[00:26:54] Hello! [00:28:18] If I have an old database on my server, and I have a .sql file of the new database, do I need to delete the database or can I just import the data with < operator? [00:29:16] So I have wikidb on the server with old content and new content should be added from .sql file. Does the wikidb need to be removed and established before importing the new content? [00:36:36] anyone here know how to fix the sidebar width bug for IE? [00:50:23] 03(NEW) Allow restriction of access to Special:Code - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=17984 enhancement; Normal; MediaWiki extensions: CodeReview; (stwalkerster) [01:08:31] 03(mod) Editnotices generating invalid XML (HTML) around unordered lists - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=17486 (10Amalthea.wikimedia) [01:35:11] Hi all, I was wondering if anybody could give me help on making a template. Currently, I have all "Room" articles (for example, Room 105) include the template {{rooms}}, which, as part of its function, puts it in the category Rooms. I instead want it to put it in the category Rooms, but indexed by number (that is, the title of the page without "Room"). I understand it is something like... [01:35:12] ...[[Category:Rooms|105]], but how do you automate this as part of the template? [01:35:47] can't you pass the number into the rooms template as a parameter? [01:35:52] if you have Extension:StringFunctions you can do it [01:35:57] otherwise just pass it manually [01:36:24] If I do have StringFunctions, how can I accomplish this? [01:36:43] Substrings? [01:36:57] yes [01:37:09] assuming all pages have a well-formed title [01:37:17] (which you could check) [01:37:18] I could check if it is valid [01:37:24] yup :) [01:37:29] Right, thanks :) I'll come back if I need any more help. [02:00:07] 03(mod) Change default locale for {{formatnum:}} parser function - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=17659 (10OnTheWhite) [02:09:07] hi guys.. could someone here tell me why the blue boxes with the dashed outlines here run off the page and make the browser scroll sidways? --- http://wiki.club-ubuntu.org/index.php/Podcast_Episode_0.1 [02:09:41] they have whitespace: preserve [02:09:47] so they will not automatically wrap [02:09:53] huh? [02:09:56] in CSS [02:10:01] well is there a better way to format then? [02:10:07] are you just wanting an indent? [02:10:10] emma, don't start lines with a space unless you want that format. [02:10:10] so that it accomplishes a similar effect? [02:10:13] in which case start the line with : [02:10:25] Id like to somehow 'set that stuff off' but i don't need it to be in a blue box like that. just something. [02:10:55] probably easiest to start the line with a colon, unless you need more [02:11:11] (in which case we can maybe help you create a template) [02:12:23] i'll see what putting a colon there does. [02:13:19] cirwin: not too bad, is there a way to color or bold it or anything? [02:13:22] You can do most HTML/CSS styling that you want directly. E.g.,
Questions: Blah blah blah ...
[02:13:42] yeah, like simetrical says [02:13:46] Bold is '''bolded text''', color you have to use CSS for (...) or whatever. [02:13:47] okay [02:14:08] if it's not asking too much could you recommend a good
for that? [02:14:25] depnds what you want [02:14:32] I just gave one, but it's probably too tacky. [02:14:33] try simetricals and then change it 'til it lokks nice [02:17:39] i like it.. thank you. [02:20:44] Take a look - http://wiki.club-ubuntu.org/index.php/Podcast_Episode_0.1 [02:20:48] I think that looks pretty good [02:20:54] thanks cirwin and Simetrical [03:28:16] I have found a bug in mediawiki [03:28:32] At the very least in the diff tool, if not even something weirder... [03:29:04] http://en.wikiquote.org/w/index.php?title=Finnish_proverbs&diff=next&oldid=903244 [03:29:09] !bugzilla | cimon [03:29:09] --mwbot-- cimon: All bugs in MediaWiki should be reported at https://bugzilla.wikimedia.org. This is also the place to request site configuration changes, new features or enhancements to existing features, although bear the following points in mind before making a feature/enhancement request: 1) If the request is specific to a Wikimedia wiki, please discuss the issue on that wiki first. 2) Consider whether a custom extension would be more appropriate. [03:29:20] Look at the bottom of the page. [03:30:28] https://bugzilla.wikimedia.org/buglist.cgi?query_format=advanced&short_desc_type=allwordssubstr&short_desc=&product=MediaWiki&bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&bug_severity=blocker&bug_severity=critical&bug_severity=major&bug_severity=normal&bug_severity=minor&bug_severity=trivial [03:30:30] even though there is no indication in the diff of anything like that happening, the article page has mystically been truncated at the sayings beginning with the letter "O" [03:30:42] 898 open non-enhancement bugs in component "MediaWiki" [03:32:56] unclosed comment, cimon [03:33:14] [03:34:30] cool, at leas t Ikno how to fix [03:34:38] that breaks the rest of the page [03:34:44] will fix [03:34:45] so, not a bug ^_^ [03:34:53] but good to find these annoying things [03:35:01] just very hard to catch on the diff [03:35:32] should have stared at it harder I guess, before jumping into conclusions. [04:16:43] anyone here know how to fix the sidebar width bug for IE? [04:18:10] Hello, I just took a 1.12 installation to 1.13.5 and I am getting the following error: Fatal error: Call to a member function matchStartToEnd() on a non-object in /webroot//includes/parser/Parser.php on line 2788 [04:18:53] I am not sure what is wrong, as it seemed to be working in the test environment [04:38:53] 03(NEW) Gzipped API results - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=17985 enhancement; Normal; MediaWiki: API; (innocentkiller) [04:43:45] eghjaytee: did you run the update script? [04:55:32] 03(mod) Gzipped API results - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=17985 +comment (10b-jorsch) [04:59:11] is there a way to get MediaWiki to display GIMP's xcf format directly? [05:02:17] Afker: https://bugzilla.wikimedia.org/show_bug.cgi?id=17959 [05:08:19] thanks Splarka [05:32:35] Anyone with commit access up? I'd like to poke you guys again about https://bugzilla.wikimedia.org/show_bug.cgi?id=17537 [05:34:15] 04(REOPENED) Opening a