW 1_34 [21:05:46] wiki says HookHandlers are new to 1.35 [21:06:02] ashley: I'll check that out. [21:06:16] grknight: Ahh [21:07:31] i'm in no way an expert, but perhaps reviewing https://www.mediawiki.org/wiki/Manual:Hooks can help [22:07:58] I fixed up the declaration in extension.json, and fixed up the code in Hooks.php so it's working now [22:08:50] https://github.com/freephile/AdobeLaunch/commit/80109b37e0abb719945fee204ed7316747f974fd [22:50:03] Now if there are any JavaScript pilots out there, could you let me know what is the proper way to do a simple method call (defined by the loaded library) after the external script is asynchronously loaded? [22:50:04] So far, what I have is https://github.com/freephile/AdobeLaunch/blob/main/resources/ext.adobeLaunch/init.js