[00:28:46] <OrigamiOlivia>	 sendmail sucks, all my homies use wgSMTP
[01:14:08] <OrigamiOlivia>	 after upgrading i get Problematic modules: {"mediawiki.legacy.commonPrint":"missing","mediawiki.legacy.shared":"missing","mediawiki.skinning.interface":"missing","skins.vector.styles":"missing"} when trying to load the CSS
[01:27:29] <Reedy>	 I'm presuming you have some extensions/skins not properly updated
[01:29:24] <OrigamiOlivia>	 ok
[01:31:51] <OrigamiOlivia>	 this whole /var/lib /usr/share split is really annoying
[01:32:54] <OrigamiOlivia>	 tempted to just nuke both of them and reinstall
[01:42:42] <Reedy>	 packaging can do weird thing
[01:42:43] <Reedy>	 s
[01:44:51] <Shaun>	 I never use the debian packages.  I get what they're trying to do, but it's too much of a diversion from the documentation for me
[01:50:10] <OrigamiOlivia>	 very important question though, should i leave the little "powered by debian" badge at the bottom if i run debian but not the debian repos
[01:50:18] <OrigamiOlivia>	 but not mediawiki from the debian repos*
[01:51:39] <Shaun>	 there's no obligation to place that either way, it's purely to taste
[02:26:21] <Platonides>	 812
[02:32:59] <p858snake>	 Platonides: was a leap year
[02:35:28] <OrigamiOlivia>	 is it possible to have e.g. the "view history" link be /w/PageName?action=history instead of $wgScript?title=PageName&action=history
[02:40:04] <p858snake>	 there is https://www.mediawiki.org/wiki/Manual:$wgActionPaths
[02:54:23] <OrigamiOlivia>	 p858snake: i noticed that but that doesn't use query params
[02:54:56] <OrigamiOlivia>	 i guess i could still set it up as $wgActionPaths[$action] = "/w/$1?action=$action"
[02:54:59] <OrigamiOlivia>	 i'll try
[02:59:32] <OrigamiOlivia>	 p858snake: i tried foreach ( $actions as $action ) { $wgActionPaths[$action] = "/w/$1?action=$action"; } $wgActionPaths['view'] = "/w/$1"; but now the main page "view history" button is just a link to "/". I have wgMainPageIsDomainRoot enabled and it only works when I turn that off, but I want it on…
[03:01:57] <OrigamiOlivia>	 this seems like a bug
[03:03:37] <p858snake>	 MainPageIsDomainRoot is brand new in 1.34.X so i'm not surprised if stuff is broken with it
[03:04:25] <p858snake>	 you can file a task in phab that wgMainPageIsDomainRoot and wgActionPaths don't work together for the main page
[03:05:44] <OrigamiOlivia>	 thank you
[03:05:52] <OrigamiOlivia>	 for now i'll just use this instead https://laxstrom.name/blag/2015/08/31/mediawiki-short-urls-with-nginx-and-main-page-without-redirect/
[03:06:00] <OrigamiOlivia>	 those hooks that they have
[19:14:53] <leshaste>	 there is some way to dump the contents of https://awoiaf.westeros.org/index.php/A_World_of_Ice_and_Fire in xml/mediawiki but  I can't find the web page again. Can anyone help?
[19:15:10] <Reedy>	 Special:Export
[19:16:14] <leshaste>	 Reedy, thanks! How did you find it?
[19:18:10] <Reedy>	 My memory :P
[19:18:20] <Reedy>	 I've been using MediaWiki regularly for a good 15 years at this point :)
[19:19:30] <leshaste>	 Reedy, awesome. If I wanted as  much of the content as possible to turn into plaintext, do you know what categories I should choose?
[19:20:02] <Reedy>	 Not really, it depends on the wiki
[19:20:08] <Reedy>	 I'd probably get a list of "all pages" from the API
[19:20:24] <Reedy>	 Use that to get the list of articles, then paste those into the box on Special:Export
[19:20:40] <leshaste>	 could you help me get a loit of all pages please?
[19:20:42] <leshaste>	 list
[19:33:28] <lessshaste>	 I managed to read in the xml and find the text with
[19:33:38] <lessshaste>	 tree = etree.parse("A+Wiki+of+Ice+and+Fire-20200802192158.xml")
[19:33:38] <lessshaste>	 expression = '/m:mediawiki/m:page/m:revision/m:text/text()'
[19:33:38] <lessshaste>	 namespaces = {"m": "http://www.mediawiki.org/xml/export-0.10/"}
[19:33:39] <lessshaste>	 texts = tree.xpath(expression, namespaces=namespaces)
[19:34:34] <lessshaste>	 but texts looks like https://bpa.st/7URA  . How can I parse this into plaintext? Is that in mediawiki format?
[19:38:54] <Vulpix>	 lessshaste: that's wikitext. You'd better get the HTML pages directly and extract text from the HTML. Since that's a very wide format, you probably will find some library able to do that
[19:40:02] <lessshaste>	 Vulpix, is it really not possible to parse the wikitext?
[19:41:29] <Vulpix>	 MediaWiki parses wikitext into HTML. Also, wikitext uses templates that fetch contents from other pages. You won't be able to parse that from a single page
[19:43:51] <lessshaste>	 maybe pandoc can do it?
[19:47:50] <lessshaste>	 it looks like that all that is needed is to find the word "Synopsis" and the remove the [[...]] intelligently
[19:48:17] <lessshaste>	 the only problem being where there is an option. I.e [[name1|name2]]
[22:41:02] <Nik60>	 HI, Anyone familiar with cargo extension in here maybe?