Wikipedia article results in attempt to download gzipped version - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=7098 +comment (10edwardzyang) [05:35:16] 03(mod) Turn off zlib.output_compression for 304 - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=17537 (10edwardzyang) [06:45:43] Anyone around? [06:47:59] I need a developer! [06:49:14] domas: Haha [06:49:21] <^demon> domas: You do? [06:49:36] thats PHP problem, not mediawiki [06:49:42] Oh yeah [06:49:44] we don't call zlib.output_compression! [06:49:46] I posted a fix to PHP too [06:49:48] *domas checks more :) [06:49:54] No one wanted to look at it [06:50:09] ^demon: dunno. what do you think about 'recursive' protection on subpages? [06:50:24] for some special, e.g. project, namespace? [06:51:18] For shallow hierarchies, it doesn't seem like that would be too much performance penalty [06:51:39] ezyang: indeed. and it is quite difficult to make deep hierarchy in 255 bytes [06:52:16] Meh, you could totaly make 122 levels [06:52:25] *ezyang can't do math [06:52:42] thats shallow! :) [06:52:52] anyway, you know whole tree from the URL [06:52:53] <^demon> domas: I feel the same way I do about recursive redirects. Useless :p [06:53:13] ^demon: well, stuff could be used for things like Editnotices and other page metadata used by templates [06:54:46] so can a combination of title blacklist and page protection? [06:54:52] (sorta) [06:55:11] ghm, you're right :)) [06:55:33] sometimes Splarka gives good stuff. [06:55:40] shhhh! don't tell my ego [06:55:42] but it wasn't my idea [06:55:47] they're discussing it on the VPT [06:56:00] (in regard to edit notices anyway) [06:58:21] =) [06:58:23] good! [06:58:39] good to have reasonable discussions [06:59:31] I dunno... an #ifexist call for every edit page load... [06:59:46] message cache points at parser and laughs [07:00:54] well, thats better than polluting message cache [07:00:59] #ifexist is not _that_ expensive [07:01:03] compared with alternatives [07:01:25] look at it the other way, we constantly do 'ifexist' for every red/blue link out there [07:01:34] the method is different, but not much [07:01:42] well, those at least get cached [07:01:53] where? [07:02:02] but it is true, every time you edit a page, you are gonna ruin the cache anyway, heh [07:02:14] where are those cached? [07:02:16] on the inky things, the squids [07:02:24] every page view doesn't #ifexist every red link [07:02:38] every page view hitting the backend does ;-) [07:03:12] but you can't cache edit pages... I hope... [07:03:44] hmm, another alternative is javascript/ajaxy, but then that gives an extra hit for every edit page load [07:04:03] so someone has to suffer: parser, message cache, or api [07:05:56] so domas: is the #ifexist your fav solution? [07:05:59] Splarka: parser costs for us because people have 100k edges [07:06:03] other than "live without it" [07:06:29] Splarka: the ifexists/etc would cost us what, 10ms? 5ms? [07:06:50] I don't think in terms of milliseconds, I think in terms of domas-curses-per-minute [07:06:55] haha [07:07:02] lol [07:07:08] I'm a nice guy! [07:07:14] I like reasonable solutions! :) [07:07:37] the only reason your name is listed as one of the top contributors to MediaWiki is because you remove features that cost too much [07:07:51] http://img.photobucket.com/albums/v358/AlexClarke/Starfighter/sarcasm_detector.jpg *BZZ* [07:07:54] *domas checks http://en.wikipedia.org/w/index.php?title=Pericardiacophrenic_veins&action=edit&forcetrace=true [07:08:19] Splarka: there were times when I used to rewrite features :)) [07:08:47] well, if you kept that up, people could just commit anything and get it optimized for free [07:10:11] :) [07:10:23] I can do that 1% of my time [07:10:30] anything beyond that crosses frustration thresholds :) [07:10:33] mmm, what was brion's thought... something like just a blob of all the editnotices outside the usual database (maybe an extension), and editable by sysops via a special page or something? [07:11:39] bah [07:12:09] brion doesn't know what to develop [07:12:13] so ends up doing such stuff :) [07:12:36] "outside the usual database"? what? why? [07:12:49] well, as in, not as subpages or MW namespace pages [07:12:56] not stored as revisions [07:13:11] why? [07:13:18] *Splarka shrugs [07:13:23] *domas shrugs too [07:13:35] http://leuksman.com/log/2009/01/29/e-mail-scams-and-out-of-band-article-notices/ [07:14:07] > What seems like a good way to allow these out-of-band notices to be easily added to a page and edited without being a core part of the page, or an overall performance drag when no notice is present? [07:14:25] have you ever seen brion measuring performance drag? :) [07:14:27] (not directly related to edit notices) [07:14:34] only on himself [07:14:41] not sure about it [07:14:42] anyway [07:14:58] well, how about a custom namespace for stuff that doesn't need to be in the message cache but has to be guaranteed to be protected.. for crap like the spam blacklist, whitelist, title blacklist, and other configuration crap (maybe even site css/js) [07:15:14] and it could have a stupid name like Config:, heh [07:15:20] Is there a free Wiki Farm that provides password protected MediaWiki hosting? [07:15:46] Is there a free Wiki Farm that provides password protected MediaWiki hosting? [07:15:52] Splarka: that should be able to live in Project: [07:16:16] hmm, like another title hack, any subpages of Project:Config/ ? [07:16:32] requring 'editconfig' right, given to sysops? [07:16:40] (ew) [07:16:57] no [07:17:21] whew [07:17:44] I like simple stuff [07:17:51] like some simple protection :) [07:24:21] *Splarka suddenly wonders if domas feels guilty for removing the editnotice feature [07:26:08] Is there anyone who will answer a hosting question? [07:26:49] nono [07:26:53] message cache is complicated solution [07:27:01] ? [07:27:09] Sk8er5000: wasn't for you! [07:27:14] oh [07:27:14] you're not the center of the world [07:27:16] nor universe [07:27:18] nor multiverse [07:27:21] well i didnt't know [07:27:26] now you know [07:27:30] OK [07:27:31] domas is [07:27:33] yup! [07:27:35] that's why he sucks so much [07:27:36] :( [07:27:37] gravity! [07:27:51] Splarka: you just made a fat joke? [07:27:55] nope [07:27:58] an ego joke [07:28:11] how long about until I will recieve help? [07:28:35] !hosting [07:28:35] --mwbot-- For a list of