[03:21:57] <coder9999>	 Hi
[03:28:42] <coder999917>	 Hello, We are a medium to large sized company who calls Wikipedia public APIs.  We have a Java application calling Wikipedia APIs. We am using the API endpoints with params, from both US and EU regions.Since the last 2-3 months, We have seen a lot of intermittent latencies and timeouts from the Wikipedia APIs being used and was wondering if there
[03:28:43] <coder999917>	 is something which can be done to reduce them. What is the official channel for our company to communicate with WikiMedia Support Services since we have more than a few million users on our system?
[07:11:00] <chek_>	 is there some quick way to just block a user and revert all of their contributions
[07:13:08] <DannyS712>	 not really, but there is rollback, and some scripts that use activate all rollback links on a user's contribs pages
[07:25:16] <chek_>	 ok so if I have a list of usernames which are all just registered bots that created spam pages what can I do to fix this without going to each contribution page and hitting revert and selecting a reason
[07:26:12] <chek_>	 I imagine this spam flood is very common on small wikis
[07:28:31] <chek_>	 I mean at some point it's easier for me to just go into sql and purge these if there isn't just an easy "obliterate this user and all of their contributions" button
[09:50:26] <RhinosF1>	 api.php?action=query&meta=oath&oathuser=<username> allows you to check whether someone has 2FA on, correct?
[10:09:03] <RhinosF1>	 ZI_Jony: do you know if api.php?action=query&meta=oath&oathuser=<username> definitely allows you to see if 2FA is on for someone else?
[10:11:20] <ZI_Jony>	 RhinosF1: I don't think
[10:11:39] <RhinosF1>	 ZI_Jony: me checking it seems to indicate so
[10:12:26] <RhinosF1>	 but I can't be sure
[10:12:28] <DannyS712>	 @RhinosF1 can you send me a link of an api query that worked for you?
[10:12:58] <RhinosF1>	 DannyS712: it's not WMF Prod
[10:13:12] <RhinosF1>	 but if you're meta sysop, you are?, it might work
[10:13:17] <DannyS712>	 oh, then it depends on what rights you have
[10:13:36] <DannyS712>	 the rights needed aren't given to anyone other than staff (and maybe stewards?)
[10:16:09] <Majavah>	 Stewards have it, not sure about staff
[10:16:10] <ZI_Jony>	 I know olny CA team of WMF can see, not all staff.
[10:18:35] <ZI_Jony>	 RhinosF1: what happen? any problem?
[10:19:03] <RhinosF1>	 ZI_Jony: not a wmf issue
[10:19:42] <ZI_Jony>	 RhinosF1: miraheze?
[11:50:11] <wolfgang>	 Hi! I have at present problems logging in on https://www.mediawiki.org/ with my firefox browser. Message is "There seems to be a problem with your login session; this action has been canceled as a precaution against session hijacking. Please resubmit the form."
[11:50:34] <wolfgang>	 With chromium browser the login works ok.
[11:50:55] <wolfgang>	 Anyone knows what this is?
[11:54:35] <wolfgang>	 Ok. Seemed to be some strange caching promblem. Solved. Bye.
[13:38:19] <kfogel>	 I hope this channel is an appropriate place for a MediaWiki-related open source job posting.  https://www.fossjobs.net/job/10221/phppython-programmer-at-open-tech-strategies-llc/
[13:40:46] <Reedy>	 Not if you keep spamming them
[13:45:02] <kfogel>	 Reedy: Tone is hard to tell in IRC, so I'll just ask directly if you were referring to the fact that I posted that last night.
[13:45:58] <Reedy>	 Yes
[13:46:22] <kfogel>	 Reedy: In channels I manage, posting the same posting 12 hours apart (when it's a low traffic channel) is just a normal technique for covering all time zones, since different people are on at different times.  However, if policies are different here, I'll gladly abide by them.  I did look for channel guidance in the chan topic about this, but didn't find any.
[13:46:49] <Reedy>	 https://www.mediawiki.org/wiki/Jobs
[13:46:54] <kfogel>	 Reedy: Thank you.
