[01:22:03] <TennaFox>	 Is it possible to create a Wiki link containing an equals sign and/or question mark, that doesn't "encode" itself to something like %3F? The fact that it's doing so is messing with some page links.
[01:31:16] <SamB>	 TennaFox: say what?
[01:31:22] <SamB>	 how is it messing with page links?
[01:55:46] <TennaFox>	 SamB: An extension uses URLs like wiki/Special:ect?id=400. If you try to link to that normally [[Special:ect?id=400]], it becoames Special:ect%3Fid%3D400
[01:56:08] <SamB>	 TennaFox: ah
[01:56:21] <SamB>	 TennaFox: there's a magicword or something for that
[02:02:24] <rohit-dua>	 i..  all tpb uploads in archive.org do not have a direct link to pdf download. rather they link to google.com PDF. why does archive not provide direct PDF downloads for these google-books uploads..?
[02:04:06] <Hello71>	 how would we know?
[02:04:58] <rohit-dua>	 https://archive.org/details/charlesdarwinsw03darwgoog  PDF has a link to google-books
[02:17:15] <TennaFox>	 SamB: fullurl *kind of" does that, only it displays the full URL as implied.
[02:17:34] <SamB>	 TennaFox: I meant, to pass parameters
[02:18:13] <SamB>	 though it might not use wiki/
[02:19:04] <SamB>	 rohit-dua: did you look in the actual download directory?
[02:22:07] <SamB>	 rohit-dua: also maybe ask in #wikisource
[02:23:04] <SamB>	 I suspect they are more familiar with coaxing things out of archive.org than this channel is ...
[02:23:54] <Quasar`>	 is there supposed to be a very long pause at the end of running refreshLinks.php
[02:27:40] <TennaFox>	 SamB: I simply cannot find anything of the sort, I'll keep searching though I'm not sure if such a thing exists.
[02:30:33] <Quasar`>	 or rather is it supposed to just stop responding midway through the links table and allow me to type things at the keyboard
[02:30:43] <Quasar`>	 it doesn't seem the script has terminated but it's not doing anything either
[02:40:42] <Quasar`>	 I opened another terminal and I see php in top using 2.6% memory but 0% CPU
[02:40:45] <Quasar`>	 so it's clearly locked up.
[02:41:07] <Quasar`>	 what do I do, is it safe to terminate?
[02:41:08] <SamB>	 TennaFox: oh, I think you're supposed to put the fullurl invocation inside the usual link markup
[02:41:22] <SamB>	 where the URL would normally go
[02:47:04] <Quasar`>	 I re-ran it starting from the id it ended at and it has locked up again
[02:47:09] <Quasar`>	 does this mean our database is corrupt?
[02:52:52] <SamB>	 Quasar`: I don't know; have you done anything to try and see what it was trying to do when you killed it?
[02:53:06] <Quasar`>	 how would I do that
[02:53:53] <Quasar`>	 it says "Refresing links tables."  "Starting from page_id of 8300 of 17966"
[02:53:56] <Quasar`>	 and then 8300
[02:53:58] <Quasar`>	 and then nothing.
[02:54:02] <Quasar`>	 just locks up.
[02:54:49] <legoktm>	 open up mysql and run `show processlist` to see if any queries are being made?
[02:56:10] <Quasar`>	 Access denied
[02:56:20] <Quasar`>	 I apparently don't have rights to run it from the command line.
[02:56:23] <legoktm>	 huh
[02:56:27] <legoktm>	 run as root? :P
[02:56:34] <Quasar`>	 don't have root.