popular wiki hosting services, see . You can also set up MediaWiki on most webspace that provides PHP and MySQL (SSH recommended). [07:29:50] everyone is the center of their own universe, self awareness is our greatest asset and greatest liability [07:30:18] I'm after a host that offers password protected wikis for private use [07:31:27] it wouldn't be free, for private use [07:31:47] own server would be my only option? [07:32:09] or pay for hosting, or find a friend to lend you server space [07:32:56] ill just get XAMPP [07:33:14] you can use any 5$ virtual hosting out there [07:33:19] or 3$ [07:33:23] whatever costs are nowadays [07:35:23] hmm, was that a threat? [07:37:53] 03rotem * r48415 10/trunk/extensions/Collection/Collection.php: Adding the first six extra namespaces ( + talk pages) to the article namespaces that can be added to collections, as they differ between wikis. More extra namespaces should be added if necessary. [08:09:23] 03(NEW) Set $wgBlockAllowsUTEdit = true for zh.wikibooks - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=17986 normal; Normal; Wikimedia: Site requests; (jwongwiki) [08:10:16] what is 'deferred' in code review? [08:12:51] hi guys ..... i suddenlly lost my wiki website [08:12:52] I thought it meant "we don't care", like for strange forks and extensions? [08:13:09] http://damaonline.net/koc/index.php5/Main_Page <--- can some one please visit and give a hint plz plz [08:14:16] where is this Main_page stored ... it says this page is missing [08:14:36] 03(NEW) Set $wgBlockAllowsUTEdit = true for zh.wikipedia - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=17987 normal; Normal; Wikimedia: Site requests; (jwongwiki) [08:19:20] http://damaonline.net/koc/index.php5/Main_Page <--- can some one please visit and give a hint plz plz [08:19:33] where is this Main_page stored ... it says this page is missing [08:23:06] *OverlordQ wishes 17837 would get fixed :( [08:24:11] 03(mod) Set $wgBlockAllowsUTEdit = true for zh.wikibooks - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=17986 +shell (10p858snake) [08:27:49] http://damaonline.net/koc/index.php5/Main_Page <--- can some one please visit and give a hint plz plz [08:27:55] where is this Main_page stored ... it says this page is missing [08:30:07] Mba7eth: ask it once, you dont need to repeat [08:30:14] if somebody knows and feels like helping, they'll answer [08:30:39] I'd love to help, but I dont have the foggiest :) [08:30:54] OverlordQ: Thanks .... but you how it feels when your site is down :( [08:31:08] OverlordQ: Plus i'm no php expert [08:31:54] do you even have php5 enabled on that server? [08:32:25] with the extenstion formatting [08:32:30] I guest so .... cuz I was working fine [08:33:49] I just want to know, where can i find the main_page [08:34:28] did you recently change anything? [08:36:19] as what the server owner says .... he recently upgraded the whole server to php5 .... but that was working with index.php5 was working fine earlier [08:36:57] OverlordQ: I already created another dummy empty site. ... http://damaonline.net/new/index.php5/Main_Page [08:37:03] this will not work [08:37:11] however if i type http://damaonline.net/new/index.php/Main_Page [08:37:30] this would work smoothly. [08:37:45] So does this mean we have no php5 on the server ? [08:38:36] shouldn't matter really, the .php5 file just calls the .php file [08:39:23] i believe you need to fix your localsettings file up [08:39:34] but i don't have a local copy with me to check [08:39:56] sounds like some weirdness with rewrite rules [08:40:25] let me show you my localsit on a pastebin [08:40:48] make sure you remove any passwords [08:41:03] also to check to see if your .htaccess still exists [08:43:03] http://pastebin.com/m3b6559a0 [08:43:38] p858snake: Is .htaccess a file. ? where can i find it ? [08:43:57] it should be in the root folder of where you have mediawiki installed [08:44:56] on line 48 of that pastebin you can change that to .php [08:45:44] using .php5 is only really needed when your running version 4 and 5 in tandem on the one server [08:47:52] p858snake: OverlordQ ..... thanks guys alot alot alot [08:48:06] my wiki is back alive :) [08:49:28] p858snake, OverlordQ ... I really don;'t know how to return this faver [08:49:29] :) [08:49:34] thanks thanks [08:49:43] no need to return any favor [09:14:38] Anyone else finding svn.wikimedia.org slow? [09:14:43] hi MinuteElectron [09:14:48] hello [09:30:44] 03(mod) Change default locale for {{formatnum:}} parser function - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=17659 (10niklas.laxstrom) [09:37:41] lol Erkkimon [09:37:42] 03(mod) Editnotices generating invalid XML (HTML) around unordered lists - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=17486 (10niklas.laxstrom) [10:03:25] Hello, friends! I have a problem, Special:Allpages doesent seem to update on a local install we have, it was recently updated to 1.14 from 1.11 but it is uncleare wether or not special:allpages was updating before the upgrade. Is there some maintenance script that should be run? [10:06:31] btw -> here is our special:version http://ubuntu-se.org/wiki/Special:Version [10:10:14] 03(NEW) Spaces before [[Category:]] link are ignored - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=17988 trivial; Normal; MediaWiki: Page rendering; (mormegil) [10:39:24] 03(NEW) Users can create other users on blocked IPs - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=17989 enhancement; Normal; MediaWiki: Blocking; (lucasbfr.bugzillaMediawiki) [10:47:38] hallo MediaWiki developers, hope you have a nice day. i'm the admin of a small mediawiki at http://socialbar.de and i have just updated from 1.12 to 1.14. now, without any other changes, the highlighting of non-existing sites is gone. i checked the css, but everything looks fine. were there any changes in the last versions concerning the highlighting of non-existing sites? Thanks for your help. [10:47:54] 03(mod) User can't login after being renamed at nlwiki - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=17982 normal->high (10erwin) [10:48:13] 03(mod) Users can create other users on blocked IPs - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=17989 +comment (10overlordq) [10:51:18] 04(REOPENED) Change default locale for {{formatnum:}} parser function - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=17659 +comment (10OnTheWhite) [11:02:34] Anybody here has any experience with large MediaWiki installations? [11:19:42] 03(NEW) trackback.php lacks a .php5 equivalent - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=17990 normal; Normal; MediaWiki: General/Unknown; (slp.45325678) [11:38:33] i am very new in mediawiki and need to install http://www.mediawiki.org/wiki/Extension:Google_Analytics#Source_code [11:38:57] it all i need create file name extensions/analytics.php and copy and past the code [11:39:00] and i am done ? [11:41:00] How can I add a translated page to my wiki pages? [11:41:49] I don't know how can I do to show the list of the languages in the left side under the search box [11:41:53] any idea? [11:45:01] SOS: Add to the page [[(langauge):(page name)]] [11:46:16] SOS: Say you've translated a "Dog" page to French, add [[fr:(Dog in French)]] to the English page and [[en:Dog]] to the French page [11:57:26] davidt, so If I write this: [[en:MainPage]] it will be correct! isn't it? [11:58:03] yeah [11:58:38] but doesn't appear nothing in the left side of the page [12:02:22] SOS: http://www.mediawiki.org/wiki/Interwiki [12:02:35] in the page appear: en:MainPage [12:05:26] it is not so clear [12:05:57] davidt I have to install a new version of the mediawiki for each language? [12:06:25] SOS: Not necessarily. [12:06:41] I need to translate only a page [12:06:50] at the moment only for a tes [12:06:52] test [12:07:17] SOS: Multilanguage Wikis aren't that easy to setup [12:07:18] If I write [[en:pagename]] a link appear [12:07:31] :( [12:07:51] so there is a straitforward to do it [12:07:57] or not? [12:10:29] SOS: Is the translated page in a different-language wiki? [12:10:44] SOS: Or just a different page in the same wiki? [12:11:04] I need to translate an Italian page into English in the same wiki [12:12:11] SOS: Well, if the Italian page is part of an Italian wiki, then the syntax is [[it:(page name)]], if you need help setting up an Italian wiki in addition to the English one, it's a different story. [12:12:57] Look at http://wiki.shinesoftware.com [12:13:16] this is a italian wiki [12:13:26] SOS: Yeah, OK [12:13:44] now I need to create the same main page into English [12:14:08] in the same wiki [12:22:05] Fatal error: Call to undefined method ParserOptions::getIsSectionPreview() in /var/www/revisions/Brion/extensions/Cite/Cite_body.php on line 699 in my 1.14 test environment ... [12:22:24] is Cite specific to 1.15 at the moment ? [12:34:34] meh, exactly same template code as in enwp, but it's spitting out diff results :-/ [12:42:40] How can I rename a page? [12:43:25] "move", SOS [12:44:10] move? [12:44:20] by shell? [12:44:43] SOS: Special:MovePage [12:45:14] ok [12:45:20] let me chek [12:46:07] there is no Special:MovePage [12:53:51] SOS: Just click on "Move" on the top [12:53:56] SOS: of the page [13:04:30] you need to be logged in and have atleast the autoconfirmed permission before you can move [13:07:30] 03(NEW) Ploticus: Binary not found - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=17991 enhancement; Normal; MediaWiki extensions: SemanticResultFormats; (dasch_87) [13:07:37] 03(mod) Ploticus: Binary not found - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=17991 15enhancement->major (10dasch_87) [13:08:37] 14(INVALID) Users can create other users on blocked IPs - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=17989 +comment (10lucasbfr.bugzillaMediawiki) [13:26:28] grrr [13:32:28] 03(NEW) r48399 Broke parser tests (maybe other stuff) - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=17992 minor; Normal; MediaWiki: General/Unknown; (overlordq) [13:32:43] GRRR [13:38:36] grr? [13:39:35] 03(mod) r37313 broke parser tests for alternative databases. - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14990 +comment (10overlordq) [13:40:26] 03(mod) r37313 broke parser tests for alternative databases. - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14990 (10overlordq) [13:40:45] bugzilla really needs comment editing :P [13:41:32] and preview :) [13:45:57] 03demon * r48416 10/trunk/phase3/includes/GlobalFunctions.php: (bug 17992) Revert crack-inspired r48399. [13:46:05] 03(FIXED) r48399 Broke parser tests (maybe other stuff) - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=17992 +comment (10innocentkiller) [14:02:16] 03(mod) ForeignAPIRepo throwing error on first page load for file - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=17835 (10overlordq) [14:04:52] OverlordQ: Hunk #1 FAILED at 643. patch unexpectedly ends in middle of line Hunk #2 FAILED at 656. for patch in bug 14990 [14:06:33] bugger, lemme remake it [14:09:17] 03(mod) r37313 broke parser tests for alternative databases. - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=14990 +comment (10overlordq) [14:09:26] i was applying manually [14:09:29] *it [14:09:56] ah [14:10:16] *^demon goes and writes more crack-infused code for people to wake him up with bugs about fixing. [14:10:25] lol :D <3 [14:12:26] I should probably actually go to sleep and/or stop looking at code. I dont have crack to blame it on. [14:13:27] *ialex stabs Xcode for highlighting .inc files as Pascal [14:17:31] <^demon> OverlordQ: I don't either. But it makes for a pretty good excuse :) People just feel sorry for you and are like "Oh...it was the crack I guess. Poor guy" [14:18:32] lol [14:22:52] <^demon> OverlordQ: I did r48399 after about 12ish hours of working with Java...wasn't really at my best. [14:23:45] meh, it took me throwing lots of echo "I got here!" [14:23:48] till I found it :P [14:24:27] <^demon> It's the only place where we call mkdir() in core now...so it makes mkdir errors harder to find :) [14:24:38] <^demon> I might start passing __METHOD__ to it, just to make life easier for debugging. [14:25:36] yea I started running the parser tests and after every test it was complaining about copy( bla blah ) file does not exist. I was like O_o [14:30:14] ^demon: hey [14:30:21] <^demon> hi. [14:30:52] ^demon: Maybe you got an idea who could know how could I keep bots from overloading my db? :) [14:30:53] 03demon * r48417 10/trunk/phase3/includes/GlobalFunctions.php: Add optional $caller param to wfMkdirParents(). Should make debugging mkdir() errors easier. [14:32:08] <^demon> davidt: No clue.... [14:32:33] ^demon: sucks... [14:32:48] davidt: what kind of