[13:46:55] <Reedy>	 >Within reason, it is acceptable to post MediaWiki-related job openings on the mediawiki-l mailing list. 
[13:47:14] <kfogel>	 Reedy: That's very helpful -- much appreciated.
[16:23:44] <cronolio>	 hello, can you describe me why a lot <translate> tags is required ?
[16:24:55] <cronolio>	 i mean 
[16:25:04] <cronolio>	 blabla.</translate>
[16:25:04] <cronolio>	 <translate>
[16:25:05] <cronolio>	 <!--
[16:28:43] <Vulpix>	 cronolio: Translate tags automatically split units when there are double newlines. However, a single paragraph with multiple sentences is not split, and requires manual split
[16:30:58] <cronolio>	 as i know my example before can be replaced with
[16:30:58] <cronolio>	 blabla.
[16:30:58] <cronolio>	 <!--
[16:31:07] <cronolio>	 ah
[16:31:25] <cronolio>	 just one new line will be enough i think
[16:33:23] <Vulpix>	 Nope. Only one line doesn't split the translation unit
[16:33:35] <Vulpix>	 You can have an entire bullet list inside a single translation unit
[16:35:47] <Vulpix>	 If you mean removing the extra <translate> tags but preserving the <!-- T:NN --> markers even if there's no double newline separation between them, I'm not sure if that's supported
[16:37:29] <cronolio>	 Vulpix: https://wiki.gentoo.org/index.php?title=NVIDIA/nvidia-drivers&action=edit there is only one translate tag
[16:38:42] <Vulpix>	 Yes, and this is a poorly way of making a page translatable
[16:39:13] <cronolio>	 only one new empty line is requered. and on marking extension adds new one with <!-- T....
[16:39:30] <Vulpix>	 You don't want translators to touch the templates, but that page is including templates inside translation units
[16:40:02] <Vulpix>	 Also you don't want translators to translate long paragraphs with several sentences. That page has several of the,
[16:40:05] <Vulpix>	 *them
[16:40:58] <cronolio>	 this is other question about avoiding templates and code
[16:41:14] <Vulpix>	 Have you tried to translate that page? Please try it. That would be a very unpleasant experience
[16:41:24] <cronolio>	 spliting long tables.. yes it is minux of gentoo wiki
[16:41:36] <cronolio>	 *minus
[16:42:18] <Vulpix>	 There's a way to avoid templates in translations: </translate>{{template|param1=<translate>parameter value that needs to be translatable</translate>}}<translate>
[16:43:31] <cronolio>	 let
[16:44:40] <cronolio>	 lets talk about https://www.mediawiki.org/w/index.php?title=Gerrit/Tutorial&action=edit there is a lot 
[16:44:40] <cronolio>	 </translate>
[16:44:41] <cronolio>	 <translate>
[16:44:41] <cronolio>	 and i still don't understand why it is required :)
[16:45:45] <Vulpix>	 Some of them are not needed
[16:47:14] <Vulpix>	 I'd say, the user that made the last edit to that page doesn't know how to properly tag a page with translate tags :)
[16:48:00] <Vulpix>	 That was a blind "wrap everything that's not between <translate> tags already in <translate> tags"
[16:48:22] <cronolio>	 yes, i do trying to fix it and also understand the style of marking which i'd describe before
[16:50:46] <ProcReader>	 who can I request a rename from (on phab)? due to typo
[16:50:52] <ProcReader>	 (missing a 't')
[16:51:14] <Reedy>	 File a task, I think
[16:52:28] <Majavah>	 for user renames, file a task in https://phabricator.wikimedia.org/project/view/5/
[16:54:14] <ProcReader>	 thanks, any particular board?
[16:54:27] <ProcReader>	 or tags
[16:59:01] <ProcReader>	 ah, okay
[16:59:03] <ProcReader>	 got it
[17:01:03] <Vulpix>	 cronolio: you'd usually want to split sentences inside a paragraph in its own units. That would prevent a large chunk of text to get invalidated if only one sentence is modified
[17:36:38] <RSY>	 Hi, I'm trying to install MediaWiki just to run on my browser on my local computer, as in, the only people who can access the site are those with access to my computer.