[02:56:40] <Quasar`>	 don't have su privs either
[02:58:44] <legoktm>	 :/
[02:59:09] <Quasar`>	 oh I need to provide -p apparently
[03:00:21] <Quasar`>	 show processlist shows two processes
[03:00:34] <Quasar`>	 aside from Query, the other command is "Sleep"
[03:00:50] <Quasar`>	 with Info "NULL"
[03:02:17] <legoktm>	 well, it shows its making a query
[03:02:31] <Quasar`>	 the Query info is "show processlist"
[03:02:58] <Quasar`>	 running it again I see the Time on the sleep increase and then eventually go back down but at no time have I seen an actual SQL query running
[03:03:58] <Quasar`>	 oh wait here we go
[03:04:00] <Quasar`>	 I caught one
[03:04:24] <Quasar`>	  SELECT /* Revision::loadText 208.43.252.200 */  old_text,old_flags  FROM `text`   WHERE old_id = '27
[03:10:00] <legoktm>	 Quasar`: so, it's just running really slowly
[03:23:50] <Quasar`>	 I see it has finished while I was looking away.
[03:24:59] <legoktm>	 :)
[03:28:51] <Quasar`>	 hmm
[03:29:34] <Quasar`>	 our page table doesn't contain any of the columns that FlaggedRevs is supposed to have created
[03:29:56] <legoktm>	 flaggedrevs doesn't alter the page table, it has it's own tables
[03:30:01] <legoktm>	 its*
[03:30:07] <Quasar`>	 so the page is outdated then
[03:30:17] <Quasar`>	 the doc page that is
[03:30:41] <Quasar`>	 do you know a lot about it? because we're currently having major problems with it.
[03:31:26] <Quasar`>	 http://doomwiki.org/w/index.php?title=Automap&action=history   <- I edited this page as an anon to trigger the issue. The information here would have you believe the page has not been reviewed since 11/24/11
[03:31:51] <legoktm>	 I don't
[03:32:06] <Quasar`>	 if I go to http://doomwiki.org/w/index.php?title=Special:ReviewedVersions&page=Automap then I can clearly see that the last reviewed revision was dated 11/21/14 
[03:32:08] <Quasar`>	 I see.
[03:32:10] <Quasar`>	 well 
[03:32:15] <Quasar`>	 we're about at our wit's end with this
[03:32:51] <Quasar`>	 every time somebody without reviewer permissions edits a page, we lose two to three years' worth of reviewed revisions off the stable revision
[03:33:06] <Quasar`>	 and this just started last week; up til then it had been working fine
[03:33:11] <Quasar`>	 so I don't understand it at all
[03:33:19] <Quasar`>	 there were no config changes made during that period
[03:41:04] <Quasar`>	 ok
[03:41:10] <Quasar`>	 I see a pattern, in the flaggedrevs table
[03:41:23] <Quasar`>	 the one it has selected is the last one with a valid fr_rev_timestamp
[03:41:31] <Quasar`>	 for some reason, everything newer than that has nothing in the field.
[03:41:37] <Quasar`>	 not even the word NULL, just, nothing
[03:42:26] <legoktm>	 empty string then
[03:43:36] <Quasar`>	 I could always update the table and set fr_rev_timestamp = fr_timestamp where fr_timestamp is null but that won't fix the reason it's not populating that field to begin with :/
[03:56:57] <Quasar`>	 when it's inserting it just calls Revision::getTimestamp() to get the value that should be in there...
[04:22:06] <Quasar`>	 apparently this is a problem with our historical data
[04:22:18] <Quasar`>	 newer revisions all have a valid fr_rev_timestamp
[04:22:30] <Quasar`>	 looks like they started populating again right about when we upgraded to MediaWiki 1.22.5
[04:22:55] <Quasar`>	 so I probably do just need to fix the table but I don't know enough mysql to know if I'm doing it right.
[04:23:27] <Quasar`>	 I know plenty of Firebird PSQL and Microsoft T-SQL but every time I think that knowledge is useful against MySQL I get slapped in the face.
[05:27:13] <Quasar`>	 got it.
[05:27:16] <Quasar`>	 I think I fixed it :>
[05:29:05] <legoktm>	 yay!
[06:14:29] <Djain>	 Hi!
[06:15:05] <Djain>	 How do I ask for a submit permission on gerrit for BookManagerv2 extension?