bots? [14:33:04] davidt: friendly, polite, impolite, or hostile? [14:33:18] Duesentrieb: I guess polite [14:33:34] Duesentrieb: How do you define friendly/polite/impolite? [14:34:43] davidt: friendly would be "some user runs it to help with the site". polite is a well-written web spider (like google or yahoo or something). impolite as a badly written spider (like msn's). hostile is spam bots, etc [14:35:05] Duesentrieb: Friendly then. There's a bot that imports a lot of data into our Wiki. [14:35:27] davidt: so just tell it to pause between imports. and use the api. [14:35:42] davidt: and consider server-side scripting. [14:35:59] Duesentrieb: It uses the API, and pauses, but somehow it caused a VERY large load on our DB. [14:36:24] Duesentrieb: And our DB server is a pretty strong machine, so it's not easy to cause huge load on it. [14:36:42] Duesentrieb: And it got me wondering - what if somebody desides to abuse that... [14:36:45] davidt: try to find out what exactly is causing the load. log slow queriey, maybe. [14:37:05] what does the bot do? just add/update pages? [14:37:15] do they have a lot of links/template? [14:37:26] what kind of caching do you have enabled? [14:37:27] Duesentrieb: Add/update category pages, in a wiki that has a lot of pages, with categories. [14:37:54] hm, editing category pages shouldn't cause much trouble by itself. [14:38:14] Duesentrieb: I guess what I'd like to know, is how could I limit bots in terms of CPU/DB load usage, and so on. [14:38:23] Duesentrieb: How would I go about logging slow queries? [14:38:26] you can't. [14:38:38] that's an mysql option, look in the manual. [14:38:42] <^demon> Hmmm... [14:38:43] Duesentrieb: Oh [14:38:47] mediawiki also has some profiling options [14:38:48] !profiling [14:38:48] --mwbot-- For information on how to enable profiling for MediaWiki, see [14:39:26] <^demon> Disabling parser extensions makes a noticable difference on import, iirc. [14:39:34] davidt: make sure to make good use of cachign though. per default, everything is cached in the database - which can cause big load. i recommend to enable apc caching. [14:39:38] !tune [14:39:38] --mwbot-- I don't know anything about "tune". You might try: !tunemysql [14:39:45] hm... not quite... [14:39:50] !tunemysql [14:39:50] --mwbot-- http://dev.mysql.com/doc/refman/5.0/en/fulltext-fine-tuning.html [14:39:58] ah, no, that'S not it. silly. [14:40:16] @search fast [14:40:16] --mwbot-- Results: [] [14:40:18] select `bridgeman_id`, `bridgeman_name`, `thesaurus_id`, 50, lcase(convert(target using utf8)), language [14:40:19] from `bridgeman_to_wikiword_/* bridgeman_multilang */` as M [14:40:21] join /* wikiword_thesaurus */langlink as L on L.concept = M.thesaurus_id; [14:40:23] oops, sorry. [14:40:31] bad paste. [14:40:45] <^demon> davidt: Do you have parser functions or anything similar? [14:40:52] ^demon: yeah [14:41:21] ^demon: prone to cause cpu load, but not really database, reight? [14:41:40] ^demon: The wiki is pretty heavily extended. [14:42:00] ^demon: Don't think that we have many parser functions. [14:42:04] <^demon> Duesentrieb: Yeah, but thought disabling it during import might help a bit, hmm. [14:42:09] well, a badly written extension can cause a lot of load. especially if it triggers db queries on parse. [14:42:17] no way to tell [14:42:51] davidt: http://dammit.lt/2007/01/26/mediawiki-performance-tuning/#more-56 [14:43:03] or better http://dammit.lt/2007/01/26/mediawiki-performance-tuning/ [14:43:14] Duesentrieb: We're using APC and MemCache [14:43:46] can somebody confirm bug 17837 for me, just so I can be sure I"m not just jinxed [14:43:59] davidt: hm, that should be ok then. unless your APC cache is spilling and has no expiration date (thius appears to be common) [14:44:20] Duesentrieb: I'm not using APC cache, I use memcached for cache. [14:44:24] Duesentrieb: APC only for opcode cache. [14:44:51] ok. apc is a bit faster, but that's unreated to db performance. the important bit is not to use the db for caching. [14:45:08] 03(mod) Editnotices generating invalid XML (HTML) around unordered lists - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=17486 (10Amalthea.wikimedia) [14:45:10] (and memcached is better if you have multiple web servers) [14:45:13] Duesentrieb: It's a bit faster if you only got one server. [14:45:21] indeed :) [14:45:24] Duesentrieb: yeah, we have 2 MediaWiki servers [14:46:05] davidt: i guess you'll have to find out what queries are hitting yxour db so hard, and where they come from. and also if it's a few heavy ones or a lot of small ones. [14:46:20] Duesentrieb: What are good pointers to look for? [14:46:34] Duesentrieb: (I currently have any MW debugging disabled [14:47:00] a) enable mw profiling b) enable logging of slow queries [14:47:06] oh, and enable miser mode! [14:47:14] !wg MiserMode [14:47:14] --mwbot-- http://www.mediawiki.org/wiki/Manual:%24wgMiserMode [14:47:40] this mostly affects special pages though, probably not so important for impoert [14:47:50] Duesentrieb: yeah [14:47:59] Duesentrieb: This particular bot is mostly using the API [14:48:58] Duesentrieb: And enabling the profiling isn't something I'd like to keep running all the time, isn't it? [14:49:07] no. [14:49:14] Duesentrieb: Because the bot already did it's job... [14:49:36] well, i don't think you can find out *anything* when it's not running. [14:49:50] you can setr profiling to take samples, though. [14:49:57] i.e. only run on every 100th request. [14:50:07] <^demon> You can disable expensive parser functions. [14:50:23] ^demon: doesn't miser mode do that too? [14:51:04] <^demon> I thought those were 2 different variables. [14:51:11] dunno, maybe [14:51:16] <^demon> I could be completely wrong. Dunno. [14:53:05] Duesentrieb: Setting it to take samples is a good thing [14:53:50] Duesentrieb: BTW, if a job from the jobqueue is running a query - will it have a comment with the user that started this request? [14:55:01] i don't think so. sounds like a good idea... you'd know who to yell at :) [14:56:20] Duesentrieb: Well, regular queries do, from what I've seen [14:56:32] Or is it a local modification? [14:56:53] they have a comment that contains the function that caused them... i have never seen a user name there. [14:57:19] generally, the job queue triggers high level functions, so the queries look like whatever they would look like