[17:36:53] <cronolio>	 Vulpix: https://www.mediawiki.org/w/index.php?title=Gerrit%2FTutorial&type=revision&diff=3913139&oldid=3913036 what do you think ?
[17:38:12] <cronolio>	 wiyhout slitting big pharagrafs yet, bet removed a lot of translate tags
[17:38:22] <RSY>	 at index.php I see "If you are reading this in your web browser, your server is probably * not configured correctly to run PHP applications!" But I don't know anything about setting up PHP
[17:42:26] <Reedy>	 RSY: There's plenty of guides online for it
[18:00:28] <kfogel>	 RSY: This often means that the PHP modules are not enabled for your webserver.  Turning them on (or obtaining them and then turning them on) is usually pretty easy.  Reedy is right: if you put your error message into a search engine, you will get plenty of results that show how to fix this.
[18:02:37] <Vulpix>	 cronolio: looks a lot better! Some inline templates and URLs should be put inside <tvar> though
[18:04:30] <Vulpix>	 Shirayuki will probably fix that soon, since he usually do a lot of translation work
[18:04:44] <Vulpix>	 I'd fix that myself but I can't do that right now
[18:07:59] <RSY>	 There are a lot of complications, I've been trying to set up Apache on its own just to work for the past hour, without even getting the "It works!" screen to pop up. I'm on a Mac
[18:08:15] <RSY>	 DocumentRoot "/Users/me/mediawiki-1.34.1"
[18:08:17] <RSY>	 I set it to that
[18:08:55] <RSY>	 and set Directory Index to index.html (a dummy HTML file I created) and localhost still comes up as nothing
[18:09:24] <RSY>	 When I type command sudo apachectl start, it says "/System/Library/LaunchDaemons/org.apache.httpd.plist: service already loaded"
[18:11:17] <RSY>	 and I have been following several of the tutorials
[18:12:46] <Reedy>	 Did you restart apache after changing paths
[18:13:08] <Reedy>	 /other apache config
[18:40:27] <poomba>	 dude what the fuck
[18:40:27] <poomba>	 i don't even want to troll anymore
[18:40:28] <poomba>	 Reception123 showed me the way
[18:40:28] <poomba>	 i don't even want to troll anymore
[18:40:29] <poomba>	 so I am just goign to lurk
[18:40:29] <poomba>	 in ##gognurk
[18:40:30] <poomba>	 with all my wikipedian friends
[18:43:38] <apergos>	 thanks for that
[18:44:15] <Reedy>	 they're ccycling through most of the channels it seems
[18:45:10] <apergos>	 meh
[19:03:35] <gagoomba>	 dude what the fuck
[19:03:35] <gagoomba>	 i don't even want to troll anymore
[19:03:36] <gagoomba>	 Reception123 showed me the way
[19:03:36] <gagoomba>	 i don't even want to troll anymore
[19:03:37] <gagoomba>	 so I am just goign to lurk
[19:03:37] <gagoomba>	 in ##gognurk
[19:03:38] <gagoomba>	 with all my wikipedian friends
[19:03:39] <gagoomba>	 :)
[21:21:31] <tumbleweed3>	 Hey! I want to make a template that shows some code passed in an argument. how do evaluate the {{{arg}}} but not the code contained?
[21:24:14] <tumbleweed3>	 e.g. {{ShowCode|arg={{PAGENAME}} }}
[21:24:29] <bd808>	 tumbleweed3: I think I might understand your question, but let me try rephrasing. You want a template with an argument that is itself wikitext and you would like that wikitext to remain unparsed?
[21:25:16] <tumbleweed3>	 yea. in my example I can get it to show {{{arg}}} or I can get it to show the pages name. but i cant get {{PAGENAME}}
[21:25:40] <tumbleweed3>	 i guess showing it unparsed is what i want
[21:26:19] <bd808>	 I went down this rabbit hole a few weeks ago myself. :)
[21:26:39] <tumbleweed3>	 any insight ?
