[02:26:44] hi [02:27:03] does anyone knows a working tool to export mediaiki pages to HTML or PDF? [02:32:09] Hi dwaan. [02:32:16] !pdf [02:32:16] There is no such key, you probably want to try: !gitcheat, !wiki2pdf, [02:32:20] !wiki2pdf [02:32:20] Google it; one example is http://toolserver.org/~magnus/pdf.php [02:32:24] Heh. [02:32:30] dwaan: Yes, both are easily possible. [02:32:42] Usually any modern operating system can generate a PDF with the print function. [02:32:48] above link is not working [02:32:54] And MediaWiki automatically converts wiki markup to HTML. [02:32:56] and sorry im talking more mass export [02:33:00] not a single page at a time [02:33:01] You view a wiki page in your browser, it's HTML. [02:33:10] Oh, there's a maintenance script to dump all the HTML. [02:33:17] It's called dumpHTML.php or something. [02:33:21] It's included with MediaWiki. [02:33:27] What are you trying to do? Archive the wiki? [02:33:40] https://www.mediawiki.org/wiki/Extension:DumpHTML [02:33:45] DumpHTML required a lot of work and is permanently broken since August 2008. [02:33:54] yes archive for later conversion to sharepoint [02:34:03] My condolences. [02:34:43] Oh, it got split from MediaWiki core. [02:35:22] dwaan: You can use api.php to dump the HTML from every page. [02:35:25] That's probably your best bet. [02:35:41] Modern developers use Parsoid and mwoffliner: [4] [5]. [02:35:45] You could use that. [02:36:03] The fact that dumpHTML is allegedly so broken is kind of amusing. [02:36:11] so many dead links [02:36:12] Like how difficult could it be to write generated HTML to a file. [02:36:13] Yeah. [02:36:14] on mediawiki's own wiki [02:36:15] lol [02:36:18] Be bold! [02:36:24] You can edit the pages and improve them. [02:37:19] I think api.php is what I would use. [02:37:31] Or you could just use wget/curl. [03:44:23] Haley dwaan: i thought DumpHTML got fixed at one point, i wonder if the documentation page is just out of date [04:16:16] Could be! [06:16:41] Haley: anyway, apparently the new fangled way from memory is that parosid can chuck out a page in html via the api so request it that way, but that involves end wikis being able to get paroside etc to work for them [10:47:11] Hi, what tables should be shared by wgSharedTables to block sharing to work? I share ipblocks and ipblocks_restrictions, but that doesn't make the blocks applied on one wiki to be applied on other wiki [10:47:35] Still, while trying to block on the other wiki, after confirming it says "Already blocked", so it probably realizes somehow the user is blocked [10:47:41] I guess some table is missing, but which [10:47:41] ? [11:36:41] Urbanecm: have you confirmed they aren't blocked, and its not just the logging thats missing (see cavearts https://www.mediawiki.org/wiki/Manual:Shared_database#The_ipblocks_table) [11:44:07] For a research project I am trying to set up a local wikidata stack. I am using https://github.com/wmde/wikibase-docker and have no problems running it. My question boils down to this: How and what data dump should I use if I am only interest in the current set of entities and their relations. Is the *-pages-articles-multistream.xml.bz2 on https://dumps.wikimedia.org/wikidatawiki/20190101/ for example a good start? Thank you [11:44:08] in advance for any advice you can give me. [11:53:48] p858snake, yes, I've confirmed they are blocked. Honestly, I've just blocked my test user and then tried to log in using that test acc on both wikis [14:02:03] hello here [14:03:30] this is the good place for talk about postmarket os [14:04:29] I am trying to port/install postmarket os on a galaxy A5 samsung [14:04:50] it's possible [14:05:27] infinity_blue, welcome to the IRC channel for MediaWiki. [14:05:34] I have no idea what "postmarket os" is. [14:05:50] What brings you to #mediawiki specifically? [14:06:05] for that [14:06:36] I want to install postmarket on my mobile [14:07:42] infinity_blue, how is that related to MediaWiki? [14:07:42] but before i dont know if i start the tutorial for install it or if i have to compil it [14:07:48] Why do you ask this in #mediawiki ? [14:08:06] Why do you think #mediawiki has anything to do with "postmarket os"? [14:08:20] heu its on the postmarket website [14:08:26] Link welcome. [14:08:37] :-D [14:08:43] i mistake [14:09:10] sorry [14:10:35] ok thx i go to find the good place [15:42:26] Hello, I have a question regarding the recommended import method for wikidata dumps. Is the utility "mwdumper" still the prefered way? I had a lot of trouble with it and I simply want a local instance of the wikidata stack. (Using https://github.com/wmde/wikibase-docker) [16:37:13] does the database need to be joined to the domain for the mediawiki site to access it? they will be on separate systems [16:37:45] Can i pull moment.js for example on mediawiki [16:38:04] like src="babalblala.j" on the head or something [16:38:22] nunyet: mediawiki will need to be able to contact the db server, wherever it lives, so the appropriate port must be open in your firewall [16:39:37] Thank you! [17:06:50] nunyet, what are you referring to by 'the domain'? [17:07:29] the database server doesn't need to know anything about the hostname that mediawiki is responding to requests for [17:41:45] biberao: yeah why not [17:42:43] Krenair: maybe nunyet means network or something like that [18:09:54] harmaahylje: thanks [18:24:44] Sorry, stepped away. I was referring to authentication [18:27:37] I realized that auth happens at the mediawiki server, not the db server [18:29:30] there's both [18:29:48] users log into mediawiki with their own credentials, and mediawiki auths to the db server with a static credential [18:30:32] for this reason, if the services aren't on the same server, I strongly encourage that you configure SSL on mysql (or whatever db server you use) with a valid certificate, to protect the db server password in transit [18:31:35] yes, and very good point [18:31:50] Thank you! [18:57:55] hi there, is this a valid place for technical questions about wikis i have installed? [19:10:37] snaxyc: absolutely! [19:20:00] aha good. i need to change a user pw, but when i go to the special page it says email is disabled, and won't give me the reset form [19:20:30] and no luck so far with the php maintenance script, though i'm slowly working through it. [19:20:36] is there an easier way? [19:29:49] Don't know about that, but the experts are here, just be patient [19:30:32] if the user doesn't have email set, perhaps there is no way to verify who is resetting the password, and therefore it is not possible? [19:31:37] snaxyc: this seems to be the thing: https://www.mediawiki.org/wiki/Manual:ChangePassword.php [19:34:31] yeah that's the one i'm poking at. thought i'd ask. [19:43:48] the maintenance script is the only way to do it if email is disabled [19:44:11] alternatively, you could block the old account and they can create a new one [20:09:53] Skizzerz: can you clarify? I enabled email, but now I find that none is set for that user. Does that mean I'm hosed? Or could I set it in the db? [20:10:55] you could, but how do you know that email address corresponds to that user? [20:24:46] Skizzerz: it's an internal wiki, i know the person [20:25:09] ok, easiest thing is still the changePassword.php maintenance script [20:25:14] run it from ssh / command line [20:25:27] then let them know the new temporary password and instruct them to change it once they're back in [20:25:39] yeah, i tried that, was getting errors, for reasons. I could track it down, but it will take a while. [20:25:43] (then they could set their email address so they can do regular password resets in the future) [20:25:46] what errors? [20:27:15] meh, i can work through it, it's particular to the fact that i'm on a synology nas & not famililar with it [20:28:06] in any case, the user has never actually been used. can i delete and then recreate that exact user? i'm attached to the name format [20:28:13] (somewhat) [20:29:44] well, if you post the error you're getting with the changepassword script I can probably let you know what's going on within a minute or three [20:29:54] without knowing that error message though I can't help much [20:30:29] chasing down other rabbit holes seems premature at this time (deleting users isn't really a supported scenario in mediawiki; there's a script that can blow off *all* unused accounts, and an extension that lets you do it per-user) [20:30:49] Skizzerz: thanks I understand. I don't mean to be "that guy" but I'm not in front of that computer