anyway... [14:57:50] though if the comment relies on evaluation wgUser, then queries run from the queue would indicate a misleading user. [15:00:36] Duesentrieb: Very weird. [15:00:52] !upload [15:00:52] --mwbot-- File uploads are disabled per default, set $wgEnableUploads=true to enable it. See for configuration details, and for using uploaded images and other files. [15:01:19] Duesentrieb: http://paste2.org/p/164869 [15:01:28] Duesentrieb: That's how I see it... [15:01:55] Duesentrieb: "Tagbot" is the username of the bot [15:02:41] maybe the api does that. would make sense. [15:03:22] Duesentrieb: Yeah, but I don't see it in the code [15:05:14] me nither [15:06:10] davidt: $commentedSql = preg_replace('/\s/', " /* $fname $userName */ ", $sql, 1); [15:06:24] either this is pretty new, or i just looked past it until now :) [15:06:43] Duesentrieb: We're using 1.13.5, so it's not very new. [15:06:47] Duesentrieb: Where did you find it? [15:07:03] Database.php, in the query method. [15:07:21] it only gets the user name if the user object is already unstubbed. so it's not always visible. [15:07:40] but it does thake the current user. so for jobs from the queue, it's misleading. [15:07:56] Duesentrieb: For jobs from the queue - what's the current user? [15:08:17] Duesentrieb: We're running jobs only through a cronjob [15:08:25] depends on how the queue is run. per default, it'S triggered on page views - so it's whoever requested taht page [15:08:40] if it's run by a cron job, then... don't know. the default user for scripts. whatever that is. [15:29:49] Duesentrieb: Can a job be added twice into the job queue? [15:30:23] yes. the second instance should get ignored when it's due for execution, though [15:30:39] Duesentrieb: Oh, why add it in the first place then? [15:31:31] Hello, I just took a 1.12 installation to 1.13.5 and I am getting the following error: Fatal error: Call to a member function matchStartToEnd() on a non-object in /webroot/includes/parser/Parser.php on line 2788 [15:31:52] I saw that some other wikis had this issue on google, but they seem to be up now, I am not sure what is wrong [15:44:54] <^demon> \o/ [15:45:07] ¯\(°_o)/¯ [15:45:43] *stwalkerster_ becomes worried by OverlordQ's mismatched eyes ;P [15:46:23] <^demon> OverlordQ: Well I did that in preparation for my commit... [15:46:31] <^demon> But then I had inconsistent line endings :-\ [15:50:03] hello [15:50:12] I need some help... [15:51:00] !ask | guerito20 [15:51:00] --mwbot-- guerito20: Don't say "I have a question", or ask "Is anyone around?" or "Can anyone help?". Just ask the question, and someone will help you if they can. Also, please read < http://workaround.org/getting-help-on-irc > for a good explanation of getting help on IRC. [15:52:10] I am about to install mediawiki on my webserver, but I need it to be multilingual, like wikipedia. The same articles in different languages..with the menu on the left side [15:52:29] also I need different titlepages, but I still need it to be one wiki [15:52:49] guerito20: mediawiki doesn't do that. [15:52:52] one wiki per language [15:53:15] and what about wikipedia? [15:53:16] guerito20: there are some extensions that help you with having multiple languages in one wiki, but it's never quite nice. [15:53:27] ehm... [15:53:30] wikipedia has one wiki per language [15:53:38] not one wiki for all languages [15:53:42] okey, thank you [15:53:48] guerito20: Wiki Family [15:54:01] !wikifamily | guerito20 [15:54:01] --mwbot-- guerito20: To run multiple wikis, you can simply install MediaWiki in different folders, with different databases or in one with database prefixes. You can also have multiple wikis using a single installation: and . [15:54:09] !e polyglot | guerito20 [15:54:09] --mwbot-- guerito20: http://www.mediawiki.org/wiki/Extension:polyglot [15:54:35] Duesentrieb: nice [15:54:54] what, polyglot? nasty hack :P [15:55:14] okey, thank you, I will try to deal with it by manner you recommanded me :) [15:55:15] Duesentrieb: :) no [15:55:20] Duesentrieb: "!e" :0 [15:55:38] ah that [15:55:48] handy yea [15:55:55] Duesentrieb: Any way to see a list? [15:56:20] of what? bot commands or extensions? [15:56:31] the !e command just generates a url [15:56:33] !e asdlfkgjadsfglkjsdfgö [15:56:33] --mwbot-- http://www.mediawiki.org/wiki/Extension:asdlfkgjadsfglkjsdfg%C3%B6 [15:57:03] Duesentrieb: Bot commands [15:57:11] !help [15:57:11] --mwbot-- Hi! I'm mwbot, a bot that was quickly whipped up by Daniel Cannon (AmiDaniel) to help out around #mediawiki. Some quick help is at < http://www.mediawiki.org/wiki/Mwbot >, you can find all my source code at < http://amidaniel.com/viewvc/trunk/MWBot/?root=svn >. A dump is now available at < http://toolserver.org/~amidaniel/botbrain.html > [15:57:32] Duesentrieb: Cool, thanks [16:00:52] <^demon> asdf >.< [16:14:19] aoeu [16:14:31] *Jhs misses his dvorak [16:16:42] 03mark * r48418 10/trunk/pybal/pybal/ (4 files in 2 dirs): Fix bug where reenabled servers would not be set back up and repooled after configuration merge [16:21:22] Jhs: Don't the regular keyboard labels confuse you when using dvorak? [16:22:02] davidt, yeah, they pretty much fuck me up :D [16:22:29] *davidt wonders if there are any dvorak keyboards for sale [16:22:41] however i almost never use them, but have to now because my laptop broke down [16:27:16] The [ ] external link tag doesn't support spaces in links, how do I remove the spaces if they are given through a template? [16:30:21] ������������! [16:30:38] svip: {{#urlencode|{{{1}}}}} (i think) [16:31:32] ��� ����� ����� ���������� �� �������� ����������� �����? [16:33:21] 03(mod) Editnotices generating invalid XML (HTML) around unordered lists - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=17486 (10davidg) [16:33:45] svip: Hello! Can you give me advice in using e-mail? [16:34:22] Duesentrieb: It was {{urlencode: }}, but thanks! :D [16:54:25] 03(mod) Editnotices generating invalid XML (HTML) around unordered lists - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=17486 (10Platonides) [16:54:26] 03(mod) XHTML compliance (tracking) - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=209 (10Platonides) [16:57:32] 03(mod) Mixing Wikitable and HTML syntax breaks with Tidy disabled - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=8948 (10Platonides) [16:57:33] 03(mod) XHTML compliance (tracking) - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=209 (10Platonides) [16:58:18] Can anyone confirm that buggy behaviour in 1.14? After moving a page the category sortkey doesn't update [17:30:29] Any chance for some profiling help? [17:49:32] 03(mod) Implement the Interlanguage extension in Wikipedia - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=15607 (10Wiki.Melancholie) [17:53:45] 03(mod) Change default locale for {{formatnum:}} parser function - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=17659 (10niklas.laxstrom) [17:54:19] <^demon> Any Java folks around? [17:54:37] ##java ? [17:54:39] ;) [17:54:59] argh [17:55:28] why do people post "this might be related to [something very different here]" on bugs [17:57:33] <^demon> Nikerabbit: What bug? [17:57:50] <^demon> stwalkerster: I had a List question. [17:58:00] *davidt is trying to figure out MediaWiki profiling [17:58:07] <^demon> When you remove() an object from it, does that decreate the size()? [17:58:12] ^demon: that bug [17:58:26] 03(mod) Mixing Wikitable and HTML syntax breaks with Tidy disabled - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=8948 (10nospam) [17:58:27] 03(mod) Parser generates invalid XHTML with a list inside a table - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=17486 (10niklas.laxstrom) [17:58:27] 03(mod) XHTML compliance (tracking) - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=209 (10nospam) [17:58:27] Can't find any information on the UDP profiling, nor how do I print timing into the debug log [18:01:52] return mkdir( $dir, $mode, true ); // PHP5 <3 [18:02:06] was that decided that it is good thing that umask modifies the mode? [18:02:51] <^demon> Dunno. Wasn't paying attention to that. [18:05:31] Hello, I just took a 1.12 installation to 1.13.5 and I am getting the following error: Fatal error: Call to a member function matchStartToEnd() on a non-object in /webroot/includes/parser/Parser.php on line 2788 [18:05:54] I saw google pick up a couple of these too, does anyone know what is going on here? [18:19:12] Can anyone confirm that buggy behaviour in 1.14? After moving a page the sortkey doesn't update [18:22:34] is it on your private wiki? [18:24:54] eghjaytee: did you install the new over the old one? [18:26:50] <^demon> #android is useless :-\ [18:29:22] saper: you ment me? Yes I have that bug on my own wiki, so before I open a bug I want to ask if anyone can confirm it or if it's caused by some own false settings [18:32:33] 03(mod) Oddly named talk pages cannot be unwatched via Special:Watchlist/edit - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=5280 +comment (10jidanni) [18:38:37] ok I absolutely have that bug but a test was working fine on mediawiki.org HMH checking settings [18:39:28] 03(mod) Implement the Interlanguage extension in Wikipedia - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=15607 (10amikeco) [18:41:45] 03(mod) Implement the Interlanguage extension in Wikipedia - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=15607 (10amir.aharoni) [18:42:25] *^demon puts a plate of cookies near VasilievVV [18:42:46] ^demon: eh? [18:43:34] <^demon> VasilievVV: Any chance you could take a look at file redirects anytime soon? There's a few outstanding issues I can't seem to fix :-\ [18:43:54] ^demon: bug numbers? [18:44:11] <^demon> !b 15008 [18:44:11] --mwbot-- https://bugzilla.wikimedia.org/show_bug.cgi?id=15008 [18:45:01] Oh, I remember it [18:45:16] <^demon> That's the one bug. But in general in playing with it: it seems kind of messy. It feels like we're loading more data than needed (we shouldn't need a full loadFile() to resolve a redirect, IMO). That, and we can't seem to have *local* redirects to *foreign* images. [18:46:01] <^demon> Which should be possible, at least in theory. We store that data in the redirect table already :-\ [18:46:57] As faras I remember, forbiding redirects from local to shared was planned behavior [18:47:13] <^demon> Any particular reason? [18:48:13] Hmmm... Why do we forbid interwiki-redirects? [18:50:15] <^demon> Foreign image sources aren't like iw redirects. [18:50:31] <^demon> IW redirects require an Http redirect. [18:50:37] I see [18:50:56] I still have no ideas why should we support shared redirects [18:51:07] <^demon> With IW redirects, you're going from local to foreign. With shared files, you're bringing data in local from foreign. [18:51:34] <^demon> It's not shared redirects I'm looking for. I'm suggesting local redirects to shared images (on the local wiki) [18:51:44] 03ashley * r48419 10/trunk/extensions/SocialProfile/UserProfile/ (4 files): SocialProfile: cleanup to UserProfile. made RemoveAvatar valid XHTML and documented wAvatar class. [18:52:31] ^demon: why would anybody use it? [18:54:04] <^demon> Same reason anyone would use image redirects to begin with. [18:54:30] <^demon> If you can view the image locally, I don't see any reason why you can't redirect to it. [18:54:50] Subfader: check your job queue settings maybe [18:55:45] where can i do that? I checked the categorylinks table and after moving pages the cl_sortkey is really still the same [18:56:17] oh i have an extension using the job queue. will disable for a test [18:56:43] <^demon> VasilievVV: Any if we really *dont* want it, then we need to stop storing the info in the redirect table. I figure if it's there, we might as well use it (plus, I think it might simplify image redirects so we don't have to keep casing into isLocal() or not) [18:59:55] saper: Thanks! It was that extension (ReplaceText) [19:00:02] will report a bug there :) [19:20:02] Hi [19:20:08] I'm trying to use Template:Languages [19:20:14] but it doesn't appear to work ... [19:20:22] I installed the ParserFunctions extension [19:20:25] but no luck [19:20:27] any other ideas? [19:20:32] this is for the PostgreSQL wiki [19:21:03] http://wiki.postgresql.org/wiki/FAQ is the english content [19:21:18] so I put the spanish translation in http://wiki.postgresql.org/wiki/FAQ/es [19:21:30] the English page has the Spanish link in it, and it works [19:21:42] but the Spanish page does not point to the English page [19:25:01] hhhi Duesentrieb [19:25:44] must be wrong with the template [19:29:48] *something [19:32:22] and you're running 1.12, I'm sure something (might) have changed between then and now wrt ParserFunctions [19:34:10] running a non-release 1.12 is