[06:15:29] <SamB>	 Djain: you need permission just to *submit*?
[06:15:37] <Djain>	 yes
[06:16:03] <Djain>	 submit as in submit changes to merge
[06:16:18] <SamB>	 I'm guessing either they screwed up their configuration, or they don't like help
[06:17:36] <Djain>	 SamB: Sorry, I didn't get you.
[06:18:05] <Djain>	 What configuration are you talking about?
[06:18:32] <SamB>	 Djain: presumably it's possible to set things up so anyone can submit stuff for review
[06:20:00] <Djain>	 and how do we merge those changes to master after the review?
[06:20:50] <Djain>	 Right now, only I am working on BookManagerv2 with my mentor
[06:21:09] <Djain>	 can I get permission to merge my changes?
[06:21:36] <SamB>	 Djain: I don't know; I would ask your mentor
[06:22:23] <Djain>	 Oh, actually Raylton himself has asked me to ask for permission on #mediawiki
[06:23:02] <p858snake|l>	 file a bug in bugzilla under Wikimedia -> Git/Gerrit is the easiest way
[06:24:35] <legoktm>	 er
[06:24:42] <legoktm>	 Djain: create a request on https://www.mediawiki.org/wiki/Gerrit/Project_ownership
[06:25:04] <Djain>	 Okay
[10:37:01] <okino>	 Hi, all.
[10:37:09] <okino>	 I just have one question.
[10:37:36] <okino>	 Can MW itself list all pages that include galleries created with a <gallery> tag?
[10:37:38] <okino>	 Thanks.
[10:38:57] <Vulpix>	 okino: no, that's not possible
[10:39:57] <okino>	 Vuplix: Well, than I have to use other tools :-) Thanks a lot for answer and have a nice day.
[10:40:14] <okino>	 uups, sorry, Vulpix :-) Blame my fingers :-)
[10:43:37] <Vulpix>	 no problem! :)
[14:49:42] <okino>	 Hi all again today :-)
[14:51:06] <okino>	 One question again: Does Mediawiki list somehow and somewhere pages, where the defaultsort is duplicate? The pages create the "MediaWiki:Duplicate-defaultsort" warning that can be found on the page, but can I find a list of pages with the problem?
[14:51:07] <okino>	 Thanks.
[15:02:35] <Krenair>	 okino, I don't believe so, no
[15:02:42] <Krenair>	 Only the overriding value is stored in the DB
[15:05:17] <okino>	 Well, I've got an idea - when asking and thinking about the answer and about how to put the question better. So thank you, in fact you helped me :-)
[15:05:21] <okino>	 Have a nice day.
[18:30:10] <Nemo_bis>	 How cute, wikihow now runs on a very recent release, MediaWiki 1.23wmf15
[18:31:10] <Krenair>	 eh
[18:31:33] <Krenair>	 If they're going to run a wmf branch I would heavily advise they keep up with the latest on enwiki at least
[18:32:42] <Krenair>	 wikipedia* even. (enwiki used to get them at a slightly different time)
[18:33:58] <Krenair>	 Nemo_bis, but yeah, 1.23 has not been used in wikimedia production since 06/03/14
[18:34:10] <Krenair>	 1.23wmf15*
[18:43:24] <Nemo_bis>	 They used to run years-old releases, I don't see how few weeks can matter :)
[18:46:29] <Krenair>	 Hmm... Maybe it's okay.
[18:47:09] <Krenair>	 I'm not quite sure how we handle security issues that don't affect tarball versions though. Obvious they get deployed to the Wikimedia cluster, and fixed on master
[18:59:50] <bawolff>	 I doubt there's that many security issues
[19:00:04] <bawolff>	 especially comparatively to what they used to run
[19:33:01] <Betacommand>	 bawolff: did you see Anomie's comment?
[19:33:21] <bawolff>	 About the api change? yes. I'm planning to address it later today
[19:33:31] <Betacommand>	 bawolff: thanks