right now, and it's not the part I want help with currently [20:31:14] k [20:31:31] aha, that last bit is very imformative, thanks. I have to do other stuff now but will be back in a bit [20:31:40] (and look at those errors) [20:31:56] *super grateful for the input* [20:32:24] well I'd suggest either 1) using that script to change password, or 2) use the resetUserEmail maintenance script to set their email address so they can use the regular password reset feature [20:33:08] if maintenance scripts really aren't working for you for whatever reason, you can also set the email and mark it as confirmed via a direct db query [20:44:37] Hi, I'm trying to export some tables in the DB, but its seems they encoded with "latin1_swedish_ci" which makes me hard struggle, there is option to query the db from webapi (or any other way to get the details)? [20:52:17] The api allows to query the most usual data [20:52:22] !api [20:52:22] The MediaWiki API provides direct, high-level access to the data contained in the MediaWiki databases. Client programs should be able to use the API to login, get data, and post changes. Find out more at < https://www.mediawiki.org/wiki/API >. For client libraries in various languages, see < https://www.mediawiki.org/wiki/API:Client_Code >. [21:11:21] I've been doing a bit of research on a common auth database for users [21:11:56] we have a mediawiki instance deployed and right now it's using it's own login db [21:12:43] we have a user db (about 20k users) in a mysql db, and would like to have mediawiki dip into this for unknown users [21:13:14] we're not needing SSO or anything, just want to keep the users in the other db [21:13:28] is there a good extention I should be looking at? [21:19:50] DarkHelmet: I don't see anything like that currently, you'll probably need to write your own extension [21:21:01] it's in the shit [21:21:26] whoops, wrong channel, I should say that's not great [21:21:48] was hoping it was just like easy to get it to use a standard auth database [21:22:09] DarkHelmet: is this other DB another mediawiki install? Of just some random table of user data? [21:22:46] just a random table [21:22:59] we could (and do) have LDAP spun up on it [21:23:04] if we can do LDAP [21:23:13] yeah, there's an ldap extension [21:23:42] https://www.mediawiki.org/wiki/Extension:LDAP_Authentication [21:23:44] I have RT for ticketing using the db directly now, it was easy [21:24:16] another route is a custom PrimaryAuthenticationProvider -- https://www.mediawiki.org/wiki/Manual:SessionManager_and_AuthManager [21:24:18] Im talking about cargo data tables, that exposed in the api also? [21:25:12] maybe? check your wiki's api.php page (you can visit it with your browser); it automatically generates documentation for everything it exposes [21:25:57] looks like the least painfull way is going to be LDAP [21:26:38] DarkHelmet: there are a ton of auth related extensions -- https://www.mediawiki.org/wiki/Category:User_identity_extensions -- https://www.mediawiki.org/wiki/Extension:PluggableAuth is one that is kind of a swiss army knife of hooking MediaWiki to external auth systems [21:26:44] :O Great! ty [21:27:06] thanks bd808 [21:43:29] maybe we should have a GSOC project for writing a bunch of well documented wrappers/boilerplates for all the common extension requests [21:44:35] I want to use my own user database, I want to use my own SSO, I want to get the username for the request from an external source, that kind of thing [21:45:24] the main problem with AuthManager is that you have to understand all its concepts to know where to extend, even though extending it is not actually hard [21:45:44] and those concepts are a lot less intuitive than AuthPlugin was [21:47:54] Well said, Tgr [22:04:24] I don't need SSO [22:04:50] really don't care if they need to login with user/pass to the wiki before they can edit [22:05:16] SSO would be cool, but it's beyond our org at this point [22:05:30] perfect is the enenmy of good :) [22:16:00] I'd love SSO, but it'd mean coordinating with far-flung parts of the company, which isn't something we tend to be very good at [22:28:15] how do i have any script on maintenance dir to have to specify a Localsettings file? [22:29:45] tgr: good idea [22:29:50] is it easy to extend?