probably non-optimal either [19:41:28] actually I tried a local 1.14 installation and it doesn't work there either [19:41:36] the templates I copied verbatim from mediawiki.org [19:47:25] I noticed that Template:Languages/Lang uses a #language function [19:47:36] is it possible that I need another extension for that one to work? [19:48:30] (I couldn't find a way to do a search for #language; the search engine seems to ignore the #, which gives a lot of spurious hits) [19:48:34] same with Google [19:48:51] #language is a magic word [19:50:06] and it's supposed to work fine in plain mediawiki with no extensions? [19:53:57] I just noticed something [19:54:11] if I create a page FAQ/de/de then a "Deutsch" link appears in FAQ/de [19:54:24] it's as if it thought that FAQ/de is the english page [19:55:33] yea, templay probably needs fixed, just give it the page param [19:55:39] {{Languages|FAQ}} [19:55:43] *template [19:57:36] interesting, yes, it works that way [19:57:41] thanks, OverlordQ [19:59:41] oh wait, I might know what's different [20:01:11] your wiki probably has subpages turned off for the main namespace [20:02:34] and (I guess) mw.org has them turned on (somebody correct me) [20:03:21] so instead of thinking FAQ/de is the de subpage of FAQ it's taking FAQ/de to be the entire name of the page [20:04:17] hi all, i'm new to mediawiki and i'm facing a problem... i want to extend the Syntax highlighting extension to fetch de code from an url instead of writing it [20:05:28] !wg NamespacesWithSubpages | alvherre [20:05:28] --mwbot-- alvherre: http://www.mediawiki.org/wiki/Manual:%24wgNamespacesWithSubpages [20:05:40] i want to have another parameter to the tag... url ... and if is present, fetch the code.. and then parse it [20:06:51] any ideas? thanks ... [20:07:27] katratxo: has it been done? dunno, is it possible? likely, is it a good idea? dunno. [20:10:18] OverlordQ: i want to maintain the code in a mercurial repo .. and reference in the wiki with e.g. [20:11:49] katratxo, maybe you want to use hgweb instead. [20:11:53] That's what it's meant for. [20:12:04] MediaWiki is not meant to display content dynamically from other sources. [20:12:16] There's the CodeReview extension, but that's Wikimedia-specific. [20:12:26] I'd imagine you'd have trouble finding a decent extension that does what you want. [20:12:34] Unless you want to use an iframe or something. [20:13:48] Simetrical: thanks for the info ... hgweb is up and running ... but i wanted to have home howtos with 'real' examples, fetched from the hg repo through hgweb (raw file) [20:14:12] I wouldn't recommend it, personally, but good luck in any case. [20:14:24] ok, thanks [20:16:11] 03nad * r48420 10/trunk/extensions/RecordAdmin/RecordAdmin_body.php: last revision broke it, fix quotes [20:58:13] Hmm, no roan [21:01:09] where can i change the size of the textarea text box on the "edit" page? [21:03:58] 03daniel * r48421 10/trunk/WikiWord/ (3 files in 2 dirs): update biography patterns [21:07:31] 03(NEW) Have result list count displayed in API output, also option to only get count - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=17993 enhancement; Normal; MediaWiki: API; (sam) [21:35:22] 03(mod) Change default locale for {{formatnum:}} parser function - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=17659 (10OnTheWhite) [21:35:42] I should really setup a MW dev/test env [22:15:41] How can I remove a particular category from being included when a user clicks on "Random Page"? [22:16:07] ie: a random page from any category EXCEPT for the ones I specify [22:16:25] mm, you can't [22:16:34] Hmm [22:16:45] Not so random pages [22:16:48] Is there a way I can make a Random Page widget that will pull from a white list then? [22:17:08] So a list of categories that I DO specify [22:19:09] I just installed a fresh MW 1.14 and read in the release notes about the magic words __INDEX__ and __NOINDEX__ that can be added in the markup. I tested it but found no resulting or "noindex" robot tags. For the magic words to function in say the primary article namespace, does it need to be switched on in localSettings and if so how? Any pointers to a relevant how-to page would be appreciated. TIA [22:43:12] I misunderstood how it works. I now read the docs and found that __INDEX__ or __NOINDEX__ does indeed function. They just overrides whatever is defined in $wgDefaultRobotPolicy [22:52:24] 03nad * r48422 10/trunk/extensions/RecordAdmin/RecordAdmin_body.php: remove spurios quote added by Purodha [23:14:58] Checking this again, I find no way of actually making __NOINDEX__ to take effect. I can only override $wgDefaultRobotPolicy='noindex,nofollow' with __INDEX__ to change existing metas to contain 'index,nofollow' but __NOINDEX__ in the markup has no effect on a per article basis when the default is 'index,follow'. Is this a missing feature or a bug? [23:34:50] where would I go to report a security vulnerability in mediawiki? [23:35:16] NoodleSoup, it works on Wikipedia . . . [23:35:22] Egeste, security@wikimedia.org, I think. [23:37:09] or you could probably find someone in here to tell [23:37:53] or report it on bugzilla [23:38:14] Not for security vulnerabilities, please. [23:38:16] don't put security bugs on bugzilla [23:38:21] We'd prefer to fix them *before* they're publicly announced. [23:38:43] although if you put it on bugzilla it will probably get fixed /really/ quickly :) [23:39:06] http://www.mediawiki.org/wiki/Security [23:40:03] i just sent an email to security@wikimedia.org [23:41:46] 03(NEW) How to specify _request's_ POST data in gzip format? - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=17994 enhancement; Normal; MediaWiki: API; (raccvet1) [23:47:05] Ok thanks Simetrical. Wikipedia is of course 1.15 so I guess it is a bug in 1.14.0 that was fixed in some later version. [23:51:22] I don't think anyone's ever changed the feature since I first wrote it. [23:51:29] It definitely worked in 1.14alpha. [23:52:07] 03(mod) Link to user and user talk page in Special:Block - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=12050 summary; +comment (10mikelifeguard) [23:54:51] 03(mod) Special:BlockList search doesn't find autoblocks - 10https://bugzilla.wikimedia.org/show_bug.cgi?id=3345 summary; +comment (10mikelifeguard) [23:57:32] Ok you should know I guess. Strange I can't make __NOINDEX__ take effect in 1.14.0.