[21:26:45] <tumbleweed3>	 or just rabbits down there?
[21:27:04] <bd808>	 There does not seem to be a way to do it without the template caller being in charge. They can wrap the argument in <nowiki></nowiki>
[21:27:24] <bd808>	 some notes about what I tried at https://wikitech.wikimedia.org/wiki/User:BryanDavis/Sandbox/Codesample
[21:28:27] <tumbleweed3>	 hmm dont get the caller being in charge part, but ill check out your code thank!
[21:29:07] <bd808>	 the caller being in charge means that "{{ShowCode|arg=<nowiki>{{PAGENAME}}</nowiki> }}" is needed.
[21:30:24] <bd808>	 the MediaWiki parser does not currently have a way to tell it "arg 2 of this template should be treated as raw text" which is what you would need to avoid the literal <nowiki/> wrapper when calling the template
[21:34:20] <tumbleweed3>	 hmm i see how passing <nowiki> works
[21:34:48] <tumbleweed3>	 but then i cannot evaluate the code later in the template can I
[21:35:00] <tumbleweed3>	 can i somehow unwrap it?
[21:35:06] <tumbleweed3>	 or do i have to pass it twice?
[21:35:40] <Platonides>	 you might be able to use something like {{#tag:poem|{{{arg}}} }}
[21:38:10] <Platonides>	 hmm, not really
[21:38:27] <Platonides>	 I think the arg may be processed before being given to the template
[21:39:33] <tumbleweed3>	 So with passing the arg containing code, i cannot get something like "{{PAGENAME}} evaluates to: 'This pages name' " ?
[21:39:48] <tumbleweed3>	 where i get both the code and the evaluated code
[21:40:01] <tumbleweed3>	 from {{ShowCode|arg={{PAGENAME}} }}
[21:41:03] <Krinkle>	 tumbleweed3: you need to pass it twice to do that, or you need to make it so that the template is in control
[21:41:29] <Platonides>	 you could do {{ShowCode|arg=PAGENAME }} and then in ShowCode {{ {{{arg }}} }}
[21:41:41] <Krinkle>	 tumbleweed3: in control can be done for example via a^^
[21:41:45] <Krinkle>	 exactly
[21:42:24] <Platonides>	 Arg is <nowiki>{{</nowiki>{{{arg}}}<nowiki>}}</nowiki> : {{ {{{arg}}} }}
[21:42:41] <Platonides>	 it's quite limited in that you must be passing templates/magic words, of course
[21:46:38] <tumbleweed3>	 ok thanks! i was going to use it debug my cargo queries, by seeing the code alongside the evaluation
[21:46:48] <tumbleweed3>	 but i guess its  a long shot
[21:47:05] <tumbleweed3>	 ill just have to type it out twice
[22:04:09] <Krinkle>	 subbu: wpt* no longer have separate parsoid deployments, is that right?
[22:04:37] <Krinkle>	 I'm noticing that a few deprecation regressions were missed because Parsoids log messages don't go to type:mediawiki
[22:04:46] <Krinkle>	 with Parsoid now bundled as libary, might make sense to change
[22:19:24] <subbu>	 Krinkle, that is right. But, we caught the deprecations as part of rt testing. That change is riding this week's train. https://www.mediawiki.org/wiki/Parsoid/Deployments#June_16_-_June_18:_(in_progress)_v0.12.0-a17_as_part_of_1.35.0-wmf.37
[22:25:13] <Krinkle>	 subbu: I filed https://phabricator.wikimedia.org/T255627
[22:25:58] <Krinkle>	 subbu: right, but the rest of engineering doesn't know to check that.  loses vertical support you could benefit from, and posibley more important, means people are sometimes making misinformed risk calculations, not realising they've also just deployed parsoid 
[22:26:18] <Krinkle>	 e.g. wmf-config changes coudl break parsoid and I wouldn't know until health checks fire (if it's bad enough)